type 1 diabetes in youth

TYPE 1 DIABETES IN YOUTH:
A psychosocial challenge
Minke Eilander
TYPE 1 DIABETES IN YOUTH:
A psychosocial challenge
Minke Eilander
VRIJE UNIVERSITEIT
TYPE 1 DIABETES IN YOUTH:
A psychosocial challenge
ACADEMISCH PROEFSCHRIFT
The research presented in this thesis was conducted within the department of Medical Psychology,
ter verkrijging van de graad Doctor aan
at the VU university medical center Amsterdam, and the Amsterdam Public Health research institute.
de Vrije Universiteit Amsterdam,
The studies discussed in this thesis were supported by the Dutch Diabetes Foundation (Grant
op gezag van de rector magnificus
2011.13.1449 and Grant 2009.80.103) and by the European Foundation for the Study of Diabetes.
prof.dr. V. Subramaniam,
The publication of this theses was kindly supported by VU university medical center Amsterdam,
in het openbaar te verdedigen
department of medical psychology and by Sanofi.
ten overstaan van de promotiecommissie
van de Faculteit der Geneeskunde
© 2017 M.M.A. Eilander
op maandag 22 mei 2017 om 11.45
All rights reserved. No part of this may be reproduced, stored or transmitted in any form
by any means, without prior permission of the author.
Illustrations:
Linda Rusconi
www.lindarusconi.nl
Cover design & lay out:
Mookje Korst
www.mookdesign.net
Printing:
Ridderprint BV
Picture:
Tim Müller
in de aula van de universiteit,
De Boelelaan 1105
door
Minke Marijke Anke Eilander
geboren te Amsterdam
promotor:
prof.dr. F.J. Snoek
leescommissie:
prof. dr. R.J.B.J. Gemke
copromotoren:
dr. M. de Wit
prof. dr. M.A. Grootenhuis
dr. J. Rotteveel
prof. dr. J.M. Koot
prof. dr. K. Luyckx
dr. M. Finken
dr. C.M. Verhaak
paranimfen:
Josine van Kesteren
Stefanie Rondags
Contents
Chapter 1
General introduction
Chapter 2
Diabetes IN develOpment (DINO) The bio-psychosocial, family functioning and parental well-being of youth
with type 1 diabetes: a longitudinal cohort study design
Chapter 3
Low self-confidence and diabetes mismanagement in youth with type 1 diabetes mediate the relationship between
behavioral problems and elevated HbA1c
Chapter 4
Disturbed eating behaviors in adolescents with type 1 diabetes
How to screen for yellow flags in clinical practice?
Chapter 5
Not parental well-being but parental diabetes behaviors and stress are related to glycemic control in youth with type 1 diabetes
Longitudinal data from the DINO study
Chapter 6
Implementation of quality of life monitoring in Dutch routine care of adolescents with type 1 diabetes: Appreciated but difficult
Chapter 7
General Discussion
9
15
31
45
61
75
91
Summary
103
Samenvatting
109
Appendixes
117
References
137
Dankwoord
153
Chapter
1
General introduction
‘Sometimes at school, when it’s really quiet in our classroom,
my insulin pump suddenly starts beeping’
9
1
General Introduction
General Introduction
Living with a chronic illness can be challenging, especially when its daily management is
in a diabetic ketoacidosis: a potentially life-threatening coma [6]. Although there is no cure for
demanding. The main focus of this thesis is the psychosocial situation of youth aged eight
T1D yet, glucose levels can be regulated with exogenous synthetic insulin supplementation.
to eighteen years with type 1 diabetes. A short definition of the terms ‘youth’ and ‘type 1
However, too high glucose levels (hyperglycemia) or too low glucose levels (hypoglycemia)
diabetes’, and the outline of the thesis are given in this prolusion.
often occur in T1D. Hyperglycemia increases the risk of diabetes ketoacidosis, and micro- and
macro-vascular complications later in life such, as nephropathy, retinopathy, cardiovascular
Youth
disease, stroke and kidney failure [7]. Hypoglycemia requires immediate treatment in order
In youth, major physical, cognitive and psychosocial developments take place in a relatively
to prevent fainting, a seizure or coma. To keep blood glucose levels close to normal ranges,
short period of time. Growing up is best characterized as becoming an independent individual,
intensive treatment is required [8]. People with T1D depend on multiple daily insulin injections
influenced by experiences gained in this first period in life [1]. Both positive and negative
(MDI) or an insulin pump (Continuous Subcutaneous Insulin Infusion, CSII), accompanied by
experiences, defined as risk and protective factors, compromise mental representations
frequent self-monitoring of blood glucose levels (several times a day), while balancing the
for understanding the world, self and others [2]. A chronic illness is a significant risk factor.
energy intake (counting carbohydrates) and physical activity.
The accumulation of risk factors (such as a pessimistic personality or parental divorce) and
its balance with protective factors (such as intelligence, a great medical team or a positive
‘Sometimes people freak out when I have to inject insulin.’
relationship with the primary care givers - from now on referred to as parents -) impact the
psychosocial situation (later) in life [3]. From roughly eight years old, the capability to think
Pediatric diabetes patients are recommended to consult their diabetes care team (in either
logical, critical and hypothetical develops, contributing to the ability to see consequences and
a hospital or a specialized diabetes center) every three months [9], where their glycated
nuances, to make autonomous choices and to be increasingly responsible for these choices [4].
Hemoglobin A1c (HbA1c) level is assessed. HbA1c reflects average glucose levels over the past
These are all characteristics that are of great importance when managing a chronic illness.
four to twelve weeks and serves as an indication of how the diabetes overall is controlled [6].
Since high HbA1c levels predict diabetes complications later in life [8], a target HbA1c level
Diabetes
below 58 mmol/mol (7.5%) is recommended in pediatric diabetes care [10].
Type 1 diabetes (T1D) is a common chronic illness in youth. The prevalence in the Netherlands
is estimated at 18.6 per 100.000 children aged 0-14 [5]. T1D is characterized by insulin secretion
Psychosocial impact
deficiency. The β cells in the pancreas are not able to produce (enough) insulin: a hormone
With recent medical developments a cure for T1D seems closer than ever. In the mean
responsible for the regulation of the blood glucose levels. Insulin enables body cells to convert
time, youngsters still get diagnosed and have to manage their diabetes. And even though
carbohydrates (glucose) in energy. This distinguishes type 1 from type 2 diabetes (in which
technologies, aimed to reduce the inconvenience of managing the illness, are constantly
life style, age and increased body weight are associated with insulin resistance and secretion).
improving, high technology does not imply better diabetes care and better HbA1c. Research
Without enough insulin, blood glucose levels rise. Consequently, other substrates are used as
showed that the introduction of new technologies, such as CSII and continuous glucose
sources of energy. This causes the production of ketones which, when left untreated, results
monitoring devises, imposes psychological adjustment and time investment that many youths
10
11
1
1
General Introduction
General Introduction
and their parents find demanding [11]. Attention towards the psychosocial aspects related to
Content of thesis
T1D self-management increased the last decades: we have witnessed a trend from a biological
As stated, many factors - either risk or protective - shape diabetes and psychosocial outcomes
perspective towards a bio-psychological perspective. The daily hassle of diabetes challenges
in life. While having T1D can be considered a risk factor in itself, within its treatment there are
the psychosocial well-being of youngsters as they are required to manage their diabetes 24/7.
many factors that contribute to the diabetes management, its outcomes and the psychosocial
It should therefore perhaps not come as a surprise that depression and low emotional well-being
functioning. To gain knowledge about some of these factors, the psychosocial development of
are common amongst youth with T1D [12] and are associated with poor diabetes outcomes
youth aged eight to eighteen years with T1D was researched.
[13]. However, as pediatric diabetes care and available technologies improve rapidly, research
In Chapter 2, we describe the design of the Diabetes IN develOpment (DINO) study, aimed
shows no impaired Quality of Life (QoL) of youngsters with T1D compared to healthy peers
to follow the psychosocial, cognitive and physical development of youth with T1D aged eight
[14]. QoL is defined by objective and subjective indicators collectively reflecting a broad range
to fifteen years at baseline and their parents for four years. In Chapter 3, we address the
of life domains [15]. Nonetheless, youth with T1D often struggle with the negative impact of
relationship between adolescents’ problem behavior and HbA1c, and the mediating roll of
diabetes on their daily functioning and have diabetes-related worries, such as passing out
diabetes management and self-confidence. Chapter 4 presents the prevalence of disturbed
due to hypoglycemia [14]. Recent research revealed that adolescents, especially girls, with
eating behaviors in youth with T1D. Also, we explain why and how clinicians should pay
T1D are more likely to develop disturbed eating behaviors. This puts them at risk of several
more attention to this phenomenon. In Chapter 5, we describe the longitudinal relationship
psychosocial as well as physical complications [16, 17]. Needless to say, it is an important part
between parental well-being and youth’s diabetes outcomes in more detail. In Chapter 6,
of pediatric diabetes care to screen for psychosocial problems and to monitor and improve the
the implementation of the assessment of QoL is addressed; an evaluation study of the DAWN
QoL of young people with diabetes [18]. Currently, this is more advised than mandatory.
MIND-Youth program. The implications of the results presented in this thesis are described
in Chapter 7, the general discussion.
‘It’s so embarrassing when my mom discusses my diabetes with strangers.’
‘It’s annoying when I go to a club or to the airport and they check my bag
Parents play an important role in the treatment of T1D, as they perform or monitor the diabetes
and I have to explain my diabetes.’
management of their children [7, 19]. In childhood and adolescence, supportive parenting and
parental monitoring of diabetes management predicts better HbA1c levels. And although a good
family climate associates with better diabetes and psychosocial outcomes in youth [20, 21], T1D
challenges the family domain as well: compared to parents of healthy youngsters, parents of
youth with T1D have a higher level of stress and a lower emotional well-being. Furthermore,
they are at risk of posttraumatic syndromes from the diagnosis of T1D, especially when other
risk factors are present, affecting psychosocial and diabetes outcomes in their children [22-25].
12
13
1
Chapter
2
Diabetes IN develOpment (DINO) The bio-psychosocial,
family functioning and parental well-being of youth
with type 1 diabetes: a longitudinal cohort study design
‘I think it’s super nice that you ask questions about my life with diabetes’
Minke Eilander
Maartje de Wit
Joost Rotteveel
Henk-Jan Aanstoot
Willie Bakker-van Waarde
Mieke Houdijk
Marjolein Luman
Roos Nuboer
Jaap Oosterlaan
Per Winterdijk
Frank Snoek
BMC pediatrics, 2015. 15:82
15
DINO study design
2
DINO study design
Abstract
Background
Background. Strict glycemic control during adolescence decreases the risk of developing
In 2009, a report in the Lancet concluded that: “If present trends continue, doubling of new
complications later in life, even if this level of control is not maintained afterwards. However,
cases of type 1 diabetes in European children younger than 5 years is predicted between
the majority of adolescents with type 1 diabetes (T1D) are in poor control and so far medical or
2005 and 2020, and prevalent cases younger than 15 years will rise by 70%. Adequate
psychological interventions have shown limited success. Adolescence is characterized by major
health-care resources to meet these children’s needs should be made available” [26]. This
biological, psychosocial, cognitive and parent-child relationship changes and the complex
clearly underscores the importance of understanding the specific (changing) needs of youth
interaction between these developmental trajectories, and its impact on health outcomes is
with type 1 diabetes (T1D) to improve quality and efficacy of pediatric diabetes care. This
still poorly understood. A specific topic of interest in this context is the timing of diagnosis.
holds in particular for adolescent diabetes care, as clinical data have shown repeatedly that
The longitudinal study DINO (Diabetes IN develOpment) aims to examine: 1) If and how the
during adolescent years patients have great difficulty reaching and maintaining optimal
onset of T1D before vs. during puberty results in different outcomes of glycemic control, self-
glycemic control [27, 28]. Less than 15% of the young patients keep constant or reach HbA1c
management, psychological functioning and diabetes-related quality of life. 2) The timing of
levels below 8% (64 mmol/mol) from pre-puberty to young adulthood [27, 29]. In contrast to
onset of disturbed eating behavior, its risk factors and its prospective course in relation to glycemic
earlier belief, puberty years provide no protection against the risk of developing microvascular
and psychological consequences. 3) If and how the onset of T1D before vs. during puberty results
complications in later years as a result of prolonged hyperglycemia. In fact the reverse is
in different family functioning and parental well-being. 4) If and how the cognitive development
true: Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and
of youth with T1D relates to glycemic control and diabetes self-management.
Complications (DCCT/EDIC) Study has convincingly shown that the better the glycemic control
during adolescence, the lower the risk of developing complications later on in life - even if
Methods/design. DINO, a longitudinal multi-center cohort study, is conducted in youth with
that level of control is not maintained afterwards [8]. Adolescence is a critical period for the
T1D in the age range 8-15 years at baseline. Participants will be divided into two subgroups:
establishment of lifelong positive and risky health-related behaviors and, importantly, such
pre-pubertal and pubertal. Both groups will be followed for 3 years with assessments based
‘programming’ apparently applies to mental health as well [30]. In what manner biological,
on a bio-psychosocial model of diabetes, scheduled at baseline, 12 months, 24 months and
psychosocial and cognitive programming interact in youth with T1D is largely unknown.
36 months examining the biological, psychosocial - including disturbed eating behaviors - and
Of interest is the question how different trajectories develop during pre- and pubertal years,
cognitive development, family functioning and parental well-being.
and to what extent these years offer a window of opportunity for early detection and targeted
interventions to improve health outcomes. With the longitudinal cohort research DINO
Discussion. A better understanding of how the different trajectories affect one another will
(Diabetes IN develOpment) the complex interaction of biological, psychosocial and cognitive
help to gain insight in the protective and risk factors for glycemic outcomes and in who needs
development, family functioning and parental well-being will be studied. In order to do so,
which support at what moment in time. First results are expected in 2016.
a bio-psychosocial approach is called for. The bio-psychosocial model appreciates these
complex interactions, the onset and the demands of diabetes (Figure 1).
16
17
2
2
DINO study design
DINO study design
Figure
1: A framework for understanding the development and outcomes from pre-puberty adherence to diabetes management over a 4 year follow-up period compared to patients
diagnosed after this age [40]. Deterioration in adherence occurred in all age ranges as duration
Error! Style not defined. into puberty. Bio-psychosocial model based on Holmbeck and Sherpa [31].
increased [35]. It is of note that these longitudinal studies used age as an indicator for pubertal
status and not the actual physical development such as Tanner stage, in which the primary and
Developmental trajectories • Biological/puberty • Psychosocial • Cognitive • Social roles & Autonomy Psychosocial Outcomes • Self-­‐management • Psychosocial functioning • Family functioning • Diabetes-­‐related QoL Diabetes onset Diabetes Outcomes • Glycemic control • Physical outcomes secondary sex characteristics are scaled [41]. Gender differences and individual variations in
puberty onset were therefore not taken into account in these previous studies. Reviews have
clearly identified the lack of prospective cohort studies in representative (pre)pubertal groups
[42, 43]. Whether developing diabetes during puberty alters the duration and termination of
Interpersonal contexts • Parents (family) • Peers • School • Sport club puberty and results in psychological risks is unknown.
Psychosocial development. In general, mid-adolescence appears the most vulnerable period
Demographic and clinical v ariables • Co-­‐morbidity • Ethnicity • Family structure for developing psychological problems [44]. In youth with diabetes, rates of depression,
anxiety and disturbed eating behavior tend to be worrisomely elevated and are associated
with poor glycemic control [45-48]. Adolescents with T1D have more emotional issues
compared to healthy peers [49]. A study on female adolescents with diabetes showed that
Biological development. There is substantial variation between individuals in the time of Biological
development.
There is substantial
variationdevelopment between individuals
in thedifferences time of onset,
onset, duration and termination of the pubertal and these have in adolescents diagnosed <3 years from menarche [50], a lower overall sense of control was
duration
andpsychological termination ofconsequences the pubertal development
and these
have
and
social and [32]. In addition, it is differences
known that the social
onset and in frequent ‘unexplainable’ (high) blood glucose values, easily inducing feelings of anger,
psychological
consequences [32]. In addition, it is known that the onset and termination of
termination of puberty is delayed in children with diabetes compared to healthy youth [33, frustration and discouragement, thereby contributing to poor adherence and subsequent
puberty
is delayed in children with diabetes compared
to healthy youth [33, 34]. However,
34]. However, the consequences for diabetes self-­‐
management and health outcomes have deterioration of glycemic control and quality of life (QoL) [51]. Ten years after diagnosis, young
not consequences
been studied. for
Research that glycemic control tends to be not
better children the
diabetesrevealed self-management
and health
outcomes
have
beenfor studied.
adults with diabetes seem in general to be psychologically well adjusted, but do report lower
with shorter diabetes duration [35-­‐
37]. One study specifically showed that in T1D patients Research
revealed that glycemic control
tends to be better for children with shorter diabetes
perceived competence, including self-worth [52]. Low (diabetes specific) self-esteem is found
with pubertal compared to specifically
pre-­‐pubertal onset, that
glycemic ontrol wwith
as better and compared
daily insulin duration
[35-37].
One study
showed
in T1Dcpatients
pubertal
to be associated with poor adherence and a predictor of deteriorating glycemic control in late
doses were lower after 6 years control
of diabetes, irrespective of ainsulin
ge-­‐related factors 38]. Pafter
erhaps to
pre-pubertal
onset,
glycemic
was better
and daily
doses were [lower
as a result of that, pre-­‐pubertal onset of diabetes is found to predict earlier onset of 6 years of diabetes, irrespective of age-related factors [38]. Perhaps as a result of that,
retinopathy [39], suggesting that youth diagnosed during or after puberty do better than pre-pubertal onset of diabetes is found to predict earlier onset of retinopathy [39], suggesting
those diagnosed early in life. Older longitudinal studies showed that patients diagnosed that youth diagnosed during or after puberty do better than those diagnosed early in life.
before the age of 13 had better adherence to diabetes management over a 4 year follow-­‐up Older longitudinal studies showed that patients diagnosed before the age of 13 had better
period compared to patients diagnosed after this age [40]. Deterioration in adherence adolescence [29, 53]. However, as stated before, the majority of these studies have not made
occurred in all age ranges as duration increased [35]. It is of note that these longitudinal 18
studies used age as an indicator for pubertal status and not the actual physical development associated with poorer metabolic control. Hormonal fluctuations due to puberty can result
a distinction between pre- and pubertal onset of diabetes and can therefore not inform us on
the relevance of timing of diabetes onset on psychosocial development.
Adolescents with T1D are at an increased risk of disturbed eating behavior (DEB) compared
to healthy peers [54-58] due to hormonal changes [28], the focus on food, issues around control
19
2
2
DINO study design
DINO study design
and autonomy in diabetes care. This ‘Diabulimia’ has been frequently reported among
Family functioning and parental well-being. The way the family of T1D youth functions is
adolescents with T1D; 33-53% reported to engage in unhealthy eating behaviors and insulin
important, both as determinant and consequent of poor diabetes control. A negotiating parent-
restriction for weight purposes was prevalent in up to 30% of patients [56, 59-64]. DEB
child environment is beneficial for children with diabetes. In addition, shared responsibility
increases the risk of poorer glycemic control, earlier complications from diabetes, particularly
for diabetes management tasks is shown to be associated with better psychological health,
retinopathy and nephropathy as well as mortality [57-62, 65-68]. In T1D, it is suggested that
self-care behavior and metabolic control [76-78]. A lack of collaboration between children and
in most cases the DEB developed after diabetes onset [69]. Although the peak of onset of DEB
parents can result in conflicts which are often associated with poor glycemic control and QoL
is in adolescence only one study assessed risk factors for the onset of DEB in adolescent girls
[79, 80], however, this seems to be related to ethnicity [81]. Shared responsibility regarding
[70]. Currently, diabetes teams are hesitant to discuss DEB with their patients [71], because
diabetes tasks between parents and adolescents (rather than complete/sudden transfer of
they are afraid they might bring the association between insulin and weight control to mind
parental control) for diabetes management may serve as a way to achieve autonomy for self-
of the adolescent. It is important to know the timing of onset of DEB, who is at risk, how to
care. Youth with an inordinate self-care autonomy relative to their psychological maturity are
address these behaviors and to be able to identify those at risk for DEB [65].
at greater risk of poor treatment adherence, worse diabetic control and more hospitalizations
[82]. Inconsistencies regarding competence and independence between parents and children
Cognitive development. Neuropsychological research shows that children with T1D, especially
with T1D is associated with poorer diabetes outcomes [83]. Furthermore, the better parents
those with early-onset diabetes (≤6 years of age), have mild impairment of cognitive functioning [72]
are able to adopt youth’s perspectives the better the glycemic control [84]. Recent research
including poorer academic achievement [73], lower verbal intelligence, and worse performance
reveals that parental involvement [85] in diabetes care and greater overall parental support
on measures of attention, executive function, mental flexibility, and psychomotor speed [72]
[86] are associated with better health [85] and service use [86], and greater parental motivation
compared to healthy controls. This challenges adolescents’ diabetes adherence behaviors, since
is related to child’s healthier diet [87]. These findings highlight the importance of parenting
these tasks are of great importance to organize and plan the diabetes management.
practices. One of the major tasks for parents is to be responsive to adolescents’ needs for
In general, adolescence is a critical period for brain maturation, essential for the development
increasing responsibility and decision making power while at the same time maintaining a high
of higher cognitive functions. Significant improvements in cognitive processing speed and
level of cohesiveness in the family. However, parental well-being influences the way this
intellectual functioning are evident throughout adolescence and mature in young adulthood,
task proceeds. Recent research reveals that parents with T1D children were more anxious
with the most dramatic improvements occurring in the development of executive functions
and perceived less family cohesion than the parents of healthy youth [88]. The diagnosis,
including abstract thought, organization, decision making, planning, and response inhibition
hypoglycemic events, as well as the chronic nature of diabetes and its demands all contribute
[44, 74]. This implies that later on in adolescence the ability to critically outweigh the costs and
to anxiety and depressive symptoms in parents [89, 90]. Importantly, worse parental well-being
benefits of (non)adherence behaviors increases [75]. However, until that time the adolescent’s
is shown to be associated with poorer glycemic control of the children [91, 92] and maternal
brain is inclined to engage in risk taking behavior and prefers immediate rather than long term
depression is found to be associated with acute hospitalization [93]. Of interest is how family
satisfaction [74]. Their stage of cognitive development and the challenges facing adolescents,
functioning and parental well-being influences adolescents’ diabetes outcomes and development,
might challenge their ability to manage their diabetes on a daily basis.
and how parental well-being influences youths’ diabetes and psychosocial outcomes.
20
21
2
DINO study design
2
DINO study design
Overall, studies integrating the biological, psychosocial and cognitive developmental trajectories,
Methods / Design
family functioning and parental well-being are lacking [43] with a few positive exceptions.
A prospective multi-center cohort study will be conducted in youth with T1D in the age
Wiebe et al. examined the relationship between self-efficacy, parental responsibilities, pubertal
range 8-15 years at baseline. For 3 years, participants’ biological, psychosocial and cognitive
maturation and adherence [94]. Luyckx et al. determined developmental classes of glycemic
development will annually be assessed (at baseline, 12 months, 24 months and 36 months),
control in young people with T1D throughout adolescence and emerging adulthood, in relation
as represented in Figure 2. During the course of the study, newly diagnosed youth will be
to general family climate and self-concept [29]. King et al. used latent growth class analysis to
included 6 months after diagnosis -when diabetes regimen is habituated- and follow the
look at trajectories of metabolic control in relation to autonomy, diabetes management and
scheduled assessments. By consequence, not all newly diagnosed will have the same number
hospitalizations [85]. These studies used a person-centered approach that is uniquely suited
of assessments. Participants will be divided into two subgroups:
to capture diversification in glycemic control, looking for meaningful subgroups characterized
A) Pre-Pubertal Onset of Diabetes (Tanner stage 1) and
by unique developmental pathways.
B) Pubertal Onset of Diabetes (Tanner stage 2-5).
Group A will provide information on the effects of puberty on the developmental trajectories in
Overall aims and research question. There is paucity of evidence with regard to the question
relation to diabetes outcomes in youth already diagnosed with diabetes. Group B will provide
if and how (living with) diabetes during pre-pubertal years and early adolescence predict
information on how the onset of diabetes affects the developmental trajectories.
glycemic control, self-management, psychosocial functioning and diabetes-related QoL during
later years. The importance of being diagnosed with diabetes before versus during puberty
Procedure and participants. Five Dutch pediatric diabetes care clinics agreed to participate.
has hardly received attention in past decades, while the mechanisms and role of puberty could
Pediatricians will recruit youth diagnosed with T1D and their parents. Exclusion criteria are:
give important information for (well-timed) future interventions.
other types of diabetes than Type 1 (e.g. type 2 or MODY), younger than 8 years or older than
The primary goal of DINO is to further our understanding how the onset of diabetes impacts
15 years at baseline, not speaking the Dutch language, and mental retardation. All medical
the biological, psychosocial and cognitive development and family functioning and parental
parameters are taken from hospital charts and no extra tests will be performed. As represented
well-being during pre-pubertal and pubertal years in Dutch youth with T1D.
in Figure 2, youth between 8 and 15 years at baseline with T1D who consent to participate
will complete an online survey regarding their psychosocial development. If a paper survey is
With DINO we will examine: 1) If and how the onset of T1D before vs. during puberty results
preferred or participants do not respond to the e-mail, a paper version is sent to their home
in different outcomes of glycemic control, self-management, psychological functioning and
address. Due to the age difference, 8-11 year olds will complete a shorter and more simple
diabetes-related quality of life. 2) The timing of onset of disturbed eating behavior, its risk
survey than participants 11 years and older. To gain better insight in the perspectives of the
factors and its prospective course in relation to glycemic and psychological consequences.
adolescents about DEB, we will conduct interviews with a selection of youth. Only youth at risk
3) If and how the onset of T1D before vs. during puberty results in different family functioning
(based on their answers on the online survey) will be invited for the interview. The interviews
and parental well-being. 4) If and how the cognitive development of youth with T1D relates to
will take place at the adolescent’s residence or at the outpatient clinic, depending on the
glycemic control and diabetes self-management.
adolescent’s preference. Parents will report on family functioning and parental well-being
22
23
2
DINO study design
2
DINO study design
by an online survey as well, unless the paper version is preferred. A neuropsychological test
battery will be used to assess the cognitive development. Most test results will be compared
to normative data, however, three measures will also be administered to a gender and age
matched sample of 100 healthy controls as reference values are not yet available for these
2
Table 1: Overview of study measures – Socio-Demographic and clinical data
Error! Style not defined. Socio-­‐Demographic data History Baseline 12 months 24 months 36 months Date of birth H Gender H Ethnicity P Education level child P+C P+C P+C P+C Socioeconomic status P Family structure C C C C Family related life events P P P P P Clinical data H H H H H Treatment regime H H H H Care consumption H H H H H Tanner stage at t ime of diagnosis [41] H Current Tanner stage [41] H H H H Blood p ressure H H H H Weight and H eight H H H H Hemoglobin A1c (HbA1c) H H H H H Number of diabetes related hospitalizations H H H H H At risk: DEB
Diabetic k etoacidosis (DKA) H H H H H questionnaires
Indicators for complications H H H H H Severe hypoglycemic episodes P P P P P tasks. These healthy controls are derived from schools in the Netherlands and will be
measured cross-sectional.
Figure 2: Flowchart annual DINO procedure
Participants
History of m edical and psychological co-­‐morbidity 8-15 years old at baseline
Medical
8-10 year olds:
≥11 year olds:
parameters
Online survey
Online survey
Online survey
parents
Cognitive tests
derived from
hospital charts
H=Hospital, P=Parent, C=Child, HC=Healthy control DEB interviews
Ethical considerations. The study protocol was approved by the medical ethical committee of
Data analyses. Using descriptive statistics, baseline data are analyzed cross-­‐
sectionally and and
Data
analyses. Using descriptive statistics, baseline data are analyzed cross-sectionally
VU University Medical Centre (date: December 19th, 2012). Youth and parents are provided
scores are compared with reference values when applicable. Uni-variate analysis ANOVA will
with written information about the study and are asked to provide written informed consent
be used to explore differences between boys and girls and Tanner stages. Pearson or Spearman
(both parents - if applicable - and youth ≥12) prior to the data collection.
correlation is used to explore associations between several outcomes (such as age, cognitive
scores are compared with reference values when applicable. Uni-­‐variate analysis ANOVA will be used to explore differences between boys and girls and Tanner stages. Pearson or Spearman correlation is used to explore associations between several outcomes (such as age, cognitive development, social-­‐emotional development, glycemic control, psychological development,
development,
glycemic
control, whether psychological
functioning,
functioning, social-emotional
QoL, diabetes management and DEB). To examine associations are Study measures. An overview of study measures is shown in Table 1, 2 and 3. A full description
QoL,
diabetesby management
andmDEB).
examineawhether
associations
are mediated by other
mediated other variables, ultiple To
mediation nalysis will be used [95]. of these measures is presented in Appendix 1.
variables,
multiple mediation analysis will be used [95].
24
25
25
DINO study design
2
DINO study design
2
Error! Style not defined. Table 2: Overview of study m easures – Psychosocial development, DEB, Cognitive development Table
2: Overview of study measures – Psychosocial development, DEB, Cognitive development
Table of study m easures – Parental assessment Table3: 3:Overview Overview
of study
measures
– Parental
assessment
Psychosocial development History Strengths and Difficulties Questionnaire (SDQ) [96] Revised Children’s Quality of Life Questionnaire (KINDL-­‐R) self-­‐esteem subscale [97] KIDSCREEN Autonomy subscale [98, 99] Diabetes Family Responsibility Questionnaire (DFRQ) [100] MIND Youth Questionnaire (MY-­‐Q) [101] & adapted version for 8-­‐10yr o lds Confidence in Diabetes Self-­‐care Youth (CIDS-­‐youth) [102] Mismanagement scale – renewed [103] Adherence H Disturbed Eating Behavior (DEB)* 2 questions regarding d ieting status and frequency Diabetes Eating Problems Scale-­‐Revised (DEPS-­‐R) [61, 68] Questions of the AHEAD study [60] DEB semi structured interview 2 MY-­‐Q subscale body and weight [101] Cognitive development Wechsler Intelligence Scale for Children III (WISC-­‐III) subtests Information; Picture Arrangement; Arithmetic; Block Design; Digit Span [104, 105] Attention Network Task (ANT) -­‐adapted version [106, 107] Eriksen Flanker Task [108, 109] Klingberg Task -­‐adapted version [107, 110, 111] Behavior Rating Inventory of Executive Functioning questionnaire (BRIEF) [112, 113] H=Hospital, P=Parent, C=Child, HC=Healthy control. Baseline P+C ≥11 12 months P+C ≥11 24 months P+C ≥11 36 months P+C ≥11 C C C C C ≥11 P+C ≥11 C ≥11 P+C ≥11 C ≥11 P+C ≥11 C ≥11 P+C ≥11 C C C C C ≥11 C ≥11 C ≥11 C ≥11 C ≥11 H C ≥11 H C ≥11 C ≥11 H C ≥11 C ≥11 H C ≥11 be administered to a gender and age matched sample of 100 healthy controls. Because of the
C ≥11 C ≥11 C ≥11 self-management, psychological functioning and diabetes related QoL. The differences between
C+HC C ≥11 C ≥11 P C C ≥11 P C C ≥11 P C the two groups on family functioning and parental well-being and the development of DEB are
Baseline 12 months 24 months 36 months P P P P P P P P P P P P Revised (PAID-­‐PR) [114, 1 15] WHO-­‐Five Well-­‐being Index (WHO-­‐5) Diabetes Family Behavior Checklist (DFBC) [119] H=Hospital, P=Parent, C=Child, HC=Healthy control With regard to the cognitive development, performance on the computerized measures will
longitudinal nature of this study, Generalized Estimation Equations (GEE) is used to examine
the differences between the two groups diagnosis pre-puberty vs puberty on glycemic control,
investigated by GEE as well. GEE adjusts for the correlation between repeated observations
taken in the same patient and has the advantage of handling longitudinal data on subjects
with varying numbers of unequally spaced observations. The latter is important, because
the assessments are scheduled within routine care and as a consequence, the time between
C+HC C C C the consultations will differ. Longitudinal linear regression analyses, using GEE, enables us to
C+HC C+HC C C C C C C examine the association between the developmental trajectories in relation to diabetes onset
P+HC P P P younger participants. Kindly note Appendix 1 in this thesis for a d escription of study m easures. 26
Problem A reas In Diabetes-­‐Parents History [116-­‐118] * DEB is assessed in a step-­‐wise manner in order to minimize the burden in adolescents with no DEB and 26
Parental assessment and outcomes.
Data are controlled for demographic and clinical variables and examined for associations
with and predictors of biological and social-emotional developmental outcomes. Latent
class growth analysis is used to identify developmental trajectories of glycemic control and
psychological, cognitive and family functioning.
27
2
DINO study design
DINO study design
For the analyses of the interviews, a framework approach is used [120, 121]: interviews are
diabetes care. This might include the use of more sensitive screening instruments, for example
transcribed verbatim, and key words and codes to extract content from the text are assigned
to assess cognitive functioning in relation to self-management, risk factors for DEB and other
by two researchers, at least for the first few transcripts using Atlas.Ti.
psychological problems. This would enable diabetes teams to better personalize their care for
adolescents with diabetes. Better understanding can contribute to the development of new
Sample size. Sample size calculations indicate that a sample of 86 patients is sufficient to detect
interventions aimed at, for example, prevention and/or treatment of depressive symptoms
a statistically clinically difference of ≤ 0.5% HbA1c (sd = 1.1%) at a significance level of 5%
and better tailoring of self-management education to the developmental phase of the child.
with a power of 80%, given three follow-up measurements using GEE analyses and taking into
First results of DINO are expected in 2016.
account the within-subject correlation (rho = 0.7). Sample size calculations for psychological
functioning (as measured with the SDQ) indicate that a sample of 40 patients is sufficient
Limitations of this comprehensive study. A selection bias and the adolescents lost to follow
to detect a difference of 20% in the proportion likely “cases” with mental health disorders
up might influence the external validity of the results. However, this is almost unavoidable
(assuming p=0.33 [122]) at a significance level of 5% with a power of 80%, given three follow-
in longitudinal research studies. In addition, the newly diagnosed youth are not followed the
up measurements using longitudinal logistic regression analyses and taking into account the
entire 3 years of the study. To assess psychological development and family functioning and
within-subject correlation (rho = 0.2). This means that in order to detect differences in HbA1c
parental well-being self-report measures are used. With regard to the cognitive development,
and likely cases of psychological dysfunction between youth with pre- vs. pubertal onset of
there can be an interviewers or observers bias as multiple test leaders will perform the
diabetes, both groups should contain at least 43 patients. Given an expected drop-out rate
neuropsychological assessments throughout the Netherlands. With a standardized training
of 10%, we will include at least an additional 4 patients in each group. Therefore, we aim to
program we try to minimize the bias. The use of neuroimaging techniques such as functional
include a minimum of 100 boys and 100 girls (N=200).
Magnetic Resonance Imaging (fMRI) might provide a more objective, additional source of
information, nonetheless this was not an option within the current budget. As seen in the
Discussion
bio-psychosocial model (Figure 1) the T1D adolescent functions in a broad social network
To the best of our knowledge, only a few studies have examined the effect of diabetes onset
School and friends for example play an important role in youth’s development, yet we did not
during pubertal vs. pre-pubertal years and little longitudinal research is available in children
include these factors because of the feasibility of the study.
and teenagers with T1D, although a lot changes during pubertal years. There is large individual
variability in glycemic and psychological trajectories. The way youth and families cope with
puberty and the developmental changes differ as well. In our research project DINO we aim to
assess the different developmental trajectories (biological, psychosocial - including disturbed
eating behaviors - and cognitive) and family functioning and parental well-being. This will provide
insight in protective and risk factors for glycemic outcomes and in who needs which support
at what moment in time. Better understanding contributes to the optimization of pediatric
28
29
2
Chapter
3
Low self-confidence and diabetes mismanagement in youth
with type 1 diabetes mediate the relationship
between behavioral problems and elevated HbA1c
‘When I was fourteen, shit hit the fan. I had all kinds of problems
I started smoking weed and was more or less depressed
My glucose levels went from bad to worse’
Minke Eilander
Maartje de Wit
Joost Rotteveel
Frank Snoek
Journal of Diabetes Research, 2015. 501:681421
31
Self-confidence, diabetes management and behavior problems
3
Self-confidence, diabetes management and behavior problems
Abstract
Introduction
Introduction. Previous studies indicated an association between behavior problems
Adolescence is characterized by major biological and psychosocial changes, which interact
(internalizing, externalizing) and glycemic control (HbA1c) in youth with type 1 diabetes (T1D).
in complex ways. This is particularly true for youth with type 1 diabetes (T1D) and probably
The aim of this study is to examine if this association is mediated by self-confidence and
explains the fact that youth with diabetes have the poorest glycemic control of all age groups,
mismanagement of diabetes.
with less than 15% of adolescents with type T1D reaching Hemoglobin A1c (HbA1c) levels
below target [27, 29, 30, 76, 77]. On top of the hormonal changes that negatively affect
Methods. Problem behavior was assessed with the Strengths and Difficulties Questionnaire.
blood glucose control, adolescents with T1D have an increased risk of developing depression,
Mediating variables were assessed using the Confidence In Diabetes Self-care-Youth and
anxiety and disturbed eating behaviors, relative to healthy youth. These problems typically
Diabetes Mismanagement Questionnaire. HbA1c was derived from hospital charts. Bootstrap
occur in mid-adolescence [44] and result in poor glycemic control [45-48, 122]. Externalizing
mediation analysis for multiple mediation was utilized.
behavior problems (e.g., attention and disruptive behavior complaints) are associated with
decreased glycemic control as well [123-126]. Although behavior problems at diagnosis
Results. 88 youths with T1D, age 11-15 yr participated. The relation between both overall
do not seem to impact lifelong poor glycemic control [127], they have found to be directly
problem behavior, externalizing behavior problems and HbA1c was mediated through
associated with hyperglycemia [126]. Adolescents showing external problem behavior seem to
confidence in diabetes self-care and Mismanagement (a₁b₁ + a₂b₂ path; point estimate = 0.50
be generally unresponsive to punishment, are often impulsive and have concentration problems
BCa CI 95% 0.25-0.85; a₁b₁ + a₂b₂ path; point estimate = 0.73BCa CI 95% 0.36-1.25).
[128]. Problematic behavior, both internal and external, frequently coincides with low self-efficacy
beliefs, low self-esteem, dysfunctional coping mechanisms, and increased risk taking behavior, all
Conclusions. Increased problem behavior in youth with T1D is associated with elevated HbA1c
complicating daily self-management of diabetes [128]. Self-efficacy beliefs, for example, low self-
and mediated by low self-confidence and diabetes mismanagement. Screening for problem
confidence, and diabetes mismanagement are likely to mediate the relationship between behavior
behavior and mismanagement, and assisting young patients in building confidence seem
problems and poor glycemic outcomes, but this hypothesis has not been previously tested.
indicated to optimize glycemic control.
Using baseline data of multicenter cohort DINO study (diabetes in development) that
examines the complex interaction between biological and psychosocial development during
adolescence [129], we examined whether overall, external, and internal problem behavior are
associated with glycemic control and whether this relationship is mediated by confidence in
diabetes self-care and mismanagement.
Materials and Methods
Participants. Youth aged 8 to 15 treated (N = 598) in 5 pediatric diabetes centers in the
Netherlands were invited to participate in the DINO study. The participating centers provide
32
33
3
3
Self-confidence, diabetes management and behavior problems
Self-confidence, diabetes management and behavior problems
secondary and tertiary clinical care to children and adolescents with T1D in their region and
used and 1 was adjusted. The recall period was changed from 10 days to the past week.
can be considered representative of youth with T1D in the Netherlands. Exclusion criteria were
Answers are given on a 5 point Likert scale (e.g., “In the past 7 days, how often did you miss
mental retardation, diabetes other than type 1 and diagnosis less than 6 months prior to the
shots/did not bolus?”). Cronbach’s α of the original version is 0.74 and 0.60, respectively [103].
start of the study. Written informed consent was obtained from all parents and adolescents 12
Cronbach’s α of the adapted version in the current study was 0.47. Higher scores indicate more
years and older. Participants completed an online survey. In view of their age, 8-11 year olds
mismanagement (range 4-16).
completed a shorter survey than participants 11 years and older. Data from the latter survey
Demographic and diabetes related data were derived from hospital charts during the same
were used for the study reported in this paper. In total, 151 children and adolescents (25.3%)
period as the completion of the survey. HbA1c was used as a marker of glycemic control over the
agreed to participate, of whom 100 were ≥ 11 years. The DINO study was approved by the
past 8-12 weeks, with recommended target <7.5%, 58 mmol/mol [10].
medical ethical committee of VU University Medical Center.
Analyses. T-tests and chi-square tests were applied in order to examine differences in HbA1c, age
Measures
and gender between responding and nonresponding adolescents. To examine whether there was
Problem behavior was assessed using the Strengths and Difficulties Questionnaire (SDQ). SDQ
a relationship between problem behavior (overall, external, and internal) and glycemic control,
[96, 130] captures emotional and behavioral functioning, and contains 25 items, rated on a
and whether this relationship is mediated by confidence in diabetes self-care and diabetes
3-point Likert scale (e.g., “Other people my age generally like me.”). The SDQ comprises five scales:
mismanagement, bootstrap mediation analysis for multiple mediation through the Indirect Macro
emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems,
was applied [95, 131], correcting for age and gender. Since we chose to use more than one possible
and pro-social behavior. The overall score of problem behavior (range 0-40) can be divided into
mediator, this method was considered more appropriate than traditional models [95, 132-134].
external (range 0-20) and internal (range 0-20) problem behavior. Cronbach’s α was 0.70 on
Analyses were performed with a bootstrap of 5000 resamples, in which random samples based on
the overall scale [96, 130], in the current study 0.60. Higher scores indicate more problematic
the original data are generated. A 95% confidence interval (CI) was calculated [131].
behavior; scores ≤13 are considered normal.
Confidence in diabetes self-care was assessed using an adapted adolescent version of the
Results
Confidence in Diabetes Self-Care Scale (CIDS) [102]. The original adult version of the CIDS consists
A total of 88 adolescents (45 boys) completed the online survey (88.0% of the 100 youths ≥ 11 y),
of 20 items on a 5 point Likert scale, Cronbach’s α=0.86 [102]. The adapted youth version consists
mean age 12.9 (±1.2) years with a mean disease duration of almost 6 years. Baseline
of 12 items (e.g., “I believe I can check my blood glucose at least 2 times a day”): 10 of the original
characteristics are shown in Table 1. There were no differences in HbA1c, age and gender
questionnaire, 2 items combined to 1 (original questions 2 and 20) and 1 additional item regarding
between responders and nonresponders. Thirteen adolescents (14.8%) reported problem
alternations in blood glucose. Cronbach’s α of the CIDS-Youth in the current study was 0.79.
behavior above the normal range (overall problem behavior score >13). Almost three-quarters
Higher scores represent higher diabetes self-confidence (range 12-60).
of adolescents (72.7%) had HbA1c levels above target.
Mismanagement in diabetes self-care was assessed using an adapted version of the
mismanagement scale [103]. The original version consists of 10 items of which 3 items were
34
35
3
Error! Style not defined. Self-confidence, diabetes management and behavior problems
Self-confidence, diabetes management and behavior problems
Error! Style not defined. Figure 1: Graphic representation of the multiple mediation model of the associations between overall
Figure 1:Graphic
representation
of the
multiple
mediation
model
the associations
between self-­‐care and
problem behavior and glycemic control with of confidence in diabetes Table 1 Table 1:
Characteristics of participating adolescents
Characteristics of participating adolescents 3
overallmismanagement problem behaviorof and
glycemic
control
diabetes
self-carewhile diabetes self-­‐
care. with
5000 confidence
resamples inwere calculated using the
and mismanagement
of diabetes
resamples were calculated while bootstrap method (Preacher self-care.
& Hayes, 5000
2008). bootstrap method (Preacher & Hayes, 2008).
using the
Boys (N / %) 45 (51.1) Age (yrs) 12.9 ±1.2 HbA1c 64.3 mmol/mol (8.0%) ±11.5 mmol/mol Age diabetes onset 7.1 ± 3 .8 Diabetes duration (yrs) 5.8 ± 3 .8 Pump / injections (%) 80.7 / 19.3 ≥4 p er day Traditional family composition (%) 83 SDQ Overall p roblem behavior (0-­‐40) 8.6 ± 4 .3 SDQ External problem b ehavior (0-­‐20) 4.9 ± 2 .8 SDQ Internal problem b ehavior (0-­‐20) 3.7 ± 2 .8 CIDS-­‐youth (12-­‐60) 51.2 ± 5.3 Mismanagement (4-­‐16) 6.4 ± 2 .0 Total effect (c path) β = 0.625; P = 0 .029 Overall problem behavior HbA1c Direct effect (c' path) β = 0.120; P = 0 .685 Direct effect (a¹ path) β = -­‐0.362; p < 0 .01 Direct effect (b¹ path) β = -­‐0.794; P < 0 .01 Data are m eans ± SD, unless otherwise indicated Overall problem behavior. Figure 1 shows the multiple mediation model of the associations
between overall problem behavior and glycemic control. A significant total effect (c path)
was found
between overall problem behavior and glycemic control (β = 0.625, p = 0.029),
Overall Problem Behavior. Figure 1 shows the multiple mediation model of the associations Direct effect (a² path) β = 0.183; P < 0 .01 indicating
that higher
problem
behavior
scores control. are associated
with higher
HbA1c.
between overall overall
problem behavior and glycemic A significant total effect (c path) Mediation
analysis showed that this
relationship was mediated by confidence in diabetes
was found between overall problem behavior and glycemic control (β = 0.625, p = 0.029), indicating that higher overall scores are associated with higher self-care
and mismanagement
as theproblem indirect behavior effect was
significant
(a₁b₁ + a₂b₂
path
pointHbA1c. Mediation this relationship mediated confidence in diabetes estimate
= 0.50, analysis BCa 95%showed 0.25 tothat 0.85)
and
the directwas effect
(c’ path)by was
not anymore
self-­‐c=are and mismanagement as the indirect effect was significant (a₁b₁ + confidence
a₂b₂ path point (β = 0.120,p
0.685).
Increased overall problem
behavior
was associated
with higher
estimate = 0.50, BCa 95% 0.25 to 0.85) and the direct effect (c' path) was not anymore in diabetes self-care (a₁ path β = -0.362, p < 0.01) and worse self-care of diabetes (a₂ path β
(β = 0.120,p = 0.685). Increased overall problem behavior was associated with higher = 0.183, p < 0.01). Lower confidence in diabetes self-care was associated with higher HbA1c
confidence in diabetes self-­‐care (a₁ path β = -­‐0.362, p < 0.01) and worse self-­‐care of diabetes (b₁ path β = -0.794, p < 0.01). The association between mismanagement of diabetes and higher
(a₂ path β = 0.183, p < 0.01). Lower confidence in diabetes self-­‐care was associated with HbA1c was borderline significant (b₂ path β = 1.189, p = 0.057).
higher HbA1c (b₁ path β = -­‐0.794, p < 0.01). The association between mismanagement of Confidence in diabetes self-­‐care Specific indirect effect (a¹b¹ path) Point estimate = 0 .29; BCa 95% CI 0.10 to 0.53 Direct effect (b² path) β = 1.189; P = 0 .057 Mismanagement Specific indirect effect (a²b² path) Point estimate = 0 .22; BCa 95% CI 0.02 to 0.51 Total indirect effect (a¹b¹+a²b² path) Point estimate = 0.50 BCa 95% 0.25 to 0 .85 diabetes and higher HbA1c was borderline significant (b₂ path β = 1.189, p = 0.057). 38 36
37 37
3
3
Self-confidence, diabetes management and behavior problems
Self-confidence, diabetes management and behavior problems
External problem behavior. Figure 2 shows the multiple mediation model of the associations
Figure 2: Graphic representation of the multiple mediation model of the associations between
between external problem behavior and glycemic control. A significant total effect (c path)
external
problem behavior and glycemic control. 5000 resamples were calculated was found between external problem behavior and glycemic control (β = 1.00, p = 0.02):
while Graphic using therepresentation bootstrap method
& Hayes,
2008).model of the associations between
of (Preacher
the multiple mediation Error! Style not defined. 3
increased external problem behavior was associated with higher HbA1c. Again, multiple
external problem behavior and glycemic control. 5000 resamples were calculated while mediation analysis showed that this relationship was mediated by confidence in diabetes
self-care and mismanagement as the indirect effect (a₁b₁ + a₂b₂ path point estimate = 0.73,
BCa 95% 0.36 to 1.25) was significant and the direct effect (c’ path β = 0.27, p = 0.56) was
External problem behavior in diabetes self-care (a₁ path β = -0.49, p = 0.02) and worse self-management of diabetes
(a₂ path β = 0.32, p < 0.01). Low confidence in diabetes self-care and worse self-management
Internal problem behavior. Figure 3 shows the multiple mediation model of the associations
between internal problem behavior and glycemic control. In contrast to the overall and
external problem behavior, the total effect between internal problem behavior and glycemic
low confidence in diabetes self-care (a₁ path β= -0.418, p = 0.042), but not with worse
Direct effect (b¹ path) β = -­‐0.770; P < 0 .01 Direct effect (a² path) β = 0 .316; P < 0 .01 control was not significant (c path β = 0.494, p = 0.270). However, multiple mediation analysis
as a result of this mediation. Increased internal problem behavior was associated with
Direct effect (a¹ path) β = -­‐0.499; P = 0 .02 0.14 to 0.96). The association of the direct effect (c’ path β = -0.01, p = 0.981) decreased
Direct effect (c' path) β = 0.267; P = 0 .56 (b₁ path β = -0.77, p < 0.01; b₂ path β = 1.12, p = 0.08).
as the indirect effect was significant (a₁b₁ + a₂b₂ path point estimate = 0.50, BCa 95%
HbA1c of diabetes were both associated with higher HbA1c, however the latter was not significant
did show a significant mediation by confidence in diabetes self-care and mismanagement
Total effect (c path) β = 0.997; P = 0 .02 not anymore. Increased external problem behavior was associated with low confidence
Confidence in diabetes self-­‐care Specific indirect effect (a¹b¹ path) Point estimate = 0 .38; BCa 95% CI 0.11 to 0.71 Mismanagement Specific indirect effect (a²b² path) Point estimate = 0 .35; BCa 95% CI 0 .02 to 0.81 self-management of diabetes (a₂ path β = 0.126, p = 0.104). Low confidence in diabetes
self-care and worse self-management were both associated with a higher HbA1c (b₁ path β =
Point estimate = 0.73 -0.820, p < 0.01; b₂ path β = 1.286, p = 0.027).
BCa 95% 0.36 to 1 .25 Total indirect effect (a¹b¹+a²b² path) 38
39
Direct effect (b² path) β = 1.120; P = 0 .08 Self-confidence, diabetes management and behavior problems
Figure 3: Graphic representation of the multiple mediation model of the associations between 3
internal problem behavior and glycemic control. 5000 resamples were calculated while using the bootstrap method (Preacher & Hayes, 2008).
Self-confidence, diabetes management and behavior problems
Discussion
The aim of the present study in adolescents with T1D was to investigate whether there is a
relationship between problem behavior and glycemic control and whether this relationship is
mediated by low confidence in diabetes self-care and mismanagement of diabetes. Increased
Error! Style not defined. overall and external problem behavior were found to be associated with increased HbA1c and
Figure 3: these relationships were mediated by confidence in diabetes self-care and self-management
of diabetes. Interestingly no total effect was found between internal problem behavior and
Total effect (c path) β = 0.494; P = 0 .270 glycemic control, and the relationship between internal problem behavior and diabetes
Internal problem behavior HbA1c interpret these findings with caution, as the relationship could be dose-dependent: the risk
Direct effect (c' path) β = -­‐0.01; P = 0 .981 of worsened illness management increases when internal problems get more serious [128].
Direct effect (a¹ path) β = -­‐0.418; P = 0 .042 Direct effect (b¹ path) β = -­‐0.820; P < 0 .01 management was not significant; however, the indirect effect was significant. We should
The adolescents participating in our study reported somewhat less problematic behavior
on all three SDQ scales (overall, external and internal problem behavior) compared to the
11-16 year old adolescents participating in SDQ validation study published in 2003 (overall
Direct effect (a² path) β = 0.126; P = 0.104 Confidence in diabetes self-­‐care Specific indirect effect (a¹b¹ path) Point estimate = 0 .34; BCa 95% CI 0.04 to 0.73 Mismanagement Specific indirect effect (a²b²path) Point estimate = 0 .16; BCa 95% CI -­‐0.004 to 0.49 Total indirect effect (a¹b¹+a²b² path) Point estimate = 0.50 BCa 95% 0.14 to 0 .96 Direct effect (b² path) β = 1.286; P = 0 .027 M=8.6 compared to 9.9, external M=4.9 compared to 5.8 and internal M=3.7 compared to
4.1) [130]. Our sample appears less problematic than previously reported in the literature
where adolescents with T1D were found to have more problem behavior compared to
healthy peers [45-48]. Possible explanations for this discrepancy could be that we included a
slightly younger group compared to the SDQ validation study, a selection bias or the fact that
previous research was conducted a decade ago. Nevertheless, almost 15% of the adolescents
in our study reported levels of problem behavior above the normal range. This underscores
the clinical relevance of our proposed model.
The relationship between more behavior problems and suboptimal HbA1c levels
has been demonstrated in other studies as well [123-125]. The present study confirms
our hypothesis that this relationship is mediated by confidence and self-management
of diabetes. More behavioral problems seem to decrease the adolescent’s confidence
in the management of their diabetes, in concordance with previous research [128].
The need to address psychosocial issues in pediatric diabetes care is recognized [79, 135].
40
41
3
3
Self-confidence, diabetes management and behavior problems
Self-confidence, diabetes management and behavior problems
Psychosocial well-being is an important outcome in and of itself, but also has clear relevance
consists of 4 rather than the 10 items in the original questionnaire. Psychometric validation of
to understanding problems in achieving satisfactory glycemic control [136, 137]. Timely
the scale warrants further research.
detection and management of psychosocial issues however, has shown to be difficult in
routine care, where time is limited and measurements of adolescents’ physical health often
Conclusion. More problem behavior in adolescents with type 1 diabetes is associated with
have priority [138]. Our findings corroborate the clinical relevance of finding practical ways to
worsened glycemic control and this relationship is mediated by low confidence in diabetes
ensure that assessment and management of behavioral problems in adolescents are in place.
self-care and poorer self-management of the diabetes. This finding has clinical implications.
Psychosocial screening should include both internal problem behavior as well as external
Limitations. Although the current study contributes to enhancing our understanding of the
problem behavior. To assist adolescents in achieving better glycemic control, it would seem
complex interactions between psychosocial and biological developmental trajectories, some
imperative to help them build their confidence and reduce diabetes mismanagement, for
limitations should be taken into account. First, our study was cross-sectional and we cannot
example, improving their self-care practices.
infer causality. The 12 year study of Northam et al. examined the relationship between
problem behavior at diagnosis and longtime poor glycemic control, but did not look at possible
mediating pathways [127]. Future longitudinal research is planned to examine this relationship
in more detail. Moreover, the relationship between the psychosocial development and diabetes
outcomes is multifaceted. In the present study we only tested the contribution of a few of the
factors involved and took HbA1c as a marker for glycemic control. In addition, we may want
to explore the momentary impact of behavior problems on blood glucose fluctuations which
is likely to exist. Conversely, high and low blood glucose values can influence the adolescents’
behavior, thereby creating a vicious cycle of events [139]. With regard to the measurements,
we chose to administer the questionnaires via the internet, for pragmatic reasons. Also online
administration of questionnaires is patient-friendly and more appealing to adolescents than
traditional paper-and-pencil. We should acknowledge that we cannot validate that respondents
have all filled in the online questionnaire without interference from others (e.g., parent,
siblings); however, several studies have shown over the years that questionnaires completed via
the internet are as reliable as paper-and-pencil [140, 141]. The internal consistency (Cronbach’s
α) of the adapted version of the mismanagement scale proved to be relatively low in our study
(α = 0.47). This may be due to the fact that management behaviors are relatively independent
of one another or due to the small number of items, as the adapted version of the questionnaire
42
43
3
Chapter
4
Disturbed eating behaviors in adolescents with type 1 diabetes
How to screen for yellow flags in clinical practice?
‘Nobody explained to me that the consequences
of high glucose levels were that serious’
Minke Eilander
Maartje de Wit
Joost Rotteveel
Henk-Jan Aanstoot
Willie Bakker-van Waarde
Mieke Houdijk
Roos Nuboer
Per Winterdijk
Frank Snoek
Pediatric Diabetes, 2016. doi:10.1111/pedi.12400
45
Disturbed eating behavior
Disturbed eating behavior
Abstract
Introduction
Background. Adolescents with type 1 diabetes are at an increased risk of disturbed eating
Adolescents with type 1 diabetes (T1D) are at risk of disturbed eating behavior (DEB) and more
behaviors (DEBs).
severe eating disorders [54, 58]. DEBs increase the risk for poor glycemic control, premature
diabetes complications, as well as mortality [58, 60, 61, 65, 67, 68]. The focus on food in
4
Objective. The aims of this study are to (i) explore the prevalence of DEBs and associated
relation to blood glucose, and insulin-related weight gain [142, 143], might be contributing
‘yellow flags’, and (ii) establish concordance between adolescents–parents and adolescents–
factors. Diabetes patients have the unique possibility to manipulate weight by under dosing
clinicians with respect to DEBs.
or omitting insulin [54], causing rapid weight loss. This ‘diabulimia’ may present in both obese
and slim patients. Precise prevalence data are deficient, due to differences in definition and
Methods. Adolescents (11–16 yr) and parents completed questionnaires. A stepwise approach
measurement [144], but it is estimated that 20-40% of T1D youth (mis)use insulin to influence
was used to assess DEBs: only adolescents whose answers raised psychological yellow flags for
their weight [144, 145] and the probability for T1D girls to develop DEBs is 60-79% [17]. DEBs
DEBs completed the Diabetes Eating Problems Scale – Revised and questions from the AHEAD
are a sincere cause of concern as it might develop into an eating disorder [146]. A total of
study. Parents and clinicians shared their observations regarding possible DEBs. Kruskal–Wallis
61% of young T1D patients with a history of DEBs still misused insulin 8–12 yr later [64], and a
tests, post hoc Mann–Whitney U test, and chi-squared tests were utilized to examine clinical
10-yr follow up study revealed that 32% of young women with T1D met the criteria for an
yellow flags. Cohen’s kappa was used to assess concordance.
eating disorder and 59% showed DEBs [17]. Previous research identified risk factors for DEBs:
(a history of) obesity, female, early menarche, body dissatisfaction, depression, reduced quality
Results. Of 103 adolescents participated (51.5% girls), answers of 47 (46.5%) raised psychological
of life (QoL), perfectionism, impulsivity, adverse parenting, family dieting, pressure to be slim,
yellow flags, indicating body and weight concerns. A total of 8% scored above cut-off for DEBs.
preoccupation with food intake/diet; low self-esteem and anxiety, and diabetes specific: poor
Clinical yellow flags were elevated glycated hemoglobin A1c (p = 0.004), older age (p = 0.034),
metabolic control; recurrent diabetic ketoacidosis (DKA); recurrent hypoglycemia (a possible
dieting frequency (p = 0.001), reduced quality of life (p = 0.007), less diabetes self-confidence
sign of induced vomiting or overdosing insulin after binge eating); missed clinical visits; refusal
(p = 0.015), worsened diabetes management (p < 0.001), and body dissatisfaction (p < 0.001).
to be weighed; T1D diagnosis close to puberty [16, 61, 147-150].
Body Mass Index (BMI) z-scores and gender were no yellow flags. Concordance between
Research showed that diabetes clinicians find it difficult to identify DEB as the signs are
parents and adolescents was slight (k = 0.126 and 0.141), and clinicians and adolescents
often unclear and not spontaneously brought forward [71]. Clinicians appear to be hesitant to
was fair (k = 0.332).
discuss DEBs with their patients, afraid they might bring the association between insulin and
weight to mind [71]. An alternative would be to assess and discuss these issues with parents.
Discussion. Half of the adolescents reported body and weight concerns, less than 1 in 10
Literature on the accuracy of parental observations with regard to potential DEB or those of
reported DEBs. Screening for yellow flags for DEBs as a part of clinical routine using a stepwise
the clinicians lacks. As the prevalence of DEBs is higher in girls than in boys (both in healthy
approach and early assistance is recommended to prevent onset or deterioration of DEBs.
and T1D adolescents) [60, 61, 151], DEBs in boys often go unnoticed by diabetes teams as they
do not screen DEBs in boys [71, 152]. To the best of our knowledge, there is little guidance
46
47
4
4
Disturbed eating behavior
Disturbed eating behavior
for pediatric diabetes teams when it comes to DEBs. To this purpose, identification of ‘yellow
the way you eat so you can lose weight)’(never, 1–4 times, 5–10 times, ≥10 times, and
flags’– both psychological (concerns about weight/looks) and clinical (potential risk factors) –
always) [60]. Endorsement of any of these items was considered a psychological yellow flag
should be helpful, particularly when these are easily accessible in routine care, for example
(Table 1). Further assessment was conducted in step 2: The direct assessment of eating and
as part of periodic QoL monitoring [153].
weight control behavior by using the Diabetes Eating Problems Scale-Revised (DEPS-R) [68]
Here we report on cross-sectional data in a sample of Dutch T1D adolescents participating
and 14 questions from the AHEAD study [60]. The following factors were examined to identify
in the larger ongoing Diabetes IN development (DINO) study [129] to answer two research
clinical yellow flags: gender, dieting frequency, age, glycated hemoglobin A1c (HbA1c), BMIz,
questions using a stepwise approach. A: What is the prevalence of DEBs and what are the
QoL, diabetes self-confidence, diabetes management, and body image. Parents and clinicians
associated yellow flags? B: What is the concordance regarding DEBs between the adolescents
were asked about possible DEB observed.
and their parents, and the adolescents and their clinicians?
Measures. Diabetes specific problematic eating and weight control behavior was assessed by
Methods
using the DEPS-R [68] supplemented by 14 questions from the AHEAD study [60]. Both measures
Participants. The DINO study [129] was approved by the medical ethical committee of VU
were not available in Dutch, therefore a translation was made, using the following protocol:
University Medical Center. T1D youth 8–15 yr treated in five Dutch diabetes centers were
two researchers from our study group made independent translations and they met in order
asked to participate in this longitudinal study into the psychosocial and biological development
to reach consensus. As part of the translation process, the measures were back-translated by
of T1D adolescents. Exclusion criteria were mental retardation; diabetes other than type 1;
a bilingual nurse, fluent in English and Dutch. This English version was approved by the author
diagnosis <6 months. Written informed consent was obtained from parents and adolescents
of the DEPS-R. The DEPS-R questionnaire measures eating problems in the context of diabetes.
≥12 yr. Participants and parents completed an online or (in case of non-response) a paper
Originally a cut-off of ≥20 has been established for being at risk [61, 68], however, one question
survey. Due to the age differences, 8-11yr olds completed a shorter survey than participants
concerned ketones. Most Dutch T1D adolescents are not familiar with the exact definition
≥11yr (N = 118). Cross-sectional data from the latter survey are reported here. Medical data
of ketones, therefore it was excluded from this Dutch version. Consequently, the new
were obtained from hospital charts.
cut-off was ≥18, range 0–75 (original 0–80), higher scores indicate more weight loss activities.
In the this study, Cronbach’s α = 0.81.
Procedure. Eating problems were assessed in a step-wised manner in order to minimize
Weight loss techniques were examined with 14 questions from the AHEAD study [60].
the burden in adolescents with no DEBs: all adolescents ≥11 yr completed a survey in
One question about smoking was excluded, as we believed the 11-13 yr participants would
which their psychosocial situation was measured (step 1). When answers raised yellow
not relate. Eating behaviors were classified as ‘very unhealthy’ if adolescents engaged in
flags for DEBs, adolescents were directed to diabetes-specific eating behavior questions
one or more of the following activities in the past 3 months in order to lose weight: took
(step 2). Step 1 involved administering: The MIND Youth Questionnaire (MY-Q) captures QoL [101].
diet pills, vomited, skipped insulin dose(s), took less insulin than prescribed, used laxatives,
Its subscale Body /weight served as a start. One question about dieting frequency was
or used diuretics. Eating behaviors were classified as ‘unhealthy’ if adolescents engaged in
asked: ‘How often have you gone on a diet during the last year? (by ‘diet’ we mean changing
one or more of the following activities: fasted, ate very little food, used food substitutes
48
49
4
4
Disturbed eating behavior
Disturbed eating behavior
(powder/special drink), or skipped meals. The following activities were classified as ‘healthy’:
the DEPS-R, and those who did, four categories were constrained: ‘No DEPS-R’; DEPS-R
exercised, ate more fruits and vegetables, ate less high-fat foods, or ate less sweets [60].
scores 0–8; 9–17; ≥18 [68]. To identify clinical yellow flags, Kruskal–Wallis tests were used for
AHEAD answers are used as descriptive data.
continuous variables. When significant, post hoc Mann–Whitney U test was utilized in which
Diabetes specific QoL was assessed by the MY-Q. Answers are given on a 5-point Likert scale.
Bonferroni correction was applied and the significance level was set at α (0.05/6) ≤0.008.
Range 0–100, higher scores indicate better QoL. MY-Q domain Body/ Weight was excluded
Regarding dieting frequency: answers were recoded into ‘never’ and ‘dieted’. For dichotomous
from the MY-Q sum score and analyzed as a separate variable; body image (Range 0–100,
variables gender and dieting frequency chi-squared tests were utilized, in which DEPS-R
higher scores indicate more body satisfaction). Without this subscale Cronbach’s α = 0.84.
9–17 and DEPS-R ≥ 18 formed one group as ≥18 had less than five counts. When significant,
Diabetes specific self-confidence was assed using an adapted youth version of the Confidence
χ² standardized residual was determined: >1.96 or <–1.96 was considered significant.
in Diabetes Self-Care Scale (CIDS) [102]. This version consists of 12 items: 10 original, 2 were
To examine agreement between adolescents’ MY-Q item answers and parent’s answers,
combined to 1 (original items 2 and 20) and 1 added regarding alternations in blood glucose.
Cohen’s kappa was determined (–1 through 1: < 0, no agreement; 0–0.20, slight; 0.21–0.40, fair;
Cronbach’s α of the CIDS-Youth in this study is 0.77. Higher scores represent higher diabetes
0.41–0.60, moderate; 0.61–0.80 substantial; 0.81–1, almost perfect agreement). Adolescent’s
self-confidence (range 12–60). Management in diabetes self-care was assessed by a renewed
answers about body satisfaction were recoded: never and seldom into ‘no’; sometimes, often
version of the diabetes mismanagement scale [103]. The original version consists of 10 items:
and always into ‘yes’. Answers about omitting insulin were recoded: 1–3, 4–7, 8–11, 12–14d
3 used and 1 adjusted. The recall period was changed from 10 days to 7. Answers are given on
into ‘yes’, no days into ‘no’. Using Cohen’s kappa, adolescents’ answers on dieting frequency
a 5-point Likert scale. Cronbach’s α of this adapted version is 0.44. Higher scores indicate more
were compared to clinicians’ answers.
mismanagement (range 4-16). Height and weight were used to calculate body mass index
(BMI) scores, and Growth analyzer4.0 was used to calculate BMIz scores [154].
Results
To measure concordance, parents’ observations with regard to possible DEBs were
Participant characteristics. In total, 103 (87.3% of invited) adolescents (11–16 yr) completed
examined by two questions: (i) Does your child seem satisfied with his/her looks? (yes, no,
the survey, 77.7% online. Table 1 shows their characteristics and answers for step 1 in the
no and this concerns me, don’t know: recoded into ‘yes’ and ‘no’. ‘Don’t know’ was excluded
stepwise approach. Mean MY-Q body/weight score was 79.25 (±13.42), boys scored higher
from analyses); (ii) Do you think your child omitted insulin in order to influence his/her weight
(M-boys: 81.9, M-girls 76.7), however not significant (T = 1.96, p = 0.053). Regarding dieting
(yes, this happens a lot, yes, this happens sometimes, no, I don’t know: recoded into ‘yes’ and
frequency, no gender differences were found (χ² = 2.587, p = 0.108).
‘no’, ‘I don’t know’ was excluded from analyses). Clinicians were asked whether they thought
adolescents were concerned about their weight and food intake. Answers ‘yes’, ‘not sure’ and ‘no’.
Prevalence of DEBs. The answers of 47 adolescents (45.6%, 26 girls) raised psychological yellow
Not sure (N = 1) was excluded from analysis.
flags (step 1) and these adolescents were directed to the DEPS-R and AHEAD questions (step 2).
Mean DEPS-R score was 10.4 (±7.59). A total of 19.4% had DEPS-R scores between 0 and 8
Statistical analyses. In order to examine the differences between the adolescents whose
(N = 20), 18.5% had DEPS-R scores between 9 and 17 (N = 19), and 7.8% scored above cut-off
answers did not raise yellow flags (were not directed to step 2) and therefore did not complete
(DEPS-R score ≥18, N = 8).
50
51
4
Disturbed eating behavior
Disturbed eating behavior
Yellow flags. Table 2 shows the association between DEBs and gender and dieting activities
in the past year. The presence of dieting activities related to DEB (χ² = 13.78, p = 0.001).
Associations between DEBs and continuous variables are shown in Table 3. Older age
(p = 0.034), increased HbA1c (p = 0.004), reduced QoL (p = 0.007), less diabetes self-confidence
(p = 0.015), poorer diabetes self-management (p < 0.001), and concerns about body image
(p < 0.001) related to DEB. Gender and BMIz did not. Table 4 shows post hoc analyses. With
4
respect to weight loss techniques, 40 adolescents engaged in healthy activities, 3 in unhealthy,
and none in very unhealthy activities such as ‘Took less insulin than prescribed’. In contrast, 16
adolescents reported on the MY-Q that they had omitted insulin intentionally in the past 14d.
Twenty adolescents (11 rarely; 6 sometimes; 2 often; 1 almost always) scored positive on
DEPS-R item ‘When I eat too much, I administer less insulin (bolus less)’. Eight adolescents
[five rarely; two sometimes; one almost always) scored on DEPS-R ‘After I ate too much, I skip
my next insulin dose (bolus)’].
Concordance between adolescent –parents, and adolescent–clinicians. With regard to adolescents’
body satisfaction, adolescents and parents showed slight agreement (k = 0.126, N = 94):
in 12 cases, the parents thought adolescents were not happy with the way they looked, but
adolescents did not report body dissatisfaction. One adolescent reported low body satisfaction,
while the parent did not. With regard to omitting insulin: adolescents and parents showed slight
agreement (k = 0.141, N = 100): 13 adolescents omitted insulin while the parents thought they
did not, and in six cases it was the other way around. Clinicians and adolescents showed fair
agreement (k = 0.332, N = 101) in terms of weight loss activities: 10 adolescents tried to lose
weight, while the clinicians thought they did not and in eight cases it was the opposite.
52
Error! Style defined. Table
1:not Demographic
and diabetes-related characteristics of participating adolescents with T1D
N=103 Age in years Girls Family structure other than t raditional (= t wo b iological parent-­‐family) Education level Primary school Lower / vocational education (VMBO/MBO) Higher / p re-­‐university (HAVO/VWO) Attendance to a special school Ethnic identification other than Dutch* Born in the Netherlands Education level parents Low Average High Treatment pump HbA1c % (mmol/mol) [SD] BMIz Age of onset d iabetes Diabetes duration in y ears I am happy with the way I look All the time Often Sometimes Very seldom† Never† I have been t rying to control my weight or shape in different ways All the time† Often† Sometimes Very seldom Never In how many of the past 14 days have you had eating b inges No days 1-­‐3 days 4-­‐7 days† 8-­‐11 days† 12-­‐14 days† In how many of the past 14 days have you omitted insulin o n purpose No days 1-­‐3 days† 4-­‐7 days† 8-­‐11 days† 12-­‐14 days† * Parents were asked with which ethnicity they identified themselves. † Psychological yellow f lag: directed to step 2 52 53
% / Mean (SD) 13.5 (1.49) 51.5 17.5 11.7 27.2 57.3 3.9 5.8 100 9.7 25.2 63.1 80.4 8.0% (64.1) [3.5%(15.8)] 0.64 (1.0) 7.0 (3.9) 6.5 (3.8) 19.4 51.5 27.2 1.9 0 2.9 12.6 20.4 19.4 44.7 42.7 36.9 11.7 2.9 4.9 84.5 9.7 4.9 1.0 0 4
Error! Style not defined. Twenty adolescents (11 rarely; 6 sometimes; 2 often; 1 almost always) scored positive on 5.4 (1.4) 5.6 (1.6) 6.9 (1.4) 8.0 (3.1) <0.001* Body Image 88.1 (8.6) 71.6 (10.0) 69.4 (9.5) 60.2 (8.1) <0.001* DEPS-­‐R, Diabetes Eating Problem Scale – Revised; HbA1c, glycated h emoglobin A1c; QoL, quality of life DEPS-­‐R item ‘When I eat too much, I administer less insulin (bolus less)’. Eight adolescents Disturbed
eating
behavior
* < = 0.05 [five rarely; two sometimes; one almost always) scored on DEPS-­‐R ‘After I ate too much, I skip my next insulin dose (bolus)’]. Table
4: Post hoc tests between DEPS-R groups
Gender boys/girls (N) Dieted in last y ear: No Count Expected Std. Residual Dieted in last y ear: Yes Count Expected Std. Residual Disturbed eating behavior
Table 4 Post hoc tests b etween DEPS-­‐R groups Table
2: Differences between DEPS-R groups
Table 2 Differences b etween DEPS-­‐R groups 4
Diabetes management No DEPS-­‐R N=56 (%) Mean (SD) 29 (58.0%)/27 (50.9%) 54 (60.7%) 48.4 0.8 2 (14.3%) 7.6 -­‐2.0 * DEPS-­‐R 0-­‐8 N=20 (%) Mean (SD) 7 (14.0%)/13 (24.5%) 17 (19.1%) 17.3 -­‐0.1 3 (21.4%) 2.7 0.2 DEPS-­‐R 9-­‐17 and DEPS-­‐R ≥18 † N=27 (%) Mean (SD) 14 (51.9%)/ 13 (48.1%) 18 (66.7%) 23.3 -­‐1.1 9 (33.3%) 3.7 2.8 * Chi-­‐square χ² = 1.82, p = 0.402 χ² = 13.78, p = 0.001 No DEPS-­‐R No DEPS-­‐R No DEPS-­‐R DEPS-­‐R 0-­‐8 DEPS-­‐R 0-­‐8 DEPS-­‐R 9-­‐17 vs vs vs vs vs vs DEPS-­‐R 0-­‐8 DEPS-­‐R 9-­‐17 DEPS-­‐R ≥18 DEPS-­‐R 9-­‐17 DEPS-­‐R ≥18 DEPS-­‐R ≥18 0.995 0.004* 0.230 0.046 0.394 0.438 Boys 0.264 0.026 0.410 0.962 0.204 0.034 Girls 0.648 0.096 0.026 0.071 0.024 0.403 HbA1c (p-­‐value) 0.057 0.041 0.034 0.016 0.002* 0.811 QoL (p-­‐value) 0.620 0.027 0.004* 0.222 0.017 0.056 Diabetes confidence (p-­‐value) 0.523 0.127 0.007* 0.087 0.008* 0.083 Diabetes management (p-­‐value) 0.649 <0.001* 0.004* 0.003* 0.020 0.484 <0.001* <0.001* <0.001* 0.454 0.009 0.025 Age (p-­‐value) Body Image (p-­‐value) DEPS-­‐R, Diabetes Eating Problem Scale – Revised. * Post hoc significant = Std. Residual >1.96 o r <-­‐1.96 DEPS-­‐R, Diabetes Eating Problem Scale – Revised; HbA1c, glycated h emoglobin A1c; QoL, quality of life † DEPS-­‐R 9-­‐17 and DEPS-­‐R ≥ 18 formed one group a s ≥18 had less than 5 counts * Significant at level=α ≤ 0.008 Error! Style not defined. Discussion
Concordance between adolescent –parents, and adolescent–clinicians. With regard to Table ifferences b etween DEPS-­‐R DEPS-R
groups groups
Table
3:3 DDifferences
adolescents’ body sbetween
atisfaction, adolescents and parents showed slight agreement (k = 0.126, N = 94): in 12 cases, the parents thought adolescents were not happy with the way they looked, but adolescents did not report body dissatisfaction. One adolescent reported low No DEPS-­‐R DEPS-­‐R 0-­‐8 DEPS-­‐R 9-­‐17 DEPS-­‐R ≥18 Kruskal-­‐Wallis N=56 N=20 N=8 adolescents p-­‐value body satisfaction, while the parent did not. With regard tN=19 o omitting insulin: and Mean (SD) 13.2 (1.5) Mean (SD) 13.3 (1.7) Mean (SD) 14.3 (1.1) Mean (SD) 13.9 (1.5) parents showed slight agreement (k = 0.141, N = 100): 13 adolescents omitted insulin while Age 0.034* The primary aim of this study was to assess the prevalence of DEB and to identify yellow
54
flags in T1D adolescents. Answers of almost half of the adolescents raised psychological
yellow flags and therefore they seemed concerned about their body/weight, however, less
than 1 in 10 scored above cut-off, indicating the presence of DEBs. This is less than reported
the did (1.4) not, and in 12.8 six (c1.4) ases it was the (1.3) other way around. Girls parents thought they 13.0 14.0 14.6 (0.9) Clinicians 0.040* by Wisting [61], in which 18.3% scored above cut-off on the DEPS-R, and Saßmann [152]
Boys adolescents showed 13.5 (0.8) 12.7 1.5) loss and fair (1.5) agreement 14.1 (k =( 2.0) 0.332, N =14.6 101) in terms o
f w(eight (in which 15.4 scored above cut-off). However, the adolescents in those studies were older.
0.078 HbA1c 62.2 (11.2) (18.1) 72.5 (21.8) thought 71.8 (12.2) activities: 10 adolescents tried to lose w58.5 eight, while the clinicians they did not 0.004* and BMIz 0.5 (1.0) 0.6 (1.1) 1.1 (0.9) 0.9 (1.2) 0.092 QoL 73.0 (9.8) 71.3 (9.6) 68.7 (9.0) 59.8 (10.5) 0.007* Diabetes confidence 52.7 (5.0) 53.5 (4.6) 50.8 (4.6) 52.0 (5.0) 0.015* Diabetes management 5.4 (1.4) 5.6 (1.6) 53 6.9 (1.4) 8.0 (3.1) <0.001* Body Image 88.1 (8.6) 71.6 (10.0) 69.4 (9.5) 60.2 (8.1) <0.001* in eight cases it was the opposite. DEPS-­‐R, Diabetes Eating Problem Scale – Revised; HbA1c, glycated h emoglobin A1c; QoL, quality of life Since the prevalence of DEBs is highest in 17–19yr old [61], it could therefore be suggested
that the prevalence increases when our participants grow older. Future research is planned to
examine the course of DEBs in our population over time.
Another explanation for the differences in prevalence could be the methodology used.
Rather than screening all participants, we applied a stepwise approach, which may have ruled
* < = 0.05 out adolescents at risk, although initial assessment did not indicate this risk. Clinical yellow
flags for DEBs include elevated HbA1c; slightly older age; dieting activities; reduced QoL; low
confidence in diabetes self-care; worsened diabetes self-management; body dissatisfaction,
Table 4 Post hoc t ests b etween DESP-­‐R groups No DESP-­‐R No DESP-­‐R No DESP-­‐R DESP-­‐R 0-­‐8 DESP-­‐R 0-­‐8 DESP-­‐R 9-­‐17 vs DESP-­‐R 54 vs DESP-­‐R vs DEPS-­‐R vs DESP-­‐R vs DESP-­‐R vs DESP-­‐R 0-­‐8 9-­‐17 9-­‐17 ≥18 ≥18 ≥18 55
4
4
Disturbed eating behavior
Disturbed eating behavior
which is in coherence with previous research [145, 151, 155]. Yet, HbA1c on its own should
Its answers provide the opportunity to discuss worries about weight, looks and food intake.
be interpreted with caution as the deviation of the mean in our study is quite high. Moreover,
In this study, half of adolescents reported these kind of concerns, making it appropriate and
lower HbA1c values do not necessarily imply less DEBs as, for example, a shift may occur from
feasible to talk about this subject. The stepwise approach used in this study provides the
binging and omitting insulin toward restricted food (carbohydrates) intake, or intentional
opportunity to screen for yellow flags in clinics as well. Further research should confirm the
hypoglycemia to control bingeing [156]. Contrary to our expectations [61, 145, 151, 155],
validity of this stepwise approach that has practical and economic advantages over screening
increased BMIz scores and being female were no yellow flags for DEB. As we examined
the whole patient population. This could be a sensitive but not specific method. In our study
a relative young population, it is possible that these differences do exist when time goes by.
adolescents were asked in different ways if they omitted insulin. The way these questions are
Nonetheless, previous studies have shown that DEBs in T1D boys often stay unnoticed [71, 152].
formulated, appeared to make a difference: none of the adolescents skipped insulin in order
In young men, insulin restriction and age are not associated with DEB, [155], which implies that
to lose weight, but 20 reported insulin omission when asked on a Likert scale if they administer
DEBs follow a different course between men and women. Our findings confirm the importance
less insulin after eating too much. We do not know the exact intention for omitting insulin; low
of attention for DEB in boys, the DEPS-R is found to be a good measure for this purpose [152].
blood sugar levels or fear of hypoglycemia can be plausible reasons as well. In the current study
Although a trend toward lower body satisfaction in girls was found, they did not differ in dieting
this was not examined. When consulting, the possibility to differentiate an answer is important
frequency in the past year.
in order to establish whether reduction of boluses/injections is in fact dysfunctional or not.
The second aim of the study was to assess whether parents’ and clinicians’ observations
Confirmation of DEBs should be followed up by education and – if necessary – treatment.
with regard to eating behaviors was in concordance with the opinions of the adolescents.
Research showed that compared with the physical assessment, monitoring and discussing QoL
We found that clinicians’ and adolescents’ agreement was not more than ‘fair’. Concordance
is often less of a priority for pediatric diabetes teams [138, 159], which might challenge the use of
between parents and adolescents was even lower. This indicates that the opinion of parents
the suggested stepwise approach. Nonetheless, it offers an opportunity to provide (preventive)
or clinicians in most cases does not represent that of the adolescent when it comes to body/
psycho-education. Further research is needed to examine whether psycho-education is feasible
weight concerns. Despite the fact that literature lacks when it comes to on the accuracy of
and useful. Since this is a young population, prevention appears worthwhile. As far as we
parental observations with regard to potential DEBs or those of the clinicians, previous studies
know, very few DEB psycho-education or prevention programs are available for T1D patients,
have shown that indeed parents and pediatric diabetes teams are not in agreement with
of which none for boys. These programs were note effective when it comes to omitting insulin
adolescents in terms of adolescents’ psychosocial well-being [157] [158]. The assessment of
and HbA1c [160, 161], however, they were beneficial for the risk factors involved, such as body
DEBs should thus preferably involve the adolescent him of herself.
dissatisfaction and self-esteem [161, 162]. Also, multidisciplinary care appears beneficial for
Clinical pediatric diabetes guidelines state that QoL should be measured annually [153].
treating DEBs in T1D patients [163]. As DEBs and eating disorders show high comorbidity and
Also the assessment of DEBs is recommended. But since pediatric diabetes teams are hesitant to
mortality rates [147], attention for eating behaviors is warranted.
do so - afraid to give adolescents new ideas about skipping insulin [71] - the assessment of QoL
Although the current study contributes to enhancing our understanding of DEB, some
could serve as an opportunity to assess and discuss DEBs as well. The usage of a questionnaire
limitations are worth mentioning. First, cross-sectional data were used and we cannot
that captures body satisfaction as part of QoL, such as the MY-Q [101], is recommended.
infer causality. Future longitudinal research is planned to examine the course of DEBs and
56
57
4
Disturbed eating behavior
Disturbed eating behavior
the yellow flags in more detail. The DEPS-R was used to examine DEB, which is a relatively
new measure and its sensitivity and specificity are not yet sufficiently studied. Also, Dutch
version used in the exploratory study has not been validated yet. Further research is
needed to investigate whether it is suitable in clinical practice as well. Adolescents were
asked to postpone the completion of the survey if their blood glucose was 12 mmol/L
(216 mg/dL) or <4 mmol/L (75 mg/dL) [139], but the completion took place without supervision.
4
Even though paper-and-pencil and internet questionnaires are equally reliable [140, 141],
4
we should acknowledge that almost one fourth completed the survey on paper. The internal
consistency of the mismanagement scale was relatively low (α = 0.44).
To conclude, half of adolescents are concerned about their body weight, food intake
or physical appearance, but in most cases this is not problematic. Being female or being
over- or underweight was not associated with DEBs in this study. DEBs are often not recognized
as such by clinicians and parents. When monitoring QoL, a stepwise approach such as used in
this study can be applied in clinical practice to assess and discuss potential body and weight
concerns with T1D adolescents.
58
59
Chapter
5
Not parental well-being but parental diabetes behaviors and stress
are related to glycemic control in youth with type 1 diabetes
Longitudinal data from the DINO study
‘The way my son copes with his illness relies for the most part on how we as parents cope
In general, to what degree do parents accept their child’s disease
and, as far as we can tell now, will have for the rest of his life’
Minke Eilander
Maartje de Wit
Joost Rotteveel
Henk-Jan Aanstoot
Willie Bakker-van Waarde
Mieke Houdijk
Roos Nuboer
Per Winterdijk
Frank Snoek
Submitted
61
Parental emotional well-being
Parental emotional well-being
Abstract
Introduction
Objective. To evaluate 1) the longitudinal relation between parental well-being and glycemic
In the treatment of youth with type 1 diabetes, the primary caregivers (in most cases parents)
control in a cohort sample of parents of youth with type 1 diabetes, and 2) if youth’s problem
play an important role as they perform or monitor the diabetes management on a daily basis
behavior, diabetes parenting behavior, and parental diabetes stress impact this relation.
[19]. Compared to parents of healthy youth, these parents have more concerns about their
child’s health: they often worry about severe hypoglycemia that might result in coma, and
5
Research Design and Methods. Parents of youth 8-15 years old (at baseline) (N = 174)
the risk of complications later in life [114, 164]. Consequently, their emotional well-being is
participating in the DINO study, completed questionnaires at three time waves (each with
challenged and depression seems more prevalent in parents of youth with type 1 diabetes
a 1 year interval). Using Generalized Estimating Equations, the relation between parental
[92, 165]. Previous cross-sectional studies showed a direct association between parental
well-being (WHO-5) and youth’s Hemoglobin A1c (HbA1c) was examined. Second, relations
depression and youth’s glycemic control (HbA1c) [92, 165]. The underlying mechanisms might
between scores on WHO-5, Strength and Difficulties Questionnaire (SDQ), Diabetes Family
be that parental depression increases diabetes related parenting stress and unsupportive
Behavior Checklist (DFBC) supportive and unsupportive scales, Problem Areas In Diabetes-
parenting behavior (e.g. criticism and nagging) [166], which associate with youth’s problem
Parent revised (PAID-Pr) and HbA1c were analyzed.
behavior (such as reduced emotional well-being or depression) [167-169] and glycemic control
[94, 123, 170]. Although depression is a frequently used outcome in research on diabetes,
Results. Low well-being was reported by 32% of parents. No relationship was found between
a recent study pointed out that questionnaires used in diabetes research to screen for depression
parents’ WHO-5 scores and youth’s HbA1c (β = -0.052, p = 0.650). Over time WHO-5 related
are not suitable for diagnostic purposes: emotional distress is often qualified as depression
to SDQ (β = -0.219, p < 0.01); DFBC unsupportive scale (β = -0.174, p < 0.01); PAID-Pr
[171]. It seems that well-being is more validated outcome, which provides information
(β = -0.666, p < 0.01). Both DFBC scales (supportive β = -0.259, p = 0.01; unsupportive
about generic subjective emotional well-being [172]. It is likely to assume that good parental
β = 0.383, p = 0.017), PAID-Pr (β = 0.276, p < 0.01) and SDQ (β = 0.424, p < 0.01) related to HbA1c.
well-being associates with better HbA1c levels in youth. Supportive parenting (e.g. positive
reinforcement) is found to have a positive influence on youth’s glycemic control [13, 173]:
Conclusions. Reduced parental well-being relates to increased problem behavior in youth,
positive communication and parenting support empower youth to manage their diabetes,
unsupportive parenting and parental stress, which in turn have a negative effect on HbA1c. Youth’s
in addition to feeling appreciated by their parents. Whether good parental well-being relates
problem behavior was also impacted by unsupportive diabetes parenting and stress. The external
to supportive diabetes parenting behavior is, however, unknown.
generalizability is questioned by the low response rate; 25% of invited families participated. Our
The above mentioned studies all separately contribute to the understanding of the impact of
findings highlight the importance of interventions that also target parents’ psychosocial functioning.
parental well-being, support and stress on youth’s diabetes and behavior outcomes. However,
longitudinal studies incorporating all these factors at once are scarce. Little is known about
the mechanisms behind psychosocial aspects of both youth and parents and their relation
with diabetes outcomes over time. Identification of these mechanisms could help to refine
and further develop family interventions to improve outcomes in youth with type 1 diabetes.
62
63
5
Parental emotional well-being
With this study we therefore aim to gain insight into the complex interactions between these
Parental emotional well-being
Figure 1: Flowchart participants
factors. First, we hypothesize that there is a longitudinal relation between better parental
T0: 2013 N=151
2014 N=14
2015 N=9
Total N=174
emotional well-being and more optimal glycemic control in youth. Secondly, we expect youth’s
problem behavior, diabetes parenting behavior and parental diabetes stress to have an impact
on this relation.
Research Design and Methods
(DINO) study [129]: a longitudinal study on the psychosocial and biological development of
T1: 2014 N=151
2015 N= 14
Total N=165 youth with type 1 diabetes. Exclusion criteria were: mental retardation; diabetes other than
type 1; diagnosis <6 months; unfamiliarity with the Dutch language. Data from three time
waves (T0 in 2013, T1 in 2014 and T2 in 2015, each with a 1 year interval) were included in the
T2: 2015: 151
study reported on in this paper. At T0, T1 and T2 parents completed an online or (in case of
Procedure and participants. Data were collected as part of the ongoing Diabetes IN development
5
Questionnaire completed: 100% N=174
Questionnaire completed: 82.78% N=125
non-response) a paper survey in which they reported on their own psychosocial functioning
and that of their children. Medical data were obtained from hospital charts. Written informed
Questionnaire completed: 92.12%
N=152
Measures
5
Participants. In total 670 youths with type 1 diabetes aged 8-15 treated in 5 Dutch diabetes
Parental well-being. WHO-Five Well-being Index (WHO-5) was used to capture the level of
Measures emotional well-being. The scale has five positively worded items, scored on a 6 point Likert scale
Parental well-­‐being. WHO-­‐Five Well-­‐being Index (WHO-­‐5) was used to capture the leve
(e.g., “I have felt cheerful and in good spirits”). Scores are summated with a range from 0-25:
emotional well-­‐being. The scale has five positively worded items, scored on a 6 point Li
≤13 indicates low well-being,
a score
of ≤7
likely
depression
[117,
118,
174]. The
WHO-5
a well
scale (e.g., “I have felt cheerful and in good spirits”). Scores are issummated with a ra
centers and their parents were asked to participate in the DINO study (at the start of the study
validated questionnaire
Internal
the current
sample
was 0.92.
from [117,
0-­‐25: 172,
≤13 175].
indicates low consistency
well-­‐being, afor
score of ≤7 likely depression [117, 118, 174]. 598 youths were invited, in the two years following another 72). In total 25.0% of the 670
invited youths and their parents participated: at the start of the study 151 were included, one
Parental diabetes stress. The Problem Areas In Diabetes-Parents Revised version
63 (PAID-Pr) was used to assess diabetes related distress in parents of youth with diabetes.
year later another 14 and in the third year 9, as shown in Figure 1. At T0 174 parents (100% of
It contains 18 items, scored on a 5 point Likert scale (e.g., “I feel upset when my child’s blood
participating parents) completed the questionnaire, at T1 152 (92.1% of 165 participating) and
sugars are out of range”). Scores are summated and range from 0-72: higher scores indicate
at T2 125 (82.8% of 151 participating). Main reasons to reject participation were inherent to the
more diabetes stress [114]. Internal consistency for the current sample was 0.90.
puberty of youngsters: they declared to be too busy with school and friends, while also managing
Diabetes parenting behavior. The frequency of supportive and unsupportive diabetes
their diabetes. In two-parent families, parents decided who completed the questionnaire.
specific parenting behaviors was assessed using the Diabetes Family Behavior Checklist (DFBC).
consent was obtained from all parents and from youth ≥12 years. The DINO study was approved
by the medical ethical committee of VU University Medical Center.
The DFBC contains 16 items on a 5 point Likert scale, of which nine tap into supportive parenting
64
65
Parental emotional well-being
Parental emotional well-being
behaviors and seven into unsupportive behaviors (e.g., “How often do you argue with your child
analyses using GEE where conducted in which the WHO-5, PAID-Pr and DFBC scores from these
about his/her diabetes self-care activities”). Scores on the supportive scale range from 9-45;
exceptional parent’s (e.g. T1 father) were excluded from analyses.
a higher total score indicates more parental support. Scores on the unsupportive scale range
5
from 7-35; a higher score indicates more criticism and less support [119]. Internal consistency
Results
the supportive subscale for the current sample was 0.64 and for the unsupportive subscale 0.66.
Participant characteristics. Baseline characteristics of the 174 parents are presented in
Youth’s problem behavior. The Strengths and Difficulties Questionnaire (SDQ) was used
Table 1. Mothers were more likely to complete the questionnaires (87.3% at T0). Average
to measure emotional and behavioral functioning. The SDQ comprises five scales: Emotional
parental education level was respectively high: 64.7%; moderate: 24.1%; low 10.4%. Mean
symptoms, Conduct problems, Hyperactivity/inattention, Peer relationship problems and
age of youth was 11.64 (±2.18) years and gender was equally distributed. Mean HbA1c was
Pro-social behavior. The parental version of the SDQ, in which parents report on their child’s
62.15 mmol/mol (±11.35) or 7.84% (±3.19). Mean diabetes duration was 5.10 years (±3.61),
behavior, was used in this study. It contains 25 items, scored on a 3 point Likert scale (e.g., “My
94.2% of youngsters were born in the Netherlands and 81.2% lived in a two-parent family.
child often lies or cheats”). Scores are summated and range from 0-50: higher scores indicate
The percentage of parents with WHO-5 scores ≤13 indicating low emotional well-being on T0
more problem behavior [130]. Only the total score was used in this study. Internal consistency
was 32.0%, on T1 25.5% and on T2 26.7%. Low emotional well-being on 2 or 3 points in time
for the current sample was 0.77.
was reported by 22.5% of parents. The percentage of parents with WHO-5 scores ≤7 indicating
likely depression on T0 was 8.1%, on T1 10.7% and on T2 10.8%. Depressive symptoms on 2 or
Analyses. Generalized estimating equations (GEE) with exchangeable correlation structure was
3 points in time were reported by 8.1% of parents.
used to examine the longitudinal relations between variables. GEE adjust for the correlation
Parental well-being, diabetes parenting behavior, parental diabetes stress and youth’s
between repeated measurements within a person and uses all available longitudinal information
glycemic control at baseline of dropouts on T1 and T2 did not differ from participating parents
in analyses. The following longitudinal relations were examined: parental well-being on youth’s
(data not shown).
HbA1c; parental well-being on youth’s problem behavior; parental well-being on diabetes
By using GEE the longitudinal relations between the variables were analyzed separately,
parenting behavior; parental well-being on parental diabetes stress; diabetes parenting behavior on
adjusted for parents’ education level and gender, youth’s age, gender, diabetes duration,
youth’s problem behavior; parental diabetes stress on youth’s problem behavior; youth’s problem
to examine: 1. Is there a longitudinal relation between parental well-being and youth’s glycemic
behavior on HbA1c; diabetes parenting behavior on HbA1c; parental diabetes stress on HbA1c.
control? 2. And do youth’s problem behavior, diabetes parenting behavior and parental
We adjusted for youth’s age, gender and diabetes duration, and parents’ gender and
diabetes stress impact this relation? Results are presented in Table 2.
education level as reported at baseline by including these variables as covariates. T-tests were
used to examine if parental well-being, diabetes parenting behavior and parental diabetes
Parental well-being. There was no longitudinal relation between parental well-being and HbA1c
stress, and youth’s glycemic control of dropouts on T1 and T2 differed from parents who did
(β = -0.052, p = 0.650), as shown in Table 2. There was no longitudinal relation between
participate on T1 and T2. Data were analyzed using SPSS 22 [176]. As parents in a two-parent
parental well-being and supportive diabetes parenting behavior (β = -0.045, p = 0.33) either.
family decided one of them completed the questionnaire on each time wave which, sensitivity
However, a longitudinal relation was found between lower parental well-being on one side and
66
67
5
Parental emotional well-being
Parental emotional well-being
more unsupportive diabetes parenting behavior (β=-0.174, p < 0.001) and higher parental
Diabetes parenting behavior and parental diabetes stress. Supportive and unsupportive
diabetes stress (β = -0.666, p < 0.001) on the otherLower parental well-being was related to
diabetes parenting behavior, as well as parental diabetes stress were related to worse glycemic
increased problem behavior in youth (β = -0.219, p < 0.001).
control (β = -0.259, p = 0.01; β = 0.383, p = 0.02; β = 0.276, p = 0.001).
Table 2: Relations b etween parental well-­‐being, HbA1c, youth’s problem b ehavior, d iabetes parenting b ehavior Table 1: Baseline characteristics
Table
2: Relations
and parental diabetes between
stress parental well-being, HbA1c, youth’s problem behavior, diabetes Error! Style not defined. Boys N (%) 95% LCI 95% UCI p-­‐value WHO-­‐5 – HbA1c -­‐0.052 -­‐0.275 0.172 0.650 WHO-­‐5 – SDQ -­‐0.219 -­‐0.312 -­‐0.125 0.000** WHO-­‐5 – DFBC supportive -­‐0.045 -­‐0.134 0.044 0.325 WHO-­‐5 – DFBC unsupportive -­‐0.174 -­‐0.268 -­‐0.081 0.000** WHO-­‐5 – PAID-­‐Pr -­‐0.666 -­‐0.858 -­‐0.474 0.000** DFBC supportive – SDQ -­‐0.013 -­‐0.123 0.097 0.821 DFBC unsupportive – SDQ 0.298 0.165 0.431 0.000** PAID-­‐Pr – SDQ 0.173 0.104 0.242 0.000** SDQ – HbA1c 0.424 0.191 0.658 0.000** DFBC supportive – HbA1c -­‐0.259 -­‐0.455 -­‐0.063 0.010* 1 (0.6) DFBC unsupportive – HbA1c 0.383 0.069 0.697 0.017* 10 (5.8) PAID-­‐Pr – HbA1c 0.276 0.117 0.436 0.001** 11.64 (2.18) HbA1c % (mmol/mol) 5
β 87 (50.0) Youth’s a ge (yrs) 7.84 (±3.19) (62.15 (±11.35)) Age diabetes onset 6.56 (3.72) Diabetes duration (yrs) 5.10 (3.61) Pump / injections N (%) 134/39 (77.5/22.4) Completion by mothers N (%) 151 (87.3) Completion online / paper N (%) 157/16 (90.8/9.2) Parental education N (%) -­‐
Low 18 (10.4) -­‐
Moderate 42 (24.1) -­‐
High 112 (64.7) -­‐
NA Ethnic identification other than Dutch* N (%) parenting behavior and parental diabetes stress
Adolescents born in the Netherlands N (%) 169 (94.2) HbA1c, H emoglobin A1c. WHO-­‐5, WHO-­‐Five Well-­‐being Index. DFBC, Diabetes Family Behavior Checklist. Traditional family composition N (%) 138 (81.2) PAID-­‐Pr, Problem A reas In Diabetes – Parents Revised version. SDQ, Strengths and Difficulties Questionnaire. WHO-­‐5 (0-­‐25) 15.49 (5.14) * Statistically significant at the p < .05 level DFBC supportive (9-­‐45) 30.15 (5.37) ** Statistically significant at the p ≤ .001 level DFBC unsupportive (7-­‐35) 14.45 (4.44) PAID-­‐Pr (0-­‐72) 13.54 (9.08) SDQ (0-­‐40) 8.20 (6.03) Data are m eans ± SD, unless otherwise indicated. HbA1c, H emoglobin A1c. WHO-­‐5, WHO-­‐Five Well-­‐being Index. DFBC, Diabetes Family Behavior Checklist. PAID-­‐
Youth’s problem behavior. Supportive diabetes parenting behavior did not relate to youth’s
problem behavior (β = -0.013, p = 0.82), however, unsupportive diabetes parenting behavior
Pr, Problem Areas In Diabetes – Parents Revised v ersion. SDQ, Strengths and Difficulties Questionnaire. and parental diabetes stress did (β = 0.298, p < 0.001; β = 0.173, p < 0.001). Youth’s problem
* Parents were asked with which ethnicity they identified themselves. behavior was related to poorer glycemic control (β = 0.424, p < 0.001).
Diabetes parenting behavior and parental diabetes stress. Supportive and unsupportive diabetes parenting behavior, as well as parental diabetes stress were related to worse 68
glycemic control (β = -­‐0.259, p = 0.01; β = 0.383, p = 0.02; β = 0.276, p = 0.001). 69
5
Parental emotional well-being
Parental emotional well-being
In 13 cases either T0, T1 or T2 was completed by the other parent (e.g. T0 mother, T1 father
psychosocial functioning. As supportive parenting was related to better glycemic control,
and T2 mother). Sensitivity analyses did not show differences in regression coefficients and
it seems a factor to intervene on in itself.
significance compared to the primary results based on all data, as described above.
Currently, several parental interventions aiming to improve youth’s HbA1c are available, including
5
Conclusions
multi system therapy, (tele-health) behavior therapy and parental stress-reduction techniques
The aim of this study was to examine the longitudinal relation between parental well-being
[177, 178]. With the current study, we identified effective aspects of these interventions, namely
and youth’s glycemic control, and the impact of youth’s problem behavior, diabetes parenting
the reduction of unsupportive parenting behavior and parental diabetes stress and improving
behavior and parental diabetes stress on this relation. In our study, about one third of parents
supportive parenting, which is in coherence with previous research [179-183]. This is especially
reported low emotional well-being. Reduced parental well-being related to increased problem
true during adolescence when youth become more responsible for their own diabetes tasks
behavior in youth, more unsupportive diabetes parenting behavior and parental diabetes
[94]; a phase in which the communication between parents and youth is challenged. Based
stress, which in turn showed a negative effect on glycemic control. Youth’s problem behavior
on the impact of parent’s psychosocial aspects on youth’s problem behavior as shown in our
was also impacted by unsupportive diabetes parenting and parental diabetes stress. These
study, it is likely to assume that the reduction of unsupportive parenting and parental diabetes
effects were expected as previous research showed that parenting stress, unsupportive
stress, youth’s problem behavior improves as well. Our results indicate that reducing youth’s
parenting and youth’s well-being impact diabetes outcomes [94, 123, 167]. However,
problem behavior is also beneficial for glycemic outcomes. Screening for problem behaviors
our results did not support the hypothesized direct relation between reduced parental
in pediatric diabetes care is recommended by international guidelines [18], although
well-being and worsened glycemic control in youth. A relation that was found in some
this seems difficult in clinical practice [138]. Assessment of parental well-being, diabetes
[92, 165], but not all cross-sectional studies [166]. Our longitudinal design allowed us to examine
parenting behaviors and stress, and youth’s problem behavior seems appropriate within
the relation over time. This provided more robust results. Also, we assessed well-being rather
clinical practice, not only to optimize diabetes outcomes but to improve the quality of life
than depression. It seems that low parental well-being in general does not impact HbA1c,
of youth and their parents as well.
but it can increase non-supportive diabetes parenting behaviors, parental diabetes stress and
Previous literature formed the foundation of the hypothesis that psychological variables
youth problem behaviors, which impact the quality and the amount of parental involvement
influence HbA1c. WHO-5 scores are a reflection of generic and subjective emotional well-being,
in diabetes care, as described by Young et al. [173], causing HbA1c levels to rise. This implies
unrelated to disease [172] and therefore are more likely to be the independent variable.
that family interventions aimed to optimize glucose levels are more likely to succeed when
However, it could be argued that there is also a reversed pathway. Law (2013) for example
the main focus is the reduction of negative diabetes parenting behavior and parental diabetes
concluded that HbA1c predicts parental distress, nevertheless, cross-sectional data were used
stress. This in addition to screening for and treating parental low mood and depression, as the
in that study [168]. The mechanisms presented in our study can be explained by the cycle
effectiveness of such interventions is likely hindered by low parental well-being.
of miscarried helping [184]: youth’s high or low glucose levels increase parental worries and
Somewhat surprising, we found no relationship between parental well-being and supportive
fear, contributing to negative parent-child communication, characterized by criticism and
parenting behavior. Parental well-being seems to impact more on negative than positive
blame. Consequently, youth feel discouraged and avoid measurement or disclosure of glucose
70
71
5
Parental emotional well-being
Parental emotional well-being
levels, resulting in poor glucose levels. The complexity of the relation between psychosocial
and diabetes outcomes is also shown in a longitudinal study revealing that HbA1c was only
influenced by parental involvement when youth did not show depressive symptoms [185].
Clearly, there are other factors that appear to predict and associate with parental functioning,
youth’s functioning and diabetes outcomes which were not evaluated in the current study.
Diabetes adherence, for example, is found to mediate the relation between critical parenting
and HbA1c [123], and less confidence in diabetes self-care and diabetes management mediate
the relation between youth’s problem behavior and HbA1c [186]. Also, we did not assess the
influences of e.g. peers [187]. A challenging task for future research is to gain more inside in
5
the complex interactions between these variables. It could be argued that Structural Equation
5
Modeling or Mixed Models would be appropriate to analyze the longitudinal data presented
in this study. However, given the sample size and the relatively low frequency of missing data,
GEE were the considered most robust to examine the data.
Limitations. Although our study contributes to the understanding of the mechanisms between
psychosocial factors and diabetes outcomes, there are limitations that we should acknowledge.
The main limitation is the low response rate: a quarter of invited families participated. Difficulties in
terms of recruiting adolescents with type 1 diabetes are described in other studies as well [79, 170].
The internal consistency (Cronbach’s α) of the DFBC proved to be questionable in our study
(supportive scale: α = 0.64; unsupportive scale: 0.66), which is, however, in coherence with other
research (0.73 and 0.43 respectively) [119]. In addition, the fact that mainly mothers completed
the questionnaire, the respectively high education level and the limited ethnic diversity of
participating parents, might question the external validity of the results and stress the relevance
of future studies in more diverse populations.
In summary, this longitudinal study showed that reduced parental well-being is associated with
unsupportive diabetes parenting behavior, parental diabetes stress and youth’s problem behavior,
which in turn relate to less optimal glycemic control. These findings highlight the importance of
interventions that not only target youth’s glycemic control but also parents’ psychosocial functioning.
72
73
Chapter
6
Implementation of quality of life monitoring
in Dutch routine care of adolescents with type 1 diabetes:
Appreciated but difficult
‘Our team struggles with the logistics. For example, where do youth
complete the questionnaire and when do we discuss the results? This in addition to
the routine medical consultations. Furthermore, who owns the implementation?’
Minke Eilander
Maartje de Wit
Joost Rotteveel
Nienke Maas-van Schaaijk
Angelique Roeleveld-Versteegh
Frank Snoek
Pediatric Diabetes, 2016. 17(2): p. 112-119
75
Implementation of QoL
Implementation of QoL
Abstract
Introduction
Objective. Monitoring quality of life (QoL) improves well-being and care satisfaction of
International Society for Pediatric and Adolescent Diabetes (ISPAD) guidelines state that
adolescents with type 1 diabetes. We set out to evaluate the implementation of the program
pediatric diabetes teams should routinely assess the quality of life (QoL) of their patients [153, 188].
DAWN (Diabetes Attitudes Wishes and Needs) MIND-Youth (Monitoring Individual Needs in
How this should be done, is not stated. Consequently, these guidelines are practiced in various
Young People With Diabetes) (DM-Y), in which Dutch adolescents’ QoL is assessed with the
of ways or not at all [159]. Research has shown that well-being of adolescents with type 1
MIND Youth Questionnaire (MY-Q) and its outcomes are discussed. Successful implementation
diabetes improves when teams periodically monitor and discuss QoL of these adolescents
of DM-Y warrants close study of experienced barriers and facilitators as experienced by diabetes
[79]. Furthermore, adolescents are more satisfied with the care they receive from teams who
care teams as well as adolescents and parents.
pay attention to their QoL [79]. In order to enhance psychosocial and treatment outcomes
in adolescents with type 1 diabetes by monitoring and discussing QoL as part of the routine
Methods. The study was conducted in 11 self-selected Dutch pediatric diabetes clinics. A mixed
pediatric diabetes care, the intervention program DAWN MIND-Youth was developed (DM-Y:
methods approach was used. Ten diabetes teams (26 members) were interviewed; 36 team
Diabetes Attitudes Wishes and Needs and Monitoring Individual Needs in Young People With
members, 29 adolescents, and 66 parents completed an online survey.
Diabetes) [189]. Annually, adolescents complete the MIND Youth-Questionnaire (MY-Q) [101]:
a (web-based) measure of diabetes related QoL prior to their periodic consultation with
6
Results. Two of 10 teams successfully implemented DM-Y. Whereas 92% of teams valued DM-Y
the diabetes team, and outcomes are discussed with the patient. Assessment of parental
as a useful addition to routine care, most clinics were not able to continue because of logistical
well-being is optional and discussed in a separate consultation. Teams are offered one day
problems (lack of time and manpower). Still, all teams had the ambition to make DM-Y integral
training on how to use DM-Y and how to discuss MY-Q outcomes with adolescents, focusing
part of routine care in the nearby future. Seventy-nine percentage of the parents and 41%
on patient-centered communication.
of the adolescents appreciated the usage of MY-Q, same percentage of adolescents neutral.
The effectiveness of an intervention depends on the level of integration in the routine
health care [190]. Previous studies reported that the implementation of a (new) intervention
Conclusions. DM-Y is highly appreciated by teams, as well as adolescents and parents, but for
requiring practice change is challenging [190]. Certain barriers appear difficult to overcome,
most clinics it is difficult to implement. More effort should be paid to resolve logistic problems
such as the practical organization and communication processes within care teams [191].
in order to facilitate dissemination of DM-Y in care nationwide.
These barriers can result in negative attitudes and demotivation to change, which, in turn, can
interfere with the implementation. So attitudes toward the routine care need to change in order
to implement an intervention [190-192]. Before the implementation is considered successful,
five stages need to be followed [193]: Orientation, Insight, Acceptance, Change, Maintenance
(Box 1). With respect to DM-Y, successful implementation implies that the assessment of the
MY-Q and the discussion of its outcome with the adolescent are anchored in the routine care.
76
77
6
Implementation of QoL
Box
1:Style Stages
and steps in the implementation process [193]
Error! not defined. Research design and methods
Procedure. The study was approved by the medical ethical committee of VU University Medical
Center. Experiences of individual team members were assessed by (i) an online survey and (ii)
Stages Steps Orientation 1 Promote awareness of innovation 2 Stimulate interest and involvement Insight 3 Create understanding 4 Develop insight into own routines Participants and measures
Acceptance 5 Develop positive attitude to change Diabetes care teams. To gain insight into their levels of Acceptance, Change, and Maintenance,
6 Create positive intentions/decision to change we used a mix-method strategy. First, of the 71 team members of 11 clinics, 65 were
7 Try out change in practice invited to complete an online survey (e-mail addresses of the remaining six members were
8 Confirm value of change unknown). The survey consisted of 65 items (Appendix 2A, supporting Information). Second,
Maintenance 9 Integrate new practice into routines 10 Embed new practice in organization Change 6
Implementation of QoL
interviews were held with each care team separately. Experiences of adolescents who
completed the MY-Q and their parents were assessed by means of an online survey.
group interviews with the teams were conducted and team members decided to participate.
In the interviews, three main topics were addressed: (i) Acceptance: the motivation to start
with DM-Y; (ii) Change: the current situation regarding the usage of DM-Y; (iii) Maintenance:
the optimization of implementation.
In 2010 13 self selected Dutch pediatric diabetes care teams signed up for the DM-Y pilot
study. These teams all offer specialized diabetes care [on average 108 (±63) children] that
isIn 2010 13 self selected Dutch pediatric diabetes care teams signed up completely reimbursed with no need for co-payment. Of these 13 teams,
12 participated
for the DM-­‐
Y pilot examine their experiences and explore whether they might function as barriers or facilitators
instudy. the 1 Tday
training,
one postponed
thecare usage
DM-Y. In
11(±63) clinicschildren] at least tsome
hese teams aof
ll owhich
ffer specialized diabetes [on ofaverage 108 hat is 11 clinics (Table 2) were asked to participate in the evaluation research, of which 150 did (41.7%).
completely reimbursed with no need for co-­‐payment. Of these 13 teams, 12 participated in adolescents
completed the MY-Q in the period since training (2010) and this study (2012). With
Written informed consent was obtained from adolescents and parents. Demographic
the 1 day training, of which usage of DM-­‐
Y. In 11 clinics at least some respect
to the five stages [193]one postponed the these 11 clinics completed
the stages
Orientation and Insight.
and medical information of these 150 participants was collected from hospital charts.
adolescents completed the MY-­‐
in the period since training (2010) and this study (2012). However,
to examine whether allQteam
members Accepted DM-Y and whether teams were
With respect to the five stages [193] these 11 clinics completed the stages Orientation and able to follow the remaining steps Change and Maintenance, and thus successfully implement
Insight. However, to examine whether all team members Accepted DM-­‐Y and whether DM-Y, we evaluated the implementation process and the usage of DM-Y. For this we assessed
teams were able to follow the remaining steps Change and Maintenance, and thus the experiences of the 11 pediatric diabetes teams and their vision on possible barriers and
successfully implement DM-­‐Y, we evaluated the implementation process and the usage of facilitators as well as the experiences of adolescents and their parents as these might serve as
DM-­‐Y. For this we assessed the experiences of the 11 pediatric diabetes teams and their barriers or facilitators in the implementation process [194].
vision on possible barriers and facilitators as well as the experiences of adolescents and Sixty-nine adolescents and 89 parents received an invitation to an online survey, the e-mail
their parents as these might serve as barriers or facilitators in the implementation process HbA1c between adolescents. Chi-squared tests were used to examine differences between MY-Q
Adolescents and parents. The aim of the surveys for adolescents and parents was to
in the implementation process. All children (N = 316) who completed the MY-Q in the
addresses from the remaining participants were unknown. The survey for adolescents consisted of
17 items (Appendix 2B), the survey for parents of 22 items (Appendix 2C).
Analyses. With regard to the surveys, statistical analyses were performed using SPSS.20. t-Test
and Mann–Whitney test were applied in order to examine differences in MY-Q outcomes and
[194]. 78
79
6
Implementation of QoL
Implementation of QoL
outcomes and gender, ethnicity, and clinic. To examine whether answers related to occupation
Overall successfulness-maintenance. We ranked the 11 participating clinics according to
(pediatrician, diabetes nurse, etc.) or clinic, Kruskall–Wallis tests were used. Template analysis
the degree in which they were successful in implementing DM-Y; in clinic 1 it had become a
was used to analyze the interviews with the teams. Interviews were transcribed verbatim
structural part of the routine care; in contrast, clinic 10 rarely used DM-Y (Table 2). Two teams
and key words and codes to extract content from the text were assigned by two researchers
implemented DM-Y in routine care as planned: adolescents in clinics 1 and 2 completed the
(M. E.
and
research
Error! Style not d
efined. MY-Q annually and outcomes were discussed; clinics 3–6 incidentally used DM-Y and it was
assistant) [195].
no structural part of the routine care; clinics 7 and 8 started with DM-Y but did not continue
Results
and clinics 9 and 10 had not actually started the usage; they based their experiences on eight
Results Diabetes care teams. Thirty-six (55.4%) team members completed the online survey (Table 1).
Diabetes care teams. Thirty-­‐six (55.4%) team members completed the online survey (Table Appendix 2A shows the frequency of answers. In general, teams consisted of 8 (±2) members
1). Appendix 2A shows the frequency of answers. In general, teams consisted of 8 (±2) (every team
contained
diabetes anurse,
pediatrician,
psychologist, and dietician).
total 10
members (every team caontained diabetes nurse, pediatrician, psychologist, and Indietician). Table 2. Maintenance: integration DM-­‐Y in routine care 2010-­‐2012 teams
(26
members)
participated
the interviews
thatinterviews lasted 30 that min (range
In total 10 team
teams (26 team members) inparticipated in the lasted 13–50)
30 min Table 2: Maintenance: integration DM-Y in routine care 2010-2012
(Table 1).13–50) One clinic
declined
the interview
becausethe of lack
of time (reorganization).
(range (Table 1). One clinic declined interview because of lack of time and four adolescents, respectively. In order to understand why the implementation was not
successful in all clinics, the stages Acceptance, Change, and Maintenance are addressed below.
(reorganization). 6
Hospital Table 1: Demographic characteristics diabetes teams
Overall* Adolescents in care Success-­‐ who completed fulness MY-­‐Q (%) Parental part Consult parent DM-­‐Y † outcome ‡ 1 A 107 (71.3) A A N % 2 A 33 (57.9) B C Team m embers survey 36 3 B 18 (64.3) C Occupation Pediatrician 16 44.4 4 B 23 (53.5) C Diabetes nurse 13 36.1 5 B 24 (36.4) B B Psychologist 6 16.7 6 B 18 (17.4) A C Dietician 1 2.8 7 C 38 (47.5) C Male/female 7/29 19.4/80.6 8 C 21 (39.6) C Interview 26 9 D 8 (4.1) B N/A Team m embers present Pediatrician 12 46.2 10 D 4 (6.1) C N/A Diabetes nurse 10 38.5 11 N/A 19 (40.4) N/A Psychologist 3 11.5 DM-­‐Y, DAWN MIND-­‐Youth; MY-­‐Q, MIND Youth Questionnaire; N/A, information not available. Dietician 1 3.8 * A, structural part of routine care. B, used but not part of routine care. C, started usage but did not continue. Male / female 4/22 15.4/84.6 D, d id not start, seldom used in the past. † A , parental part is structural part of care. B, parental part depended on parents’ motivation and whether their Dutch was sufficient. C, did not use parental part. Overall successfulness-­‐maintenance. We ranked the 11 participating clinics according to the ‡ A , outcomes d iscussed with parents. B, only discussed when outcome was a larming o r on parents’ initiative. degree in which they were successful in implementing DM-­‐Y; in clinic 1 it had become a structural part of the routine care; in contrast, clinic 10 rarely used DM-­‐Y (Table 2). Two 80
teams implemented DM-­‐Y in routine care as planned: adolescents in clinics 1 and 2 C, not d iscussed. 81
6
6
Implementation of QoL
Implementation of QoL
Acceptance: motivations to change. According to 88.9%, colleagues within the teams were
Change: organization. In eight teams the conversation with the adolescent was led by one
motivated to start with DM-Y. Stated motives are grouped into four categories (Table 3). First,
professional; in clinics 3 and 4 two professionals consulted with the child simultaneously.
psychosocial aspects: nine teams mentioned their focus on the broader context of the child,
Who led the conversation (Table 4) depended for six teams (1, 2, 4, 6, 9, 10) on practical and
in addition to physical aspects. Moreover, clinics 1 and 8 referred to interactions between
organizational considerations and for four teams (2, 5, 7, 9) on the expected MY-Q outcome.
physical and psychosocial factors (e.g. the influence of stress on glycemic control). Second,
In addition, three teams (1, 7, 8) mentioned the level of comfort adolescents feel in presence
external motivation: ISPAD guidelines advise routine assessment of psychological functioning
of diabetes nurses compared to pediatricians, therefore the nurse discussed QoL. Clinic 5 used
[188, 196], which contributed to the start of DM-Y according to seven clinics. In addition, the
the conversation as an opportunity to improve adolescents’ comfort level while being alone
Dutch
Association of Pediatricians endorsed DM-Y, which motivated four clinics. Partnerships
Error! Style not defined. with the pediatrician. In clinic 6 the pediatrician felt more competent than the nurse to consult.
between
clinics was another external motive, which corresponded with the survey responses
the survey responses which showed that the nationwide usage of DM-­‐Y was appealing to which showed that the nationwide usage of DM-Y was appealing to 86.1% (13.9% neutral).
86.1% (13.9% neutral). In addition, clinic 1 thought the MY-­‐Q outcomes would help finance In addition, clinic 1 thought the MY-Q outcomes would help finance the assessment of QoL by
the assessment of QoL by insurance companies. Third motivation, objectify QoL: three insurance companies. Third motivation, objectify QoL: three teams wanted to objectify their
teams wanted to objectify their clinical impression of QoL of their patients. Fourth clinical impression of QoL of their patients. Fourth motivation, usability: five teams previously
motivation, usability: five teams previously used other QoL measures but these were not used
other QoL
measures but these were not convenient in practice.
convenient in practice. With regard to the parental part of DM-Y, well-being of parents was examined by five teams
outcomes contributed to the adolescents’ health. 75.0% would recommend DM-Y to other
Table 33:
. A cceptance and and
Change: overview of motivation to change ato
nd change
value of cand
hange Table
Acceptance
Change:
overview
of motivation
value
of change
specialists and 91.7% thought DM-Y is an additional value to routine care. This was further
investigated in the interviews (Table 3). According to eight teams the MY-Q gained insight in
Motivation Value Hospital Psychosocial External Objectify Usability aspects QoL 1 √ √ √ √ 2 √ √ √ √ 3 √ √ 4 √ 5 √ √ √ 6 √ √ √ 7 √ √ 8 √ √ 9 √ √ 10 √ √ QoL, quality of life; MY-­‐Q, MIND Youth-­‐Questionnaire. Barriers MY-­‐Q Consult Logistics √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ Web-­‐based MY-­‐Q √ √ √ √ √ √ √ (1, 2, 5, 6, 9), of which clinic 1 discussed every outcome with the parents and clinic 5 only
discussed outcomes when they were considered alarming or on parents’ initiative (Table 2).
Whether clinic 9 discussed outcomes with parents has not been addressed in the interview.
Change: consultation and the value of DM-Y. In the survey, 66.7% stated that discussing MY-Q
all life-domains and in areas in which problems occurred. The subsequent consultation was
a contribution to routine care for all teams. None of the members felt incapable to discuss
outcomes with the adolescents (58.3% felt capable, 33.3% neutral). Nine teams (except clinic
9) used the outcomes as a way to start the conversation about QoL. According to five clinics
(1, 3, 7, 8, 10) these conversations differed from routine conversations as adolescents opened
up and provided more detailed information. Seven clinics (1–4, 6–8) stated DM-Y gave insight
into psychosocial aspects and helped to concrete and objectify their instinctive thoughts. In the
survey, 58.3% mentioned that the conversation did not concord with their prior expectations
and 86.1% thought DM-Y helped to talk about QoL in more detail. Moreover, 94.4% found that
DM-Y contributed to the recognition of psychosocial problems. In the interviews, six teams
Change: organization. In eight teams the conversation with the adolescent was led by one (1, 2, 4, 6, 8, 9) concluded that screening for problems took place in the subsequent conversation:
professional; in clinics 3 and 4 two professionals consulted with the child simultaneously. Who led the conversation (Table 4) depended for six teams (1, 2, 4, 6, 9, 10) on practical and 82
organizational considerations and for four teams (2, 5, 7, 9) on the expected MY-­‐Q outcome. 83
6
Implementation of QoL
Implementation of QoL
by discussing the outcome it became clear whether referral or extra help was required.
80.6% appreciated the web-based MY-Q, for 41.7% difficulties with the website hindered DM-Y.
According to five teams (1, 3, 6–8) the consultation gave insight into problems in parental
diabetes management, and offered the opportunity to provide parental education. In three
Maintenance: internal and external amendments required to optimize further usage. All interviewed
clinics (1, 3, 7) the subsequent conversation provided an opportunity to talk with adolescents
teams wanted to pursue DM-Y or start again. Clinics 1 and 4 planned to maintain their current
without parents present. Furthermore, clinics 1 and 8 explicitly appreciated MY-Q questions
usage of DM-Y. Logistical amendments would optimize the usage, according to all interviewed
regarding satisfaction with care, which should be assessed on a routine basis, according to
teams (Table 3). Clinics 5–10 believed they needed to construct a schedule in order to organize
ISPAD guidelines [196]. According to 58.3%, DM-Y led to more referrals (22.2% did not know),
proceedings, such as the availability of a private computer room in order to complete the MY-Q.
generally to a psychologist or social worker. This was not related to occupation or clinic. 66.7%
Although the web-based MY-Q caused problems, a paper version was no alternative for six
found the parental part beneficial (19.0% neutral) and 47.6% thought it gave more insight into the
clinics (3, 5, 7–10) because they lacked time to score the answers and perceived a paper-and-
child’s life (28.6% neutral). 47.1% felt capable to discuss outcomes with parents (29.4% neutral).
pencil questionnaire outdated. Clinics 2, 4, and 6 would use a paper version to overcome the
The remaining five clinics found it too difficult to schedule; logistical difficulties and no
computer-related barriers. Teams experienced uncertainties regarding the free accessibility of
possibility to discuss the outcome.
the website in the nearby future. Clinics 4 and 8 created their own computerized MY-Q to avoid
web-based difficulties. Clinic 4 just started the usage of this software program, clinic 8 did not
6
Maintenance: facilitators and barriers for the integration of DM-Y into routine care. All teams
yet. The future usage of DM-Y in clinics 2, 3, 8, and 11 is determined by a merger between these
struggled with the logistics of DM-Y, which hindered the implementation in four clinics (7–10); this
teams and other pediatric diabetes clinics. Five teams (4–8, 10) addressed their need for more
included room to complete the MY-Q, requesting adolescents to come earlier for this purpose,
collaboration, external input, clearance, and feedback between participating clinics and The
problems with the web-based MY-Q, take notice of the outcome prior to consult and discuss within
Dutch Association of Pediatrics to stimulate DM-Y: insight into best practices would help overcome
the regular time. Three teams (5, 9, 10) claimed to be demotivated because of these barriers.
barriers. 44.7% would like to have an annual meeting (19.4% not applicable) and 61.1% thinks extra
Additional tasks in order to use DM-Y caused time pressure according to clinics 1, 3, and 6–10.
information, such as a newsletter, would stimulate the implementation of DM-Y (27.8% neutral).
Mandatory physical measurements were prioritized and challenged the implementation. Teams 1,
2, and 4 had a DM-Y schedule in which the tasks of each member were clear. This served as a facilitator
Adolescents and parents. Twenty-nine (42.0%) adolescents completed the online survey, 72.4%
for embedding DM-Y. In the survey, only 22.2% experienced time pressure (38.9% neutral, 36.1%
girls, mean age 15 (±2.2), 65.5% treated in clinic 1. No differences in age, ethnicity, and MY-Q
did not experience time pressure). This was not related to occupation or clinic. According to 75.0%,
outcomes were found between those who completed the survey and those who did not. However,
DM-Y did not interfere with regular medical care for adolescents (19.4% neutral). Collaboration
participants did have lower HbA1c [62.8 mmol/mol (7.9%) vs. 75.0 mmol/mol (9.0%), T = 3.53,
within teams with regard to DM-Y was considered good by 61.1%. Disagreeing 22.2% came from
p < 0.001] and were more likely to be girls (χ2 = 4.283, p = 0.038). Appendix 2B shows the frequency
clinic 3, 10, and 11 (Kuskall–Wallis χ2 = 21.751, p = 0.01). This was unrelated to occupation. Other
of answers on the survey. The subsequent conversation was appreciated most by adolescents
questionnaires besides MY-Q were used by 36.1%, of which 66.7% thought this did not hinder
from clinic 1 (χ2 = 8.250, p = 0.041). 41.4% found it beneficial to complete the MY-Q, same
DM-Y. According to seven teams the web-based MY-Q frequently caused problems. Even though
percentage neutral. According to 62.1% more type 1 diabetes adolescents should complete the
84
85
6
6
Implementation of QoL
Implementation of QoL
MY-Q (27.6% neutral). With regard to the subsequent conversation, 51.9% was neutral and 25.9%
diabetes teams, adolescents, and parents, but difficult to implement. Eight of ten interviewed
was positive about talking about their QoL. 85.2% felt themselves heard. 6.9% of the adolescents
pediatric diabetes teams did not use the program to its complete extent, yet according to all
found it hard to talk about QoL with the team member. Sixty-six (74.2%) of the parents completed
interviewed teams, DM-Y contributed to routine care. The MY-Q was considered a convenient
the survey, most of them mothers (84.8%). 45.5% was from clinic 1 but no differences were found
tool to investigate the adolescents’ well-being, to screen for problems, and to facilitate the
between their answers and those from other clinics. Appendix 2C shows the frequency of answers
conversation with the adolescents. All teams wanted to pursue the usage of DM-Y or start again.
on the survey. 78.8% of parents found DM-Y beneficial to their child. 80.3% did not consider it a
In the two clinics that successfully implemented DM-Y, MY-Q assessment, and discussion
waste of time. 61.5% completed the parental part of DM-Y (23.1% could not remember). 80.0% of
is anchored in their routine pediatric diabetes care. Four teams also used DM-Y, but not as part
them believed it to have added value to the pediatric diabetes care. 90.0% considered it a good
of their routine care. The remaining four teams no longer used DM-Y. Because all participants
idea if the parental part of DM-Y was introduced to more parents. The outcome was discussed
valued DM-Y, the question remains what it is that makes implementing DM-Y difficult.
with 23.1%. All of these parents found this beneficial and all felt themselves heard by the team
It is well-known that the implementation of an intervention requiring practice change
member. Of those who did not complete the parental part, 20.0% would not have liked to.
in health care is challenging [190-193]. The eight teams that did not use DM-Y to its complete
extent lacked time and manpower and failed to make a schedule regarding the roles and tasks
Table
4: Change: overview of team members discussing MY-Q outcomes with adolescents
in order to use DM-Y. Most teams struggled with logistics and the majority of teams prioritized
Hospital 1 2 3 4 5 6 7 8 9 10 Nr of Team members 1 1 2 2 1 1 1 1 1 1 Nurse Pediatrician Psychologist √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ physical measurements. The fact that adolescents and their parents valued DM-Y appeared
not to facilitate the implementation process.
With regard to the five stages of implementation [193], the teams Oriented themselves
and gained Insight into DM-Y because they were self selected and in all clinics at least some
adolescents completed the MY-Q. In order to accept change of routine care, team members
had to develop a positive attitude toward DM-Y [190-193] and prioritize implementation [193].
At first, teams were motivated to start and decided to change their routine care. Nonetheless,
the positive attitude did not sustain in most teams. The stages Change and Maintenance were
MY-­‐Q, MIND Youth-­‐Questionnaire. difficult to accomplish for the majority of teams. Possibly, the start of DM-Y was probably
Conclusions
a stage of opportunism, rather than complete acceptance.
Assessment
and discussion of QoL should be conducted in routine pediatric diabetes care,
All teams offer specialized pediatric diabetes care and are homogeneous with regard to their
according to ISPAD guidelines [188, 196], based on the evidence that systematic monitoring
reimbursement system. However, time appears to be a crucial factor, where clinics perform
of QoL improves well-being of adolescents and satisfaction with care [79]. In order to continue
physical assessments on a routine basis, at the expense of QoL. As the motivation for teams to
and expand the usage of DM-Y, this study aimed to evaluate the implementation process
start DM-Y was often external, a quality of care indicator, such as stated by the Better control in
and the actual usage of the intervention. Results show that DM-Y is highly appreciated by
Pediatric and Adolescent diabeteS: Working to crEate CEnTers of Reference (SWEET) group [197],
86
87
6
collaboration, communication regarding DM-­‐Y, and clarity about the availability of the web-­‐
based MY-­‐Q could contribute to its further usage. With best practices in mind, it seems Implementation of QoL
worthwhile to assist teams with the organization of DM-­‐Y. For the two teams who Implementation
of QoL
maintained DM-­‐Y a schedule helped to facilitate the implementation. Two other teams attempted to solve web-­‐based problems by creating their own MY-­‐Q software. A summary could help anchor DM-Y. In addition, it should be investigated whether legislation is helpful.
of facilitate helped
dissemination of the
DM-­‐implementation.
Y in routine care is whorecommendations maintained DM-Yto a schedule
to facilitate
Twonationwide other teams
Nonetheless, we should keep in mind that the teams participating in this study were self-selected;
represented Table web-based
5. attempted toin solve
problems by creating their own MY-Q software. A summary
when DM-Y becomes mandatory new non-self-selected teams might experience even more
of recommendations to facilitate dissemination of DM-Y in routine care nationwide is
represented in Table 5.
challenges with regard to the implementation.
For the majority of teams, the consultation based on MY-Q outcomes did not concord with
their prior expectations. Even though most members found themselves capable to discuss QoL
Table 5: Summary of recommendations to facilitate dissemination of DM-Y.
with adolescents, research has shown that pediatric diabetes professionals find it easier to discuss
Within teams Between teams / National level Prioritize DM-­‐Y Prioritize DM-­‐Y Schedule roles and tasks Build quality of care indicators Additional training Interactive, continuous and evidence based education period of time [198], consequently insufficient skills could form a barrier in the implementation
MY-­‐Q at home / create MY-­‐Q software Strategies for future usage process. A possibility to stimulate the use of DM-Y could therefore be additional training. One could
Consensus, collaboration and f eedback Collaboration and exchange of best practices medical issues than psychosocial issues [198]. In addition to a questionnaire, such as the MY-Q,
communication skills seem of great importance [199]. Trained skills often do not last over a longer
argue that another possibility could be to use DM-Y on only those who seem at risk, as not the
6
Table 5. Summary of recommendations to facilitate dissemination of DM-­‐Y. DM-­‐Y, DAWN MIND-­‐Youth; MY-­‐Q, MIND Youth-­‐Questionnaire consultation itself but the time spend to organize proceedings for all adolescents was considered
Our study contributes to a greater understanding of the implementation process in a practice
time consuming. But, for instance, glycemic control has not been found to be a good predictor of
setting from the users point of view, with the use of both quantitative and qualitative measures
QoL [200, 201], therefore it is difficult to determine who is at risk. Addressing psychosocial problems
monitoring of QoL in all children is to recognize life is not only about diabetes. Because the MY-Q can
[203]. Nonetheless, some limitations should be taken into account. With regard to the experiences
86 of teams, there could be an interviewers bias. There was some discrepancy between the survey
results and the interviews concerning experienced level of time pressure. Perhaps members
be completed online, a solution for the lack of space and time is to complete it at home; however,
who were more positive about DM-Y completed the survey, while those who experienced time
as research pointed out that adolescents feared they had to discuss answers with their parents
pressure did not, which underscores the value of our mixed-methods approach. The fact that only
[101] it was preferred to complete the MY-Q in the clinic. Difficulties with the implementation of
little over half of the members completed the online survey reflects on the mentioned barriers of
DAWN MIND procedure are also seen in adult care[202]. Even though satisfied, lack of time and
the implementation process. Regarding the experiences of adolescents and parents, there might
manpower and logistical issues were barriers here as well [202]. Blair recently revealed that in
be a recall bias as some participants received care in clinics where DM-Y was no (longer) part of
order to change routine pediatric care, interactive, continuous, evidence based education, and
the routine care. Less than half of the invited adolescents completed the online survey and mainly
a certain amount of local consensus and feedback are required. Actions and measures at the
those with lower HbA1c and female gender, which might limit the external validity.
team level are essential when developing plans for change in routine care [194]. Teams express
The teams participating in this study already followed the first stages of implementation
their need for more collaboration, communication regarding DM-Y, and clarity about the
prior to the research. New teams starting with DM-Y might face extra barriers, which makes
availability of the web-based MY-Q could contribute to its further usage. With best practices
the implementation even more challenging. Therefore, more effort should be paid to facilitate
in mind, it seems worthwhile to assist teams with the organization of DM-Y. For the two teams
dissemination of DM-Y in routine care nationwide.
at an early stage, prior to the label ‘at risk’, prevents progression of problems. The strength of
88
89
6
Chapter
7
General Discussion
‘It’s challenging to take care of all my diabetes tasks
I feel proud when my parents or diabetes team tell me I’m doing a good job’
91
General discussion
7
General discussion
Reflection on main findings
Discussion
The previous Chapters confirm the understanding that type 1 diabetes (T1D) impacts many
Risk and protective factors
aspects of the daily lives of youth with T1D. Its treatment builds, to a large extent, on the
A chronic illness such as T1D is a significant risk factor for psychosocial outcomes (later) in life
self-management of these youngsters and their parents. The daily hassle of diabetes in
[3]. In this thesis, we identified psychosocial risk factors and protective factors that associate
addition to normal developmental tasks of youth (becoming an independent individual [1]),
with diabetes outcomes: problem behavior, disturbed eating behavior, parental stress and
challenges the psychosocial well-being of these youngsters and that of their parents. Within
unsupportive parenting are related to poorer glycemic outcomes, and supportive parenting
the treatment of T1D there are many factors that contribute to the diabetes management and
behavior is associated with more optimal HbA1c. In previous research other factors are
its outcomes. The aim of this thesis was to gain knowledge about some of these psychosocial
found to predict poor glycemic control too, including negative affects about blood glucose
factors and their relation with diabetes outcomes, in order to improve pediatric diabetes care.
monitoring and negative attribution of peers [204-206]. In coherence with literature, most
For this purpose the DINO study was conducted, as described in Chapter 2. Based on the
of the psychosocial risk factors and protective factors identified in this study associated with
results of this study, we conclude that A) there is a relation between youth’s problem behavior
one another as well. For example, the study reported on in Chapter 5, showed that parental
and worsened glycemic control, mediated by less confidence in diabetes self-care and diabetes
well-being related to youth’s problem behavior and in turn, this is associated with glycemic
mismanagement, as described in Chapter 3; B) disturbed eating behavior is a sincere cause
control. But in Chapter 3, we concluded that the relation between problem behavior and
of concern in pediatric diabetes care, which highlights the importance of monitoring and
glycemic control is mediated by youth’s confidence in, and management of diabetes self-care
discussing this issue with youth themselves, as described in Chapter 4; C) general parental
(although this was based on youth’s reports while the study reported on in Chapter 5, was
well-being influences youth’s problem behavior as well as diabetes parenting behavior and
based on parental reports). In addition, it can be assumed that worse diabetes outcomes have
parental diabetes related stress. In turn, these factors influence glycemic control, as described
a negative impact on youth’s confidence in diabetes self-care. Youth’s perceptions about T1D
in Chapter 5. These results provide insight into the psychosocial mechanisms and challenges
(that associate with diabetes outcomes [207, 208]) are shaped by diabetes related risk factors
that youth with T1D face. Also, the results clearly show that the psychosocial well-being of
and protective factors (e.g. how T1D can be managed at school, satisfaction with treatment
these youngsters is an important outcome, in addition to diabetes outcomes. This highlights
by the care team and the self-management) [207]. This highlights the importance of the
the importance of monitoring and discussing their quality of life (QoL). However, we are drawn
domains in which these factors exist (such as family, peers, school, diabetes team, but also the
to the conclusion that, D) it is difficult to implement the structural assessment of QoL in routine
community and the cultural believes). The family domain is one of great importance (as shown
pediatric diabetes care, as described in Chapter 6. Luckily, all diabetes teams participating in
in Chapter 5), since parents perform or monitor the diabetes tasks [7, 19]. A good parent-child
our study acknowledge the importance of the psychosocial aspects of T1D in their pediatric
relationship is a significant protective factor in diabetes care: we found that supportive
care. Not only for the sake of people with diabetes: literature shows that more attention to
parenting and reduced parental stress have a positive effect on youth’s glycemic control. Which
these aspects is cost-effective [135].
is consistent with previous research that demonstrated that parent’s acceptance and diabetes
In this final Chapter, the results of this thesis are discussed and suggestions are given on
specific knowledge increases youth’s disclosure in terms of diabetes management, which
how to incorporate them in clinical care.
predicts better diabetes outcomes [20, 21, 209]. The parent-child relationship also reflects on
92
93
7
General discussion
General discussion
disturbed eating behavior (DEB): a negative family climate relates to a negative body image and,
of youngsters with T1D promises better diabetes and psychosocial outcomes [213, 214], which
consequently, with problematic eating [16, 210, 211]. When parents restrict their youngster
makes a positive approach in clinical care worthwhile.
with T1D from food and when parents criticize their own and their child’s body and weight,
the risk of DEB and more severe eating disorders in youth with T1D increases [210, 211].
Figure 1:Children’s circle tool based on the ecological model Bronfenbrenner and adapted
But, as shown in Chapter 4, parents know little of the body and weight concerns of their youth.
by Prof. Barbara Anderson on behalf of DAWN Youth.
Thus, the family domain impacts diabetes outcomes, however, T1D impacts the psychosocial
functioning of parents as well: previous research revealed that the diagnosis of T1D is often
traumatic for parents and parents of youth with T1D have impaired levels of stress and low
emotional well-being, which affects psychosocial and diabetes outcomes of their children [22-25].
‘If you’d have asked me these questions a few years ago, the answers would be different.
Perhaps as a parent, I became immune to the never ending worries and the feeling of failure
towards my child, no matter how hard I try. In diabetes, a 100% good result
is just not going to happen. That’s quite frustrating.’
Obviously, the interactions between psychosocial and diabetes specific aspects are complex
and there seems to be a two-way street [212]. The bio-psychological model shown in Chapter 2,
7
7
gives a schematic impression of these interactions, however, this model presents only univariate
directions. Given the associations between many factors involved, multivariate relations are
more likely to exists. The accumulation of risk factors and its balance with protective factors
predict psychosocial situation (later) in life [3] and, consequently, diabetes management and
its outcomes. The children’s circle tool, presented in Figure 1, is a diabetes specific visualization
of the ecological model of Bronfenbrenner [2] and shows the interactions between a youngster
Clinical practice
and several environmental domains. With respect to the individual youngster with T1D, it is
Although progress is being made towards the cure for T1D, currently, its daily management
important to acknowledge that the occurrence and combination of protective factors and
remains a hassle. In this thesis we identified several factors that are related to glycemic
risk factors differs. It is therefore a challenging task in clinical practice to construe research
outcomes and the psychosocial situation. In clinical practice attention towards these factors
(based on mean outcomes of participants) to individuals, with their unique set of risk factors
remains relevant, as suggested in previous literature [215]. Not necessarily by including all
and protective factors. Building resilience by expanding the protective factors in the life
factors specifically in guidelines, but as an addition to the recommendation to monitor QoL.
94
95
General discussion
7
General discussion
The MIND Youth Questionnaire (MY-Q) assesses QoL on several life domains, including eating
Reflection on research methods
behavior, family functioning and emotional well-being. Its outcome can be used to examine
DINO study design
(or serve as an indication to examine) the identified factors as well. In order to screen on external
Within the DINO study, one of the primary questions we wanted to answer was whether
problems, an amendment of the MY-Q by the addition of an external problem behavior scale is
youngsters diagnosed with T1D before puberty differed from those diagnosed during puberty.
recommended. Other measures used in this thesis are available for screening purposes in clinical
By including participants aged eight to fifteen years at baseline, an equal amount of both
practice as well (including the Dutch version of the DEPS-R and the AHEAD for the assessment
groups was expected. However, the majority of participants were diagnosed before puberty
of eating problems, translated by the DINO study group). It is shown that youngsters are more
(87.2% for girls and 95.1% for boys). As stated in the DINO study design, Chapter 2, a minimum
satisfied with the care they receive when pediatric teams monitor their QoL. Also, their QoL
of 43 youngsters in each group was required to detect differences in HbA1c and likely cases of
increases [79]. This potential protective aspect of pediatric diabetes care deserves to be expanded.
psychological dysfunction between youth with pre vs. pubertal onset of diabetes. Therefore,
In addition to the care directly aimed at youth, attention towards protective and risk factors
with only fourteen youngsters with pubertal onset of diabetes in total, assumptions about
in the family domain is recommended too. As shown in Chapter 6, parents of youth with T1D
dissimilarities between the two groups are unreliable. By including adolescents older than
appreciate it when pediatric diabetes teams take their emotional well-being and diabetes related
fifteen years at baseline we might have been able to answer this question.
stress into consideration. Currently, several diabetes specific parental interventions are available,
In Chapter 3 we described the influence of problem behavior on HbA1c, in youth with a mean
including multi system therapy, (tele-health) behavior therapy and parental stress reduction
age of thirteen years. Since problem behavior - both internal and external - is more prevalent in
techniques [177, 178]. However, despite the appreciation of parents and the availability
older adolescents [128], the prevalence of problem behavior in our study is likely to increase over
of effective interventions, it remains difficult to actually intervene on parenting behavior:
time. Future research is needed to determine if this impacts the results presented in this thesis.
in general, parents are content with their own parenting skills [216]; advice from the medical
Within the DINO study we did not take the influence of physiological aspects- such as
team can be interpreted as criticism [217]. After all, the main focus of the diabetes care team
adrenaline an cortisol caused by stress - on HbA1c into account, although this relationship has
is to take care of the diabetes. In order to assess if assistance or advice in parenting is feasible
been well described [221, 222]. Future research is needed to examine if the relation between
without implying insufficient parenting, the Structured Problem Analysis of Raising Kids (SPARK)
problem behavior and HbA1c is not only mediated by self-management and self-confidence,
can be used [218]. Although it would benefit from amendment for this specific group, it might be
but also by stress.
interesting to use this validated measure for the detection of parenting issues or struggles [219].
In general, when parenting issues occur, a communication style defined by partnership (aimed at
External validity
the same interest, namely their child as a healthy and happy individual), compassion and respect
In contrast to other countries (Scandinavian countries for example), there is no national
for the parenting style and knowledge is successful, such as motivational interviewing [217, 220].
diabetes registry in the Netherlands from which we could extract data. Unfortunately within
To summarize, in clinical practice it is worthwhile to monitor QoL in order to identify potential risk
the DINO study we did not include as many participants as we anticipated: one-fourth of
factors for poor psychosocial and diabetes outcomes in both youth and their parents. Interventions
the approached youngsters gave their informed consent. Difficulties in terms of recruiting
should not only target risk factors but build on protective factors too.
adolescents with T1D are described in other studies as well [79, 170]. Main reasons to reject
96
97
7
General discussion
General discussion
participation were inherent to the puberty of youngsters: they declared to be too busy
Measures
with friends and school in order to complete questionnaires or undergo cognitive tests.
HbA1c: Consistent with the majority of studies on diabetes, we used HbA1c as primary diabetes
Consultations with adolescents and parents a priori about the content of the research and
outcome. Although it is considered the golden standard [233], HbA1c does not reflect the daily
by approaching them personally rather than by mail [223], might have boosted the number
fluctuations of blood glucose levels. Since hyper- and hypoglycemia affect functioning [6, 234]
of participating youth. However, it would be unethical if youth’s contact information was
and daily fluctuations are an independent risk factor for diabetes complications [235],
shared with the researchers prior to youth’s participation. Clearly, posters and flyers at the
it would be interesting to use sensor glucose measurements in future research to determine
clinics were insufficient to increase the number of participants. For future research it would
the influence of daily variation in glucose levels rather than HbA1c.
therefore be interesting to use social media in order to recruit youth. Youth’s Facebook
Adherence was measured with an adapted version of the mismanagement scale. Since the
posts, Tweets (Twitter posts), YouTube videos, or other social media content could be
external validity of the mismanagement scale in our study was relatively low, as shown in
used as data sources for qualitative research as well [224]. Also, the education level of the
Chapter 3, it would be interesting to further examine the relations presented in this thesis by
majority of DINO-participants was high, and all youths were born in the Netherlands.
using the Self-Care Inventory [236], a youth and parents report commonly used in diabetes
research to measure adherence.
7
The factors mentioned above question the generalizability of the results. Previous research
Parental proxy bias: Even though youth’s behavior is frequently assessed by parental proxy
showed that higher socioeconomic status and being part of an ethnic majority are
reports, a recent study showed that parents’ self-reported well-being associates with their
associated with better diabetes outcomes and less depressive symptoms [48, 225-227].
assumptions of their children’s well-being [158]. In Chapter 5, we concluded that optimizing
Since diversity in socioeconomic status and ethnicity was underrepresented in our study,
parental well-being can be beneficial for both psychosocial and diabetes outcomes in youth.
there is a chance that we underestimate the prevalence of psychosocial problems and poor
We chose to use the parental reports, because not all youngsters completed the Strengths
diabetes outcomes, which highlights the importance of screening for these problems in
and Difficulties Questionnaire with which we assessed problem behavior: only youngsters
clinical care. Future research should aim to include more heterogeneous samples in order
eleven years and older. Based on the parental proxy bias, it seems worthwhile to examine
to provide results that are generalizable to all ethnicities and education levels. Barriers to
if the relation between parental well-being and problem behavior diminishes when youth’s
participate in health research exists in two areas: the manner on which research is designed
reporting of their own functioning is used.
and the perception of minority groups on the research process [228]. Therefore improving
Evaluation of the implementation of the QoL method: In order to evaluate the implementation
relationships between potential participants and medical teams (including social marketing
of DAWN MIND-Youth, health care professionals completed a questionnaire specific for this
and community outreach), recruiting participants personally (face to face instead of mail)
evaluation study. It could be suggested that the external validity of the results improves
and provide linguistically appropriate materials might be beneficial in terms of research
when a structured questionnaire was used to evaluate a health care innovation, such as
participation [228-232].
the Measurement Instrument for Determinants of Innovations (MIDI) [237]. However, both
questionnaires were founded on the implementation stages described in Chapter 6 [193],
and at the time of our evaluation, the publication of the MIDI took another year.
98
99
7
General discussion
General discussion
Future perspectives
Research
With the DINO study, the cognitive development of youth was examined for four years,
as described in Chapter 2. Currently, it is planned to analyze these cognitive data, along
with other collected data that were not included in this thesis. The DINO study group is
in the process of grant applications in order to continue its research. By including more
To conclude, T1D is a psychosocial challenge for both youth and their parents. Both risk
factors and protective factors influence psychosocial and diabetes outcomes. Pediatric
diabetes teams can be a significant protective factor when they screen, monitor and discuss
not only diabetes outcomes but psychosocial outcomes too.
participants, we expect to answer the question whether youth diagnosed with T1D before
puberty differ from those diagnosed during puberty nonetheless. This in addition to the
further investigation of the psychosocial, cognitive and physical development of youth with T1D.
Within the studies reported on in this thesis, we did not include a healthy control group.
Future research is encouraged to investigate the differences in risk and protective factors
between youth with T1D and their parents, and the average Dutch population. Also, the
Dutch version of the DEPS-R has not been validated yet. And even though its validity in
other languages is well described [152, 238], it would be a contribution to research and
clinical proposes to examine the validity of this translated version as well.
Implementation of QoL
7
7
As described in Chapter 6, diabetes teams wanted to use the MY-Q as assessment tool for
the QoL of youth with T1D in routine pediatric care, however, logistical barriers hindered
this process. To facilitate the usage of the MY-Q, the VU university medical center and the
Academic Medical Center now provide KLIK-diabetes. KLIK (Kwaliteit van Leven In Kaart),
developed by Academic Medical Center [239], is an online web-based platform supporting
the measurement of the QoL of children with medical conditions and those of their parents.
Youth and /or parents complete questionnaires online and the results are discussed by the
healthcare team. By the inclusion of the MY-Q and by the adjustment of the KLIK training for
diabetes professionals, the barriers diabetes teams faced in the earlier version of the MY-Q are
likely to be declined. Evaluation is planned to examine whether KLIK diabetes is indeed feasible.
100
101
Summary
‘It’s nice that you investigate all of these things in order to optimize diabetes care
and I do appreciate that, but ehmm ...
why didn’t you before, so that it would be of any use for me too?’
S
103
Summary
Summary
Chapter 1: General introduction
Chapter 3: Low Self-Confidence and Diabetes Mismanagement in Youth with Type 1
In this introductory Chapter, a short definition of both youth and type 1 diabetes (T1D), and the
Diabetes Mediate the Relationship between Behavioral Problems and Elevated HbA1c
outline of the thesis are given. In youth, mental representations for understanding the world,
In Chapter 3, the relation between behavior problems (internalizing, externalizing) and
self and others are formed. These representations are based on experiences, including risk
glycemic control (HbA1c) in youth with type T1D is described. We wondered if the association
and protective factors. The accumulation of risk factors and its balance with protective factors
between problem behavior and HbA1c is mediated by self-confidence and mismanagement of
predict the psychosocial situation (later) in life. Also, characteristics that will help to manage a
diabetes. In this sub-study, 88 youth with T1D, aged eleven to fifteen years participating in the
chronic illness are developed, such as the capability to think logical, critical and hypothetical,
DINO study completed a (online) survey. Problem behavior was assessed with the Strengths
which enables youngsters to oversee consequences and nuances. One of the main chronic
and Difficulties Questionnaire; mediating variables were assessed using the Confidence In
youth illnesses is T1D. The daily hassle of T1D challenges the psychosocial and emotional
Diabetes Self-care-Youth and Diabetes Mismanagement Questionnaire; HbA1c was derived
well-being of youth and their parents, as it requires intensive self-management. While having
from hospital charts. To analyze the data, bootstrap mediation analysis for multiple mediation
T1D can be considered a risk factor in itself, within its treatment there are many factors that
was utilized. Both overall problem behavior and externalizing behavior problems are related to
contribute to the diabetes management, its outcomes and the psychosocial situation. To gain
worsened HbA1c. This relation is mediated through less confidence in diabetes self-care and
knowledge about some of these factors, the psychosocial development of youth aged eight to
mismanagement of diabetes. A direct relation between internal problem behavior and HbA1c
eighteen years with T1D was assessed in this thesis.
is not found, however, more internal problem behavior did result in less confidence and more
mismanagement, which are associated with increased HbA1c. Screening for problem behavior
Chapter 2: Diabetes IN develOpment (DINO): the bio-psychosocial, family functioning and
and mismanagement, and assisting youth with T1D in building confidence seem indicated in
parental well-being of youth with type 1 diabetes: a longitudinal cohort study design
order to optimize glycemic control.
The DINO study protocol is described in Chapter 2. The DINO (Diabetes IN develOpment) study
S
is a longitudinal multi-center cohort study in which the biological, psychosocial - including
Chapter 4: Disturbed eating behaviors in adolescents with type 1 diabetes. How to screen
disturbed eating behaviors-and cognitive development of youth with T1D aged eight to fifteen
for yellow flags in clinical practice?
years at baseline was researched. In Chapter 2, four research questions are proposed: 1) If and
Adolescents with T1D are at an increased risk of disturbed eating behaviors (DEBs). In
how the onset of T1D before vs. during puberty results in different outcomes of glycemic control,
Chapter 4, the prevalence of these behaviors and the associated psychological and clinical
self-management, psychological functioning and diabetes-related quality of life (QoL). 2) The
‘yellow flags’ are described. Also, the concordance between adolescents and parents, and
timing of onset of disturbed eating behavior, its risk factors and its prospective course in relation
adolescents and clinicians with respect to disturbed eating behaviors are presented. In this
to glycemic and psychological consequences. 3) If and how the onset of T1D before vs. during
sub-study, 103 adolescents aged eleven to sixteen years and their parents participating
puberty results in different family functioning and parental well-being. 4) If and how the cognitive
in the DINO study completed questionnaires. A stepwise approach was used to assess DEBs:
development of youth with T1D relates to glycemic control and diabetes self-management.
only adolescents whose answers raised psychological yellow flags for DEBs completed the
104
105
S
Summary
Summary
Diabetes Eating Problems Scale-Revised and questions from the AHEAD study. Kruskal-Wallis
behavior and parental diabetes stress, which in turn showed a negative effect on glycemic control.
tests, post hoc Mann-Whitney U test and chi-squared tests were utilized to examine clinical
Youth’s problem behavior is also impacted by unsupportive diabetes parenting and parental
yellow flags. Parents and clinicians shared their observations with regard to possible disturbed
diabetes stress. Supportive parenting is related to better glycemic outcomes. However, our
eating behaviors. Cohen’s kappa was used to assess concordance. The concordance between
results do not support the hypothesized direct relation between reduced parental well-being
parents and adolescents was slight and that between clinicians and adolescents was fair. The
and worsened glycemic control in youth. Family interventions, aimed to improve youth’s HbA1c,
answers from half of the adolescents raised psychological yellow flags, indicating concerns
are more likely to succeed when they target negative diabetes parenting behavior, parental
about their body weight and physical appearance. In most cases this was not problematic:
diabetes stress and youth’s problem behavior, and stimulate supportive parenting.
less than one in ten adolescents reported DEBs. Clinical yellow flags included elevated HbA1c,
older age, increased dieting frequency, reduced QoL, less diabetes self-confidence, worsened
Chapter 6: Implementation of quality of life monitoring in Dutch routine care of adolescents
diabetes management and body dissatisfaction. Contrary to what we expected, gender and
with type 1 diabetes: appreciated but difficult
body mass index (BMI) are not associated with DEBs in this study. In order to prevent the onset
In Chapter 6, the evaluation of the implementation of the DAWN MIND-Youth program is
or deterioration of DEBs, a stepwise approach to screen for these behaviors such as described
described. This program aims to improve pediatric care for youth with T1D: youth complete
in Chapter 4 is recommended. Pediatric guidelines instruct diabetes teams to assess the QoL of
the MIND Youth Questionnaire (MY-Q) regarding their QoL and its outcomes are discussed
patients annually. This mandatory assessment provides the opportunity to assess DEBs as well.
with a diabetes clinician. Additionally parental well-being and parental diabetes stress can be
examined. In the Netherlands, eleven hospitals started the usage of the method in 2010. In order
S
Chapter 5: Not parental well-being but parental diabetes behaviors and stress are related
to evaluate its implementation, 36 diabetes team members, 29 adolescents and 66 parents
to glycemic control in youth with type 1 diabetes. Longitudinal data from the DINO study
completed an online survey regarding their overall experience with the DAWN MIND-Youth
In Chapter 5, we examined whether there is a longitudinal relation between parental well-being
program. In addition, a semi structured interview was conducted with ten diabetes teams.
and youth’s glycemic control. And if youth’s problem behavior, diabetes parenting behavior and
Template analysis was used to analyze the interviews. The results indicated that the program
parental diabetes stress impact this relation. In this sub-study, 174 parents of youth with T1D
served as a screening tool; teams appreciated the possibility to concretize their feelings about
participating in the DINO study completed questionnaires at three time waves, each with a one
adolescents. Even though the method was a high valued addition to the routine care, only
year interval. Using Generalized Estimating Equations analyses the relation between parents’
two out of ten interviewed teams successfully implemented the DAWN MIND-Youth program.
well-being (WHO-5 scores) and youth’s HbA1c was examined. Second, the relations between
Logistical barriers hindered the implementation. Nonetheless, all interviewed teams wanted to
parents’ well-being, youth’s problem behavior (Strength and Difficulties Questionnaire), diabetes
include the program in their routine care in the nearby future. With respect to the adolescents
parenting behavior (Diabetes Family Behavior Checklist supportive and unsupportive scales),
and their parents: parents highly appreciated the usage of the MY-Q and the parental
parental diabetes stress (Problem Areas In Diabetes-Parent revised) and HbA1c were analyzed.
questionnaires, adolescents provided a neutral to positive opinion. It seems worthwhile
Low emotional well-being is reported by 32% of participating parents. Reduced parental well-
to continue the usage and further implementation of the DAWN MIND-Youth program,
being is related to increased problem behavior in youth and more unsupportive diabetes parenting
in which more effort should be made to overcome logistical barriers.
106
107
S
Summary
NL samenvatting
Chapter 7: General discussion
In this final Chapter, the main findings of the thesis and its implications for clinical practice
and future research are discussed. There are many factors that contribute to the diabetes
management, its outcomes and youth’s psychosocial functioning. Within this thesis, we
identified psychosocial risk factors and protective factors that are associated with diabetes
outcomes. Problem behavior, disturbed eating behavior, parental stress and unsupportive
Diabetes type 1 in de kindertijd en adolescentie:
een psychosociale uitdaging
parenting are related to poorer glycemic outcomes. Supportive parenting behavior is associated
with more optimal HbA1c. Since most of these psychosocial aspects interact with each other
as well, a multivariate direction is likely to exist. The children’s circle tool (Figure 1 in Chapter 7)
visualizes the interactions between a youngster and several environmental domains
‘Ik ben trots op mijn diabetes als ik terugkijk op wat ik er allemaal voor moet doen’
(e.g. family, peers, school, diabetes team, but also the community and the cultural believes).
We were drawn to the conclusion that T1D is a psychosocial challenge for both youth and their
parents. Several risk factors and their balance with protective factors influence psychosocial
and diabetes outcomes. Pediatric diabetes teams can be a significant protective factor when
they screen, monitor and discuss not only diabetes outcomes, but also psychosocial outcomes,
including their QoL. Structural attention for youth’s QoL is, however, challenging to implement
in routine diabetes care. The external validity of the results presented in this thesis is questioned
by the low response rate and the lack of socioeconomic and ethnic diversity. Suggestions
on how to include a heterogenic population in research are made.
S
NL
108
109
Samenvatting
Samenvatting
Hoofdstuk 1: Algemene inleiding
Hoofdstuk 2: Diabetes IN Ontwikkeling (DINO): de bio-psychosociale ontwikkeling, het
In het eerste hoofdstuk van dit proefschrift zijn een algemene introductie en de opzet van het
gezinsfunctioneren en het ouderlijk welbevinden van kinderen en adolescenten met
proefschrift beschreven. De kindertijd en adolescentie zijn meestal relatief korte periodes in
diabetes type 1: een longitudinaal cohortstudiedesign
het leven. Toch zijn ze gevuld met ervaringen waarmee kinderen en jongeren hun omgeving
In hoofdstuk 2 is het design van de DINO-studie (Diabetes In Ontwikkeling) beschreven. Aangezien
betekenis geven en waarmee ze ontdekken hoe ze zich verhouden tot die omgeving.
er nog weinig onderzoek gedaan was naar de wijze waarop de biologische, psychosociale, cognitieve
Sommige van deze ervaringen kunnen beschouwd worden als risicofactoren of protectieve
ontwikkelingen en veranderingen in de ouder-kind relatie in de kindertijd en adolescentie
factoren. De optelsom van risicofactoren en de balans met protectieve factoren voorspellen
interacteren met diabetesuitkomsten, is het DINO-onderzoek opgezet. In deze longitudinale
het psychosociale functioneren (op latere leeftijd). Het hebben van diabetes type 1 (DT1)
multi-center cohortstudie zijn kinderen en adolescenten met DT1 in de leeftijd van acht tot
is zo’n risicofactor. Bij mensen met deze chronische ziekte zijn de insuline producerende cellen
en met vijftien (bij aanvang van de studie) gevolgd, met metingen op baseline, na 12 maanden,
in de pancreas niet meer instaat dit hormoon te produceren. Hierdoor wordt de glucose in
na 24 maanden en na 36 maanden. Op deze meetmomenten is de biologische, de psychosociale
de bloedbaan niet opgenomen in de lichaamscellen en zodoende stijgt het glucoseniveau.
(inclusief eetproblematiek) en cognitieve ontwikkeling evenals het gezinsfunctioneren en het
De HbA1c-waarde in het bloed geeft een indicatie van de gemiddelde bloedglucose van
ouderlijk welbevinden in kaart gebracht. Het doel daarvan is het vinden van een antwoord op de
de voorgaande vier tot twaalf weken. Een verhoogd HbA1c verhoogt de kans op diverse
volgende onderzoeksvragen: 1) Wat zijn de verschillen tussen de deelnemers die vóór hun puberteit
lichamelijke complicaties. Mensen met DT1 zijn afhankelijk van exogene insuline, die
gediagnosticeerd werden met DT1 en hen die tijdens hun puberteit gediagnosticeerd werden met
ze meerdere malen per dag moeten spuiten (of toedienen via een insulinepomp). Ook
DT1, als het gaat om glycemische controle (HbA1c), zelfmanagement, psychologisch functioneren
dienen ze hun bloedglucose een aantal keer per dag te meten, hun koolhydraatinname bij
en diabetes-gerelateerde kwaliteit van leven (KvL)? 2) Wat is de prevalentie eetproblematiek in
te houden en hun lichamelijke activiteit te monitoren om zo de kans op complicaties te
deze groep, wanneer ontstaat het, welke factoren zijn daarmee geassocieerd en hoe beïnvloedt
verkleinen. Dit vergt dagelijks veel inzet, waardoor de kans op psychosociale problematiek,
deze problematiek de diabetes- en psychosociale uitkomsten? 3) Wat zijn de verschillen tussen
zoals een verminderd welbevinden en eetproblematiek, toeneemt. Aangezien de diagnose
de deelnemers die vóór hun puberteit gediagnosticeerd werden met DT1 en hen die tijdens
DT1 meestal in de kindertijd of adolescentie gesteld wordt, spelen ouders een grote rol in
hun puberteit gediagnosticeerd werden met DT1, als het gaat om het gezinsfunctioneren en het
het dagelijkse management. Om meer zicht te krijgen op de verschillende risicofactoren en
ouderlijk welbevinden? 4) Relateert de cognitieve ontwikkeling van kinderen en adolescenten
protectieve factoren die een rol spelen bij het management van DT1 en zo invloed hebben
met T1D aan de glycemische controle en het diabetesmanagement en hoe?
op diabetes en psychosociale uitkomsten, is de psychosociale ontwikkeling van kinderen en
adolescenten in de leeftijd van acht tot achttien met DT1 onderzocht.
Hoofdstuk 3: Verminderd zelfvertrouwen en diabetes mismanagement bij adolescenten met
diabetes type 1 medieert de relatie tussen probleemgedrag en verhoogd HbA1c
NL
In hoofdstuk 3 is de relatie tussen probleemgedrag (totaal, intern en extern) en glycemische
controle (HbA1c) beschreven. Onderzocht is of deze relatie gemedieerd wordt door de mate
zelfvertrouwen in de diabetes en mismanagement van de diabetes. In deze sub-studie namen
110
111
NL
Samenvatting
Samenvatting
88 adolescenten tussen de elf tot en met vijftien jaar deel. Data van het DINO-onderzoek
en professionals onderzocht. De overeenkomsten blijken in beide gevallen niet optimaal
zijn gebruikt: probleemgedrag is gemeten met de Strengths and Difficulties Questionnaire,
(beperkte overeenkomst tussen ouders en adolescenten, en redelijke overeenkomst tussen
mediërende variabelen zijn gemeten met de Confidence In Diabetes Self-care-Youth en de
professionals en adolescenten). De helft van de adolescenten gaf aan een risico te lopen op
Diabetes Mismanagement Questionnaire, en HbA1c-waardes zijn uit de medische dossiers
eet- en gewichtsproblematiek (psychologische ‘gele vlaggen’). Echter, minder dan één van
gehaald. Om de data te analyseren, is gebruikt gemaakt van bootstrap mediation analysis
de tien adolescenten rapporteerden daadwerkelijk eetproblemen. Klinische ‘gele vlaggen’
for multiple mediation. Uit de resultaten blijkt dat zowel totaal probleemgedrag als extern
zijn een verhoogd HbA1c, hogere leeftijd, op dieet zijn geweest, verlaagde KvL, verminderd
probleemgedrag geassocieerd zijn met een slechter HbA1c, en dat deze relatie wordt
zelfvertrouwen in de diabetes, diabetes mismanagement en ontevredenheid met het eigen
gemedieerd door minder zelfvertrouwen in de diabetes en door mismanagement van de
lichaam. In tegenstelling tot de verwachtingen blijken Body Mass Index (BMI) waardes en sekse
diabetes. Een directe relatie tussen intern probleemgedrag en HbA1c is niet gevonden, maar
niet geassocieerd met eetproblematiek en zijn dus geen klinische ‘gele vlaggen’. De conclusie
intern probleemgedrag relateerde wel aan minder zelfvertrouwen en meer mismanagement,
van dit hoofdstuk is dat het onderwerp eetproblematiek en de geassocieerde ‘gele vlaggen’
die wel associeerden met verhoogd HbA1c. De conclusie van dit hoofdstuk is dat screenen op
aandacht verdienen in de diabeteszorg om zo het ontstaan of de verergering van eetproblemen
probleemgedrag en mismanagement, en het verhogen van zelfvertrouwen in de diabetes een
te voorkomen. Aangezien zowel ouders als diabetesprofessionals eetproblemen niet adequaat
positieve invloed kan hebben op de diabetesuitkomsten.
lijken te signaleren, is het aan te bevelen op deze problematiek te screenen. Een stapsgewijze
procedure zoals gebruikt in dit onderzoek zou ook in de praktijk ingezet kunnen worden.
Hoofdstuk 4: Eetproblemen bij adolescenten met diabetes type 1. Op welke ‘gele vlaggen’
NL
moeten we alert zijn?
Hoofdstuk 5: Niet het ouderlijk welbevinden maar opvoedgedrag en opvoedstress zijn
In hoofdstuk 4 zijn de prevalentie van eetproblemen en de geassocieerde ‘gele vlaggen’
gerelateerd aan de glycemische uitkomsten van kinderen en adolescenten met diabetes type 1
beschreven. Daarnaast is de mate van overeenkomst over het bestaan van eetproblematiek
Longitudinale data van de DINO-studie
tussen adolescenten en hun ouders, en adolescenten en diabetesprofessionals vermeld. Om
In hoofdstuk 5 is de longitudinale relatie tussen ouderlijk emotioneel welbevinden en de
dit te onderzoeken zijn data van het DINO-onderzoek gebruikt. 103 adolescenten van elf tot
glycemische uitkomsten van kinderen en adolescenten met DT1 beschreven. Daarnaast is
en met zestien jaar en hun ouders hebben vragenlijsten ingevuld waarin een stapsgewijze
de impact onderzocht van probleemgedrag, ouderlijk diabetes-opvoedgedrag en ouderlijke
procedure is gevolgd: alleen de adolescenten die op een KvL-vragenlijst (MY-Q) aangaven
diabetesstress op deze relatie. In deze sub-studie hebben 174 ouders van kinderen en
een risico te lopen op eet- of gewichtsproblematiek (psychologische ‘gele vlaggen’), vulden
adolescenten met DT1 participerend aan de DINO-studie vragenlijsten ingevuld op drie
de Diabetes Eating Problems Scale - Revised en vragen van de AHEAD studie in. Door gebruik
verschillende momenten, elk met een interval van een jaar. Met behulp van Generalized
te maken van Kruskal-Wallis tests, post hoc Mann-Whitney U tests en chi-squared tests zijn
Estimating Equations analyses is de relatie tussen ouderlijk emotioneel welbevinden (WHO
klinische ‘gele vlaggen’ in kaart gebracht. De ouders en diabetesprofessionals zijn gevraagd
Five Well-being Index scores) en het HbA1c onderzocht. Vervolgens zijn de relaties tussen
naar hun observaties over de aan- of afwezigheid van mogelijke eetproblematiek. Met Cohen’s
het ouderlijk welbevinden, probleemgedrag van kinderen en adolescenten (Strength and
kappa is de mate van overeenstemming tussen adolescenten en ouders, en tussen adolescenten
Difficulties Questionnaire), ouderlijk diabetes-opvoedgedrag (Diabetes Family Behavior
112
113
NL
Samenvatting
Samenvatting
Checklist: positieve en negatieve schalen), ouderlijke diabetesstress (Problem Areas In
Het blijkt voor de teams echter lastig het programma in te bedden in de routinezorg: twee van de
Diabetes-Parent revised) en het HbA1c geanalyseerd. Een derde van de ouders rapporteerde
tien teams (20%) is dit gelukt. De bijkomende logistiek vormt een barrière en daarom wordt de MY-Q
verminderd emotioneel welbevinden. Dit blijkt gerelateerd aan meer probleemgedrag bij
in de meeste ziekenhuizen niet structureel afgenomen. Toch achten de teams, de adolescenten
kinderen en adolescenten, meer negatief opvoedgedrag en meer ouderlijke diabetesstress,
en de ouders het waardevol om het gebruik van het programma te continueren en verder
en deze factoren zijn geassocieerd met slechtere glycemische controle. Negatief opvoedgedrag
te implementeren. Om dit te realiseren zal er geïnvesteerd moeten worden in het oplossen van
en ouderlijke diabetesstress zijn gerelateerd aan meer probleemgedrag van kinderen en
de logistieke barrières.
adolescenten. In tegenstelling tot onze verwachtingen is er geen directe relatie gevonden
tussen verminderd ouderlijk welbevinden en slechtere glycemische uitkomsten. Positief
Hoofdstuk 7: Algemene discussie
opvoedgedrag is geassocieerd met betere glycemische controle. Op basis van deze resultaten
In dit afsluitende hoofdstuk worden de belangrijkste bevindingen en de implicaties voor
wordt geconcludeerd dat gezinsinterventies gericht op het verbeteren van het HbA1c van kinderen
de klinische praktijk beschreven. Het doel was meer zicht te krijgen op de verschillende
en adolescenten, een grotere kans van slagen hebben als ze eveneens interveniëren op negatief
risicofactoren en protectieve factoren die een rol spelen bij het management van DT1. In dit
opvoedgedrag, ouderlijke stress en probleemgedrag, en juist positief opvoedgedrag stimuleren.
proefschrift zijn diverse factoren geïdentificeerd die interacteren met diabetesuitkomsten.
Probleemgedrag, eetproblematiek, ouderlijke stress en negatief opvoedgedrag associëren met
Hoofdstuk 6: De implementatie van het monitoren van de kwaliteit van leven in de
slechtere glycemische controle bij kinderen en adolescenten met DT1. Positief opvoedgedrag
Nederlandse pediatrische diabeteszorg: waardevol maar complex
relateert aan een beter HbA1c. Aangezien de meeste van deze psychosociale aspecten ook
Hoofdstuk 6 beschrijft de evaluatie van de implementatie van het DAWN MIND-Youth programma:
elkaar beïnvloeden, is het bestaan van multivariabele relaties zeer aannemelijk. Echter, per kind
het meten en bespreken van de KvL van adolescenten van tien jaar en ouder met DT1.
en adolescent verschilt de optelsom van risicofactoren en de balans met protectieve factoren.
Ten behoeve van dit programma vullen adolescenten jaarlijks de MIND Youth Questionnaire in
Professionals binnen het kinderdiabetesteam kunnen een belangrijke protectieve factor
(MY-Q): een diabetes-specifieke KvL-vragenlijst waarin verschillende levensdomeinen aan bod
zijn wanneer zij naast de diabetesuitkomsten ook aandacht besteden aan de psychosociale
komen. De uitkomst wordt met hen besproken. Facultatief kan het ouderlijk welbevinden en
aspecten. Het blijkt echter een uitdaging om op structurele wijze aandacht te besteden aan de
de ouderlijke diabetesstress worden gemeten en besproken. Elf ziekenhuizen in Nederland
KvL van kinderen en adolescenten met DT1. Suggesties over de wijze waarop de conclusies van
zijn in 2011 met het DAWN MIND-Youth programma gestart. Om de implementatie ervan te
dit proefschrift kunnen bijdragen aan de verbetering van de pediatrische diabeteszorg, als ook
evalueren, hebben 36 diabetes professionals, 29 adolescenten en 66 ouders een online vragenlijst
de beperkingen van dit onderzoek staan beschreven in hoofdstuk 7.
ingevuld over hun ervaring met het programma. Met tien kinderdiabetesteams hebben
NL
semi-gestructureerde interviews plaats gevonden die zijn geanalyseerd met template analysis.
NL
Het blijkt dat de teams het DAWN MIND-Youth programma beschouwden als een handig hulpmiddel
om meer te weten te komen over de verschillende levensdomeinen van de adolescenten.
Door deze screenende functie kunnen problemen in een vroeg stadium gesignaleerd worden.
114
115
A ppendixes
1.
Study measures DINO study (Chapter 2)
2.
Evaluation DAWN MIND-Youth (Chapter 6)
A.
Diabetes care teams’ survey and answers
B.
Adolescents’ survey and answers
C.
Parents’ survey and answers
117
A
Appendixes
Appendixes
Appendix 1: Study measures DINO study
Emotional and behavioral functioning: The Strengths and Difficulties Questionnaire (SDQ)
Biological development
[96, 130] is used to measure emotional and behavioral functioning and contains 25 items, rated
To assess biological development, in addition to information provided by the pediatric diabetes
on a 3-point Likert scale. The SDQ comprises five scales: Emotional symptoms, Conduct problems,
team self-report measures are used. The biological development is assessed both with respect
Hyperactivity/inattention, Peer relationship problems and Pro-social behavior. Cronbach’s α for
to socio-demographic and clinical data.
parents is 0.81, for self-report 0.70 [96, 130]. Both adolescents and parents complete the SDQ.
Socio-demographic data cover: date of birth and gender are documented at baseline by
Self-esteem: The KINDL-R self-esteem subscale (4 items on a 5 point Likert scale) assesses
the diabetes teams. Ethnicity, family structure, education (school level) and family-related life
overall feelings of self-worth and self-acceptance [97]. Cronbach’s α for the total KINDL-R
events are evaluated every year by the parents. The latter refers to either positive or negative
is 0.82 [240].
life events that can impact the adolescent and his/her family (e.g. birth, marriage, divorce,
Autonomy: The Autonomy subscale of the KIDSCREEN asks about generic autonomy with
moving house, death of a relative etc.). By classifying the highest level of education in a
5 items on a 5 point Likert scale. Cronbach’s α for this subscale is 0.84 [98, 99].
household, social economic status is determined.
Deviation responsibility diabetes tasks: Diabetes Family Responsibility Questionnaire
Clinical data (annually documented by the diabetes care team) include: history of medical
(DFRQ) (17 items) is used to examine the transfer of responsibility for diabetes management
and psychological (contact with mental health services) co-morbidity; treatment regimen
tasks from parents to youth [100]. On each item is asked who initiates the responsibility:
(number of injections/pump, insulin units); care consumption (referrals, nr. of contacts with
the parent, both parent and child equally or the child. Both adolescents and parents complete
pediatrician, nurse, psychologist etc.); Tanner stage (pubertal stages will be determined by
the DFRQ. Internal consistency has been reported and considered acceptable [100]
visual inspection, using the criteria and definitions described by Tanner [41] and testes volume
Quality of Life: Diabetes related QoL and Diabetes Self-management are assessed by the
of boys will be determined by palpation); blood pressure; weight; height; glycemic control
MIND Youth Questionnaire (MY-Q) [101], in which the following domains are covered: social
(determined by means of glycosylated hemoglobin (HbA1c)); number of diabetes-related
impact (friends, school, free time), parents, responsibility, body image and eating behavior,
hospitalizations; diabetes ketoacidosis (DKA); indicators for complications (micro-albumin
control perceptions, worries, treatment satisfaction and emotional well-being. Cronbach’s α
excretion rate, lipid profile and retina screening). Although we do not measure the quality
of the MY-Q is 0.8 [101]
of care, we gain insight in the care consumption as an indicator for what is optimal care. Since
Self-efficacy: an adapted version of the Confidence in Diabetes Self-Care Scale (CIDS)
all assessments are scheduled in routine care, no extra assessments need to be scheduled.
suitable for adolescents is used to measure diabetes specific self-efficacy [102]. The original
The presence or absence of severe hypoglycemic episodes (defined as needing more assistance
adult version of the CIDS consists of 20 items on a 5 point Likert scale and Cronbach’s α=0.86
from others than usual) are evaluated every year by the parents.
[102]. This adapted youth version consists of 12 items: 10 of the original, 2 items combined to
1 (original question 2 and 20) and 1 additional item regarding alternations in blood glucose.
A
Study measures – Psychosocial development
Management in diabetes self-care: This is assessed by using a renewed version of the
Psychosocial functioning of youth 11 years and older is assessed annually by using 7 questionnaires
diabetes mismanagement scale [103]. The original version consists of 10 items, 3 items were
combined in one online survey.
used and 1 was adjusted, recall is changed from 10 days into 7 days. Answers are given on a
118
119
A
Appendixes
A
Appendixes
5 point Likert scale. Cronbach’s α of the original version is 0.74 and 0.60, respectively [103].
Study measures – Cognitive development
Adherence: Diabetes teams are annually asked about their opinion with regard to
Cognitive development is assessed annually using a short form of the Wechsler Intelligence
adolescent’s overall diabetes adherence and diabetes instructions.
Scale for Children-III (WISC-III) [104], three computerized tasks, and an executive functioning
Psychological functioning of children 8-10 is assessed by a selection of these questionnaires
questionnaire completed by parents. Ideally assessment takes place prior or following the
(KINDL, adapted (shorter and easier to read) version of the MY-Q and KIDSCREEN), as not
adolescent’s regular assessment with the diabetes team in the diabetes clinic. Prior, during and
all are validated for younger children.
after cognitive assessment, blood glucose values are measured. If the blood glucose level is
Disturbed eating behavior: DEB is assessed in adolescents 11 years and older in a step-wise
found below or above the adolescent’s personal average (ideally between 5-10) and participant’s
manner in order to minimize the burden in adolescents with no DEB and younger participants.
concentration is therefore impaired, examination is postponed until blood glucose is considered
Diabetes specific measures are used as it has been shown that generic measures tend to
normal. In statistical analyses we will correct for extreme low or high blood glucose values. Tests
inflate estimates of eating problems in a population with T1D [58, 65].
were administered in fixed order by trained examiners using standardized instructions.
Step 1: The MY-Q subscale Body image & eating behavior serves as a start asking general
Full scale IQ will be estimated using a well-established short version of the Wechsler
questions about trying to control weight, body image, eating binges and skipping insulin.
Intelligence Scale for Children (WISC-III), using five subtests: Information; Picture arrangement;
In addition, two questions about dieting status and dieting frequency are asked [60]. Step
Arithmetic; Block design; Digit Span [104, 105, 241].
2: If any of these items are endorsed, the next step is to assess diabetes specific DEB and
With respect to the computerized tasks, an adapted version of the Posner’s Attention
weight control behavior by using the Diabetes Eating Problems Scale-Revised (DEPS-R) [61,
Network Task (ANT) [106, 107] suitable for children is used to assess orienting, alerting and
68] and questions of the AHEAD study [60]. DEPS-R is specifically designed to measure with
executive attention, as well as speed and consistency in speed of information processing. The
16 items on a 6 point Likert scale eating problems in the context of diabetes, including
Eriksen Flanker Task [108, 109] is used to assess interference control. An adapted version of
insulin restriction. Cronbach’s α = 0.86 [61, 68]. Adolescents can endorse on AHEAD weight
the Klingberg task [107, 110, 111] is used to measure visuo-spatial working memory.
control items whether certain behavior (e.g. ate less high-fat foods) occurred in the past
Executive functioning is additionally measured using the Behavior Rating Inventory of
3 months [60]. Step 3: Additionally, adolescents enrolled in step 2 are invited to participate
Executive Functioning questionnaire (BRIEF) [112, 113], an 86 item 3-point Likert scale parent
in interviews. Our aim is to gain better insight in initiation of DEB, pro’s and con’s, insulin
report questionnaire. The BRIEF allow calculation of two indices: the Behavioral Regulation index
restriction (maladaptive, treatment (un-)related), course of DEB and receptiveness to help.
(subdomains Inhibition, Shift and Emotional Control) and Metacognition index (subdomains
In order to assess whether parents think their child shows disturbed eating behavior, the
Initiate, Working Memory, Plan/Organize, Organization of materials, and Monitor) [113].
parental online survey includes two adjusted questions from the MY-Q regarding weight and
Performance on the WISC-III and BRIEF will be compared to normative data. Computerized
looks (Do you have the impression your child is happy with the way he/she looks? Do you have
measures will be administered to a gender and age matched sample of 100 healthy controls as
the impression your child’s tries to control her weight by eating less or omitting insulin?).
there is currently no Dutch normative sample is available for these tasks. Healthy controls are
120
121
A
Appendixes
derived from primary and secondary schools in the Netherlands and will complete identical
Appendixes
Appendix 2A: Diabetes care teams’ survey and answers
cognitive tests as the diabetes group (including the subtasks of the WISC III) and parents will
complete the BRIEF. Controls will be measured cross-sectional, at baseline.
Study measures – Parental assessment
Online survey: N = 36
In order to evaluate DM-Y we would like to invite you to share your experiences. Since your
answers contribute to the optimization of the program and to the further implementation
into more hospitals, your help is of great importance. Completion takes approx. 15 minutes.
If necessary you can close the questionnaire and continue later at all times. Good luck!
To measure family functioning and parental well-being, parents annually complete an online
survey in which they are asked about their child’s development and their own well-being and
parenting style. The latter is assessed by using the following measures.
Parental distress: Problem Areas In Diabetes-Parents Revised (PAID-PR) assesses diabetes
related distress in parents of children with diabetes in 20 items, scored on a 5 point Likert scale.
Cronbach’s α=0.87 [114, 115].
Parental well-being: WHO-Five Well-being Index (WHO-5) captures emotional well-being with
five positively worded items, scored on a 6 point Likert scale. Cronbach’s α=0.91 [116-118].
Diabetes specific parenting style: The Diabetes Family Behavior Checklist (DFBC) [119]
assesses the amount of supportive and non-supportive diabetes specific parenting behavior.
DFBC contains 16 items on a 5 point Likert scale, of which nine regarding positive parenting
behaviors and seven negative. Cronbach’s α for the positive items is 0.73, for the negative 0.43 [119].
Parents decide themselves whether the mother or the father complete the online survey.
We expect that the majority will be mothers.
A
Demographics
1. Your function
o
o
o
o
o
o
o
o
o
Diabetes nurse 36.1%
Pediatrician 44.4%
Psychologist 16.7%
Pedagogue 0%
Dietician 2.8%
Assistant 0%
Pedagogical worker 0%
Social worker 0%
Other 0%
2. In which hospital do you work?
1 19.4%
2 8.3%
3 8.3%
4 8.3%
5 8.3%
6 13.9%
7 11.1%
8 8.3%
9 0%
10 11.1%
11 2.8%
3. Gender
o Male 19.4%
o Female 80.6%
4. What is your age?
o Younger than 30 0%
o 30−39 22.2%
o 40−49 44.4%
122
123
A
Appendixes
o 50−60 30.6%
o Older than 60 2.8%
5. How many members does your team consist (including yourself)? M=8 ±2
6. Which occupation does your team consists of? Multiple options possible.
o Diabetes nurse 100%
o Pediatrician 100%
o Psychologist 94.4%
o Pedagogue 11.1%
o Dietician 100%
o Assistant 50%
o Pedagogical worker 11.1%
o Social worker 44.4%
o Other 0%
7. How many children with type 1 diabetes does your team treat?
o 0−25 0%
o 25−50 8.3%
o 50−100 44.4%
o 100−200 36.1%
o over 200 11.1%
o Don’t know 0%
8. How many 10 years and older?
o Less than 25% 0%
o 25% − 50% 2.8%
o 50% − 75% 52.8%
o More than 75% 25.0%
o Don’t know 19.4%
o
o
o
o
o
o
Working with DM-Y takes too much of my time.
D 36.1%, N 38.9%, A 22.2%, NA 2.8%
DM-Y is no addition to the routine care.
D 91.7%, N 5.6%, A 2.8%, NA 0%
DM-Y contributes to the screening for psychosocial problems.
D 0%, N 5.6%, A 94.4%, NA 0%
It appeals to me that the MY-Q can be completed on a computer.
D 11.1%, N 5.6%, A 80.6%, NA 2.8%
It appeals to me that more Dutch hospitals use DM-Y.
D 0%, N 13.9%, A 86.1%, NA 0%
Working with DM-Y gives me satisfies me in my job.
D 19.4%, N 41.7%, A 38.9%, NA 0%
10. Whom consults with the adolescents regarding the MY-Q outcome?
Multiple options possible.
o Diabetes nurse 75.0%
o Pediatrician 88.9%
o Psychologist 13.9%
o Pedagogue 2.8%
o Dietician 0%
o Assistant 0%
o Pedagogical worker 0%
o Social worker 0%
o Other 5.6%
o Don’t know 0%
11. Does your team use other questionnaires besides the MY-Q?
o Yes, purpose: 36.1% (mainly for the transition to adult care)
o No> question 13. 58.3%
o Don’t know > question 13. 5.6%
9. Kindly indicate the degree to which you agree on the following statements
regarding you experiences with DM-Y.
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o The purpose of DM-Y is clear to me.
D 0%, N 0%, A 100%, NA 0%
o I am satisfied with DM-Y.
D 5.6%. N 25.0%, A 69.4%. NA 0%
o I understand how to use DM-Y.
D 2.8%, N 16.7%, A 80.6%, NA 0%
o I want to continue the usage of DM-Y.
D 0%, N 25.0%, A 75.0%, NA 0%
A
Appendixes
124
12. Kindly indicate the degree to which you agree on. (N=15)
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o The usage of other questionnaires hindrances the usage of DM-Y.
D 66.7%, N 13.3%, A 6.7%, NA 13.3%
13. The next statements concern the adolescents who completed the MY-Q. Kindly indicate the degree to which you agree on.
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o Adolescents are satisfied with DM-Y.
0%, N 58.3%, A 30.6%, NA 11.1%
125
A
Appendixes
o
o
o
o
o
o
o
o
o
o
DM-Y interferes with the regular medical care for adolescents.
D 75.0%, N 19.4%, A 0%, NA 5.6%
DM-Y helps to talk about the quality of life of adolescents in more detail.
D 0%, N 5.6%, A 86.1%, NA 8.3%
I’m satisfied with the way the MY-Q outcomes are consulted
with the adolescents.
D 5.6%, N 33.3%, A 52.8%, NA 8.3%
I feel capable to talk with adolescents about their quality of life.
D 0%, N 33.3%, A 58.3%, NA 8.3%
When I notice problems, I quickly offer a solution.
D 16.7%, N 47.2%, A 27.8%, NA 8.3%
While consulting the outcomes with the adolescents,
I only discuss topics marked by a flag.
D 30.6%, N 11.1%, A 41.7%, NA 16.7%
The conversation based on the MY-Q outcomes sometimes
does not reconcile with my prior expectation.
D 19.4%, N 8.3%, A 58.3%, NA 13.9%
Discussing the MY-Q outcomes contributes to the adolescents’ health.
D 2.8%, N 27.8%, A 66.7%, NA 2.8%
The adolescents give politic correct answers to the MY-Q.
D 16.7%, N 36.1%, A 38.9%, NA 8.3%
Parents think DM-Y is beneficial to their children.
D 0%, N 50.0%, A 36.1%, NA 13.9%
14. Do the adolescents experience troubles while completing the MY-Q?
Multiple options possible.
o No. 0%
o Yes, the reason why is unclear to them. 2.8%
o Yes, the MY-Q is too difficult (language/grammar). 30.6%
o Yes, computer related problems. 41.7%
o Yes, lack of room / space available. 25.0%
o Yes, the completion takes too much time. 16.7%
o Yes, the addressed domains do not suit the adolescents’ lives. 5.6%
o es, other. 11.1%
o Not applicable. 13.9%
15. Did the MY-Q outcomes led to more referrals?
o Yes. 19.4%
o No, equal >question 17. 58.3%
o No, less >question 17. 0%
o Don’t know >question 17. 22.2%
A
126
Appendixes
16. To whom was referred to? – Open ended (N=7)
Psychologist / social work.
17. Did the parents complete the parental-part of DM-Y?
o Yes. 58.3%
o No>question 22. 30.6%
o Don’t know> question 22. 11.1%
18. Do the parents experience problems while completing the parental questionnaire? (N=21)
o No. 0%
o Yes, the reason why is unclear to them.9.5%
o Yes, the MY-Q is too difficult (language/grammar). 14.3%
o Yes, computer related problems. 33.3%
o Yes, lack of room / space available. 14.3%
o Yes, the completion takes too much time. 9.5%
o Yes, the domains do not suit the parents’ lives. 0%
o Yes, other. 33.3%
o Not applicable 9.5%
19. Kindly indicate the degree to which you agree on the following statements. (N=21)
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o The parents are satisfied with the parental-part of DM-Y.
D 0%, N 61.9%, A 23.8%, NA 14.3%
o The parents give politic correct answers.
D 42.9%, N 33.3%, A 9.5%, NA 14.3%
o Parents’ answers gave more insight into the child’s life.
D 4.8%, N 28.6%, A 47.6%, NA 19.0%
o The completion of the parental-part is beneficial.
D 4.8%, N 19.0%, A 66.7%, NA 9.5%
20. Are the outcomes of the parental-part of DM-Y discussed with the parents. (N=21)
o No> question 22. 19.0%
o Yes, almost always. 33.3%
o Yes, if the answers required a consultation. 28.6%
o Don’t know> question 22. 19.0%
21. Kindly indicate the degree to which you agree on the following statements. (N=13)
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o I am satisfied with the way outcomes are discussed with parents.
D 11.8%, N 17.6%, A 35.3%, NA 35.3%
o I feel capable to discuss the outcomes with parents.
D 5.9%, N 29.4%, A 47.1%, NA 17.6%
127
A
Appendixes
The next questions concern the training and the implementation of DM-Y.
22. Did you attend the DM-Y training?
o Yes. 52.8%
o No> question 24. 47.2%
23. Kindly indicate the degree to which you agree on the following statements. (N=19)
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o I have learned a lot from this training > question 25.
D 0%, N 36.8%, A 63.2%, NA 0%
24. You did not attend the training. How were you prepared for the usage of DM-Y? (N=17)
Mainly through colleagues who did attend the training.
25. After how many months after the training (attend by yourself or colleague) was the first MY-Q completed by an adolescent?
o 1−3 months. 52.8%
o 3−6 months. 25.0%
o 6−9 months. 8.3%
o 9−12 months. 8.3%
o Over 12 months. 5.6%
26. Kindly indicate the degree to which you agree on the following statements.
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o I am satisfied with the DM-Y manual.
D 8.3%, N 36.1%, A 47.2%, NA 8.3%
o My colleagues were motivated to start with DM-Y.
D 0%, N 11.1%, A 88.9%, NA 0%
o The coherence within my team regarding DM-Y is upright.
D 22.2%, N 16.7%, A 61.1%, NA 0%
27. Are the MY-Q outcomes discussed in a team meeting?
o Yes. 69.4%
o No> question 29. 30.6%
28. Which outcomes are discussed in these meetings?
o All patients who completed MY-Q. 44.0%
o Only problematic outcomes. 52.0%
o Other. 4.0%
A
29. Would you like to have an extra DM-Y meeting/training?
o Yes, annually. 41.7%
128
Appendixes
o
o
o
o
Yes, every six months. 0%
No, the current proceeding is sufficient. 33.3% No, I use DM-Y in a way that suits me better. 5.6%
Not applicable. 19.4%
30. Have you contacted e.g. school or sport club on the occasion of the MY-Q outcome
or consultation?
o No. 58.3%
o Yes, more often than without the usage of DM-Y. 0%
o Yes, as often as without the usage of DM-Y. 19.4 %
o Yes, less often than without DM-Y. 0%
o Not applicable. 22.2%
31. Within your team, is there a member responsible for DM-Y?
o No. 11.1%
o Yes, diabetes nurse. 50.0%
o Yes, pediatrician. 22.2%
o Yes, psychologist. 5.6%
o Yes, dietician. 0%
o Yes, assistant. 5.6%
o Ja, Pedagogical worker. 0%
o Yes, social worker. 0%
o Yes, other. 5.6%
o Not applicable. 0%
32. In the (recent) past, did you participate in a communication training?
o No. 0%
o Yes, as part of my education. 47.2%
o Yes, I followed a course. 55.6%
o Not applicable. 0%
33. Kindly indicate the degree to which you agree on the following statements.
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o Difficulties within the management (such as the lack of financial compensation or protocols) hindrance DM-Y.
D 50.0%, N 25.0%, A 25.0%, NA 0%
o My team contacted/made arrangements with health care insurances regarding DM-Y.
D 55.6%, N 19.4%, A 0%, NA 25.0%
o DM-Y fits the vision and policy of our hospital.
D 5.6%, N 25.0%, A 69.4%, NA 0%
129
A
Appendixes
o
o
o
o
o
We inform adolescents and/or parents with regards
to the development of DM-Y.
D 36.1%, N 36.1%, A 16.7%, NA 11.1%
Additional information (e.g. newsletter or phone contact) with regards to DM-Y stimulates the implementation and embedding of the program.
D 5.6%, N 27.8%, A 61.1%, NA 5.6%.
The annual completion of the MY-Q is too little.
D 86.1%, N 13.9%, A 0%, NA 0%
It is feasible to have a team member responsible for DM-Y.
D 2.8%, N 11.1%, A 86.1%, NA 0%
I would recommend DM-Y to a co-specialist.
D 0%, N 25.0%, A 75.0%, NA 0%
Thank you for your time and effort!
Appendixes
Appendix 2B: Adolescents’ survey and answers
Online survey: N=29
Thank you for your help! The questions below regard the questionnaire about your quality of life,
it’s called the MY-Q. You completed the MY-Q on the computer in your hospital. Maybe this was
a while ago. If it is difficult to answer a question because you do not remember, you can press
‘Don’t know’. Nobody except the researcher finds out which answers are yours! Good luck!
1. The statements below concern the MY-Q (the questionnaire you completed in the hospital
regarding your quality of life). Kindly indicate the degree to which you agree on.
Disagree (D), Neither agree nor disagree (N), Agree (A), Don’t know (DK)
o I found it beneficial to complete the MY-Q.
D 10.3%, N 41.4%, A 41.1%, DK 6.9%
o Completing the MY-Q was a waste of my time.
D 86.2%, N 13.8%, A 0%, DK 0%
o It is a good idea if more adolescents with type 1 diabetes
would complete the MY-Q.
D 3.4%, N 27.6%, A 62.1%, DK 6.9%
2. Is there something which made it difficult to complete the MY-Q?
Multiple answers possible.
o No. 0%
o Yes, I did not understand why I had to. 6.9%
o Yes, the MY-Q was too difficult (language/grammar). 3.4%
o Yes, the computer did not work sufficient. 3.4%
o Yes, there was no room / space available. 3.4%
o Yes, the completion took too much time. 3.4%
o Yes, the domains did not suit my life. 13.8%
o Yes, other. 6.9%
o Don’t know. 6.9%
3. Was there something missing in the MY-Q? Which topics were unaddressed while they seemed important to you?- Open ended
Family 6.9%, worries regarding diabetes 3.4%, sport 3.4%
4. The following statements concern the conversation after you completed the MY-Q.
You’ve talked with your doctor or nurse about some of your answers.
Disagree (D), Neither agree nor disagree (N), Agree (A), Not applicable (NA)
o I’d liked talking about my answers.
D 11.1%, N 51.9%, A 25.9%. DK 11.1%
o I was not listened to.
D 85.2%, N 7.4%, A 0%, DK 7.4%.
A
130
131
A
Appendixes
o
o
My doctor or nurse now knows better how I’m doing.
D 14.8%, N 44.4%, A 18.5%, DK 22.2%.
I discussed certain topics I normally wouldn’t talk about.
D 29.6%, N 37.0%, A 22.2%, DK 11.1%
5. What did you appreciated in the subsequent conversation? -Open ended.
Don’t know. 20.7%, No conversation. 6.9%, Understanding. 34.5%, Addressed topics. 6.9%, Other. 13.8%
6. What did you find disappointing in the subsequent conversation?
Of course the person whom talked to you will not read this! -Open ended.
Don’t know 17.2, No conversation 10.3%, Difficult to talk about 6.9%, Nothing 27.6%,No understanding 10.3%, Other 13.8%
7. Did you talk with your parents about the MY-Q?
o Yes. 26.9%
o No. 53.8%
o Don’t know. 19.2%
8. If so, were did you talk about? –Open ended
Content 20.7%, Don’t remember 6.9%
9. How many times in the past 2 years did you talk with a psychologist about the way you feel?
o 0 times > question 11. 65.4%
o 1−2 times. 11.5%
o Over 2 times. 23.1%
o Don’t know. > question 11. 0%
10. Can you remember when this happened? –Open ended (N=8)
Answers range from past week to past year.
11. We would like to ask a few youngsters if they have any suggestions with regards
to the MY-Q and the subsequent conversation. Could we contact you about this?
o Yes, by phone. 7.7%
o Yes, by e-mail. 57.7%
o No. 34.6%
12. Do you have any suggestions or remarks concerning the MY-Q or the subsequent conversation? (N=2) In my case, the MY-Q questions were not suitable.
Thank you very much for your help! Your answers help us improve the program!
A
132
Appendixes
Appendix 2C: Parents’ survey and answers
Online survey N=66
Thank you for participating in this survey. This survey regards the computerized-questionnaire
filled in by your child in the hospital regarding their quality of life (it’s called MY-Q). Maybe this
happened a while ago. If it is difficult to answer a question because you do not remember, you
can press ‘Don’t know’. Good luck!
1. Gender
o Male 14.3%
o Female 85.7%
2. Kindly indicate the degree to which you agree on the following statements regarding
the MY-Q (quality of life questionnaire completed by your child).
Disagree (D), Neither agree nor disagree (N), Agree (A), Don’t know (DK)
o My child thought completing the MY-Q was beneficial.
D 12.1%, N 48.5%, A 22.7%, DK 0%
o I think the completion of the MY-Q was beneficial for my child.
D 0%, N 18.2%, A 78.8%, DK 3.0%
o It is a good idea if more adolescents with type 1 diabetes would complete the
MY-Q. D 0%, N 15.2%, A 78.8%, DK 6.1%
o Completing the MY-Q was a waste of time.
D 80.3%, N 12.1%, A 1.5%, DK 6.1%
3. Did you complete the parental- questionnaire in the hospital? This questionnaire regarded your state of mind and whether certain aspects of the diabetes caused problems.
o Yes > question 5. 61.5%
o No. 15.4%
o Don’t know. 23.1%
4. Would you have liked to complete the parental-questionnaire? (N=10)
o Yes > question 10. 40.0%
o No > question 10. 20.0%
o Don’t know > question 10. 40.0%
5. Kindly indicate the degree to which you agree on the following statements regarding the parental- questionnaire you completed in the hospital. (N=40)
Disagree (D), Neither agree nor disagree (N), Agree (A), Don’t know (DK).
o I found it beneficial to complete the parental-questionnaire.
D 5.0%, N 7.5%, A 87.5%, DK 0%
o The completion was a valued addition to the pediatric diabetes care.
D 2.5%, N 15.0%, A 80.0% DK 2.5%
133
A
Appendixes
o It is a good idea if the parental questionnaire was introduced to more parents.
D 2.5%, N 5.0%, A 90.0%, DK 2.5%
o The diabetes team of my child now knows more about me than I’d like to.
D 62.5%, N 25.0%, A 5.0%, DK 7.5%
6. Is there something which made it difficult to complete the parental-questionnaire? Multiple answers possible. (N=40)
o No. 92.3%
o Yes, I did not understand why I had to. 0%
o Yes, the parental-questionnaire was too difficult (language/grammar). 0%
o Yes, the computer did not work sufficient.0%
o Yes, there was no room / space available. 0%
o Yes, the completion took too much time. 0%
o Yes, the domains did not suit my life. 2.6%
o Yes, other. 7.7%
Appendixes
11. If so, could you tell us something about that? Did you discussed things you usually wouldn’t talk about? – Open ended (N=12)
Mainly nothing surprising / nice topic to talk about.
12. Maybe you could advise us how to optimize the MY-Q, the parental-questionnaire
and the subsequent conversation. Could we contact you about this?
o No. 69.4%
o Yes, by e-mail. 25.8%
o Yes, by phone. 4.8%
13. Do you have suggestions or remarks he MY-Q, the parental-questionnaire
and the subsequent conversation? –Open ended (N=6)
Communication with parents from diabetes team 50.0%, Often same question
within questionnaire 33.3%, Attention for changes 16.7%.
Thank you very much for your time and effort. Your answers help us improve the program.
7. Did you discuss the outcomes with the diabetes nurse of pediatrician? (N=39)
o Yes. 23.1%
o No > question 10. 76.9%
8. Kindly indicate the degree to which you agree on the following statements regarding this subsequent conversation. (N=9)
Disagree (D), Neither agree nor disagree (N), Agree (A), Don’t know (DK)
o It was beneficial to discuss the outcomes with the nurse / pediatrician.
D 0%, N 0%, A 100%, DK 0%
o I was not listened to.
D 100%, N 0%, A 0%, DK 0%
o The outcome was consistent with the way a felt.
D 0%, N 33.3%, A 55.6%, DK 11.1%
o I’d rather had no conversation.
D 88.9%, N 11.1%, A 0%, DK 0%
9. Is there something you did not appreciate in this conversation?
If so, what exactly? – Open ended.
Mainly nothing
10. Did your child discussed the MY-Q with you?
o Yes. 44.1%
o No. 41.2%
o Don’t know. 14.7%
A
134
135
A
References
137
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Erikson EH. Childhood and society: WW Norton & Company; 1993 1993.
Bronfenbrenner U, Bronfenbrenner U. The ecology of human development: Experiments by nature and design: Harvard university press; 2009 2009.
Sameroff AJ. Environmental factors in the early screening of children at risk.
Early identification of children at risk: Springer; 1985. p. 21-44.
Piaget J. The theory of stages in cognitive development. 1971.
Van Wouwe JP, Mattiazzo GF, El Mokadem N, Reeser HM, Hirasing RA. De incidentie en de eerste symptomen van diabetes mellitus type I bij 0-14-jarigen in Nederland, I996-I999. Nederlands tijdschrift voor geneeskunde. 2004;148(37):1824-9.
Heine RJ, Tack CJ. Handboek diabetes mellitus. Utrecht: De Tijdstroom. 2004.
Silverstein J, Klingensmith G, Copeland K, Plotnick L, Kaufman F, Laffel L, et al. Care of children and adolescents with type 1 diabetes a statement of the American Diabetes Association. Diabetes Care. 2005;28(1):186-212.
White NH, Cleary PA, Dahms W, Goldstein D, Malone J, Tamborlane WV.
Beneficial effects of intensive therapy of diabetes during adolescence: outcomes after the
conclusion of the Diabetes Control and Complications Trial (DCCT). J Pediatr.
2001;139(6):804-12.
Nederlandse Diabetes Federatie (NDF). NDF Zorgstandaard Diabetes type 1 Kinderen en adolescenten. Amersfoort2015 2015.
Rewers MJ, Pillay K, de Beaufort C, Craig ME, Hanas R, Acerini CL, et al. Assessment and monitoring of glycemic control in children and adolescents with diabetes.
Pediatric Diabetes. 2014;15(S20):102-14.
Acerini C. The rise of technology in diabetes care. Not all that is new is necessarily better.
Pediatric Diabetes. 2016.
Reynolds KA, Helgeson VS. Children with diabetes compared to peers: depressed?
Distressed? A meta-analytic review. Ann Behav Med. 2011;42(1):29-41.
Helgeson VS, Siminerio L, Escobar O, Becker D. Predictors of metabolic control among adolescents with diabetes: a 4-year longitudinal study. Journal of Pediatric Psychology.
2009;34(3):254-70.
Nieuwesteeg A, Pouwer F, van der Kamp R, van Bakel H, Aanstoot H-J, Hartman E.
Quality of life of children with type 1 diabetes: a systematic review. Curr Diabetes Rev. 2012.
Felce D, Perry J. Quality of life: Its definition and measurement. Research in developmental
disabilities. 1995;16(1):51-74.
Striegel-Moore RH, Bulik CM. Risk factors for eating disorders. American Psychologist. 2007;62(3):181.
Colton PA, Olmsted MP, Daneman D, Farquhar JC, Wong H, Muskat S, et al. Eating Disorders
in Girls and Women With Type 1 Diabetes: A Longitudinal Study of Prevalence, Onset, Remission, and Recurrence. Diabetes Care. 2015:dc142646.
Delamater AM, de Wit M, McDarby V, Malik J, Acerini CL. Psychological care of children and adolescents with type 1 diabetes. Pediatric Diabetes. 2014;15(S20):232-44.
Wysocki T, Buckloh LM, Lochrie AS, Antal H. The psychologic context of pediatric diabetes.
Pediatr Clin North Am. 2005;52(6):1755-78.
Main A, Wiebe DJ, Van Bogart K, Turner SL, Tucker C, Butner JE, et al. Secrecy From Parents
and Type 1 Diabetes Management in Late Adolescence. Journal of Pediatric Psychology.
2015;40(10):1075-84.
138
References
21.
22.
23.
24. 25. 26.
27. 28. 29.
30.
31.
32.
33.
34. 35. 36. Osborn P, Berg CA, Hughes AE, Pham P, Wiebe DJ. What mom and dad don't know can hurt you: adolescent disclosure to and secrecy from parents about type 1 diabetes. Journal of Pediatric Psychology. 2013;38(2):141-50.
Hannonen R, Eklund K, Tolvanen A, Komulainen J, Riikonen R, Delamater AM, et al. Psychological distress of children with early-onset type 1 diabetes and their mothers' well-being. Acta Paediatrica. 2015;104(11):1144-9.
Landolt MA, Ystrom E, Sennhauser FH, Gnehm HE, Vollrath ME. The mutual prospective
influence of child and parental post-traumatic stress symptoms in pediatric patients. Journal of Child Psychology and Psychiatry. 2012;53(7):767-74.
Hessler D, Fisher L, Polonsky W, Johnson N. Understanding the Areas and Correlates of Diabetes-Related Distress in Parents of Teens With Type 1 Diabetes. Journal of Pediatric Psychology. 2016.
Whittemore R, Jaser S, Chao A, Jang M, Grey M. Psychological experience of parents
of children with type 1 diabetes: a systematic mixed-studies review. Diabetes Educ. 2012;38(4):562-79.
Patterson CC, Dahlquist GG, Gyurus E, Green A, Soltesz G. Incidence trends for childhood
type 1 diabetes in Europe during 1989-2003 and predicted new cases 2005-20: a multicentre prospective registration study. Lancet. 2009;373(9680):2027-33.
Dabadghao P, Vidmar S, Cameron FJ. Deteriorating diabetic control through adolescencedo the origins lie in childhood? Diabetic medicine. 2001;18(11):889-94.
Amiel SA, Sherwin RS, Simonson DC, Lauritano AA, Tamborlane WV. Impaired insulin action in puberty. A contributing factor to poor glycemic control in adolescents with diabetes. N Engl J Med. 1986;315(4):215-9.
Luyckx K, Seiffge-Krenke I. Continuity and change in glycemic control trajectories from adolescence to emerging adulthood: relationships with family climate and self-concept
in type 1 diabetes. Diabetes Care. 2009;32(5):797-801.
Bryden KS, Peveler RC, Stein A, Neil A, Mayou RA, Dunger DB. Clinical and psychological
course of diabetes from adolescence to young adulthood: a longitudinal cohort study. Diabetes Care. 2001;24(9):1536-40.
Holmbeck GN, Shapera W. Research methods with adolescents. In: Kendall PC, Butcher JN,
Holmbeck GN, editors. Handbook of research methods in clinical psychology. New York:
Wiley; 1999. p. 634-61.
Tanner JM. Sequence, tempo, and individual variation in the growth and development of boys and girls aged twelve to sixteen. Daedalus. 1971:907-30.
Elamin A, Hussein O, Tuvemo T. Growth, puberty, and final height in children with type 1
diabetes. Journal of Diabetes and its Complications. 2006;20(4):252-6.
Rohrer T, Stierkorb E, Heger S, Karges B, Raile K, Schwab KO, et al. Delayed pubertal onset
and development in German children and adolescents with type 1 diabetes: cross-sectional
analysis of recent data from the DPV diabetes documentation and quality management
system. European Journal of Endocrinology. 2007;157(5):647-53.
Kovacs M, Goldston D, Obrosky DS, Iyengar S. Prevalence and predictors of pervasive noncompliance with medical treatment among youths with insulin-dependent diabetes
mellitus. J Am Acad Child Adolesc Psychiatry. 1992;31(6):1112-9.
Alan MJ, Hauser ST, Lavori P, Wolfsdorf JI, Herskowitz RD, Milley JE, et al.
Adherence Among Children and Adolescents With Insulin-Dependent
Diabetes Mellitus Over a Four-Year Longitudinal Follow-Up: I. The Influence of
Patient Coping and Adjustment. Journal of Pediatric Psychology. 1990;15(4):511-26.
139
References
37. 38. 39. 40. 41. 42.
43. 44. 45. 46. 47. 48. 49. 50.
51.
52.
53.
54.
Rosenbauer J, Dost A, Karges B, Hungele A, Stahl A, Bachle C, et al. Improved Metabolic
Control in Children and Adolescents With Type 1 Diabetes A trend analysis using prospective multicenter data from Germany and Austria. Diabetes Care. 2012;35(1):80-6.
Kordonouri O, Danne T, Enders I, Weber B. Does the long-term clinical course of type I diabetes mellitus differ in patients with prepubertal and pubertal onset? Results of the
Berlin Retinopathy Study. Eur J Pediatr. 1998;157(3):202-7.
Donaghue KC, Fairchild JM, Craig ME, Chan AK, Hing S, Cutler LR, et al. Do All Prepubertal
Years of Diabetes Duration Contribute Equally to Diabetes Complications? Diabetes Care.
2003;26(4):1224-9.
Jacobson AM, Hauser ST, Lavori P, Wolfsdorf JI, Herskowitz RD, Milley JE, et al. Adherence
among children and adolescents with insulin-dependent diabetes mellitus over a four-year
longitudinal follow-up: I. The influence of patient coping and adjustment. J Pediatr Psychol.
1990;15(4):511-26.
Tanner JM. Normal growth and techniques of growth assessment. Clin Endocrinol Metab.
1986;15(3):411-51.
Chida Y, Hamer M. An association of adverse psychosocial factors with diabetes mellitus:
a meta-analytic review of longitudinal cohort studies. Diabetologia. 2008;51(12):2168-78.
Sawyer SM, Drew S, Yeo MS, Britto MT. Adolescents with a chronic condition: challenges
living, challenges treating. The Lancet. 2007;369(9571):1481-9.
Holmbeck GN, Abad M, Friedman D, Jandasek B. Development and Psychopathology in
Adolescents. In: Wolfe DA, Mash EJ, editors. Behavioral and Emotional Disorders in Adolescents: Nature, assessment, and treatment. New York: Guilford Press; 2006. p. 21-55.
Dantzer C, Swendsen J, Maurice-Tison S, Salamon R. Anxiety and depression in juvenile
diabetes: a critical review. Clinical psychology review. 2003;23(6):787-800.
Blanz BJ, Rensch-Riemann BS, Fritz-Sigmund DI, Schmidt MH. IDDM is a risk factor for adolescent psychiatric disorders. Diabetes Care. 1993;16(12):1579-87.
Lawrence JM, Standiford DA, Loots B, Klingensmith GJ, Williams DE, Ruggiero A, et al. Prevalence and correlates of depressed mood among youth with diabetes: the SEARCH
for Diabetes in Youth study. Pediatrics. 2006;117(4):1348-58.
Hassan K, Loar R, Anderson BJ, Heptulla RA. The role of socioeconomic status, depression,
quality of life, and glycemic control in type 1 diabetes mellitus. J Pediatr. 2006;149(4):526-31.
Zenlea IS, Mednick L, Rein J, Quinn M, Wolfsdorf J, Rhodes ET. Routine behavioral and mental
health screening in young children with type 1 diabetes mellitus. Pediatric Diabetes. 2013.
Schwartz SA, Weissberg-Benchell J, Perlmuter LC. Personal control and disordered eating
in female adolescents with type 1 diabetes. Diabetes Care. 2002;25(11):1987-91.
Anderson BJ. Family conflict and diabetes management in youth: clinical lessons from child development and diabetes research. Diabetes Spectrum. 2004;17(1):22-6.
Jacobson AM, Hauser ST, Willett JB, Wolfsdorf JI, Dvorak R, Herman L, et al. Psychological
Adjustment to IDDM: 10-Year Follow-Up of an Onset Cohort of Child and Adolescent Patients. Diabetes Care. 1997;20(5):811-8.
Schneider S, Iannotti RJ, Nansel TR, Haynie DL, Sobel DO, Simons-Morton B. Assessment
of an illness-specific dimension of self-esteem in youths with type 1 diabetes. Journal of Pediatric Psychology. 2009;34(3):283-93.
Goebel-Fabbri A. Disturbed eating behaviors and eating disorders in type 1 diabetes: Clinical significance and treatment recommendations. Curr Diab Rep. 2009;9(2):133-9.
140
References
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
Rodin G, Olmsted MP, Rydall AC, Maharaj SI, Colton PA, Jones JM, et al. Eating disorders
in young women with type 1 diabetes mellitus. Journal of psychosomatic research.
02;53(4):943-9.
Colton P, Olmsted M, Daneman D, Rydall A, Rodin G. Disturbed Eating Behavior and
Eating Disorders in Preteen and Early Teenage Girls With Type 1 Diabetes: A casecontrolled study. Diabetes Care. 2004;27(7):1654-9.
Mannucci E, Rotella F, Ricca V, Moretti S, Placidi GF, Rotella CM. Eating disorders in
patients with type 1 diabetes: a meta-analysis. J Endocrinol Invest. 2005;28(5):417-9.
Young V, Eiser C, Johnson B, Brierley S, Epton T, Elliott J, et al. Eating problems in
adolescents with Type 1 diabetes: a systematic review with meta-analysis. Diabetic
medicine. 2013;30(2):189-98.
Rydall AC, Rodin GM, Olmsted MP, Devenyi RG, Daneman D. Disordered Eating Behavior
and Microvascular Complications in Young Women with Insulin-Dependent Diabetes
Mellitus. New England Journal of Medicine. 1997;336(26):1849-54.
Neumark-Sztainer D, Patterson J, Mellin A, Ackard DM, Utter J, Story M, et al. Weight
Control Practices and Disordered Eating Behaviors Among Adolescent Females and
Males With Type 1 Diabetes Associations with sociodemographics, weight concerns,
familial factors, and metabolic outcomes. Diabetes Care. 2002;25(8):1289-96.
Wisting L, Froisland DH, Skrivarhaug T, Dahl-Jorgensen K, Ro O. Disturbed Eating
Behavior and Omission of Insulin in Adolescents Receiving Intensified Insulin Treatment A
nationwide population-based study. Diabetes Care. 2013;36(11):3382-7.
Goebel-Fabbri AE, Fikkan J, Franko DL, Pearson K, Anderson BJ, Weinger K. Insulin
Restriction and Associated Morbidity and Mortality in Women with Type 1 Diabetes.
Diabetes Care. 2008;31(3):415-9.
Bryden KS, Neil A, Mayou RA, Peveler RC, Fairburn CG, Dunger DB. Eating habits, body
weight, and insulin misuse. A longitudinal study of teenagers and young adults with
type 1 diabetes. Diabetes Care. 1999;22(12):1956-60.
Peveler RC, Bryden KS, Neil HA, Fairburn CG, Mayou RA, Dunger DB, et al. The
Relationship of Disordered Eating Habits and Attitudes to Clinical Outcomes in Young
Adult Females With Type 1 Diabetes. Diabetes Care. 2005;28(1):84-8.
Young-Hyman DL, Davis CL. Disordered Eating Behavior in Individuals With Diabetes:
Importance of context, evaluation, and classification. Diabetes Care. 2010;33(3):683-9.
Takii M, Uchigata Y, Tokunaga S, Amemiya N, Kinukawa N, Nozaki T, et al. The duration
of severe insulin omission is the factor most closely associated with the microvascular
complications of Type 1 diabetic females with clinical eating disorders. International
Journal of Eating Disorders. 2008;41(3):259-64.
d'Emden H, Holden L, McDermott B, Harris M, Gibbons K, Gledhill A, et al. Disturbed
eating behaviours and thoughts in Australian adolescents with type 1 diabetes. Journal
of Paediatrics and Child Health. 2013;49(4):E317-E23.
Markowitz JT, Butler DA, Volkening LK, Antisdel JE, Anderson BJ, Laffel LM. Brief
screening tool for disordered eating in diabetes internal consistency and external
validity in a contemporary sample of pediatric patients with type 1 diabetes. Diabetes Care
2010;33(3):495-500.
Cameron FJ, Northam EA, Ambler GR, Daneman D. Routine Psychological Screening in
Youth With Type 1 Diabetes and Their Parents A notion whose time has come? Diabetes Care
2007;30(10):2716-24.
141
References
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
Olmsted MP, Colton PA, Daneman D, Rydall AC, Rodin GM. Prediction of the Onset
of Disturbed Eating Behavior in Adolescent Girls With Type 1 Diabetes. Diabetes Care. 2008;31(10):1978-82.
Tierney S, Deaton C, Whitehead J. Caring for people with type 1 diabetes mellitus
engaging in disturbed eating or weight control: a qualitative study of practitioners'
attitudes and practices. Journal of Clinical Nursing. 2009;18(3):384-90.
Gaudieri PA, Chen R, Greer TF, Holmes CS. Cognitive function in children with type 1
diabetes: a meta-analysis. Diabetes Care. 2008;31(9):1892-7.
Dahlquist G, Kallen B. School marks for Swedish children whose mothers had diabetes
during pregnancy: a population-based study. Diabetologia. 2007;50(9):1826-31.
Schiebener J, Garcia-Arias M, Garcia-Villamisar D, Cabanyes-Truffino J, Brand M.
Developmental changes in decision making under risk: The role of executive functions
and reasoning abilities in 8-to 19-year-old decision makers. Child Neuropsychology.
2014 (ahead-of-print):1-20.
Holmbeck GN. A developmental perspective on adolescent health and illness: an
introduction to the special issues. J Pediatr Psychol. 2002;27(5):409-16.
Helgeson VS, Reynolds KA, Siminerio L, Escobar O, Becker D. Parent and adolescent
distribution of responsibility for diabetes self-care: Links to health outcomes. Journal of
Pediatric Psychology. 2008;33(5):497-508.
Anderson BJ, Brackett J, Ho J, Laffel LM. An office-based intervention to maintain
parent-adolescent teamwork in diabetes management. Impact on parent involvement,
family conflict, and subsequent glycemic control. Diabetes Care. 1999;22(5):713-21.
Wiebe DJ, Berg CA, Korbel C, Palmer DL, Beveridge RM, Upchurch R, et al. Children's
appraisals of maternal involvement in coping with diabetes: Enhancing our
understanding of adherence, metabolic control, and quality of life across adolescence.
Journal of Pediatric Psychology. 2005;30(2):167-78.
de Wit M, Delemarre-van de Waal H, Bokma JA, Haasnoot K, Houdijk MC, Gemke RJ,
et al. Monitoring and discussing health-related quality of life in adolescents with type 1
diabetes improve psychosocial well-being: a randomized controlled trial. Diabetes Care
2008;31(8):1521-6.
Anderson BJ, Brackett J. Diabetes in Children. In: Snoek FJ, Skinner TC, editors. Psychology
in Diabetes Care, 2nd edition. West Sussex: John Wiley & Sons Ltd.; 2005. p. 1-23.
Main A, Wiebe DJ, Croom AR, Sardone K, Godbey E, Tucker C, et al. Associations of
Parent-Adolescent Relationship Quality With Type 1 Diabetes Management and
Depressive Symptoms in Latino and Caucasian Youth. Journal of Pediatric Psychology.
2014:jsu062.
Wysocki T, Taylor A, Hough BS, Linscheid TR, Yeates KO, Naglieri JA. Deviation from
Developmentally Appropriate Self-Care Autonomy: Association with diabetes outcomes.
Diabetes Care. 1996;19(2):119-25.
Butner J, Berg CA, Osborn P, Butler JM, Godri C, Fortenberry KT, et al. Parent-adolescent
discrepancies in adolescents' competence and the balance of adolescent autonomy
and adolescent and parent well-being in the context of Type 1 diabetes. Dev Psychol.
2009;45(3):835-49.
Blicke M, Korner U, Nixon P, Salgin B, Meissner T, Pollok B. The relation between
awareness of personal resources and metabolic control in children and adolescents
with type 1 diabetes. Pediatr Diabetes. 2014.
142
References
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
King PS, Berg CA, Butner J, Butler JM, Wiebe DJ. Longitudinal Trajectories of Parental
Involvement in Type 1 Diabetes and Adolescents' Adherence. 2013.
Serbin LA, Hubert M, Hastings PD, Stack DM, Schwartzman AE. The Influence of
Parenting on Early Childhood Health and Health Care Utilization. Journal of Pediatric
Psychology. 2014:jsu050.
Van Allen J, Kuhl ES, Filigno SS, Clifford LM, Connor JM, Stark LJ. Changes in Parent
Motivation Predicts Changes in Body Mass Index z-Score (zBMI) and Dietary Intake
Among Preschoolers Enrolled in a Family-Based Obesity Intervention. Journal of
Pediatric Psychology. 2014:jsu052.
Moreira H, Frontini R, Bullinger M, Canavarro MC. Caring for a child with Type 1
diabetes: Links between family cohesion, perceived impact, and parental adjustment.
Journal of Family Psychology. 2013;27(5):731.
Horsch A, McManus F, Kennedy P, Edge J. Anxiety, depressive, and posttraumatic
stress symptoms in mothers of children with type 1 diabetes. Journal of traumatic
stress. 2007;20(5):881-91.
Ziegler R, Wiedebusch S, Muthny FA. 'Chronic sorrow' in parents of children with type-1
diabetes. Pediatric Diabetes. 2007;8(7):24-83.
Faulkner MS, Clark FS. Quality of life for parents of children and adolescents with type 1
diabetes. The Diabetes Educator. 1998;24(6):721-7.
Mackey ER, Struemph K, Powell PW, Chen R, Streisand R, Holmes CS. Maternal
depressive symptoms and disease care status in youth with type 1 diabetes. Health
Psychology. 2014;33(8):783.
Butwicka A, Zalepa A, Fendler W, Szadkowska A, Mlynarski W. Maternal depressive
symptoms predict acute hospitalization among children with type 1 diabetes. Pediatric
Diabetes. 2013;14(4):288-94.
Wiebe DJ, Chow CM, Palmer DL, Butner J, Butler JM, Osborn P, et al. Developmental
Processes Associated With Longitudinal Declines in Parental Responsibility and
Adherence to Type 1 Diabetes Management Across Adolescence. J Pediatr Psychol. 2014.
Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing
indirect effects in multiple mediator models. Behavior research methods.
2008;40(3):879-91.
Goodman R. The Strengths and Difficulties Questionnaire: a research note. J Child
Psychol Psychiatry. 1997;38(5):581-6.
Ravens-Sieberer U, Bullinger M. The revised KINDL-R: Final results on reliability, validity
and responsiveness of a modular HRQOL instrument for children and adolescents.
Quality of Life Research. 2001:199-.
Ravens-Sieberer U, Gosch A, Rajmil L, Erhart M, Bruil J, Power M, et al. The KIDSCREEN-52
Quality of Life Measure for Children and Adolescents: Psychometric Results from a
Cross-Cultural Survey in 13 European Countries. Value in Health. 2008;11(4):645-58.
Ravens-Sieberer U, Herdman M, Devine J, Otto C, Bullinger M, Rose M, et al. The
European KIDSCREEN approach to measure quality of life and well-being in children:
development, current application, and future advances. Quality of Life Research. 2013:1-13
Anderson BJ, Auslander WF, Jung KC, Miller JP, Santiago JV. Assessing family sharing of
diabetes responsibilities. J Pediatr Psychol. 1990;15(4):477-92.
de Wit M, Winterdijk P, Aanstoot HJ, Anderson BJ, Danne T, Deeb L, et al. Assessing
diabetes-related quality of life of youth with type 1 diabetes in routine clinical care: the
MIND Youth Questionnaire (MY-Q). Pediatric Diabetes. 2012;13(7):1-9.
143
References
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
Van Der Ven N, Weinger K, Yi J, Pouwer F, Ader H, Van Der Ploeg HM, et al. The confidence
in diabetes self-care scale: psychometric properties of a new measure of diabetes-specific
self-efficacy in Dutch and US patients with type 1 diabetes. Diabetes Care. 2003;26(3):713-8.
Weissberg-Benchell J, Glasgow AM, Tynan WD, Wirtz P, Turek J, Ward J. Adolescent
diabetes management and mismanagement. Diabetes Care. 1995;18(1):77-82.
Wechsler D. WISC-III: Wechsler intelligence scale for children: Psychological Corporation
San Antonio, TX; 1991 1991.
Wechsler D. WISC III Handleiding. London, United Kingdom: The Psychological
Corporation; 2002 2002.
Fan J, McCandliss BD, Fossella J, Flombaum JI, Posner MI. The activation of attentional
networks. NeuroImage. 2005;26(2):471-9.
de Kieviet JF, van Elburg RM, Lafeber HN, Oosterlaan J. Attention problems of very
preterm children compared with age-matched term controls at school-age. The Journal
of pediatrics. 2012;161(5):824-9.
Eriksen CW. The flankers task and response competition: A useful tool for investigating
a variety of cognitive problems. Visual Cognition. 1995;2(2-3):101-18.
Aarnoudse-Moens CS, Duivenvoorden HJ, Weisglas-Kuperus NYNK, Van Goudoever JB,
Oosterlaan J. The profile of executive function in very preterm children at 4 to 12 years.
Developmental Medicine & Child Neurology. 2012;54(3):247-53.
Westerberg H, Hirvikoski T, Forssberg H, Klingberg T. Visuo-Spatial Working Memory Span:
A Sensitive Measure of Cognitive Deficits in Children With ADHD. Child Neuropsychology.
2004;10(3):155-61.
Nutley SB, Soderqvist S, Bryde S, Humphreys K, Klingberg T. Measuring working memory
capacity with greater precision in the lower capacity ranges. Developmental
neuropsychology. 2010;35(1):81-95.
Gioia GA, Isquith PK, Retzlaff PD, Espy KA. Confirmatory Factor Analysis of the Behavior
Rating Inventory of Executive Function (BRIEF) in a Clinical Sample. Child
Neuropsychology. 2002;8(4):249-57.
Gioia GA, Isquith PK, Guy SC, Kenworthy L. Behavior Rating Inventory of Executive
Function. Child Neuropsychology. 2000;6(3):235-8.
Markowitz JT, Volkening LK, Antisdel-Lomaglio J, Anderson BJ, Laffel LM. Re-examining
a measure of diabetes-related burden in parents of young people with Type 1 diabetes:
the Problem Areas in Diabetes Survey - Parent Revised version (PAID-PR). Diabet Med.
2012;29(4):526-30.
Antisdel JE. Diabetes-specific distress among parents of youth with type 1 diabetes
(Abstract). Dissertation Abstracts International. 2001;61:6123.
de Wit M, Pouwer F, Gemke RJBJ, Delemarre-van de Waal H, Snoek FJ. Validation of the
WHO-5 Well-Being Index in adolescents with type 1 diabetes. Diabetes Care.
2007;30(8):2003-6.
Lowe B, Spitzer RL, Grafe K, Kroenke K, Quenter A, Zipfel S, et al. Comparative validity
of three screening questionnaires for DSM-IV depressive disorders and physicians'
diagnoses. Journal of Affective Disorders. 2004;78(2):131-40.
Bech P, Gudex C, Johansen KS. The WHO (Ten) Well-Being Index: validation in diabetes.
Psychother Psychosom. 1996;65(4):183-90.
Schafer LC, McCaul KD, Glasgow RE. Supportive and Nonsupportive Family Behaviors:
Relationships to Adherence and Metabolic Control in Persons with Type I Diabetes.
Diabetes Care. 1986;9(2):179-85.
144
References
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
Pope C, Ziebland S, Mays N. Analysing qualitative data. BMJ. 2000;320(7227):114-6.
Gale NK, Heath G, Cameron E, Rashid S, Redwood S. Using the framework method for
the analysis of qualitative data in multi-disciplinary health research. BMC medical
research methodology. 2013;13(1):117.
Northam EA, Matthews LK, Anderson PJ, Cameron FJ, Werther GA. Psychiatric morbidity
and health outcome in type 1 diabetes-perspectives from a prospective longitudinal
study. Diabetic medicine. 2005;22(2):152-7.
Duke DC, Geffken GR, Lewin AB, Williams LB, Storch EA, Silverstein JH. Glycemic
control in youth with type 1 diabetes: family predictors and mediators. J Pediatr Psychol.
2008;33(7):719-27.
Cohen DM, Lumley MA, Naar-King S, Partridge T, Cakan N. Child behavior problems
and family functioning as predictors of adherence and glycemic control in economically
disadvantaged children with type 1 diabetes: a prospective study. Journal of Pediatric
Psychology. 2004;29(3):171-84.
Kovacs M, Charron-Prochownik D, Obrosky DS. A Longitudinal Study of Biomedical
and Psychosocial Predictors of Multiple Hospitalizations Among Young People with
Insulin-dependent Diabetes Mellitus. Diabetic medicine. 1995;12(2):142-8.
McDonnell CM, Northam EA, Donath SM, Werther GA, Cameron FJ. Hyperglycemia and
externalizing behavior in children with type 1 diabetes. Diabetes Care. 2007;30(9):2211-5.
Northam EA, Lin A, Finch S, Werther GA, Cameron FJ. Psychosocial well-being and
functional outcomes in youth with type 1 diabetes 12 years after disease onset.
Diabetes Care. 2010;33(7):1430-7.
Carr A. The Handbook of Child and Adolescent Clinical Psychology. A Contextual
Approach. Londen: Routledge; 2006 2006.
Eilander M, de Wit M, Rotteveel J, Aanstoot HJ, Bakker-van Waarde W, Houdijk E, et al.
Diabetes IN develOpment (DINO): the bio-psychosocial, family functioning and parental
well-being of youth with type 1 diabetes: a longitudinal cohort study design. BMC
pediatrics. 2015;15(1):82.
van Widenfelt BM, Goedhart AW, Treffers PD, Goodman R. Dutch version of the
Strengths and Difficulties Questionnaire (SDQ). Eur Child Adolesc Psychiatry.
2003;12(6):281-9.
Hayes AF. Introduction to mediation, moderation, and conditional process analysis.
New York: Guilford Press; 2013 2013.
Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological
research: Conceptual, strategic, and statistical considerations. Journal of personality
and social psychology. 1986;51(6):1173.
Hayes AF. Beyond Baron and Kenny: Statistical mediation analysis in the new millennium.
Communication Monographs. 2009;76(4):408-20.
Shrout PE, Bolger N. Mediation in experimental and nonexperimental studies: new
procedures and recommendations. Psychological methods. 2002;7(4):422.
Jones A, Vallis M, Pouwer F. If it does not significantly change HbA1c levels why should
we waste time on it? A plea for the prioritization of psychological well-being in people
with diabetes. Diabetic medicine. 2015;32(2):155-63.
Cameron FJ, Smidts D, Hesketh K, Wake M, Northam EA. Early detection of emotional and
behavioural problems in children with diabetes: the validity of the Child Health
Questionnaire as a screening instrument. Diabetic medicine. 2003;20(8):646-50.
145
References
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
Wake M, Hesketh K, Cameron F. The Child Health Questionnaire in children with
diabetes: cross-sectional survey of parent and adolescent-reported functional health
status. Diabet Med. 2000;17(10):700-7.
Eilander M, de Wit M, Rotteveel J, Maas-van Schaaijk N, Roeleveld-Versteegh A, Snoek F.
Implementation of quality of life monitoring in Dutch routine care of adolescents with
type 1 diabetes: appreciated but difficult. Pediatric Diabetes. 2016;17(2):112-9.
Gonder-Frederick LA, Cox DJ, Bobbitt SA, Pennebaker JW. Mood changes associated
with blood glucose fluctuations in insulin-dependent diabetes mellitus. Health
Psychology. 1989;8(1):45.
Young NL, Varni JW, Snider L, McCormick A, Sawatzky B, Scott M, et al. The Internet is
valid and reliable for child-report: an example using the Activities Scale for Kids
(ASK) and the Pediatric Quality of Life Inventory (PedsQL). Journal of clinical
epidemiology. 2009;62(3):314-20.
Varni JW, Limbers CA, Burwinkle TM, Bryant WP, Wilson DP. The ePedsQL in Type 1 and
Type 2 Diabetes Feasibility, reliability, and validity of the Pediatric Quality of Life
Inventory Internet administration. Diabetes Care. 2008;31(4):672-7.
Holl RW, Swift PG, Mortensen HB, Lynggaard H, Hougaard P, Aanstoot HJ, et al. Insulin
injection regimens and metabolic control in an international survey of adolescents
with type 1 diabetes over 3 years: results from the Hvidore study group. European
journal of pediatrics. 2003;162(1):22-9.
Dunger DB, Ahmed L, Ong K. Growth and body composition in type 1 diabetes mellitus.
Hormone Research in Paediatrics. 2002;58(Suppl. 1):66-71.
Pinhas-Hamiel O, Hamiel U, Levy-Shraga Y. Eating disorders in adolescents with type 1
diabetes: Challenges in diagnosis and treatment. World journal of diabetes.
2015;6(3):517.
Peterson CM, Fischer S, Young-Hyman D. Topical Review: A Comprehensive Risk
Model for Disordered Eating in Youth With Type 1 Diabetes. Journal of Pediatric
Psychology. 2014:jsu106.
Taylor CB, Bryson S, Luce KH, Cunning D, Doyle AC, Abascal LB, et al. Prevention of
eating disorders in at-risk college-age women. Archives of General Psychiatry.
2006;63(8):881-8.
Fairburn CG, Harrison PJ. Eating disorders. The Lancet. 2003;361(9355):407-16.
Wilson CE, Smith EL, Coker SE, Hobbis IC, Acerini CL. Testing an integrated model of
eating disorders in paediatric type 1 diabetes mellitus. Pediatric Diabetes.
2014;16(7):521-8.
Baiano M, Salvo P, Righetti P, Cereser L, Baldissera E, Camponogara I, et al. Exploring
health-related quality of life in eating disorders by a cross-sectional study and a
comprehensive review. BMC psychiatry. 2014;14(1):165.
Takii M, Uchigata Y, Kishimoto J, Morita C, Hata T, Nozaki T, et al. The relationship
between the age of onset of type 1 diabetes and the subsequent development of a
severe eating disorder by female patients. Pediatric Diabetes. 2011;12(4pt2):396-401.
Micali N, De Stavola B, Ploubidis G, Simonoff E, Treasure J, Field AE. Adolescent eating
disorder behaviours and cognitions: gender-specific effects of child, maternal and
family risk factors. The British Journal of Psychiatry. 2015:bjp-bp.
146
References
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
Saßmann H, Albrecht C, Busse-Widmann P, Hevelke LK, Kranz J, Markowitz JT, et al.
Psychometric properties of the German version of the Diabetes Eating Problem SurveyRevised: additional benefit of disease-specific screening in adolescents with Type 1
diabetes. Diabetic medicine. 2015;32(12):1641-7.
Delameter A. ISPAD Clinical Practice Consensus Guidelines 2009 Compendium.
Psychological care of children and adolescents with diabetes. Pediatric Diabetes.
2007;10 (Suppl 12):175-84.
Talma H, Schonbeck Y, Bakker B, Hirasing RA, van Buuren S. Groeidiagrammen 2010.
Leiden: TNO; 2010 2010.
Wisting L, Bang L, Skrivarhaug T, Dahl-Jorgensen K, Ro O. Adolescents with Type 1
Diabetes-The Impact of Gender, Age, and Health-Related Functioning on Eating Disorder
Psychopathology. PloS one. 2015;10(11):e0141386.
Moosavi M, Kreisman S, Hall L. Intentional Hypoglycemia to Control Bingeing in a Patient
with Type 1 Diabetes and Bulimia Nervosa. Canadian journal of diabetes. 2015;39(1):16-7.
Boogerd EA, Damhuis AM, van Alfen-van der Velden J, Steeghs MC, Noordam C,
Verhaak CM, et al. Assessment of psychosocial problems in children with type 1 diabetes
and their families: the added value of using standardised questionnaires in addition
to clinical estimations of nurses and paediatricians. Journal of Clinical Nursing.
2015;24(15-16):2143-51.
Lopez-Perez B, Wilson EL. Parent-child discrepancies in the assessment of children's
and adolescents' happiness. Journal of experimental child psychology. 2015;139:249-55.
de Wit M, Pulgaron ER, Pattino-Fernandez AM, Delamater AM. Psychological Support
for Children with Diabetes: Are the Guidelines Being Met? Journal of clinical psychology
in medical settings. 2014;21(2):190-9.
Alloway SC, Toth EL, McCargar LJ. Effectiveness of a group psychoeducation program
for the treatment of subclinical disordered eating in women with type 1 diabetes.
Canadian Journal of Dietetic Practice and Research. 2001;62(4):188.
Olmsted MP, Daneman D, Rydall AC, Lawson ML, Rodin G. The effects of psychoeducation
on disturbed eating attitudes and behavior in young women with type 1 diabetes
mellitus. International Journal of Eating Disorders. 2002;32(2):230-9.
Wilksch SM, Starkey K, Gannoni A, Kelly T, Wade TD. Interactive programme to enhance
protective factors for eating disorders in girls with type 1 diabetes. Early intervention in
psychiatry. 2013;7(3):315-21.
Dickens YL, Haynos AF, Nunnemaker S, Platka-Bird L, Dolores J. Multidisciplinary
Residential Treatment of Type 1 Diabetes Mellitus and Co-Occurring Eating Disorders.
Eating disorders. 2015;23(2):134-43.
Driscoll KA, Raymond J, Naranjo D, Patton SR. Fear of Hypoglycemia in Children and
Adolescents and Their Parents with Type 1 Diabetes. Current diabetes reports.
2016;16(8):1-9.
Rumburg TM, Lord JH, Savin KL, Jaser SS. Maternal diabetes distress is linked to maternal
depressive symptoms and adolescents' glycemic control. Pediatric Diabetes. 2015.
Jaser SS, Linsky R, Grey M. Coping and psychological distress in mothers of adolescents
with type 1 diabetes. Maternal and child health journal. 2014;18(1):101-8.
Jaser SS, Whittemore R, Ambrosino JM, Lindemann E, Grey M. Mediators of depressive
symptoms in children with type 1 diabetes and their mothers. J Pediatr Psychol.
2008;33(5):509-19.
147
References
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
Law GU, Walsh J, Queralt V, Nouwen A. Adolescent and parent diabetes distress in
type 1 diabetes: The role of self-efficacy, perceived consequences, family responsibility
and adolescent-parent discrepancies. Journal of psychosomatic research.
2013;74(4):334-9.
Armstrong B, Mackey ER, Streisand R. Parenting behavior, child functioning, and
health behaviors in preadolescents with type 1 diabetes. Journal of Pediatric Psychology.
2011;36(9):1052-61.
Maas-van Schaaijk N, Roeleveld-Versteegh A, van Baar A. The Interrelationships Among
Paternal and Maternal Parenting Stress, Metabolic Control, and Depressive Symptoms
in Adolescents With Type 1 Diabetes Mellitus. J Pediatr Psychol. 2012.
Fisher L, Hessler DM, Polonsky WH, Masharani U, Peters AL, Blumer I, et al. Prevalence
of depression in Type 1 diabetes and the problem of over-diagnosis. Diabetic medicine. 2016.
Topp CW, Ostergaard SD, Sondergaard S, Bech P. The WHO-5 Well-Being Index: a
systematic review of the literature. Psychotherapy and psychosomatics. 2015;84(3):167-76.
Young MT, Lord JH, Patel NJ, Gruhn MA, Jaser SS. Good cop, bad cop: quality of parental
involvement in type 1 diabetes management in youth. Current diabetes reports.
2014;14(11):1-12.
Bech P, Olsen LR, Kjoller M, Rasmussen NK. Measuring well-being rather than the
absence of distress symptoms: a comparison of the SF-36 Mental Health subscale and
the WHO-Five Well-Being Scale. Int J Methods Psychiatr Res. 2003;12(2):85-91.
Hajos TRS, Pouwer F, Skovlund SE, Den Oudsten BL, Geelhoed-Duijvestijn PHLM, Tack CJ,
et al. Psychometric and screening properties of the WHO-5 well-being index in adult
outpatients with Type 1 or Type 2 diabetes mellitus. Diabetic medicine. 2013;30(2):e63-e9.
IBM Corp. IBM SPSS Statistics for Windows,. In: Corp. I, editor. 22.0 ed. Armonk, NY:
IBM Corp; 2013.
Eccleston C, Fisher E, Law E, Bartlett J, Palermo TM. Psychological interventions for
parents of children and adolescents with chronic illness. The Cochrane Library. 2015.
Lohan A, Morawska A, Mitchell A. A systematic review of parenting interventions for
parents of children with type 1 diabetes. Child: care, health and development.
2015;41(6):803-17.
Laffel LMB, Vangsness L, Connell A, Goebel-Fabbri A, Butler D, Anderson BJ. Impact of
ambulatory, family-focused teamwork intervention on glycemic control in youth with
type 1 diabetes. J Pediatr. 2003;142(4):409-16.
Wysocki T, Harris MA, Buckloh LM, Mertlich D, Lochrie AS, Mauras N, et al. Randomized
Trial of Behavioral Family Systems Therapy for Diabetes Maintenance of effects on diabetes
outcomes in adolescents. Diabetes Care. 2007;30(3):555-60.
Ellis DA, Naar-King S, Chen X, Moltz K, Cunningham PB, Idalski-Carcone A. Multisystemic
therapy compared to telephone support for youth with poorly controlled diabetes:
findings from a randomized controlled trial. Annals of Behavioral Medicine.
2012;44(2):207-15.
Saßmann H, de Hair M, Danne T, Lange K. Reducing stress and supporting positive
relations in families of young children with type 1 diabetes: A randomized controlled
study for evaluating the effects of the DELFIN parenting program. BMC pediatrics.
2012;12(1):1.
Doherty FM, Calam R, Sanders MR. Positive parenting program (triple P) for families of
adolescents with type 1 diabetes: a randomized controlled trial of self-directed teen
triple P. Journal of Pediatric Psychology. 2013:jst046.
148
References
184.
185.
186.
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.
202.
Anderson BJ, Coyne JC. Advances in child health psychology. University of Florida Press
Gainesville, FL; 1991.
Wu YP, Hilliard ME, Rausch J, Dolan LM, Hood KK. Family involvement with the diabetes
regimen in young people: the role of adolescent depressive symptoms. Diabetic medicine.
2013;30(5):596-602.
Eilander M, de Wit M, Rotteveel J, Snoek FJ. Low Self-Confidence and Diabetes
Mismanagement in Youth with Type 1 Diabetes Mediate the Relationship between
Behavioral Problems and Elevated HbA1c. Journal of Diabetes Research. 2015;501:681421.
Hains AA, Berlin KS, Davies WH, Smothers MK, Sato AF, Alemzadeh R. Attributions
of adolescents with type 1 diabetes related to performing diabetes care around
friends and peers: the moderating role of friend support. Journal of Pediatric Psychology.
2007;32(5):561-70.
International Diabetes Federation. Global IDF/ISPAD Guideline for Diabetes in Childhood
and Adolescence. Brussels: 2011 2011. Report No.
deWitM,SnoekFJ.TheDAWNMINDYouthprogram.PediatricDiabetes.2009;10Suppl13:46-9.
Drewes HW, Steuten LMG, Lemmens LC, Baan CA, Boshuizen HC, Elissen AMJ, et al.
The effectiveness of chronic care management for heart failure: meta-regression
analyses to explain the heterogeneity in outcomes. Health Serv Res. 2012;47(5):1926-59.
Grol R, Wensing M. Implementatie, effectieve verbetering van de patiëntenzorg.
Amsterdam: Reed Business; 2011 2011.
Berwick DM. The science of improvement. JAMA. 2008;299(10):1182-4.
Grol R, Wensing M. What drives change? Barriers to and incentives for achieving
evidence-based practice. Med J Aust. 2004;180(6 Suppl):S57-S60.
Blair M. Getting evidence into practice--implementation science for paediatricians.
Arch Dis Child. 2014;99(4):307-9.
Miles MB, Huberman AM. Qualitative data analysis: An expanded sourcebook. 1994.
Beverly Hills: Sage Publications. 1994.
Delamater AM. Psychological care of children and adolescents with diabetes. Pediatr
Diabetes. 2007;8(5):340-8.
SWEET pediatric diabetes working to create centres of reference. Recommendations
for Diabetes Care and Treatment in Pediatric Centres of Reference in the EU 2009. 2010.
Hambly H, Robling M, Crowne E, Hood K, Gregory JW. Communication skills of healthcare
professionals in paediatric diabetes services. Diabet Med. 2009;26(5):502-9.
Hawthorne K, Bennert K, Lowes L, Channon S, Robling M, Gregory JW. The experiences
of children and their parents in paediatric diabetes services should inform the
development of communication skills for healthcare staff (the DEPICTED Study). Diabet
Med. 2011;28(9):1103-8.
Cameron FJ. The impact of diabetes on health-related quality of life in children and
adolescents. Pediatric Diabetes. 2003;4(3):132-6.
Grey M, Boland EA, Yu C, Sullivan-Bolyai S, Tamborlane WV. Personal and Family Factors
Associated With Quality of Life in Adolescents With Diabetes. Diabetes Care.
1998;21(6):909-14.
Snoek FJ, Kersch NY, Eldrup E, Harman-Boehm I, Hermanns N, Kokoszka A, et al.
Monitoring of Individual Needs in Diabetes (MIND)-2 Follow-up data from the
cross-national Diabetes Attitudes, Wishes, and Needs (DAWN) MIND study. Diabetes
Care. 2012;35(11):2128-32.
149
References
203.
204.
205.
206.
207.
208.
209.
210.
211.
212.
213.
214.
215.
216.
217.
218.
219.
220.
Peters DH, Adam T, Alonge O, Agyepong IA, Tran N. Implementation research: what it is
and how to do it. BMJ. 2013;347:f6753.
Hilliard ME, Wu YP, Rausch J, Dolan LM, Hood KK. Predictors of deteriorations in
diabetes management and control in adolescents with type 1 diabetes. Journal of
Adolescent Health. 2013;52(1):28-34.
Berlin KS, Hains AA, Kamody RC, Kichler JC, Davies WH. Differentiating Peer and Friend
Social Information-Processing Effects on Stress and Glycemic Control Among Youth
With Type 1 Diabetes. Journal of Pediatric Psychology. 2015:jsu111.
Wisting L, Bang L, Natvig H, Skrivarhaug T, Dahl-Jorgensen K, Lask B, et al. Metabolic
Control and Illness Perceptions in Adolescents with Type 1 Diabetes. Journal of Diabetes
Research. 2015;501:456340.
Fortenberry KT, Berg CA, King PS, Stump T, Butler JM, Pham PK, et al. Longitudinal
Trajectories of Illness Perceptions Among Adolescents With Type 1 Diabetes. Journal of
Pediatric Psychology. 2014:jsu043.
Stahl-Pele A, Strassburger K, Castillo K, Bachle C, Holl RW, Lange K, et al. Quality of
life in intensively treated youths with early-onset type 1 diabetes: a population-based
survey. Pediatric Diabetes. 2014;15(6):436-43.
Berg CA, Queen T, Butner JE, Turner SL, Lansing AH, Main A, et al. Adolescent Disclosure
to Parents and Daily Management of Type 1 Diabetes. Journal of Pediatric Psychology.
2016:jsw056.
Hartl AC, Seiffge-Krenke I, Laursen B. Body image mediates negative family climate and
deteriorating glycemic control for single adolescents with type 1 diabetes. Families,
Systems, & Health. 2015;33(4):363.
Caccavale LJ, Nansel TR, Quick V, Lipsky LM, Laffel LM, Mehta SN. Associations of
disordered eating behavior with the family diabetes environment in adolescents with
Type 1 diabetes. Journal of developmental and behavioral pediatrics: JDBP. 2015;36(1):8-13.
Wysocki T. Associations between affect and glycemia: a two-way street? Diabetes
technology & therapeutics. 2012;14(4):301-2.
Hilliard ME, Harris MA, Weissberg-Benchell J. Diabetes resilience: a model of risk and
protection in type 1 diabetes. Current diabetes reports. 2012;12(6):739-48.
Masten AS. Ordinary magic: Resilience processes in development. American
Psychologist. 2001;56(3):227.
Hoey H, Aanstoot HJ, Chiarelli F, Daneman D, Danne T, Dorchy H, et al. Good Metabolic
Control Is Associated With Better Quality of Life in 2,101 Adolescents With Type 1
Diabetes. Diabetes Care. 2001;24(11):1923-8.
Zeijl E, Crone M, Wiefferink K, Keuzenkamp S, Reijneveld M. Kinderen
in Nederland. Den Haag/Leiden: SCP/TNO Kwaliteit van Leven.; 2005 2005.
Blokland G. Over opvoeden gesproken: methodiekboek pedagogisch adviseren.
Amsterdam: Nederlands Jeugdinstituut Utrecht / SWP; 2010 2010.
Staal IIE, Van den Brink HAG, Hermanns JMA, Schrijvers AJP, Van Stel HF. Assessment of
parenting and developmental problems in toddlers: development and feasibility of a
structured interview. Child: care, health and development. 2011;37(4):503-11.
Bertrand R, Hermanns JMA, Leseman P. Behoefte aan opvoedingsondersteuning
in Nederlandse, Marokkaanse en Turkse gezinnen met kinderen van 0-6 jaar. Nederlands
tijdschrift voor opvoeding, vorming en onderwijs. 1998;14:50-71.
Miller WR, Rollnick S. Meeting in the middle: motivational interviewing and selfdetermination theory. Int J Behav Nutr Phys Act. 2012;9:25.
221.
222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
Kiecolt-Glaser JK, McGuire L, Robles TF, Glaser R. Psychoneuroimmunology:
psychological influences on immune function and health. Journal of consulting and
clinical psychology. 2002;70(3):537.
Netterstrom B, Danborg L, Olesen H. Glycated hemoglobin as a measure of physiological
stress. Behavioral Medicine. 1988;14(1):13-6.
Crane S, editor Understanding Ethical Issues of Research Participation From the
Perspective of Participating Children and Adolescents: STTI; 2016.
Snelson CL. Qualitative and Mixed Methods Social Media Research A Review of the
Literature. International Journal of Qualitative Methods. 2016;15(1):1609406915624574.
Coulon SJ, Velasco-Gonzalez C, Scribner R, Park CL, Gomez R, Vargas A, et al. Racial
differences in neighborhood disadvantage, inflammation and metabolic control in
black and white pediatric type 1 diabetes patients. Pediatric Diabetes. 2016.
Hoey H. Psychosocial factors are associated with metabolic control in adolescents:
research from the Hvidoere Study Group on Childhood Diabetes. Pediatr Diabetes.
2009;10 Suppl 13:9-14.
Cameron FJ, de Beaufort C, Aanstoot HJ, Hoey H, Lange K, Castano L, et al. Lessons from
the Hvidoere International Study Group on childhood diabetes: be dogmatic about
outcome and flexible in approach. Pediatric Diabetes. 2013;14(7):473-80.
Robinson JM, Trochim WM. An examination of community members', researchers' and
health professionals' perceptions of barriers to minority participation in medical
research: an application of concept mapping. Ethnicity and Health.2007;12(5):521-39.
George S, Duran N, Norris K. A systematic review of barriers and facilitators to minority
research participation among African Americans, Latinos, Asian Americans, and Pacific
Islanders. American journal of public health. 2014;104(2):e16-e31.
UyBico SJ, Pavel S, Gross CP. Recruiting vulnerable populations into research: a systematic
reviewofrecruitmentinterventions.Journalofgeneralinternalmedicine.2007;22(6):852-63.
Sheikh A. Why are ethnic minorities under-represented in US research studies? PLoS
Med. 2005;3(2):e49.
Bonevski B, Randell M, Paul C, Chapman K, Twyman L, Bryant J, et al. Reaching the
hard-to-reach: a systematic review of strategies for improving health and medical
research with socially disadvantaged groups. BMC medical research methodology.
2014;14(1):1.
Hanas R, John G. 2010 consensus statement on the worldwide standardization of the
hemoglobin A1C measurement. Clinical Chemistry and Laboratory Medicine.
2010;48(6):775-6.
Perantie DC, Lim A, Wu J, Weaver P, Warren SL, Sadler M, et al. Effects of prior
hypoglycemia and hyperglycemia on cognition in children with type 1 diabetes mellitus.
Pediatric Diabetes. 2008;9(2):87-95.
Tay J, Thompson CH, Brinkworth GD. Glycemic variability: assessing glycemia differently
and the implications for dietary management of diabetes. Annual review of nutrition.
2015;35:389-424.
Lewin AB, LaGreca AM, Geffken GR, Williams LB, Duke DC, Storch EA, et al. Validity and
reliability of an adolescent and parent rating scale of type 1 diabetes adherence behaviors:
The Self-Care Inventory (SCI). Journal of Pediatric Psychology. 2009;34(9):999-1007.
Fleuren MA, Paulussen TG, Van Dommelen P, Van Buuren S. Towards a measurement
instrument for determinants of innovations. International Journal for Quality in Health
Care. 2014;26(5):501-10.
151
References
238.
239.
240.
241.
Wisting L, Froisland DH, Skrivarhaug T, Dahl Jorgensen K, Ro O. Psychometric Properties, Norms, and Factor Structure of the Diabetes Eating Problem Survey-Revised in a Large
SampleofChildrenandAdolescentsWithType1Diabetes.DiabetesCare.2013;36(8):2198-202.
Engelen V, Detmar S, Koopman H, Maurice-Stam H, Caron H, Hoogerbrugge P, et al.
Reporting health-related quality of life scores to physicians during routine follow-up
visits of pediatric oncology patients: is it effective? Pediatric Blood Cancer. 2012;58(5):766-74.
Ravens-Sieberer U, Bettge S, Erhart M. Lebensqualitat von Kindern und
Jugendlichen-Ergebnisse aus der Pilotphase des Kinder-und Jugendgesundheitssurveys.
Bundesgesundheitsblatt-Gesundheitsforschung-Gesundheitsschutz.2003;46(4):340-5.
Sattler JM. Assessment of children's intelligence. Philadelphia, PA: WB Saunders. 1974.
152
Dankwoord
153
Dankwoord
Dankwoord
Onderzoek doe je niet alleen, zo sprak mijn copromotor vijf jaar geleden en ze had gelijk.
hartelijk dank voor de tijd en moeite die jullie hebben genomen om dit proefschrift te lezen en
Dat alleen mijn naam op de voorkant van dit proefschrift staat, voelt dan ook een beetje gek.
te beoordelen, en de bereidheid om over de inhoud te discussiëren.
Allereerst wil ik de deelnemers en hun ouders ontzettend bedanken voor hun participatie.
Lieve dipsy’s, ik zal jullie als betrokken onderzoeksteam gaan missen. Wilmy en Anne, dank
Het was niet niks, de jaarlijkse vragenlijsten en de testrondes. Ik hoop dat dit proefschrift
jullie wel voor jullie hulp met de data. Thomas, dank je wel voor je bereidheid mee te denken
een bijdrage levert aan de kwaliteit van de diabeteszorg.
en te helpen. Oud dipsy’s Michael, Wieke, Eelco, Emine, Gulçan, Tibor, Suzanne v E, Suzanne B
en Susanne T, bedankt voor de prettige samenwerking.
Dankzij de deskundigheid, het analytisch vermogen en het vertrouwen van Frank Snoek,
Ook collega’s van de D-vleugel en andere collega’s van het VUmc: Doortje, Jessica, Jay, Lianne,
mijn promotor, is het gelukt dit proefschrift samen te stellen. Hartelijk dank, Frank.
Albert, Louise, Karin, Brahim, Martin, Dick, Ilona, Maaike, Ilse, Albert, Ted, Johannes, Karen, Noami
Maartje de Wit, mijn copromotor, ontzettend bedankt voor je ondersteuning, het delen van je
en Patricia, hartelijk dank voor de gezelligheid en de verjaardagstraktaties.
kennis en je liefde voor de wetenschap. Ik heb het getroffen met jouw dagelijkse begeleiding.
Daarnaast wil ik de testleiders bedanken die de jaarlijkse neurocognitieve taken hebben
Joost Rotteveel, mijn tweede copromotor, hartelijk dank voor je betrokkenheid en expertise. Je
afgenomen. Aletta, Anne, Annemiek, Cathelijn, Daniëlle, Eline, Emma, Joyce, Kris, Lotte P, Lotte E,
informatieve en snelle reacties heb ik als erg prettig ervaren.
Madelon, Malinda, Marleen, Marjolein R, Marjolein H, Maxime, Monica, Nienke, Paulien, Pietsje,
Rozemarijn, Theresia en Thomas, bedankt voor de prettige samenwerking.
De artikelen in dit proefschrift zouden niet tot stand gekomen zijn zonder de kinderartsen
Kees de Boer en Marc Heuvelmans, dank jullie wel voor het opzetten van de digitale omgeving
betrokken bij het DINO onderzoek: Henk-Jan Aanstoot, Willie Bakker-van Waarde, Mieke Houdijk,
ten behoeve van het DINO onderzoek.
Roos Nuboer en Per Winterdijk. Hartelijk dank voor jullie hulp en tijd. Roos, bedankt dat je me
verschillende keren een platform hebt gegeven voor het delen van de resultaten.
Lieve Josine en Stefanie, wat is het fijn om jullie als paranimfen aan mijn zijde te hebben. Hartelijk
dank voor het delen van de hoogte- en dieptepunten de afgelopen vijf jaar. Een speciaal tienjarig
Het jaarlijks verzamelen van de medische gegevens was geen makkelijke opgave. Ik wil de
jubileum zo, Josine! Stefanie, wat heb ik genoten van je humor en relativeringsvermogen.
betrokken diabetesteams en de diabetesverpleegkundigen dan ook bedanken voor hun hulp.
Marlies, Pien en Anne-Marije, dank jullie wel dat ik meer dan eens aan heb kunnen schuiven om te
Ook in een leven zonder diabetes is een sociaal netwerk een belangrijke protectieve factor.
ervaren wat diabetes in de pediatrische praktijk betekent. Ook Kim Oostrom, Jaap Oosterlaan en
Lieve vrienden en familie, ontzettend bedankt voor het luisteren, meedenken, adviseren
Marjolein Luman wil ik bedanken voor hun bijdrage aan de totstandkoming van de neurocognitieve
en sparren over de wonderenwereld van de wetenschap. Vellah, Dorette, Mick en Loekie,
testbatterij. Jos Twisk, hartelijk dank voor het advies met betrekking tot de statistiek.
dank jullie wel voor de puntjes op de i. En wat is de voorkant mooi geworden, Linda.
Kim, wat is het heerlijk om jou als zus te hebben! Bedankt dat ik jouw gezicht heb mogen
Prof.dr. Reinoud Gemke, dr. Jennifer van Dijk, dr. Martijn Finken, prof.dr. Martha Grootenhuis,
gebruiken voor veel van mijn presentatie- en wervingsmateriaal. Je was een informatief en
prof.dr. Hans Koot, prof.dr. Koen Luyckx, prof.dr. Arne Popma en dr. Chris Verhaak: allen
humoristisch klankbord.
154
155
Dankwoord
Wat tref ik het met zulke schoonouders! Dank jullie wel voor jullie betrokkenheid en het zijn
van een zorgzame en flexibele oppas.
Lieve pap, mam, mam en mam, ontzettend bedankt voor jullie ondersteuning de afgelopen
jaren en het vertrouwen in het feit dat een jaar stoppen met studeren na de HvA een goed
idee was. Wisten wij veel dat het zou leiden tot een promotieplechtigheid! En wat is het fijn
om een creatief talent als moeder te hebben. Meer dan eens heb je me geholpen met de
vormgeving van (wervings)materiaal, hartelijk dank.
Gideon en Fija, liefdes van mijn leven, ontzettend bedankt voor jullie inspiratie en motivatie.
Lieve Gi, dit proefschrift was er niet geweest zonder jouw grenzeloze support. Waar dan ook
vind ik bij jullie mijn thuis. Wat een geluk!
156
Minke Eilander was born on August 16, 1985
in Amsterdam. After het graduation from
Het Amsterdams Lyceum, she studied Social
Work at the Hogeschool van Amsterdam.
In 2006 she started Child Development and
Education at the University of Amsterdam,
and obtained her Master of Science degree
in 2011. While completing her master thesis,
that focused on the general practitioners
contribution to parenting questions, Minke
became interested in scientific research
and in medical psychology. After working
as a research assistant at the mental health
clinic GGZ inGeest, she started at the VU
university medical center as junior researcher
in 2012 and in 2013 as PhD student. First, she
evaluated the implementation of the quality
of life assessment DAWN MIND-Youth in the
routine pediatric diabetes care. A year later
the research project DINO was funded and
under the direction of Prof. dr. F. J. Snoek and
dr. M. de Wit, Minke studied the psychosocial,
biological, cognitive and family functioning
of youth diagnosed with type 1 diabetes.
Diabetes IN Ontwikkeling