Using the Health Belief Model to test factors affecting patient retention in diabetes-related pharmaceutical care services

Abstract

Background Diabetes is one of the deadliest and most costly diseases. Attrition rates among patients in diabetes management programs may preclude optimal success. Theoretical models, such as the Health Belief Model, may be useful for identifying factors responsible for patients' continued enrollment in such programs. Objectives (1) To design and test a reliable and valid survey instrument for assessing patients' perceptions of diabetes-related pharmaceutical care services. (2) To determine factors affecting patient retention in pharmaceutical care services. Methods This cross-sectional exploratory study used convenience sampling to survey type 1 and 2 diabetes patients receiving pharmaceutical care at 25 Eckerd Pharmacy sites in Florida. Survey items were designed using constructs from the Health Belief Model. Reliability (Cronbach's alpha) and validity (exploratory factor analysis) were assessed. Independent t tests, Pearson's correlation coefficients, and multiple regression analysis were conducted. Results Seventy usable surveys were returned (47%). In the absence of pharmaceutical care, patients felt susceptible to at least one of 8 diabetes-related conditions (hypoglycemia, hyperglycemia, diabetic foot infections, eye problems, cardiovascular risks, blood pressure problems, cholesterol problems, and kidney diseases). For these conditions, about 48% to 95% of patients perceived that their threat had reduced because of pharmaceutical care. Accordingly, more than half perceived the services as beneficial, with counseling for blood sugar monitoring rated as the most beneficial. All respondents rated the services as helpful, and 64 intended to continue regular utilization. Overall helpfulness of the service and patient retention were positively correlated ( r = 0.33, P < .00). Perceived susceptibility predicted perceived threat reduction ( R 2 = 0.22, P < .01). Perceived threat reduction, blood sugar monitoring, and overall helpfulness of the service predicted patient retention in the service ( R 2 = 0.41, P < .00). Factor analysis extracted 4 factors: perceived threat reduction, perceived susceptibility, perceived benefits, and blood sugar monitoring. Cronbach's alphas for the scales ranged from 0.63 to 0.91. Conclusion Two key constructs of the Health Belief Model influence use of diabetes-related pharmaceutical care services: perceived susceptibility and threat reduction. In an effort to increase patient retention, pharmacists need to assess patient perceptions and structure their services to address patient perceptions and concerns.