Adjusting primary-care funding by deprivation: a cross-sectional study of Lower layer Super Output Areas in England

Previous research has called for general practice funding to be adjusted by deprivation data. However, there is no evidence that this adjustment would better meet clinical need.

To assess (1) how accurately the capitation formula (Carr-Hill), and total general practice funding predicts clinical need and (2) whether adjusting by the Index of Multiple Deprivation (IMD) score improves accuracy.

A cross-sectional analysis of 32 844 Lower layer Super Output Areas (LSOAs) in England in 2021-2022. Sensitivity analysis used data from 2015-2019.

Weighted average Carr-Hill Index (CHI), total general practice funding, and five measures of clinical need were calculated for each LSOA. For both CHI and total funding, four sets of generalised linear models were calculated for each outcome measure: unadjusted; adjusted for age; adjusted for IMD; and adjusted for age and IMD. Adjusted R ² assessed model accuracy.

In unadjusted models, CHI was a better predictor than total funding of combined morbidity index (CMI) (R ² = 49.81%, 29.31%, respectively), combined diagnosed and undiagnosed morbidity (R ² = 43.52%, 21.39%) and emergency admissions (R ² = 32.75%, 16.95%). Total funding was a better predictor than CHI of GP appointments per patient (R ² = 28.5%, 22.5%, respectively) and age and sex standardised mortality rates (R ² = 0.42%, 0.37%). Adjusting for age and IMD improved all 10 models (R ² = 62.15%, 53.15%, 48.57%, 38.47%, 40.53%, 32.84%, 29.11%, 34.58%, 25.21%, 25.23%, respectively). All age and IMD adjusted models significantly outperformed age-adjusted models (P<0.001). Sensitivity analysis confirmed findings.

Adjusting capitation or total funding by IMD would increase funding efficiency, especially for long-term outcomes such as mortality. However, adjusting for IMD without age could have unwanted consequences.