A cross-sectional study exploring the relationship between GP practice funding and practice achievement

Problem

In the last decade, funding for general practice has decreased from 11% to 8% of total NHS spending (Appleby, 2014), while GP workload has increased by 19% (HSCIC, 2014). Recent work in Scotland has shown a mismatch between general practice funding and clinical need (McLean et al, 2015). Starfield (2005) demonstrated that greater investment in primary healthcare is associated with improved population health outcomes. Although the quality of clinical care improved after the introduction of the QOF scheme, there has been little evidence that this was the direct result of incentive payments. To date, the impact of broader aspects of practice funding on performance has not been examined. In early 2015, detailed primary care financial data were released by the Department of Health. Using these data, we aimed to explore the relationship between non-QOF NHS payments made to general practices in England and primary care performance.

Approach

Practice funding data were extracted from the National Health Applications and Infrastructure Services. We confined our analysis to practices with GMS contracts (n = 4252); data were not available for the locally determined contracts offered to PMS practices. We constructed regression models to explore the relationship between practice funding (‘global sum’ plus ‘MPIG’) and QOF outcomes, secondary care usage (outpatient, A&E and in-patient rates per 1000 registered patients) and patient satisfaction, adjusted for practice and demographic variables. We then conducted financial modelling to predict the impact of a hypothetical 10% funding increase on secondary care costs, for which we used standard cost estimates (Department of Health, 2014).

Findings

The mean funding was £68.62 per patient (5th centile: £59.83; 95th centile: £81.11). Higher funding was significantly associated with lower emergency admissions (regression coefficient B, -0.23), lower Accident and Emergency (A&E) attendances (B, -0.67) and higher patient satisfaction (overall satisfaction: B, 0.1; satisfaction with access: B, 0.1; nurse satisfaction: B, 0.5; doctor satisfaction: B, 0.2). We found no significant association with outpatient attendance or QOF performance.

In our financial model, a 10% increase in primary care funding would cost an additional £6862 per 1000 registered patients which would be offset by a £3725 reduction in emergency admissions and £576 reduction in A&E attendances.

Consequences

GMS practices with higher levels of funding had lower secondary care usage (emergency admissions and A&E attendances) and higher patient satisfaction. The lack of association between funding and QOF achievement may be attributable to the different funding stream and incentivisation for QOF. Our findings support the case for greater investment in primary care which would be substantially offset by reduced secondary care costs and would be associated with higher levels of patient satisfaction.