Framework and indicator testing protocol for developing and piloting quality indicators for the UK quality and outcomes framework

Background: Quality measures should be subjected to a testing protocol before being used in practice using key attributes such as acceptability, feasibility and reliability, as well as identifying issues derived from actual implementation and unintended consequences. We describe the methodologies and results of an indicator testing protocol (ITP) using data from proposed quality indicators for the United Kingdom Quality and Outcomes Framework (QOF). Methods: The indicator testing protocol involved a multi-step and methodological process: 1) The RAND/UCLA Appropriateness Method, to test clarity and necessity, 2) data extraction from patients' medical records, to test technical feasibility and reliability, 3) diaries, to test workload, 4) cost-effectiveness modelling, and 5) semi-structured interviews, to test acceptability, implementation issues and unintended consequences. Testing was conducted in a sample of representative family practices in England. These methods were combined into an overall recommendation for each tested indicator. Results: Using an indicator testing protocol as part of piloting was seen as a valuable way of testing potential indicators in 'real world' settings. Pilot 1 (October 2009-March 2010) involved thirteen indicators across six clinical domains and twelve indicators passed the indicator testing protocol. However, the indicator testing protocol identified a number of implementation issues and unintended consequences that can be rectified or removed prior to national roll out. A palliative care indicator is used as an exemplar of the value of piloting using a multiple attribute indicator testing protocol - while technically feasible and reliable, it was unacceptable to practice staff and raised concerns about potentially causing actual patient harm. Conclusions: This indicator testing protocol is one example of a protocol that may be useful in assessing potential quality indicators when adapted to specific country health care settings and may be of use to policy-makers and researchers worldwide to test the likely effect of implementing indicators prior to roll out. It builds on and codifies existing literature and other testing protocols to create a field testing methodology that can be used to produce country specific quality indicators for pay-for-performance or quality improvement schemes.