Results: Purpose: To quantify provider use of 21 specific treatment strategies for diabetes patients, across the three clinical domains of glucose, lipid and blood pressure control, and to correlate use of these strategies with measures of clinical inertia, defined as failure to take pharmacologic action for patients not at clinical goal.
Methods: The study involved a retrospective examination of the treatment patterns of diabetes patients by 118 primary care physicians from 3/1/05 to 2/28/06. Expert consensus identified 10 glycemic, 7 blood pressure, and 4 lipid treatment strategies (such as initiation or titration of basal and prandial insulins, initiation and titration of statins, and use of combination drugs for blood pressure). Encounter and patient level data was also assessed to determine overall use of the treatment strategies in the provider’s patient population. Analysis correlated use of specific strategies with patient level clinical inertia rates relative to A1c, BP, and LDL levels.
Results: Six 10 glucose treatment strategies (range of r .29 to .76), 7 of 7 BP treatment strategies (range of r .26 to .90), and 3 of 4 lipid treatment strategies (range of r .41 to .79) were correlated significantly (p<0.05) with domain-specific clinical inertia (the proportion of patients not at goal on an encounter without a medication intensification.
Conclusions: The ability of primary care physician to take clinical action for patients not achieving clinical goals among their panel of diabetes patients can be related to a the use of 21 specific identified clinical drug treatment strategies. EMR-derived clinical data can identify physician-specific treatment strategies and identify targets for interventions to improve diabetes care.