Using the IPQ and PMDI to predict regular diabetes care-seeking among patients with Type 1 diabetes

TLDR: Comparing the performance of the Illness Perception Questionnaire (IPQ) and a modified version of the Personal Models of Diabetes Interview (PMDI) in predicting care-seeking showed that the most important construct was treatment effectiveness.

Abstract: OBJECTIVES: The objectives were, first, to identify factors associated with regular diabetes care-seeking and, second, to compare the performance of the Illness Perception Questionnaire (IPQ) and a modified version of the Personal Models of Diabetes Interview (PMDI) in predicting care-seeking. METHOD: This was a cross-sectional study involving 42 patients who had not attended hospital diabetes clinic for a period of 18 months or more and 42 matched controls receiving specialist care. Differences in illness representations between clinic attenders and non-attenders were examined. Due to the variability in care-seeking between non-attenders (ranging from no contact with health professionals to regular general practitioners care), participants were then reclassified for further analyses into those receiving regular care from either hospital diabetes clinic or general practice (n = 52) and those receiving no regular care (n = 32). RESULTS: Patients not seeking regular care held more negative views of the control, course and consequences of diabetes than those who received regular care. Regression analyses showed that the most important construct was treatment effectiveness. Treatment effectiveness (PMDI) and control (IPQ) dimensions were associated with clinic attendance. PMDI constructs of treatment effectiveness, threat and worries were associated with regular care-seeking. CONCLUSIONS: IPQ and PMDI results were generally consistent. In order to understand diabetes care-seeking behaviour it is important to measure beliefs about the benefits of treatment recommendations in addition to those of the disease itself. Emotional and cognitive responses to diabetes should be assessed.