How do clinical quality and patient satisfaction vary with provider size in primary care? Evidence from English general practice panel data

We examine the relationship between general practice list size and measures of clinical quality and patient satisfaction. Using data on all English practices from 2005/6 to 2016/17, we estimate practice level models with rich data on patient demographics, deprivation, and morbidity. We use lagged list size to allow for potential simultaneity bias from the effect of quality on list size. We compare results from three different estimation methods: pooled ordinary least squares, random practice effects, fixed practice effects. With all three estimation methods increased list size is associated with reductions in all four measures of patient satisfaction. Increases in list size are associated with worse performance on three clinical quality indicators and better performance on three, though the precision and size of the associations varies with the estimation method. The absolute values of the elasticities of the ten quality indicators with respect to list size are small: in all cases a 10% change in list size would change quality by less than 1%. The lack of evidence that large practices have markedly better quality suggests that encouraging practices to form larger, but looser, groupings, may not, in itself, improve their performance.