Using routinely gathered data to evaluate locally led service improvements.

Background Between 2009 and 2010 NHS Ealing tested the feasibility of a) combining data from more than one data-domain at the same time to quantify patient movement across the primary care/acute hospital boundary, and b) establishing online analyses so they can be constantly updated with near real-time data to compare different subsets of patients. The reports allowed us to see: changes in hospital admissions before and after referral to community matrons of patients with complex conditions from one practice-based commissioning (PBC) groupchanges in hospital bed-days of all patients from one practice or PBC group during a complex intervention designed to assist inter-disciplinary collaboration. Results The teams leading the projects found that the reports gave them confidence in the projects and helped to influence local policy. Discussion GP consortia need to evaluate complex service improvements in order to contain costs and improve quality. They will find such reports helpful to give ongoing feedback and this may help to keep people engaged. Present plans for data warehousing in London do not have this ability - they do not combine data from across the whole health economy, and are focused either on claims validation or risk stratification.