Were NHS payments to English general practices in 2014-2015 associated with predictors of healthcare need?

Problem

In the new General Practitioner (GP) contract (2004) National Health Service (NHS) payments to English GPs aim to compensate general practices for additional workload and to reward achievement. The Global Sum Allocation Formula addresses the former by weighting payments, including adjustments for age, sex, morbidity and mortality, list turnover, patients living in nursing and residential homes, and rurality. Specific measures of deprivation and population ethnicity are not included. We examined the relationships between NHS payments and socio-economic deprivation, multimorbidity, ethnicity, and other factors associated with healthcare need in English general practices.

Approach

We used data of NHS payments to individual practices in England for 2014-2015. Categories of payment included Global Sum, Minimum Practice Income Guarantee, and Quality Outcomes Framework (QOF) achievement. Index of Multiple Deprivation (IMD) 2015 scores for individual general practices (higher scores mean greater deprivation), age structure, QOF register data for 2014-2015 (a practice’s multimorbidity index was calculated from the total number of patients on nine QOF registers divided by the number of patients in the denominator of QOF indicator SMOK002), and staff numbers were published by the Health & Social Care Information Centre. Data estimating appointment non-availability, practice patients’ ethnicity, and smoking status were obtained from the 2014-2015 General Practice Patient Survey.Of the 7,959 practices whose payments were documented, we excluded the following: those without an IMD score, with fewer than 500 patients, and/or the 1% highest and lowest paid, leaving 7,722 practices.Analyses used STATA 14.

Findings

Payments per weighted patient had a positively skewed distribution, median £131.09 (IQR: £118.22- £150.11). The Spearman correlation coefficients for payments with other indicators were: IMD -0.21, multimorbidity index -0.04, white ethnicity 0.13, aged 75 years plus 0.05, smokers -0.16, GP numbers 0.27, and reporting unable to obtain an appointment -0.15 (all p<0.002).When payments per weighted patient were divided into deciles, the median values in the lowest (median payment £102.40) and highest (median payment £227.46) deciles were, respectively, IMD 29.33 and 15.04, multimorbidity index 1.63 and 1.61, white ethnicity 89.63% and 98.30%, aged 75 years plus 7.04% and 9.28%, GPs per 10,000 population 2.28 and 3.89, and reporting being unable to obtain an appointment 11.59% and 5.97%.

Consequences

Payment had only small correlations with some healthcare need predictors: negatively with deprivation, non-white ethnicity and multimorbidity, and positively with age. Payment had a small negative correlation with a measurement of poor access. However, a limitation is that all of the analyses were univariable.Our findings suggest that the current funding formula is reactive to workload rather than anticipating a population’s healthcare needs, as set out in NHS England’s Five Year Forward View. If health inequalities are to be addressed, then it is worth reviewing the current funding formula.