Background and aims: Existing psychometric instruments to assess diabetes self-management often reveal weak or inconsistent associations with ‚hard data‘ such as medical outcomes and HbA1c. To fill this gap, the Diabetes SelfManagement Questionnaire (DSMQ) was developed, focussing on self-care activities which directly impact medical diabetes outcomes. This study evaluates the questionnaire’s practical utility in detecting high-risk patients at a tertiary diabetes
Materials and methods: 226 people with diabetes (age 43 ± 15 y.; 55% female; BMI 29 ± 7; 70% type 1 diabetes; illness duration 15 ± 10 y.; 92% treated with insulin; HbA1c 8.9 ± 1.6%) were assessed with the DSMQ and further questionnaires regarding diabetes acceptance (AADQ), coping with illness (FQCI), treatment satisfaction (DTSQ), diabetes distress (PAID), and depressive symptoms (CES-D); additional data (demographic variables, self-monitoring of blood glucose, HbA1c, and long-term complications) were gained from electronic patient records. People were then categorized by a median split of the DSMQ total score into groups performing ‘adequate’ (n = 107) versus ‘inadequate’ diabetes self-care (n = 119); the groups were compared regarding relevant outcomes using multivariate ANOVA (subsequently presented data are M ± SD, F statistic, and effect size Cohen’s d).
Results: After adjusting for sex, age, BMI, diabetes type, diabetes duration, and type of treatment, people performing ‘inadequate self-care’ compared to those with ‘adequate self-care’ showed stronger diabetes non-acceptance (31 ± 8 vs. 22 ± 6, F = 90.5, d = 1.24), less active coping with diabetes (2.7 ± 0.9 vs. 3.4 ± 0.8, F = 18.1, d = 0.82), lower diabetes treatment satisfaction (20 ± 7 vs. 24 ± 6, F = 21.3, d = 0.60), higher diabetes distress (43 ± 21 vs. 33 ± 19, F = 17.0, d = 0.85), and more depressive symptoms (24 ± 11 vs. 20 ± 11, F = 4.8, d = 0.36). Moreover, they performed fewer blood glucose self-tests (3.3 ± 3.5 vs. 5.6 ± 2.4 times per day, F = 24.7, d = 0.74), consulted their diabetologist less often (1.9 ± 1.8 vs. 2.6 ± 2.3 times per half-year, F = 6.6, d = 0.34), had a higher HbA1c value (9.5 ± 1.5 vs. 8.2 ± 1.4%, F = 34.6, d = 0.87), and showed a higher prevalence of retinopathy (28% vs. 14%, F = 6.0, d = 0.35).
Conclusion: The DSMQ yields excellent distinction between people with diabetes performing adequate versus insufficient diabetes self-care, thus enabling detection of people at high risk of a negative diabetes prognosis. The 16-item questionnaire is an efficient tool which may be used for screening and diagnostic purposes or clinical diabetes research.