BACKGROUND: Diabetic retinopathy is a costly and prevalent complication of diabetes mellitus. OBJECTIVE: To assess primary care physicians' self-reported practice patterns for the screening and detection of diabetic retinopathy relative to published guidelines. PARTICIPANTS AND METHODS: All primary care physicians (defined as general internists, family practitioners, and general practitioners) in Indiana were identified and surveyed using a mailed questionnaire. Of 2390 physicians, 1508 (63%) responded and were determined to be eligible. Of these 1508 physicians, 1058 (70%) completed all or some of the eye care-related questions. For each eye care practice, physicians were asked to specify the proportion of patients to which the practice was applied and the frequency (eg, every 3 months) with which the behavior was performed, if appropriate. Physicians were also asked to distinguish between patients with type I (insulin-dependent) and type II (non-insulin-dependent) diabetes mellitus for each practice behavior. RESULTS: Physicians' responses were classified as "high," "moderate," or "low" based on the American Diabetes Association guidelines. Forty-five percent of the physicians' responses were classified as high for referring all of their patients with type I diabetes mellitus to an eye care specialist annually as were 35% of the physicians' responses for referring their patients with type II diabetes mellitus. Fewer physicians reported high levels of routine in-office funduscopic examination. No relationship was found between the extent to which physicians performed an in-office funduscopic examination and patterns of referral to eye care specialists. Logistic regression analysis suggested that recent graduates and general internists are most likely to report behavior that is considered high (P < .05). CONCLUSION: The diabetic retinopathy-related practice patterns of primary care physicians in Indiana differ significantly from published guidelines.
BACKGROUND:Diabetic retinopathy is a costly and prevalent complication of diabetes mellitus. OBJECTIVE: To assess primary care physicians' self-reported practice patterns for the screening and detection of diabetic retinopathy relative to published guidelines. PARTICIPANTS AND METHODS: All primary care physicians (defined as general internists, family practitioners, and general practitioners) in Indiana were identified and surveyed using a mailed questionnaire. Of 2390 physicians, 1508 (63%) responded and were determined to be eligible. Of these 1508 physicians, 1058 (70%) completed all or some of the eye care-related questions. For each eye care practice, physicians were asked to specify the proportion of patients to which the practice was applied and the frequency (eg, every 3 months) with which the behavior was performed, if appropriate. Physicians were also asked to distinguish between patients with type I (insulin-dependent) and type II (non-insulin-dependent) diabetes mellitus for each practice behavior. RESULTS: Physicians' responses were classified as "high," "moderate," or "low" based on the American Diabetes Association guidelines. Forty-five percent of the physicians' responses were classified as high for referring all of their patients with type I diabetes mellitus to an eye care specialist annually as were 35% of the physicians' responses for referring their patients with type II diabetes mellitus. Fewer physicians reported high levels of routine in-office funduscopic examination. No relationship was found between the extent to which physicians performed an in-office funduscopic examination and patterns of referral to eye care specialists. Logistic regression analysis suggested that recent graduates and general internists are most likely to report behavior that is considered high (P < .05). CONCLUSION: The diabetic retinopathy-related practice patterns of primary care physicians in Indiana differ significantly from published guidelines.
Authors: Frank A Sloan; Derek S Brown; Emily Streyer Carlisle; Gabriel A Picone; Paul P Lee Journal: Health Serv Res Date: 2004-10 Impact factor: 3.402