Objective: Using a measure of provider type that includes "shared care" to determine the contribution of provider type on receipt of general and diabetes-related preventive health services.
Methods: Automated clinical and administrative data were used to identify adult patients with type 1 and 2 diabetes receiving care from a multispecialty, salaried group practice and enrolled in a large health maintenance organization between 3/97 and 2/98 (n = 10,991). Logistic regression models were fit using generalized estimating equation approaches to evaluate the contribution of provider type on service receipt.
Measures: Preventive service receipt included receipt of glycated hemoglobin and lipid testing, retinal examinations, pneumococcal vaccines, Papanicolaou (Pap) smears, and mammograms. Multivariable analyses adjusted for age, sex, race, marital status, household income, diabetes-related comorbidities and complications, prescription drug use, laboratory testing results, and frequency of medical care contact.
Results: Patients seeing an endocrinologist and primary care physician (PCP) were more likely than those seeing endocrinologists alone to receive glycated hemoglobin testing (OR, 1.42), lipid testing (OR, 1.72), mammograms (OR, 2.12), and Pap smears (OR, 2.36), and more likely than those seeing PCPs alone to receive glycated hemoglobin testing (OR, 1.79), lipid testing (OR, 1.54), retinal examinations (OR, 1.33), and mammograms (OR, 1.43). Compared with patients seeing PCPs only, patient's seeing endocrinologists only were more likely to receive retinal examinations (OR, 1.37) and less likely to receive Pap smears (OR, 0.46).
Conclusions: Care delivered by no one single provider type is associated with greater receipt of all recommended services. Instead, patients seeing both an endocrinologist and a PCP are most likely to receive recommended services.