Provider Response to Computer-Based Care Suggestions for Chronic Heart Failure

Objectives: We sought to assess the responses of providers to recommendations generated by a computer-management system for chronic heart failure (CHF). Methods: This study is an analysis of primary care providers' responses to evidence-based computer-generated suggestions regarding patients with CHF at one center of a randomized trial. The trial randomized primary care providers from 2 VA Medical Centers to receive care suggestions regarding patients with CHF, with or without inclusion of patient symptom data obtained from previsit questionnaires. At one center, providers were asked to respond to the suggestions with hand-written comments and a numerical agreement scale. Results: Providers responded to 774 care suggestions (62% of the 1246 delivered). They agreed with 41%, had major disagreements with 12%, and had minor disagreements with 22%. For 7% of the care suggestions, providers asked to not see it again for that patient. The most common reasons for major or minor disagreements were a belief that the suggestion was wrong or unnecessary (45%) or would not be tolerated by the patient (32%). External barriers to implementation of guidelines, lack of guideline awareness, or disagreement with guidelines were uncommon reasons cited by providers in this study. Conclusions: Providers agreed with less than half of computer-generated care suggestions from evidence-based CHF guidelines, most often because the suggestions were felt to be inapplicable to their patients or unlikely to be tolerated.