The usage of data in NHS primary care commissioning: a realist evaluation

Background: To improve health outcomes and address mounting costs pressures, policy-makers have encouraged primary care commissioners in the British National Health Service (NHS) to increase the usage of data in decision-making. However, there exists limited research on this topic. In this study, we aimed to understand how and why primary care commissioners use data (i.e. quantitative, statistical information) to inform commissioning, and what outcomes this leads to. Methods: A realist evaluation was completed to create context-mechanism-outcome configurations (CMOs) relating to the contexts influencing the usage of data in primary care commissioning. Using a realist logic of analysis and drawing on substantive theories, we analysed qualitative content from 30 interviews and 51 meetings (51 recordings and 19 accompanying meeting minutes) to develop CMOs. Purposive sampling was used to recruit interviewees from diverse backgrounds. Results: Thirty-five CMOs were formed, resulting in an overarching realist programme theory. Thirteen CMOs were identical and 3 were truncated versions of those formed in an existing realist synthesis on the same topic. Seven entirely new CMOs, and 12 refined and enhanced CMOs vis-à-vis the synthesis were created. The findings included CMOs containing contexts which facilitated the usage of data, including the presence of a data champion and commissioners’ perceptions that external providers offered new skillsets and types of data. Other CMOs included contexts presenting barriers to using data, such as data not being presented in an interoperable way with consistent definitions, or financial pressures inhibiting commissioners’ abilities to make evidence-based decisions. Conclusions: Commissioners are enthusiastic about using data as a source of information, a tool to stimulate improvements, and a warrant for decision-making. However, they also face considerable challenges when using them. There are replicable contexts available to facilitate commissioners’ usage of data, which we used to inform policy recommendations. The findings of this study and our recommendations are pertinent in light of governments’ increasing commitment to data-driven commissioning and health policy-making.