Impact of physicians beliefs and practices on cholesterol levels in patients with type 2 diabetes: A longitudinal assessment

Background Clinical trials demonstrate significant benefit from cholesterol management for patients with type 2 diabetes. The aim of this work was to explore the correlates of lipid management in patients with type 2 diabetes, including the subjective beliefs of physicians, setting of care, and patient-related factors. Methods This longitudinal outcomes research study involved 2359 patients with type 2 diabetes recruited by I I I general practitioners and 214 physicians practicing in diabetes clinics. Physicians' beliefs were assessed through a questionnaire administered when the study started in 1998., Main outcome measures were total cholesterol (TC) and LDL cholesterol (LDL-C) levels over 3 years and the proportion of patients treated with lipid-lowering drugs (LLDs). Results Less than one-third of the physicians (27%) stated that they routinely started pharmacologic therapy for TC values greater than or equal to200 mg/dL (more aggressive), whereas 46% considered a TC level greater than or equal to240 mg/dL as the threshold for the initiation of treatment (less aggressive). During 3 years of observation,. mean TC and LDL-C levels decreased from 2 15 +/- 40 mg/dL to 203 +/- 37 mg/dL and from 135 +/- 36 mg/dL to 126 +/- 35 mg/dL respectively, while the proportion of patients treated with LLPs increased from 13.2% to 24.6%; in particular, among individuals cared for by the more aggressive physicians, 30.0% were taking LLDs after 3 years, while only 17.7% of those followed by the less aggressive physicians and 18.1 % of those followed by > 1 physician were being treated with LLDs. Multilevel analysis showed that physicians' beliefs were an independent predictor of TC levels over the 3-year period. In patients treated with LLDs, TC levels decreased on average by 14%, and LDL-C levels decreased by 20%. Conclusion Our data show that physicians' beliefs in more aggressive management strategies will result in better mean TC values over a 3-year period.