 
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
05-March-2010, Vol 123 No 1310
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Exploring physical and
psychological wellbeing among adults with Type 2 diabetes in New Zealand:
identifying a need to improve the experiences of Pacific peoples
The aims of diabetes treatment are to promote good glycaemic
control, and quality of life.1,2 Research shows
that while an HbA1c <7% is
recommended,3,4 many, perhaps even the
majority5 of people with Type 2 diabetes do not
achieve this goal. However there is considerable variation between individuals.
Previous research has shown that differences in health outcomes among people
with Type 2 diabetes are related to clinical and demographic characteristics
including ethnicity.6-8
Research shows that many people experience diabetes-related
distress.9-12 However there have been very few
studies to date exploring individual variation in diabetes-related psychological
outcomes. As a consequence we know little about how diabetes distress might vary
between individuals. For example, although we know that differences between
ethnic groups have been shown for metabolic control, we do not know if the
psychological burden of diabetes shows a similar pattern of association with
ethnicity. Health services providing psychological support for people with
long-term conditions including diabetes are now funded in some regions in New
Zealand and so it is very timely that research now identifies the predictors of
poor illness-related psychological outcomes so that we may learn who is
most at risk, and how best to intervene, in order to promote wellbeing among
people with diabetes.
In addition to clinical and demographic characteristics,
psychological variables are also related to health outcomes. For example the
widely-used Common Sense Model of illness
self-regulation13 suggests that psychological
variables including beliefs about treatment can play a salient role in
determining coping behaviour and subsequent health
outcomes.14 Further, because treatment beliefs
are amenable to change in a way that age and ethnicity for example are not,
these psychological variables are of particular interest to researchers who may
want to identify targets for interventions that aim to improve health
outcomes.
Helping people with diabetes to stay well is clearly an
important, and challenging, goal. The evidence suggests many people with
diabetes experience suboptimal physical and psychological health outcomes but
few studies have examined how outcomes – particularly illness-related
psychological outcomes—vary across individuals. The aims of this study are
to:
Research design and methodsThe research sample (N=1015) was randomly selected from a
database of primary care records for people with diabetes (N=4857) held in
Wellington, New Zealand. This database showed an under representation of
Māori and Pacific groups compared with population-based diabetes
estimates.15 Therefore, Māori and Pacific
groups were purposefully over-sampled in the research design so that the
ethnicity of participants in this study might closely reflect the demographic
profile reported by the Ministry of Health15
for people with Type 2 diabetes in New Zealand. Inclusion criteria for this
study were: diagnosis of Type 2 diabetes by a physician in accordance with
national guidelines4; and aged ≥18.
Psychological data were collected using a mailed
questionnaire survey, with a response rate of 62% providing 629 completed
questionnaires. Potential participants were mailed a letter of introduction
informing them of the study and inviting their participation; two weeks later
they received an information sheet, a consent form, and a copy of the research
questionnaire. For consenting participants’ (N=615, 98%) relevant
physician-assessed clinical data including glycosylated haemoglobin
(HbA1c) were obtained from electronic medical
records and released to the researcher in non-identifiable form. Clinical data
and questionnaire data were linked by unique identifier codes; patient anonymity
was protected at all times during the research. Ethical approval for this study
was obtained from the Massey University Human Ethics Committee, protocol
02/140.
Participants—Table 1 provides a
summary of participants’ demographic and clinical characteristics.
Participants’ mean age was 63 years (SD=11.6), and forty-seven
percent were female. The ethnic composition of participants (57% New Zealand
European, 30% Māori, 6.4% Pacific Island) matched closely with proportional
representation of the three main ethnic groups that comprise the New Zealand
diabetes population as reported by the New Zealand Ministry of
Health.15 On average, length of diagnosis with
diabetes was 8.1 years (SD=5.8); mean
HbA1c was 7.5% (SD=1.5), with a range
from 4.5 to 13.5. Nine percent of participants were currently using insulin,
almost half (49%) had a treatment regimen that included ACE inhibitors, and 25%
had been prescribed statins. Mean body mass index (BMI) was 31
(SD=7.1), and 15% were current smokers.
Table 1. Demographic and clinical
characteristics of participants (N=615)
 Measures—Length of diabetes
diagnosis, current treatments, BMI, and most recently recorded
HbA1c were extracted from electronic medical
records. General self-reported mental health and physical health were assessed
using the SF-12 MCS subscale, and SF-12 PCS subscale,
respectively.16 Diabetes-related psychological
distress was measured using the Problem Areas In Diabetes (PAID)
scale.17 The Beliefs about Medications
Questionnaire (BMQ)18 was used to assess
concern about diabetes medicines (six items), and perceived medication necessity
(five items), with responses provided on a 5-point Likert scale: (1) strongly
disagree; (2) disagree; (3) uncertain; (4) agree; (5) strongly agree. The
average score across items in each BMQ subscale were calculated providing two
summary scores (medication concern; medication necessity) with a range from 1 to
5.
Statistical analyses—Mean scores were
first calculated for the four primary outcome variables of interest in this
study: HbA1c, diabetes-related distress (PAID),
self-reported general physical health (SF-12 PCS), and self-reported general
mental health (SF-12 MCS), and their relationships with clinical, demographic,
and psychological variables were assessed using Pearson’s correlation
coefficients. Second, ANCOVAs were used to test for between-groups differences
in outcome variables across ethnicity. In all ANCOVA tests age, length of
diagnosis, BMI, and treatment (insulin use) were entered as co-variates in the
model to control for their effects. SPSS/PC version 15.0 was used for all
statistical analyses with alpha set at .05.
ResultsLevels of physical and psychological
wellbeing—Mean HbA1c was 7.5
(SD=1.5) and, as shown in Table 1, 257 participants (42.1%) met the recommended
goal for Type 2 diabetes in New Zealand4 of an
HbA1c less than 7%. The average score on the
Problem Areas In Diabetes (PAID) scale was 19.6 (SD=19.1) and this is
in line with average scores reported in previous
research.12 Mean scores on the SF-12 PCS
(physical health) and SF-12 MCS (mental health) subscales were 44.0
(SD=10.0), and 50.5 (SD=9.3), respectively.
Relationships between clinical and demographic
characteristics and health outcomes—Table 2 shows the
relationships between demographic and clinical characteristics and the four
outcome variables of primary interest in this study
(HbA1c, diabetes-related psychological distress,
self-reported general physical health, and self-reported general mental health).
Greater age was associated with lower self-reported physical health, and higher
self-reported mental health, as is consistent with previous
research.16 Higher age was also related to
lower HbA1c, and fewer worries about diabetes.
Longer length of diagnosis was associated with higher
HbA1c, and lower self-reported general physical
wellbeing; length of diagnosis was unrelated to measures of general (SF-12 MCS)
or diabetes-specific (PAID) emotional wellbeing. There was a positive
relationship between prescribed insulin treatment, and
HbA1c.
Relationships between treatment beliefs and health
outcomes—Greater concern about diabetes medication was associated
with higher HbA1c, higher scores on the PAID, and
lower scores on the SF-12 MCS subscale as shown in Table 2. Perceived need for
diabetes medication did not show significant relationships with any of the four
outcome measures.
Table 2. Pearson’s correlations between
health outcomes and clinical, demographic, and psychological variables
(N=615)
Differences in diabetes-related outcomes by
ethnicity—Figure 1 displays means for
HbA1c across four ethnic group categories: New
Zealand European; Māori; Pacific; and Other. Figure 2 shows average scores
on the Problem Areas in Diabetes scale across these same four ethnic groups. On
average HbA1c was lowest (showing better
metabolic control) among New Zealand European participants, and highest among
Pacific participants. Pacific participants also showed the highest scores on the
PAID, as displayed in Figure 2.
Figure 1. Glycaemic control by ethnicity
showing means and 95% confidence intervals
 Figure 2. Scores on the Problem Areas in
Diabetes (PAID) scale by ethnicity showing means and 95% confidence
intervals
 Testing between-group differences—Table 3 displays means (SE), and p-values from ANCOVAs
testing for between-group differences in physical wellbeing, psychological
wellbeing, and medication beliefs across four ethnic groups. Analyses
controlling for the effects of age, length of diagnosis, BMI, and insulin use
showed that there were significant between-group differences for:
HbA1c (p=.001); scores on the PAID
(p=.001); SF-12 MCS (p=.030); and the medication concern
subscale of the BMQ (p=.001). Results show that, on average, Pacific
peoples showed the highest HbA1c, and reported
the greatest worry about diabetes; and they also showed the highest levels of
concern about prescribed medication. An additional ANCOVA controlling for scores
on the SF-12 MCS was run for the PAID and this did not change the reported
results. This suggests that the higher levels of diabetes-related distress found
among Pacific people were not accounted for by any cultural differences in
self-reported general mental health.
Comparison of respondents and
non-respondents—T-tests and
χ2 tests examining potential differences
between respondents and non-respondents showed no significant difference in
gender, or length of diabetes diagnosis, but there was a difference between
respondents and non-respondents for HbA1c,
F(1,1015) = 13.67, p < .001, and ethnicity,
χ2 (3, N = 1275) = 59.83,
p<.001. On average, non-respondents had a higher
HbA1c (M = 7.79, SD = 1.67)
than respondents (M = 7.45, SD = 1.52), and those who returned
a completed questionnaire were more likely to be of New Zealand European
ethnicity.
DiscussionOn average, people with Type 2 diabetes in New Zealand
appear to show similar levels of self-reported general physical health to peers
with diabetes in other countries,16 and similar
levels of general emotional functioning to same-age peers without
diabetes.16,19 In this study the majority (58%)
of participants did not meet the recommended
goal4 of an HbA1c
< 7%. These results are congruent with international
research5 and show that many people with Type 2
diabetes experience difficulty achieving good metabolic control. However
physical and psychological health outcomes do vary across individuals and this
variation was found to be associated with clinical, demographic, and
psychological factors.
Clinical and demographic factors associated with
differences in wellbeing among adults with Type 2 diabetes—Higher
BMI, younger age, greater length of diagnosis, and insulin use were associated
with higher HbA1c, and this is in keeping with
observations that: being overweight exacerbates difficulties with blood glucose
regulation; insulin resistance increases over time among people with Type 2
diabetes; and poor metabolic control leads over time to more intensive
treatment, including the prescription of insulin.
Results are consistent with research showing that age has a
protective effect on general psychological
wellbeing,16 and suggest that this effect may
also extend to diabetes-related worries.10
Older adults may have had more exposure to health problems (as the risk of
common chronic illnesses such as coronary heart disease increases with age), and
as a result could have developed greater skills for coping successfully with the
emotional impact of illness, in comparison with younger
adults.11 Interestingly—and in contrast
with some previous research10—insulin use
was not associated with higher diabetes-related distress.
Relationships between treatment beliefs and physical
and psychological wellbeing—Our findings show that people who are
more concerned about using prescribed diabetes medication (e.g. worried about
harmful side-effects, or possible dependency) are more likely to experience poor
metabolic control and higher diabetes-related distress than those who have less
concerns about pharmacological treatments. It is possible that this relationship
between medication concern and health outcomes is mediated by medication use:
this is in line with research showing that greater concern about prescribed
medication is associated with lower adherence to the prescribed treatment
regimen in other chronic conditions.20
Perceived need for diabetes medication was not related to differences in any of
the four primary outcome variables: this contrasts with results reported by
Barnes et al.7 but is consistent with the
absence of a relationship between medication necessity and self-reported
adherence among renal haemodialysis
patients.20
Differences in diabetes-related health outcomes
across ethnicity—In comparison with New Zealand European
participants in our study, on average, Māori and Pacific peoples
experienced lower glycaemic control. Our findings are in line with research from
New Zealand7,8 and
internationally6,21,22 showing that physical
health among people with diabetes varies by ethnicity and that indigenous
peoples and recent immigrants are often worst off. There are many possible
reasons for this and they include: delays in
diagnosis23 and
treatment;8,24 differences in lifestyle
behaviours,24 rates of
obesity,8,25 and diabetes
education;26 and
genetics.27
Findings extend our understandings of the way
diabetes-related health outcomes vary by showing that psychological
outcomes such as diabetes distress show a similar pattern of differences across
ethnicity to that previously reported for physical health. In our study the
observed pattern was very consistent across both diabetes-related physical and
psychological wellbeing, with highest metabolic control and lowest
diabetes-related distress among New Zealand Europeans, and the reverse (lowest
metabolic control and highest diabetes-related distress) reported among Pacific
peoples as Figure 1 and Figure 2 illustrate. Our findings are interesting partly
because previous international research has not shown a relationship between
ethnicity and diabetes-related distress.10
While it is useful to better understand how health
outcomes—including illness-related psychological outcomes—vary by
ethnicity, we emphasise the need to consider how this information can be used to
improve population health outcomes. For example: to identify who is most at risk
for adverse outcomes; consider how best to intervene; and to provide added
impetus for the development of interventions that are acceptable, effective, and
culturally appropriate for the New Zealand diabetes population.
Identifying who is most at risk for adverse
outcomes—Among adults with Type 2 diabetes in New Zealand
overweight individuals; those on insulin; younger adults; and those who have had
diabetes for a longer time, are most at risk for adverse physical health
outcomes including poor glycaemic control. Younger adults, and those who are
overweight, also appear to be most at risk for experiencing diabetes-related
psychological distress. Two further characteristics identified as important in
showing who is most at risk for adverse health outcomes are: high concerns about
diabetes medication, and Pacific ethnicity.
Clinical implications and opportunities for
intervention—Patients who have high concerns about diabetes
medications could be identified using a brief screening tool (for example, the
6-item medication concern subscale from the BMQ) and invited to talk in detail
about their concerns. This gives clinicians an opportunity to clarify any
misperceptions (e.g. regarding potential harms), provide reassurance, and
explain the benefits of the prescribed treatments. As previous research has
shown that lower medication concern is associated with higher self-reported
medication adherence20,28 any success in
reducing concern about diabetes medication may benefit psychological wellbeing
but could also help to increase medication adherence and therefore metabolic
control.
Younger adults and those who are overweight appear to be at
greater risk of both diabetes-related psychological distress and poor metabolic
control and this suggests another possible target group for clinical
intervention. For example, practice records could be used to identify young
overweight patients with diabetes and extra support could be provided to help
these patients manage the emotional consequences of diabetes and improve
self-care behaviour, particularly as Type 2 diabetes in young people is
associated with the early onset of significant complications such as
nephropathy.29 Interventions that are engaging
to younger adults—for example those that are text or
internet-based—may be most useful.
One important clinical implication of our findings is the
need to better assist Māori and Pacific people in managing diabetes.
Primary prevention programmes that seek to prevent the unnecessary excess burden
of morbidity through early intervention, include wider family networks,
and seek to improve both metabolic control and diabetes-related psychological
wellbeing (particularly given the importance placed on physical and
psychological wellbeing—te taha tinana, and te taha hinengaro—in
Māori health frameworks30), are strongly
encouraged. Evidence suggests that Pacific people with diabetes may be less
intensively treated than New Zealand Europeans, for example they are less likely
to be prescribed statins,8 and this may also
contribute to poor health outcomes and should be addressed in primary care
practice.
We are not aware of any other New Zealand research assessing
cultural differences in diabetes-related psychological wellbeing and this is an
important original contribution of the present study. Findings show that among
Pacific people in particular, there is a pressing need to reduce concerns about
diabetes medication and emotional distress about diabetes while continuing to
focus on improving glycaemic control. Diabetes nurse educators, diabetes
specialist nurses, and health psychologists could play an important role in
delivering this type of care in collaboration with primary care and specialist
physicians. The recent development of clinics that provide psychological support
for people with an ongoing physical illness including diabetes is very timely,
and we encourage staff at these clinics to carefully consider the needs of
Pacific peoples with diabetes.
Limitations and directions for future
research—The cross-sectional design of the current study
precludes the interpretation of causal relationships and some caution in the
generalisability of results is suggested as there were small numbers of
participants in some subgroups (e.g., Pacific peoples). The response rate is
modest (62%) and could be improved in future research. Two identifiable
strengths of this study are the inclusion of both clinical and psychological
data; and the close match between the cultural composition of the research
sample and the population of interest (people with Type 2 diabetes in New
Zealand).
There are reasons to interpret the results of this study
with some caution. First, many people may identify with more than one ethnic
group. However in the primary care database used to identify potential
participants ethnicity was recorded as a single variable with mutually exclusive
categories (e.g., New Zealand European; Māori; Pacific; or Other). Second,
differential non-response was associated with ethnicity in this study and may
represent a potential compromise to the external validity of the study. The use
of an English-language questionnaire may have reduced the response rate among
those for whom English is a second language. Further the absolute number of
Pacific Island participants in this study (n = 39) is small. We recommend that
further research is undertaken to test the validity of our findings and to
determine whether the results that we report are representative of the broader
experiences of Pacific peoples with Type 2 diabetes in New Zealand.
Findings from the present study may be useful to guide the
direction of future research. Results highlight the poor health outcomes
experienced by Pacific people with diabetes; however further research is needed
to help us understand how these come about. For example, we need to examine the
causal relationships between HbA1c and
diabetes-related distress. Poor metabolic control (with the associated increase
in complications) may cause an increase in diabetes-related distress:
alternatively, high distress and concern about medication could reduce
medication adherence, and thus lead to lower metabolic control. Prospective
cohort studies will help to disentangle the direction of causality. Future
research could also examine variation in health outcomes between Pacific
subgroups (e.g. Samoan, Tongan, and Fijian) as this has been shown to be
important in studies of cardiovascular disease
risk31 but was not feasible in our study due to
small group sizes.
Competing interests: The author has no
competing interests to declare.
Author information: Charlotte A M Paddison,
Senior Research Associate, General Practice and Primary Care Research Unit,
Department of Public Health and Primary Care, University of Cambridge,
Cambridge, United Kingdom
Acknowledgements: I am very grateful to the
participants and thank them for their time in assisting with this research. The
following organisations are acknowledged for their support: Wellington Regional
Diabetes Trust; Palmerston North Medical Research Foundation; New Zealand
Society for the Study of Diabetes; and the Economic and Social Research Council
(UK). Particular thanks are expressed to Dr Fiona Alpass and Dr Christine
Stephens (Massey University, New Zealand) for their excellent supervision and
professional guidance; and to Rebekah Tuileto'a, Fiasili Vaeau, and Mandy
Faimalie for their support.
Correspondence: Charlotte Paddison, General
Practice and Primary Care Research Unit, Institute of Public Health, University
Forvie Site, Robinson Way, University of Cambridge, Cambridge CB2 0SR, United
Kingdom. Email: camp3@medschl.cam.ac.uk
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|