Do marginal investments made by NHS healthcare commissioners in the UK produce the outcomes they hope to achieve? Observational study

OBJECTIVE: To investigate the effect of targeted marginal annual investments by local healthcare commissioners on the outcomes they expected to achieve with these investments. DESIGN: Controlled before and after study. SETTING: 152 commissioning organisations (primary care trusts) in England. METHODS: National surveys of commissioning managers in 2009 and 2010 to identify: the largest marginal investments made in four key conditions/services (diabetes, coronary heart disease, chronic pulmonary airways disease and emergency and urgent care) in 2008/2009 and 2009/2010; the outcomes commissioners expected to achieve with these investments; and the processes commissioners used to develop these investments. Collation of routinely available data on outcomes commissioners expected from these investments over the period 2007/2008 to 2010/2011. RESULTS: 51% (77/152) of commissioners agreed to participate in the survey in 2009 and 60% (91/152) in 2010. Around half reported targeted marginal investments in each condition/service each year. Routine data on many of the outcomes they expected to achieve through these investments were not available. Also, commissioners expected some outcomes to be achieved beyond the time scale of our study. Therefore, only a limited number of outcomes of investments were tested. Outcomes included directly standardised emergency admission rates for the four conditions/services, and the percentage of patients with diabetes with glycated haemoglobin <7. There was no evidence that targeted marginal investments reduced emergency admission rates. There was evidence of an improvement in blood glucose management for diabetes for commissioners investing to improve diabetes care but this was compromised by a change in how the outcome was measured in different years. This investment was unlikely to be cost-effective. CONCLUSIONS: Commissioners made marginal investments in specific health conditions and services with the aim of improving a wide range of outcomes. There was little evidence of impact on the limited number of outcomes measured.