Record rewards: the effect on risk factor monitoring of new financial incentives for UK general practices

An innovative and expensive performance-related pay scheme was introduced for general practices across the UK in 2004. It was not piloted and baseline performance data were not collected prior to its introduction. We estimate the impact of this Quality and Outcomes Framework (QOF) by analysing annual rates of recording of blood pressure, smoking status, cholesterol, body mass index and alcohol consumption based on individual patient records from 315 general practices over the period 2000/1 to 2005/6. The recording of each risk factor is designated as incentivised or unincentivised for each individual based on whether they have one of the diagnoses targeted by the QOF. The estimated impact is sensitive to the dynamic specification of the recording process and was substantially larger on the targeted patient groups (+19.9 percentage points) than the untargeted groups (+5.3). We also find positive spillovers of (+10.9) for the targeted groups onto unincentivised factors. We propose that the intended rewards per additional record were under-estimated, because account was not taken of substantial multiple-payment for co-morbid patients, levels of pre-QOF recording and the additional rewards available for risk factor control that would be achieved by measurement alone. Based on naïve assumptions, we estimate the intended financial reward per additional risk factor record to be £4.40. Allowing for co-morbidity, pre-QOF performance and the additional ‘control’ rewards, increases this average reward eleven-fold, to £48.90. Taking account of the positive spillovers reduces this figure to £25.10, but it remains substantially larger than what appears to have been intended.