Background Since 2004, policy reforms in the English NHS have allowed new contracting mechanisms for provision of general practice services. A wider variety of provider organisations were allowed under the alternative provider of medical services (APMS) contract. In addition, Primary Care Trusts have been able to directly provide general practice services via the PCTMS contract. These new contracts were introduced to challenge existing models of care and encourage innovation to meet changing healthcare needs. Concerns have been expressed about the increasing diversity of providers and commercialisation on primary care performance. We looked at whether there is variation in performance amongst practices according to the type of contract they have.
Methods Fifteen commonly reported performance indicators were selected to represent a broad spectrum of clinical quality covering clinical effectiveness, patient experience, efficiency and access. Practice-level performance data, demographic data and contract-type for all general practices in England with more than 1000 patients for the year 2011/12 were extracted from open Internet sources, including: the GP Patient Survey, the Quality Outcomes Framework, the NHS Information Centre Indicator Portal and NHS Comparators. We compared performance across these indicators by contract type with adjustments for age, gender, ethnicity and deprivation.
Results In 2011/12, there were 4517 GMS practices, 3313 PMS practices, 217 APMS practices and 72 PCTMS practice. When compared to GMS practices, APMS performed worse across 9 of the 15 indicators, and better on 2 indicators (P < 0.05). APMS practices had greater levels of satisfaction with opening hours and higher rates of generic prescription for statins, but lower performance for patient experience and clinical QOF indicators. PCTMS practices performed worse than GMS on 9 indicators, and better on none. In particular, PCTMS and APMS practices had lower patient experience, higher levels of diabetes exception reporting and higher levels of ambulatory care sensitive admissions to hospitals. (P < 0.01 for all).
Conclusion There is variation in performance for a number of measures of primary care quality that occurs with contract type, although the cause is not clear. Both APMS and PCTMS practices were consistently more poorly performing in certain areas. It is not clear whether these variations in performance are related to the contract type, or to historical contextual factors (including previous poor management). However these differences remain after adjusting for age, sex, ethnicity and multiple deprivations at practice-level. Further work to identify consequences of increased provider diversity in primary care would be useful.