Are quality improvements in the get with the guidelines: stroke program related to better care or better data documentation?

BACKGROUND- Increased compliance with performance measures could reflect better care or better data documentation. We examined trends in the documentation of eligibility criteria, treatment contraindications, and missing data in the Get With The Guidelines-Stroke program to quantify their contribution to increased performance measure compliance. METHODS AND RESULTS- Data on 569 883 ischemic stroke admissions to 1028 GWTG-Stroke hospitals between April 2003 and September 2009 were obtained. Seven measures were examined: intravenous recombinant tissue plasminogen activator therapy, early antithrombotics, deep vein thrombosis prophylaxis, anticoagulants for atrial fibrillation/flutter, discharge antithrombotics, lipid therapy, and smoking cessation. Within each target population, the proportion of subjects treated, not treated, not treated because of contraindications, or with missing data were generated by calendar year. There were minimal changes in the size of the target populations for 6 of the measures; however, the size of the deep vein thrombosis prophylaxis population was reduced ≈5% in 2008 because of a format change to the data collection form. All measures showed significant increases in the proportion of eligible subjects treated across the study period. These increases occurred without major shifts in contraindications or missing data, with the exception of anticoagulation for atrial fibrillation/flutter where the increase occurred in conjunction with a decline in contraindications. Similar findings were seen when the data were examined by the duration of hospital participation in the program. CONCLUSIONS- These findings suggest that the majority of performance improvement in the Get With The Guidelines-Stroke program represent an increase in the number of patients with stroke treated and not changes to the underlying target populations or documentation of contraindications or missing data.