11 / 56
VOLUME 13 NUMBER 1 • JULY 2016
9
SA JOURNAL OF DIABETES & VASCULAR DISEASE
RESEARCH ARTICLE
problematic to understand for some patients: ‘feeling constantly
burned out by constant effort to manage diabetes’ and ‘not having
a clear and concrete goal for managing diabetes care’.
Some adolescents and adults had challenges with specific words
in the items although they were able to deduce the meaning of
the entire question. The words and concepts that the patients
found challenging were ‘overwhelmed’, ‘regimens’, ‘unsatisfied’,
‘burnout’, ‘physicians’ and ‘concrete goal’. The patients were also
able to suggest some replacements to the words they initially had
challenges with to include ‘treatment plan’ for ‘regimens’, ‘unhappy’
or ‘happy’ for ‘unsatisfied’, ‘doctor’ for ‘physicians’. In some cases
it was difficult to recall occurrences of certain events evoked by the
questions such as when patients were angry, guilty or anxious and
uncomfortable. These challenges were mostly noted among young
people. Table 8 outlines common problems that were identified, as
indicated above.
Discussion
The aim of this study was to examine the latent structure, reliability
and validity of the PAID among individuals with type 1 or 2 diabetes
in Zambia. The results of our study strongly support a one-factor
solution of the Zambian translation of the PAID, although four items
‘concerned with complications’, ‘feelings of deprivation regarding
food’, ‘coping with complications’ and ‘feeling overwhelmed by
diabetes’ had factor loadings less than 0.30. These low loadings
may result from the fact that all our subjects were out-patients
without any serious complications. It remains unclear why the item
‘concerned about food’ had a low loading considering that initial
interviews by the first author with the patients showed that it was
a major concern.
17
Our data rejected the two-factor model found in Iceland by
Sigurdardottir and Benediktsson,
6
the three-factor model found in
Sweden by Amsberg and colleagues,
7
and the four-factor model
found in the USA/the Netherlands.
18,19
A one-factor model was also
found in the USA/the Netherlands.
18
Originally, the PAID was conceptualised as a unidimensional
scale;
2
therefore, our one-factor structure using all 20 items
remains plausible. Moreover, in studies among Chinese,
20
Dutch
and USA
18
individuals with diabetes, the one-factor solution was
also supported. The Zambian translation of the PAID showed high
internal consistency with Cronbach’s alpha and lambda, which has
recently been recommended in the literature because it shows the
least amount of bias.
36
Consequently, a total of at least 16 items
from the EFA results seems useful for clinical assessment to detect
diabetes-related distress and suggest psychological help to Zambian
patients with such distress, with possibly some word changes, as
suggested by the cognitive interview study.
Our study also found that most patients endorsed, ‘worrying
about low blood sugar reactions’, ‘feeling that diabetes is taking up
too much mental and physical energy’, ‘feeling guilty/anxious when
you get off track with your diabetes management’, ‘worrying about
future and possible serious complications’ and ‘feeling depressed
when you think about living with diabetes’ as the most bothersome
diabetes-specific problems, which is consistent with findings by
Sigurdardottir and Benediktsson.
6
‘Worrying about the future and
possible complications’, and ‘feeling guilty when you get off track
with your diabetes management’ were also found to be the most
commonly endorsed by Snoek and associates.
18
It was not surprising that our patients endorsed these items,
considering that the mean score for fear for hypoglycaemic episode
was relatively high (58 ± 12) and the diabetes self-care score was
below average (48 ± 9). Moreover, having a hypoglycaemic episode
in Zambia might be more stressful, as medical care is less available,
compared to Europe or the USA. Furthermore, all the participants
Table 6.
Stepwise multiple regression analyses predicting PAID by
demographic and clinical characteristics in 157 patients with T1DM and
T2DM
Models
Beta
t
p
-value
Model 1: demographic characteristics
Age
0.122
1.286
ns
Being T2DM
0.029
0.313
ns
Being female
0.011
0.138
ns
Socio-economic status
0.005
0.054
ns
R
2
= 0.020
Adj
R
2
= –0.007
p
≥ 0.005
Model 2: clinical characteristics
Body mass index
–0.107 –1.348
ns
Depression (MDI)
0.268
3.122
***
Fear for hypoglycaemia
0.289
3.456
***
Diabetes self-care
0.252
3.249
***
R
2
= 0.335
Adj
R
2
= 0.315
p
≤ 0.001***
*
p
< 0.05, **
p
< 0.01, ***
p
< 0.001.
Table 7.
Cognitive interview questions
Warm up question/instruction clarity
Tell me what this introduction is telling you?
Comprehension (question intent and meaning of term)
Can you tell me in your own words what this question was asking?
What does the [word/term] mean to you as it has been used in this question?
Tell me what you were thinking when I asked about [topic]?
Assumption
How well does this question apply to you?
Can you tell me more about that?
Knowledge/memory
How much would you say you know about [topic]?
How much though would you say you have given to this?
How easy or difficult is it to remember [event]?
You said [answer]. How sure are you about that?
How did you come up with that answer?
Sensitivity/social desirability
Is it ok to talk about this diabetes problem area or is it uncomfortable?
The question uses the [word/term]. Does that sound ok or would you choose
something different?
Specific and general probes
Why do you think that [topic] is the most serious diabetes problem?
How did you arrive at that answer?
Was it easy or hard to answer?
I noticed that you hesitated. Tell me what you were thinking?
Table 5.
Correlations between the total PAID, the four PAID factors
(based on previous research) and other variables of interest
Diabetes
Fear for
self-care hypoglycaemia Age SES
¶
BMI
MDI
#
PAID total
0.30**
0.35**
0.12 0.00 –0.14 0.39**
**
p
≤ 0.01, *
p
≤ 0.05;
¶
Socio-economic status;
#
Major depression inventory
total score.