Moving on from response rates: linking patient level ambulance data to ED, hospital and survival data to assess quality and performance
The problem The ambulance service has no information about what happens beyond the prehospital phase of care. This leads to process measures, e.g. response rates, being used as a proxy for quality of care. Research aim To develop better ways of measuring the quality and performance of ambulance service care. Methods Prehospital Outcomes for Evidence Based Evaluation (PhOEBE) is a five year National Institute for Health Research (NIHR) programme to develop better ways of measuring the performance, quality and impact of ambulance service care by: identifying and prioritising ambulance related outcome measures; creating a new information source by linking together routinely collected data; using the linked dataset to build predictive models to assess what aspects of care can predict good or poor outcomes (mortality and non-mortality) and measure the effectiveness and quality of ambulance service care; assessing the practical use of the measures and the linked data as a way to support quality improvement in the NHS; in order to provide better information about effectiveness and quality of care. We developed a data linking methodology that was acceptable to patients and complied with information legislation. Data from two participating ambulance services were processed and linked to other routinely collected data through the NHS Health and Social Care Information Centre's Trusted Data Linkage Service, which is a designated safe haven for health information. Conclusions It is possible to link ambulance service data to subsequent care information. The new database can be used to support audit and research and to measure the impact of any new changes and innovations in how ambulance services provide care, to ensure continued improvements for patients. Allied to this we used consensus methods to develop a set of outcome measures for measuring the quality and performance of ambulance service care, which will be tested within the linked dataset using predictive models.