Abstract
Background: Unplanned hospital admissions are costly and disproportionately affect people who are socioeconomically disadvantaged and from an ethnic minority group. A national primary care pay-for-performance scheme, the Quality and Outcomes Framework (QOF), was introduced in England in 2004 to financially incentivise general practices to meet a range of performance indicators, but the QOF's impact on unplanned hospital admissions remains unclear. We examined the association between unplanned hospital admissions for cardiovascular disease (CVD), individual-level characteristics and achievement of key QOF indicators for CVD at the patients' registered general practice. Methods: This study used the Connected Bradford dataset, which links individual-level primary and secondary care data. Our analytical sample included 508,977 patients registered with a Bradford District general practice from 2017 to 2019. Logistic regression was used to estimate associations between achievement of relevant QOF indicators and unplanned admissions for cardiovascular diseases, adjusting for individual-level differences in age, sex, ethnicity, socioeconomic status and pre-existing health conditions. Results: Significantly reduced odds of unplanned CVD hospital admissions were associated with attending a practice with higher achievement rates for QOF indicators relating to atrial fibrillation management (OR 0.97, p < 0.001), diabetes management (OR 0.98, p = 0.002), and smoking cessation (OR 0.98, p = 0.038). Conversely, increased odds of unplanned admission were associated with higher achievement for QOF indicators relating to antiplatelet or anticoagulation medication (OR 1.06, p < 0.001) and blood pressure control for diabetic patients (OR 1.02, p = 0.03). Individual-level characteristics significantly associated with increased risk of unplanned admission included living in the most deprived fifth of neighbourhoods (OR 2.00, p < 0.001) and having Pakistani ethnicity (OR 1.65, p < 0.001). Primary care diagnoses of hypertension (OR 1.79, p < 0.001), diabetes (OR 1.56, p < 0.001), chronic cardiac disease (OR 2.79, p < 0.001), and stroke (OR 1.6, p < 0.001) were all statistically significant and associated with higher odds of unplanned admissions for CVD. Conclusions: We found mixed evidence for an association between practice-level QOF achievement and unplanned hospital admissions for CVD. There were large ethnic and socioeconomic inequalities in unplanned admissions for cardiovascular disease. Supporting general practices to appropriately improve their achievement of key cardiovascular disease related QOF indicators and reducing socioeconomic inequalities might likely reduce the number of unplanned hospital admissions.