Understanding the gap between good processes of diabetes care and poor intermediate outcomes: Translating Research into Action for Diabetes (TRIAD).

BACKGROUND: Performance of diabetes clinical care processes has improved recently, but control of hemoglobin A1c (A1c) and other vascular disease risk factors has improved more slowly.

OBJECTIVES: To identify patient factors associated with control of vascular disease risk factors among diabetes patients receiving recommended care processes.

POPULATION: Managed care enrollees who participated in the TRIAD (Translating Research into Action for Diabetes) Study and received at least 5 of 7 recommended care processes during the 12 months before the second survey (2002-2003).

METHODS: Comparison of 1003 patients with good control of A1c (<8%), systolic blood pressure (<140 mm Hg) and LDL-cholesterol (<130 mg/dL) versus 812 patients with poor control for at least 2 of these factors.

RESULTS: Poorly controlled patients were younger, more frequently female, African American, with lower education and income (P < 0.001 for each). General health status was lower, body mass index higher, and insulin treatment more frequent; history of prior coronary heart disease was less frequent. They were more likely to indicate depression and hopelessness and to identify costs as a barrier to self-care; less likely to report trust in their regular physician; and more likely to smoke cigarettes and be physically inactive. Adjusting for demographic and clinical variables, concerns about costs, low trust in one's physician, current smoking, and physical inactivity remained associated with poor control. However, inclusion of these 4 variables in a single model did not diminish associations of race/ethnicity or education with control.

CONCLUSIONS: Clinical, socioeconomic, psychosocial, and behavioral factors were independently associated with poor control. However, these factors did not fully explain observed racial and socioeconomic disparities in control.