Impact of a Program to Improve Adherence to Diabetes Guidelines by Primary Care Physicians

OBJECTIVES: Previous studies have shown that primary care physician (PCP) adherence to diabetes guidelines is suboptimal. We sought to determine the state of diabetes care given by independently practicing PCPs in a rural county in Indiana and whether a multifaceted intervention targeting PCPs, patients, and the health care system would improve adherence to diabetes guidelines. RESEARCH DESIGN AND METHODS: Baseline audits to assess adherence to diabetes guidelines were done on charts of the seven PCPs in the county. Audits were repeated after development of local consensus guidelines and feedback of baseline performance and after implementation of various interventions (practice aids, physician detailing, patient education sessions, and implementation of computerized individual meal planning). RESULTS: Before any intervention, rates of adherence to guidelines were low (15% for foot exams, 20% for HbA(1c) measurement, 23% for eye exam referrals, 33% for urine protein screening, 44% for lipid profiles, 73% for home glucose monitoring, and 78% for blood pressure measurements). One year after development of local consensus guidelines and feedback of baseline performance, significant improvements were seen in blood pressure measurements (71 vs. 83%; P = 0.002), foot exams (19 vs. 42%; P < 0.001), HbA(1c) measurements (26 vs. 37%; P = 0.012), and PCP eye exams (38 vs. 46%; P = 0.043); a trend toward improvement was seen in referral to eye specialists (25 vs. 33%; P = 0.059). After a second year of multiple interventions, only blood pressure measurements (70 vs. 92%; P < 0.001) and foot exams (22 vs. 47%; P < 0.001) remained significantly improved; all other areas returned to rates indistinguishable from baseline. CONCLUSIONS: In busy primary care practices lacking organizational support and computerized tracking systems, sustained improvements in diabetes care are difficult to attain using traditional physician-targeted approaches.