ABSTRACT
INTRODUCTION: The purpose was to contribute to the development of a validated instrument for identification of patients with poor compliance in the use of hypertension medicine in primary health care. MATERIAL AND METHODS: The analysis included questionnaire data from 1,421 hypertensive patients, data from the Danish register "Patient Medical Profile" (PEM, n = 182) and electronic medicine containers (MEMS, n = 33). The analysis included preliminary scale validation resulting in five scales. The scales were used in a subsequent multivariate regression analysis which involved a total of 29 variables. The key variables were established via the correlations between variables in the graphic model and included in the instrument. RESULTS: The scale "Unintentional non-compliance" (concerning the patient's self-assessed tendency to forget to take the medicine) is central in the graphic network in terms of its relation to the scales "Intentional non-compliance" (when patients opt not to follow the prescriptions) and "Medical concern" (concerns related to medicine intake). "Unintentional non-compliance", moreover, correlates with PEM. The item concerning the patient's systolic blood pressure proved to be crucial because of its correlation with both PEM and MEMS. The scales "Medical concern" and "Intentional non-compliance" are, moreover, assessed to measure different dimensions of compliance. CONCLUSION: A group of items has been developed and validated. Collectively they can identify hypertensive patients with poor compliance in primary health care.
