Accepted Article
Received Date : 19-Oct-2014
Revised Date : 13-Mar-2015
Accepted Date : 26-May-2015
Article type
: Research Article
Research: Education and psychological aspects
Effectiveness of tailored support for people with Type 2
diabetes after a first acute coronary event: a multicentre
randomized controlled trial (the Diacourse-ACE study)
Short title: Tailored support for people with Type 2 diabetes following an acute coronary event
M.J. Kasteleyn1, R.C. Vos1, M. Rijken2, F.G. Schellevis2 and G.E.H.M. Rutten1
1
Julius Centre for Health Sciences and Primary Care, University Medical Centre Utrecht and
2
NIVEL, Netherlands Institute for Health Services Research, Utrecht, The Netherlands
Accepted
Correspondence to: M.J. Kasteleyn [email protected]
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/dme.12816
This article is protected by copyright. All rights reserved.
Accepted Article
What's new?
•
People with Type 2 diabetes experienced a low level of diabetes-related distress after a
first acute coronary event; distress remained low at 5 months and was not influenced by a
tailored supportive intervention.
•
Although people with Type 2 diabetes had a less favourable health status after hospital
discharge, a tailored support intervention improved health status to an adequate level.
•
People with Type 2 diabetes viewed an acute coronary event as a comorbidity rather than
as a complication of their diabetes.
Abstract
Aims To evaluate the effectiveness of a tailored, supportive intervention strategy in
influencing diabetes-related distress, health status, well-being and clinical outcomes in people
with Type 2 diabetes shortly after a first acute coronary event.
Methods: People with Type 2 diabetes and a recent first acute coronary event (n=201) were
randomized to the intervention group (three home visits by a diabetes nurse) or the attention
control group (one telephone consultation). Outcomes were measured after discharge
(baseline) and at 5 months (follow-up) using validated questionnaires for diabetes-related
distress (Problem Areas in Diabetes), well-being (WHO Well-Being Index) and health status
(Euroqol 5 Dimensions; Euroqol Visual Analogue Scale).
ANCOVA
was used to analyse
change-over-time differences between groups.
Results: Follow-up data were available for 81 participants in the intervention group (mean ±
SD
age 66.0±9.3 years, 76% male) and 80 in the control group (mean ±
SD
age 65.6±9.4
years, 75% male) participants. Diabetes-related distress was low after hospital discharge
(mean ± SD score intervention group: 8.2±10.1; control group: 9.2±12.4) and did not change
This article is protected by copyright. All rights reserved.
Accepted Article
illness/death of a close relative) or the demanding nature of the questionnaires included in the
study (questionnaires require ~1 h to complete). Another limitation was the small sample size
for subgroup analyses; therefore, no robust conclusions could be drawn, although they do
provide insight into gender differences. An additional conceivable limitation was that, as we
generally depended on hospital-based nurses to invite patients, these nurses might have
excluded patients whom they considered unsuitable for the study. The participants in the
present study, however, were similar to other patients with regard to treatment (percutaneous
transluminal coronary angioplasty, coronary artery bypass graft or non-invasive intervention)
after ACE [32], suggesting that participants in our study were, at least in terms of severity of
ACE, representative of those who had experienced an ACE in the Netherlands. A further
shortcoming of the study was that the control group was not matched to the intervention
group in terms of exposure to healthcare providers, with the control group receiving only one
telephone consultation compared with three home visits, and can therefore not be considered
a true attention control group in the strict meaning of the term. This shortcoming was deemed
acceptable because performing three home visits without giving support would be both
impractical and would constitute an intervention of itself. Our attention control group
therefore reflected some, but not all, of the characteristics of the intervention group. The short
follow-up period might be considered a limitation, making it difficult to determine the
duration of any treatment effect. Consequently, we were not able to determine whether
patients with an improved health status were less prone to future events or whether patients
with a poor health status show long-term deterioration and an increased risk of mortality and
morbidity [29].
In conclusion, the present study showed that people with Type 2 diabetes with a recent first
ACE show low levels of diabetes-related distress 5 months after hospital discharge. Wellbeing and health status were clearly less favourable after discharge from hospital and a
This article is protected by copyright. All rights reserved.
Accepted Article
efficacious individuals are less likely to perceive their illnesses as controllable and are less
able to maintain adequate self-management [6]. Because of these problems, people with Type
2 diabetes may experience an increase in diabetes-related distress and a decrease in health
status and well-being after an ACE. It has therefore been suggested that interventions are
needed that can improve outcomes in these patients [7], especially as patients and their
partners often lack information on the combination of Type 2 diabetes and ACE, generally
receive insufficient professional healthcare support, and would welcome a tailored
intervention [8]. We hypothesized that an intervention of this type could reduce distress,
improve self-management, and enhance psychological well-being. The aim of the present
study, therefore, was to evaluate the effectiveness of a supportive intervention in reducing
diabetes-related distress and improving both well-being and health status in people with Type
2 diabetes and a recent first ACE.
Participants and methods
Participants and setting
People with Type 2 diabetes and a first ACE were recruited, via their cardiologists, at 13
hospitals in three regions across the Netherlands, between October 2011 and August 2013.
An ACE was defined as a percutaneous transluminal coronary angioplasty, a coronary artery
bypass graft procedure and/or a myocardial infarction. Eligible patients were invited to
participate within 2 weeks after hospital discharge if: 1) they had Type 2 diabetes (>1 year);
2) were > 35 years; 3) had sufficient knowledge of the Dutch language; 4) were able to fill in
questionnaires; and 5) had no other serious illnesses or conditions that would prevent full
participation. To ensure that participants were unaware that the study had two branches (see
below), a modified informed consent procedure was used [9]: participants were told that a
This article is protected by copyright. All rights reserved.
Accepted Article
programme was being developed to support patients with Type 2 diabetes and an ACE after
hospital discharge, but were not given any further details. Once consent had been received,
the participants were randomized in a 1:1 ratio to either the intervention or the control group.
Randomization stratified by region was carried out at the participant level using a
computerized random-number generator at the research centre. The study was approved by
the Medical Ethical Committee of the University Medical Centre Utrecht and registered at
Clinicaltrials.gov (NCT01801631).
Intervention and attention control group
Full details of the design, rationale and theoretical framework of this randomized controlled
trial have previously been described [10]. The intervention focused on increasing selfefficacy and improving illness perceptions, both of which are expected to bring about
improvements in participants’ self-management which, in turn, is expected to lower diabetesrelated distress. The occurrence of an ACE also affects partners, and because partners
subsequently influence the way patients cope with their conditions, they were also involved
in the intervention [4,11]. Seven trained diabetes nurses visited participants in the
intervention group in their homes three times to discuss illness perceptions and to use
strategies of motivational interviewing to increase self-efficacy (Table 1). Participants in the
attention control group [12] received one consultation (of ~15 min) by telephone, within 3
weeks after discharge, to discuss how they were feeling and functioning.
This article is protected by copyright. All rights reserved.
Accepted Article
Outcomes
Participants completed a set of questionnaires at home, after hospital discharge but before the
first home visits/phone consultation (baseline), and 5 months later (follow-up). The primary
outcome was diabetes-related distress, which was measured using the Problem Areas in
Diabetes (PAID) scale. This is a self-reported questionnaire measuring common negative
emotions related to living with diabetes (e.g. 'feeling discouraged by your diabetes regimen')
[13]. Each item was rated on a five-point Likert scale, ranging from 0 ('not a problem') to 4
('a serious problem'). The total score was transformed to a 0–100 scale, with higher scores
representing greater distress. The questionnaire is well validated as responsive to change in a
heterogeneous group of people with diabetes [14]. Secondary outcomes were well-being
[WHO Well-Being Index (WHO-5) questionnaire: five items, range 0–100] [15], health
status [Euroqol 5 Dimensions questionnaire (EQ-5D); five items, range -0.594 to 1.00, and
Euroqol Visual Analogue Scale (EQ-VAS); one item, range 0–100] [16], anxiety and
depression [Hospital Anxiety and Depression Scale (HADS): seven items measuring anxiety,
seven items measuring depression, range 0–21 for each scale) [17] and clinical variables
(HbA1c, blood pressure and cholesterol levels). The questionnaires are described in detail in
Appendix 1. Clinical variables were retrieved from the hospital discharge letter at baseline
and, using a case report form, from the primary care physician 5 months later. Mediating
variables were self-efficacy [Confidence in Diabetes Self-Care scale (CIDS); 20 items, range
0–100] [18], illness perceptions [Illness Perception Questionnaire (IPQ)-short version] for
diabetes and ACE separately (eight items, range 0–80, higher scores represent a more
threatening view of the condition) [19] and self-management [Summary of the Diabetes SelfCare Activities Measure (SDSCA); 10 items, range 1–7] [20] (Appendix S1). Partner support
[Active Engagement, Protective Buffering and Overprotection (ABO); 19 items, range for
each scale 1–5) was also assessed [21]. The home-visit discussion topics considered
This article is protected by copyright. All rights reserved.
Accepted Article
important by the participants in the intervention group were examined, and the extent to
which participants rated the intervention as having added value in relation to the hospital
cardiac rehabilitation programme was evaluated (on a four-point Likert scale, ranging from
'not at all' to 'very much').
Statistical analysis
To detect a clinically relevant change in diabetes-related distress at a two-sided significance
level of 5%, a power of 80% and a correlation of ρ=0.3 ('medium' effect size) between
baseline and follow-up PAID scores, we calculated that a sample size of 77 per group was
required [10]. Data were analysed according to the intention-to-treat principle, including all
participants for whom a follow-up assessment was available. For participants with only
baseline scores (20%), the 'last value carried forward' principle was not used because we
assumed that perceptions are not stable over time and might fluctuate, particularly in the
dynamic period following an ACE. In addition, no data imputation was performed because
complete case analysis with covariate adjustment will give similar estimates in the event of
missing outcomes when the same predictors of missingness are included [22]. Using ANCOVA
with change scores, change from baseline to 5 months after discharge was analysed for the
primary and secondary outcomes and for the mediating variables. In the model, treatment
allocation (intervention or control group) was included as factor and the baseline score on the
questionnaire as covariate (basic model). If an intervention effect on distress was found in the
basic
ANCOVA,
the change score of the mediating variable was entered to the model to
explore whether the change in the mediating variable explained the change in distress. The
Holm-Bonferroni correction was used to adjust for multiple testing [23]. Prespecified
exploratory subgroup analyses [10] were performed to assess the impact of age (young vs.
This article is protected by copyright. All rights reserved.
Accepted Article
old, cut-off: median), gender (man vs. woman), home situation (living alone vs. not living
alone) and type of ACE (group 1: myocardial infarction + coronary artery bypass graft/only
coronary artery bypass graft; group 2: myocardial infarction + percutaneous transluminal
coronary angioplasty /only percutaneous transluminal coronary angioplasty; group 3:
myocardial infarction/unstable angina pectoris without invasive intervention) on the
outcomes. Furthermore, an exploratory subgroup analysis was performed to assess the impact
of the presence of microvascular complications at baseline (present vs. not present) on the
outcomes. SPSS version 20.0 was used for all analyses.
Results
Of the 264 individuals with both Type 2 diabetes and an ACE who were invited to
participate, 63 (24%) declined (Figure 1). The remaining participants (n=201) were randomly
allocated to either the intervention or the control group. Follow-up data were available for 81
participants in the intervention group and 80 in the control group. The mean ±SD age of
participants with follow-up was 65.8±12.9 and the mean diabetes duration was 8.0±9.5 years.
Further characteristics are shown in Table 2. A total of 40 participants failed to complete the
follow-up questionnaire, with dropout rates similar for the intervention and control groups
(intervention group: 19%; control group: 20%). Compared with those who continued in the
study, participants who dropped out (mean ±
SD
age 65.8±12.9; mean diabetes duration
10.9±9.5 years) were more often female (50 vs. 25%; P=0.001), more likely to be living
alone (35 vs. 18%; P=0.015), more likely to have microvascular complications (41 vs. 24%;
P=0.045) and had higher scores on the HADS depression scale (HADS depression: median
8.0 vs. 3.0; P=0.006). These variables were not influenced by the status of a dropout
This article is protected by copyright. All rights reserved.
Accepted Article
(intervention or control group). No significant differences were found between dropouts and
full participants regarding any other variable.
The discussion topics most important to participants during the home visits were
physical activity, nutrition, (depressive) feelings and coping with the consequence of the
ACE in daily life. Of the participants in a cardiac rehabilitation programme (54%: median
start 41 days after hospital discharge), 62% of intervention group members rated our
intervention as having added value, compared with 30% of the control group. The
individualized and timely character of the intervention received particular praise.
Distress
Low levels of diabetes-related distress were experienced at baseline (mean ±SD score:
intervention group 8.2±10.1, control group 9.2±12.4). In both groups, diabetes-related
distress remained low and no significant group effect was found (F=0.52, P=0.470; Table 3).
Subgroup analyses did not show differences between men and women, older and younger
participants, participants living alone or living together, or based on type of ACE (data not
shown).
Secondary outcomes
Feelings of anxiety and depression were low in all patients after hospital discharge and did
not change. At baseline, participants experienced relatively low levels of well-being and a
poor health status; however, the intervention group showed significant improvements in wellbeing (mean ±SD score: baseline 58.5±28.0, follow-up 65.5±23.7; P=0.005), in contrast to the
This article is protected by copyright. All rights reserved.
Accepted Article
control group (mean ±SD score: baseline 57.5±25.2, follow-up 59.6±24.4; P=0.481). No
significant group effect on well-being was found (F=3.56, P=0.061). Health status measured
on the EQ-VAS also improved in the intervention group (mean ±SD score: baseline
69.9±17.3, follow-up 76.8±15.6; P<0.001) but not in the control group (mean ±SD score:
baseline 68.6±15.9, follow-up 69.9±16.7; P=0.470), with a significant group effect in favour
of the intervention group at 5 months (F=8.80, P=0.004). Health status assessed with the EQ5D improved in both groups, but no group effect was found (F=0.66, P=0.417; Table 3).
Subgroup analyses showed differences between men (n=122) and women (n=39)
regarding well-being and health status (EQ-VAS). A trend in well-being (F=3.233, P=0.075)
was found in men, with a significant improvement in the intervention group (mean ±SD score:
baseline 75.6±28.2, follow-up 65.6±22.9; P=0.006), but no change in control group (mean
±SD score: baseline 59.1±23.9; follow-up 60.2±23.2; P=0.727). In women (mean ±SD
baseline score: intervention group 61.5±27.9, control group 53.0±28.8), no group effect was
found on well-being (mean ±SD follow-up score: intervention group 65.1±27.0, control group
58.0±28.1; F=0.146, P=0.705). In men, a group effect was found on health status (F=9.57,
P=0.003), with the intervention group improving significantly (mean ±SD score: baseline
70.4±16.5, follow-up 78.5±15.5; P<0.001), while the control group did not change (mean
±SD score: baseline 70.8±12.0, follow-up 71.8±14.8; P=0.610). In women (mean ±SD
baseline score: intervention group 68.1±19.6, control group 61.4±23.7), no group effect was
found on health status (mean ±SD follow-up score: intervention group 71.6±15.4, control
group 64.0±28.1; F=0.691, P=0.412). Differences in health status were apparent in
participants with percutaneous transluminal coronary angioplasty with/without myocardial
infarction (n=100), with coronary artery bypass graft with/without myocardial infarction
(n=30) and myocardial infarction without invasive intervention (n=31). A group effect, in
favour of the intervention group (baseline: 68.3±16.0; follow-up: 76.6±16.0), was found for
This article is protected by copyright. All rights reserved.
Accepted Article
health status in patients with percutaneous transluminal coronary angioplasty (F=6.111,
P=0.015), while a trend was found in participants with coronary artery bypass graft (mean ±
SD
score: baseline 69.2±17.2, follow-up 79.4±15.7; F=3.89, P=0.058). Participants with
myocardial infarction without invasive intervention, however, showed no group effect
(F=0.087, P=0.770). Different effects on health status were found between participants with
and without microvascular complications at baseline. A group effect in favour of the
intervention group was found for participants without microvascular complications (mean
±SD scores in the intervention group: baseline 70.0±15.2, follow-up 78.1±18.3; F=9.466,
P=0.003), while no group effect was found for participants with microvascular complications
at baseline (mean ±SD scores in the intervention group: baseline 69.5±22.2, follow-up
73.6±18.3; F=0.643, P=0.420). No differences between subgroups were found on the other
outcomes or in the other subgroups.
No significant differences between baseline and follow-up were found in either the
intervention or the control group with regard to clinical variables (HbA1c, blood pressure and
cholesterol; Table 4).
Mediating variables
Mean self-efficacy scores were relatively high at baseline and did not change after 5 months.
Baseline scores on self-management were similar for the intervention and the control group
and did not change. Regarding illness perceptions, the intervention group considered Type 2
diabetes less threatening at 5 months than at baseline (Table 3). In contrast, the control group
perceived their Type 2 diabetes as more threatening at 5 months than at baseline. A group
effect was found in favour of the intervention group (F=3.93, P=0.049). At baseline,
This article is protected by copyright. All rights reserved.
Accepted Article
participants in both the intervention and control groups considered their ACE more
threatening than their Type 2 diabetes (mean ± SD IPQ score for ACE in intervention group:
35.6±13.7 vs. IPQ score for diabetes in intervention group: 32.6±10.8, P=0.004; IPQ score
for ACE in control group: 37.9±13.5 vs. IPQ score for diabetes in control group: 32.5±12.3,
P=0.002). Five months later, this within-group difference was no longer significant for either
group (IPQ score for ACE in intervention group: 33.1±15.5 vs. IPQ score for diabetes in
intervention group: 31.2±11.4, P=0.052; IPQ score for ACE in control group: 36.8±13.6 vs.
IPQ score for diabetes in control group: 33.7±12.7, P=0.056). Of the participants, only 22%
pointed to diabetes as one of the three most important causes of the ACE. Regarding partner
support, the scale measuring overprotection showed an improvement in the intervention
group (baseline: 2.1±0.6 vs. follow-up:1.8±0.6, P<0.001), while no differences were found in
the control group. Other measures of partner support such as active engagement and
protective buffering showed no differences in either group (Table 3).
Discussion
The present study assessed the effectiveness of an intervention aimed at people with Type 2
diabetes who had recently experienced their first ACE. We hypothesized that, after
experiencing an ACE, people with Type 2 diabetes would show an increase in diabetesrelated distress and a decline in health status and well-being. Our theoretical framework
predicts that increased self-efficacy and improved illness perceptions should lead to
improvements in self-management and result in less diabetes-related distress; however, we
found that people with Type 2 diabetes already experience low levels of diabetes-related
distress after hospital discharge, which is maintained for at least 5 months. In contrast, while
well-being and health status were clearly less favourable after hospital discharge, they could
This article is protected by copyright. All rights reserved.
Accepted Article
both be improved through intervention. Rather unexpectedly, levels of diabetes-related
distress in our study group were no higher than those of other people with Type 2 diabetes in
the Netherlands [24,25]. Distress is more likely to increase from a low baseline distress level
than to decrease. One possible explanation for the low diabetes-related distress post-discharge
may be the impact of the ACE overshadowing diabetes-related concerns. In the present study,
participants did indeed perceive their ACE as more threatening than Type 2 diabetes.
Surprisingly, only one in five participants cited their diabetes as a possible cause of the ACE,
suggesting that people with Type 2 diabetes see an ACE as unrelated to their diabetes. In
retrospect, the problems people with Type 2 diabetes encounter after a first ACE are probably
not captured by diabetes-specific questionnaires such as PAID, so general distress, well-being
or health status would presumably have been better primary outcome measures.
The health status of the participants in the present study was less favourable when
compared with people with Type 2 diabetes without a recent ACE (mean scores of between
75.3 and 78.7 for EQ-VAS and between 0.82 and 0.86 for EQ-5D) [26,27]. In addition,
patients with a history of myocardial infarction, angina pectoris or a less recent coronary
artery bypass graft reported a better health status than our study participants with their very
recent ACE; however, the health status of people with Type 2 diabetes with a less recent
ACE was poorer than when a comorbidity was absent [3]. This indicates that a recent ACE
has the greatest impact on health status. After our intervention, participants reported a health
status that was similar to that of patients without a comorbidity [3]. Given that the health
status of the participants in our attention control group remained low after 5 months, one
could question whether their health status would improve spontaneously over time. As low
self-reported health status is an important predictor of mortality risk, future events and other
complications in people with Type 2 diabetes [28], our intervention supports those who might
otherwise face a high risk because of a lack of improvement in health status. Involving the
This article is protected by copyright. All rights reserved.
Accepted Article
partners, with a decrease in overprotection as a result, may have contributed to the improved
health status. A divergence in the illness perceptions of a patient and their partner has an
impact on how they each cope with the illness in daily life. It is therefore important to involve
close relatives in the management of disease [11].
The present results show certain similarities to a study by Frasure-Smith et al. [29]
that assessed the effectiveness of tailored, home-based support aimed at reducing patient
distress after a myocardial infarction. These authors reported a marginal reduction in
depression and anxiety in men, and no effect in women. As in the present study, FrasureSmith et al. included fewer female than male participants and they showed that women were
more difficult to treat than men [29]. The authors also suggested that men and women might
benefit from different supportive approaches, with men requiring more directive approaches,
while women derive greater benefit from listening [30]. Although our intervention was
tailored to the specific situation of the patient, and patients were invited to describe their
expectations of the diabetes nurse, we did not differentiate between men and women in our
intervention protocol for the diabetes nurses.
Small improvements in blood pressure, HbA1c and cholesterol levels were found. Ho
et al. [31] showed that a multifaceted intervention was successful in increasing the proportion
of patients who have experienced an ACE adhering to medication regimens 1 year after
hospital discharge, although this had no impact on blood pressure and LDL cholesterol levels.
Likewise, the intervention in the present study might also reduce the risk of discontinuation
of medication.
A limitation of the present study was the 20% dropout rate; however, this was
anticipated in the sample size calculation. Dropout rates were similar in both groups and the
reasons for dropout seemed to be independent of the intervention. The most often indicated
reasons were being too busy, lack of interest, change in personal circumstances (e.g.
This article is protected by copyright. All rights reserved.
Accepted Article
illness/death of a close relative) or the demanding nature of the questionnaires included in the
study (questionnaires require ~1 h to complete). Another limitation was the small sample size
for subgroup analyses; therefore, no robust conclusions could be drawn, although they do
provide insight into gender differences. An additional conceivable limitation was that, as we
generally depended on hospital-based nurses to invite patients, these nurses might have
excluded patients whom they considered unsuitable for the study. The participants in the
present study, however, were similar to other patients with regard to treatment (percutaneous
transluminal coronary angioplasty, coronary artery bypass graft or non-invasive intervention)
after ACE [32], suggesting that participants in our study were, at least in terms of severity of
ACE, representative of those who had experienced an ACE in the Netherlands. A further
shortcoming of the study was that the control group was not matched to the intervention
group in terms of exposure to healthcare providers, with the control group receiving only one
telephone consultation compared with three home visits, and can therefore not be considered
a true attention control group in the strict meaning of the term. This shortcoming was deemed
acceptable because performing three home visits without giving support would be both
impractical and would constitute an intervention of itself. Our attention control group
therefore reflected some, but not all, of the characteristics of the intervention group. The short
follow-up period might be considered a limitation, making it difficult to determine the
duration of any treatment effect. Consequently, we were not able to determine whether
patients with an improved health status were less prone to future events or whether patients
with a poor health status show long-term deterioration and an increased risk of mortality and
morbidity [29].
In conclusion, the present study showed that people with Type 2 diabetes with a recent first
ACE show low levels of diabetes-related distress 5 months after hospital discharge. Wellbeing and health status were clearly less favourable after discharge from hospital and a
This article is protected by copyright. All rights reserved.
Accepted Article
tailored supportive intervention resulted in an adequately improved health status at 5 months.
Well-being also improved in the intervention group compared with the control group,
although the group effect was not significant.
Funding sources
This study was funded by the Dutch Diabetes Research Foundation (Diabetes Fonds)
2009.70. The foundation’s international review committee critically reviewed the design of
the Dutch Diacourse Study and discussed it with the authors before grant assignment. The
authors are entirely responsible for the conduct of the study, all study analyses, and the
drafting and editing of the manuscript.
Competing interests
None declared.
Acknowledgements
The authors wish to thank all participants in the study, the diabetes nurses at the Stichting
Huisartsen Laboratorium and the Diabetes Zorgsysteem, the diabetes nurses who conducted
the home visits, and the 13 hospitals involved in the recruitment of participants. Other
members of the Diacourse study group are also acknowledged: Kees J. Gorter, MD, PhD
(University Medical Center Utrecht, Utrecht); Anne L. van Puffelen, MSc; Monique J. W. M.
Heijmans, PhD (NIVEL, Utrecht); Lianne de Vries, MSc; Amber A. W. A. van der Heijden,
PhD; Caroline A. Baan, PhD; Giel Nijpels, MD, PhD (VU University Medical Center,
Amsterdam).
This article is protected by copyright. All rights reserved.
Accepted Article
References
[1] Watkins KW, Connell CM, Fitzgerald JT, Klem L, Hickey T, Ingersoll-Dayton B. Effect of
adults' self-regulation of diabetes on quality-of-life outcomes. Diabetes Care 2000; 23: 1511–
1515.
[2] Wu CJ, Chang AM, McDowell J. Perspectives of patients with type 2 diabetes following a
critical cardiac event - an interpretative approach. J Clin Nurs 2008; 17(5A): 16–24.
[3] Wermeling PR, Gorter KJ, van Stel HF, Rutten GE. Both cardiovascular and noncardiovascular comorbidity are related to health status in well-controlled type 2 diabetes
patients: a cross-sectional analysis. Cardiovasc Diabetol 2012; 11: 121.
[4] Dalteg T, Benzein E, Fridlund B, Malm D. Cardiac disease and its consequences on the
partner relationship: a systematic review. Eur J Cardiovasc Nurs 2011; 10: 140–149.
[5] Leventhal H, Nerenz D, Steele D. Illness representatations and coping with health threats. In
Baum A, Taylor S, Singer J eds. Handbook of Psychology and Health, Volume VI: social
psychological aspects of health. Hillsdale, NJ: Erbaum, 2008: 219–252.
[6] Schuz B, Wurm S, Warner LM, Ziegelmann JP. Self-efficacy and multiple illness
representations in older adults: a multilevel approach. Psychol Health 2012; 27: 13–29.
[7] Peterson PN, Spertus JA, Magid DJ, Masoudi FA, Reid K, Hamman RF et al. The impact of
diabetes on one-year health status outcomes following acute coronary syndromes. BMC
Cardiovasc Disord 2006; 6: 41.
[8] Kasteleyn MJ, Gorter KJ, van Puffelen AL, Heijmans M, Vos RC, Jansen H et al. What
follow-up care and self-management support do patients with type 2 diabetes want after their
first acute coronary event? A qualitative study. Prim Care Diabetes 2014; 8: 195–206.
This article is protected by copyright. All rights reserved.
Accepted Article
[9] Boter H, van Delden JJ, de Haan RJ, Rinkel GJ. Modified informed consent procedure:
consent to postponed information. BMJ 2003; 327: 284–285.
[10] Kasteleyn MJ, Gorter KJ, Stellato RK, Rijken M, Nijpels G, Rutten GE. Tailored support for
type 2 diabetes patients with an acute coronary event after discharge from hospital - design
and development of a randomised controlled trial. Diabetol Metab Syndr 2014; 6: 5.
[11] Rutten GE. What effect does diabetes have on the family–do you know? Lancet Diabetes
Endocrinol 2014; 2: 191–192.
[12] Lindquist R, Wyman JF, Talley KM, Findorff MJ, Gross CR. Design of control-group
conditions in clinical trials of behavioral interventions. J Nurs Scholarsh 2007; 39: 214–221.
[13] Polonsky WH, Anderson BJ, Lohrer PA, Welch G, Jacobson AM, Aponte JE et al.
Assessment of diabetes-related distress. Diabetes Care 1995; 18: 754–760.
[14] Welch G, Weinger K, Anderson B, Polonsky WH. Responsiveness of the Problem Areas In
Diabetes (PAID) questionnaire. Diabet Med 2003; 20: 69–72.
[15] Bech P, Gudex C, Johansen KS. The WHO (Ten) Well-Being Index: validation in diabetes.
Psychother Psychosom 1996; 65: 183–190.
[16] Rabin R, de CF. EQ-5D: a measure of health status from the EuroQol Group. Ann Med 2001;
33: 337–343.
[17] Spinhoven P, Ormel J, Sloekers PP, Kempen GI, Speckens AE, van Hemert AM. A validation
study of the Hospital Anxiety and Depression Scale (HADS) in different groups of Dutch
subjects. Psychol Med 1997; 27: 363–370.
[18] Ven Nv, 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 selfefficacy in Dutch and US patients with type 1 diabetes. Diabetes Care 2003; 26: 713–718.
This article is protected by copyright. All rights reserved.
Accepted Article
[19] de Raaij EJ, Schroder C, Maissan FJ, Pool JJ, Wittink H. Cross-cultural adaptation and
measurement properties of the Brief Illness Perception Questionnaire-Dutch Language
Version. Man Ther 2012; 17: 330–335.
[20] Toobert DJ, Hampson SE, Glasgow RE. The summary of diabetes self-care activities
measure: results from 7 studies and a revised scale. Diabetes Care 2000; 23: 943–950.
[21] Buunk AP, Berkhuysen MA, Sanderman R, Nieuwland W, Ranchor AV. Actieve
betrokkenheid, beschermend bufferen en overbescherming. Meetinstrumenten voor de rol van
de partner bij hartrevalidatie. [Active engagement, protective behavior, and overprotection.
Assessment of the spouse's role in heart revalidation.]. Gedrag Gezondheid 1996; 24: 304–
313. [Article in Dutch.]
[22] Groenewold RHH, Donders ART, Roes KCB, Harrell FE, Moons KGM. Dealing with
missing outcome data in randomized controlled trials and observational studies. Am J
Epidemiol 2012; 175: 210–217.
[23] Holm S. A simple sequentially rejective multiple test procedure. Scand J Stat 1979; 6: 65–70.
[24] Wermeling PR, Janssen J, Gorter KJ, Beulens JW, Rutten GE. Satisfaction of well-controlled
type 2 diabetes patients with three-monthly and six-monthly monitoring. BMC Fam Pract
2013; 14: 107.
[25] Stoop CH, Nefs G, Pop VJ, Wijnands-van Gent CJ, Tack CJ, Geelhoed-Duijvestijn PH et al.
Diabetes-specific emotional distress in people with Type 2 diabetes: a comparison between
primary and secondary care. Diabet Med 2014; 31:1252–1259.
[26] Cleveringa FG, Minkman MH, Gorter KJ, van den DM, Rutten GE. Diabetes Care Protocol:
effects on patient-important outcomes. A cluster randomized, non-inferiority trial in primary
care. Diabet Med 2010; 27: 442–450.
This article is protected by copyright. All rights reserved.
Accepted Article
[27] van den Donk M, Griffin SJ, Stellato RK, Simmons RK, Sandbaek A, Lauritzen T et al.
Effect of early intensive multifactorial therapy compared with routine care on self-reported
health status, general well-being, diabetes-specific quality of life and treatment satisfaction in
screen-detected type 2 diabetes mellitus patients (ADDITION-Europe): a cluster-randomised
trial. Diabetologia 2013; 56: 2367–2377.
[28] Wennberg P, Rolandsson O, Jerden L, Boeing H, Sluik D, Kaaks R et al. Self-rated health
and mortality in individuals with diabetes mellitus: prospective cohort study. BMJ Open
2012; 2: e000760.
[29] Frasure-Smith N, Lesperance F, Prince RH, Verrier P, Garber RA, Juneau M et al.
Randomised trial of home-based psychosocial nursing intervention for patients recovering
from myocardial infarction. Lancet 1997; 350: 473–479.
[30] Cossette S, Frasure-Smith N, Lesperance F. Nursing approaches to reducing psychological
distress in men and women recovering from myocardial infarction. Int J Nurs Stud 2002; 39:
479–494.
[31] Ho PM, Lambert-Kerzner A, Carey EP, Fahdi IE, Bryson CL, Melnyk SD et al. Multifaceted
intervention to improve medication adherence and secondary prevention measures after acute
coronary syndrome hospital discharge: a randomized clinical trial. JAMA Intern Med 2014;
174: 186–193.
[32] Boersma E, Manini M, Wood DA, Bassand J, Simoons ML. Cardiovascular diseases in
Europe. Euro Heart Survey and National Registries of Cardiovascular Diseases and Patient
Management. Sophia Antipolis: European Society of Cardiology, 2002,
This article is protected by copyright. All rights reserved.
Accepted Article
Table 1 Characteristics of the tailored supportive intervention
Visit1
Components
Description
Problem mapping
From a list of 10 topics, the patient indicated to what extent problems
Timing: <3 weeks after
were
experienced:
1)
physical
activity;
2)
sexuality;
3)
discharge
pharmacotherapy; 4) monitoring scheme with different healthcare
Aimed duration: 65 min
professionals; 5) coping together with the partner; 6) coping with
diabetes in daily life; 7) coping with the ACE in daily life; 8)
(depressive) feelings; 9) nutrition/diet; 10) other problems
In depth discussion
The nurse, patient and partner discussed, in depth, the three topics that
were considered by the patient to be most important
Goal setting
The patient set goals he/she wished to achieve in the following 2 weeks
Homework
The patient was asked to keep a daily log to track strategies for coping
with events related to the topics discussed. To adjust possible
misperceptions about diabetes, ACE and their relationship, patients
were asked to fill in questions assessing their perceptions of diabetes
and of ACE
In depth discussion
Visit 2
The daily log was discussed to explore the strategies the patient used to
cope with problems
Timing: 2 weeks after
first visit
Discussion of illness
The reported illness perceptions were discussed in order to challenge
Aimed duration: 45 min
perceptions
misconceptions of diabetes and ACE
Goal setting
The goals formulated during the first visit were evaluated and new
goals were formulated
Homework
The patient was asked to use a weekly log to track strategies for coping
with difficulties
In depth discussion
Visit 3
Problems and illness perceptions that were discussed during the second
Timing: 2 months after
visit were reviewed in order to explore the remaining difficulties of
second visit
these topics. The weekly log was discussed in order to explore the
Aimed duration: 45 min
strategies used by patients to cope with problems
Problem mapping
Using the list of topics from the first visit, the patient again indicated to
what extent problems were encountered and strategies to cope with
these topics were discussed
Goal setting
The goals formulated during the second visit were evaluated, and new
goals were formulated
Discussion about the
At the end of the visit uncertainties for the future regarding coping with
future
the ACE and Type 2 diabetes were discussed
ACE, acute coronary event.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 2 Baseline characteristics of intervention and control group and dropout participants
Intervention group
Control group
All participants
Dropout
(n = 81)
(n = 80)
with follow-up
participants
(n = 161)
(n = 40)
Male gender, n (%)
62 (76.5)
60 (75.0)
122 (75.8)
20 (50)
Mean (SD) age, years
66.0 (9.3)
65.6 (9.4)
65.8 (9.3)
65.8 (12.9)
High
9 (11.1)
15 (18.8)
24 (14.9)
0 (0.0)
Middle
67 (82.7)
55 (68.8)
122 (75.8)
16 (64.0)
Low
5 (6.2)
10 (12.5)
15 (9.3)
9 (36.0)
Living alone, n (%)
14 (17.3)
16 (20.0)
30 (18.6)
11 (40.7)
Mean (IQR) duration of Type 2
7.0 (2.8–16.0)
8.5 (5–15)
8.0 (4–15)
10.9 (5–13)
67 (82.7)
66 (82.5)
133 (82.6)
24 (82.8)
Use of insulin, n (%)
28 (34.6)
23 (28.8)
51 (31.7)
11 (37.9)
Microvascular complications, n
23 (28.4)
15 (18.8)
38 (23.6)
12 (41.4)
Never
19 (23.5)
20 (25.0)
39 (24.2)
5 (19.2)
Former
55 (67.9)
53 (66.3)
108 (67.1)
15 (57.7)
Current
7 (8.6)
7 (8.8)
14 (8.7)
6 (23.1)
coronary artery bypass graft
14 (17.3)
16 (20.0)
30 (18.6)
10 (24.1)
with/without myocardial
53 (65.4)
47 (58.8)
100 (62.1)
12 (41.4)
infarction
14 (17.3)
17 (21.3)
31 (19.2)
7 (34.5)
41 (50.1)
46 (57.5)
87 (54.0)
12 (44.4)
Education level, n (%)
diabetes, years
Use of oral diabetes medication, n
(%)
(%)
Smoking status, n (%)
Type of ACE, n (%)
percutaneous transluminal
coronary angioplasty
with/without myocardial
infarction
myocardial infarction or
unstable angina pectoris
without invasive intervention
Followed cardiac rehabilitation, n
(%)
IQR, interquartile range; ACE, acute coronary event.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 3 Baseline and follow-up scores on questionnaires
Intervention group (n = 81)
Control group (n = 80)
Mean (SD) score
Mean (SD) score
Baseline
Follow-up
Baseline
Follow-up
F
P
8.2 (10.1)
9.0 (11.0)
9.2 (12.4)
9.0 (11.2)
0.52
0.470
58.5 (28.0)
65.5 (23.7)*
57.5 (25.2)
59.6 (24.4)
3.56
0.061
0.73 (0.27)
0.81 (0.19)*
0.74 (0.28)
0.79 (0.25)
0.66
0.417
69.9 (17.3)
76.8 (15.6)*
68.6 (15.9)
69.9 (16.7)
8.80
0.004
4.0 (3.7)
3.8 (3.5)
4.4 (3.9)
3.9 (3.6)
0.09
0.764
4.0 (3.6)
3.8 (3.5)
3.6 (3.3)
3.5 (3.4)
0.04
0.852
64.9 (10.5)
65.7 (10.6)
64.9 (11.4)
64.4 (11.4)
1.82
0.179
32.6 (10.8)
31.2 (11.4)
32.5 (12.3)
33.7 (12.7)
3.93
0.049
35.6 (13.7)
33.1 (15.5)
37.9 (13.5)
36.8 (13.6)
1.38
0.240
3.3 (1.3)
3.5 (1.1)
3.5 (1.3)
3.6 (1.1)
0.02
0.880
Active engagement
3.9 (0.8)
3.9 (0.7)
3.8 (0.8)
3.7 (0.9)
1.70
0.195
Protective buffering
2.2 (0.7)
2.2 (0.6)
2.3 (0.7)
2.2 (0.8)
0.10
0.749
Overprotection
2.1 (0.6)
1.8 (0.6)*
1.9 (0.7)
1.8 (0.7)
1.76
0.187
Diabetes-related
distress
(PAID)
Intervention effect
(Range 0–100)
Well-being (WHO-5)
(Range 0–100)
Heath status (EQ-5D)
(Range -0.594 to 1.00)
Health status (EQ-VAS)
(Range 0–100)
Anxiety (HADS-anxiety scale)
(Range 0–21)
Depression (HADS-depression)
(Range 0–21)
Self-efficacy (CIDS)
(Range 0–100)
Illness perceptions (IPQ) Diabetes
(Range 0-80)
Illness perceptions (IPQ) ACE
(Range 0–80)
Self-management (SDSCA)
(Range 1–7)
Partner support (ABO) (Range 1–5)
PAID, Problem Areas in Diabetes; WHO-5, WHO Well-Being Index; EQ-5D, Euroqol 5 Dimensions; EQ-VAS, Euroqol
Visual Analogue Scale; HADS, Hospital Anxiety and Depression Scale; CIDS, Confidence in Diabetes Self-Care scale; IPQ,
Illness Perception Questionnaire; SDSCA, Summary of the Diabetes Self-Care Activities Measure; ABO, Active
Engagement, Protective Buffering and Overprotection.
* Significant change over time within group.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 4 Baseline and follow-up scores for clinical variables
Intervention group (n = 81)
Control group (n = 80)
Mean (SD) score
Mean (SD) score
Intervention effect
Baseline
Follow-up
Baseline
Follow-up
F
P
Systolic blood pressure, mmHg
142.07 (15.86)
138.88 (12.21)
140.95 (18.96)
135.71 (13.68)
1.99
0.162
Diastolic blood pressure, mmHg
79.12 (9.42)
77.53 (8.52)
77.33 (9.22)
76.26 (9.45)
0.06
0.801
HbA1c, mmol/mol
55.01 (15.38)
52.33 (11.78)
51.00 (10.55)
50.32 (9.87)
0.11
0.745
HbA1c, %
7.2 (3.5)
6.9 (3.2)
6.8 (3.1)
6.7 (3.0)
0.11
0.745
Total cholesterol, mmol/l
4.37 (1.30)
3.73 (1.23)
4.37 (1.30)
4.08 (0.94)
0.80
0.378
HDL cholesterol, mmol/l
1.03 (0.31)
1.04 (0.32)
1.11 (0.31)
1.06 (0.28)
0.30
0.588
LDL cholesterol, mmol/l
2.16 (0.72)
2.07 (0.51)
2.42 (1.21)
2.24 (0.97)
0.12
0.735
This article is protected by copyright. All rights reserved.
Accepted Article
Appendix 1 Description of questionnaires
Questionnaire
Description
Score range
Problem Areas in
Self-reported questionnaire consisting of 20
Each item is rated on a five-point
Diabetes
statements identified as common negative
Likert scale, ranging from 0 ('not a
questionnaire (PAID)
emotions related to living with diabetes.
problem') to 4 ('a serious problem').
The total score is transformed to a 0–
100 scale, with higher score
representing higher distress.
WHO-Five Well-
The five items covering positive mood (good
The degree to which these feelings
being Index (WHO-
spirits, relaxation), vitality (being active and
were present is rated on a six-point
5)
waking up fresh and rested), and general interests
Likert scale, ranging from 0 ('not
(being interested in things) in the past two weeks.
present') to 5 ('constantly present')
The scores are transformed to a 0–100
scale, with higher score representing
better well-being.
Euroqol 5
The EQ-5D measures general health status on five
The EQ-5D scores was computed
Dimensions (EQ-
dimensions:
using the MVH-A1 algorithm
1) Mobility
Range -0.594 to + 1.00
2) Self-care
0: (equal to) death
3) Usual activities
1: full health
4) Pain/discomfort
Negative values: a health score worse
5D)/Euroqol Visual
Analogue Scale (EQVAS)
than death
5) Anxiety/depression
Range 0 to 100
The EQ-VAS measures the overall health state on
a graded, vertical line.
0: worst imaginable health state
100: best imaginable health state.
Hospital Anxiety and
A questionnaire measuring anxiety (seven items)
Each item is rated on a four-point
Depression Scale
and depression (seven items)
Likert scale, ranging from 'Most of
(HADS)
the time' to 'not at all'.
Sum scores for each subscale 0–21,
higher score indicate more severe
anxiety/depression.
This article is protected by copyright. All rights reserved.
Accepted Article
Summary of the
Eleven items assessing several aspects of the
Ten items are rated on an 8-point
Diabetes Self-Care
diabetes regimen: general diet, specific diet,
Likert scale, measuring how many
Activities Measure
exercise, blood glucose testing, foot care, and
days an activity is performed in the
(SDSCA)
smoking. Items measure how many days a patient
last week. Sum score 0-7.
has performed self-care activities in the last seven
days.
Confidence in
Questionnaire adapted to type 2 diabetes patients.
Each item is scored on a 5-point
Diabetes Self-care
Twenty items measure diabetes-specific self-
Likert scale ranging from 1 ('No, I
questionnaire (CIDS)
efficacy.
don’t believe I’m able to do this') to 5
('Yes, I’m sure I’m able to do this').
The total score is transformed to a 0100 scale, with higher score
indicating higher self-efficacy.
Illness Perception
Questionnaire assessing the cognitive
Eight questions answered on an 11-
Questionnaire (IPQ)
representation of illness, focuses on
point Likert scale, ranging from 0 to
– short version
physical/social/emotional consequences, duration,
10. A higher score indicates a higher
controllability, concerns, coherence and the causes
impact of the disease.
of the illness.
Active Engagement,
Measuring overprotection by the partner. Five
Each items is scored on a 5-point
Protective Buffering
items measure active engagement, eight items
Likert scale, ranging from 1 ('never')
and Overprotection
measure protective buffering and six items
to 5 ('very often'). Scores for each
(ABO)
measure overprotection.
scale 1-5.
FIGURE 1 Flow diagram of patient enrolment, allocation and analysis.
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 1.
Invited (n= 264)
Declined to participate (n= 63)
Randomized (n= 201 )
Intervention group
(n = 101)
Control group
(n = 100)
Dropout (n = 20)
• Died (n = 2)
• Illness (n = 7)
• Too busy/no
interest (n = 9)
• Unknown (n = 2)
Dropout (n = 20)
• Illness (n = 2)
• Too busy/no
interest (n = 9)
• Unknown (n = 9)
Included in analysis (n = 81)
• Did not receive all three
home visits (but followup available) (n = 12)
Included in analysis (n = 80)
• Did not receive the phone
consultation (follow-up
available) (n = 19)
This article is protected by copyright. All rights reserved.