Abstract
Background/Aims Diabetes Mellitus (DM) is a common condition associated with increased risk of microvascular and macrovascular complications.
Aim To determine if a primary care model for DM using a 9 component all-or-none bundle (Bundle) of measures improved microvascular and macrovascular complications compared to usual care (Non-Bundle).
Methods In 2006 a diabetes system of care using a Bundle of measures was implemented for some members of the Geisinger Health Plan. A cohort of 4097 primary care patients using the diabetes bundle system of care is matched with a cohort of Non-Bundle care patients to balance the baseline characteristics. The cumulative risk of incident microvascular and macrovascular complications is obtained by a weighted Cox proportional hazard model, in which individual post-bundle exposure is adjusted by inverse probability weighting (IPW).
Results 4097 patients with Bundle care were compared to Non-Bundle patients for the first three years after implementation of bundle management. The adjusted hazard ratios for myocardial infarction (MI) (HR: 0.78, 95% CI: 0.67–0.92), stroke (HR: 0.78, 95% CI: 0.64–0.95), and retinopathy (HR: 0.82, 95% CI: 0.69–0.98) were all significantly lower among patients in the bundled care model. Too few incident amputation cases (11 for Bundle, and 6 for Non-Bundle) limit the power to detect significant hazard difference for amputation (HR: 1.32, 95% CI: 0.45–3.85). Above findings are based on an observational design and the population is limited to those enrolled in a health plan in central and northeastern Pennsylvania.
Discussion A primary care all-or-none bundle of measures (Bundle) for management of patients with diabetes may reduce the risk of microvascular and macrovascular events over time. The effectiveness of the Bundle management could be observed in as early as two years.
