Matt Sutton: reduced mortality with hospital Pay for Performance in England
Download as PPTX, PDF1 like784 views
The document discusses a hospital pay-for-performance (P4P) scheme implemented in England, which aimed to improve healthcare outcomes by linking provider payments to quality indicators. An evaluation showed that this initiative resulted in a significant reduction in mortality rates by 1.3 percentage points over 18 months, equating to approximately 890 fewer deaths among patients with specific incentivized conditions. Overall, the scheme demonstrated cost-effectiveness and highlights the potential for P4P models to improve healthcare quality when designed and implemented effectively.
Matt Sutton: reduced mortality with hospital Pay for Performance in England
- 1. Reduced mortality with Hospital Pay for Performance in England Matt Sutton Professor of Health Economics Payment reform: Moving beyond Payment by Results The King’s Fund, 17 January 2013 Project funded by NIHR Health Services & Delivery Research Programme
- 2. Pay for Performance (P4P) • Health care commissioners can pay providers on the basis of: – an agreed service specification – population coverage (capitation) – volume – performance • Internationally, more third-party payers are linking a proportion of provider revenue to achievement of quality indicators • Examples in England: the Quality and Outcomes Framework, Best Practice Tariffs, and the Commissioning for Quality and Innovation framework
- 3. P4P and health outcomes • Increased adoption of P4P is occurring despite a scant evidence base – by 2009, few schemes had been evaluated at all – evaluations show at best modest and temporary effects on quality – recent Cochrane review (Flodgren, 2011) found no evidence that financial incentives lead to improvements in health outcomes • More inclusive review (van Herck, BMCHSR, 2010) highlighted that several aspects of P4P may be important: – the design of schemes – their mode of implementation – the context in which they are introduced
- 4. Advancing Quality • First hospital P4P scheme to be introduced in the UK (October 2008) • Based on Hospital Quality Incentive Demonstration (HQID) from the US • Adopted by all 24 NHS Acute Trusts in the North West SHA • Covered five patient groups: pneumonia, CABG, AMI, heart failure, hip/knee • Performance on 28 quality indicators was reported by participating Trusts – collected and fed back quarterly and published annually • Tournament scheme (for first 12 months) – top 6 Trusts received a 4% bonus on their tariff payments – next 6 Trusts received a 2% bonus on their tariff payments • Bonuses allocated internally to clinical teams for investment in care
- 5. Our evaluation • Independent evaluation funded by the NIHR Health Services & Delivery Research Programme • Collaboration between Universities of Nottingham, Manchester, Cambridge and Birmingham • Five-year study: April 2009 – March 2014 • Combination of qualitative and quantitative research
- 6. Estimation of effect on mortality • Data from national Hospital Episode Statistics • Deaths within 30 days of admission (in any hospital in England) • For patients admitted for: – three incentivised conditions (AMI, heart failure and pneumonia) – six reference conditions • Periods: 18 months before and first 18 months after introduction • Comparison of 24 North West Trusts with 132 Trusts in rest of England • Risk-adjustment using age and sex, primary diagnosis, 31 co-existing conditions, type of admission, residential location on admission
- 7. Changes in unadjusted mortality rates North West Rest of England Before After Change Before After Change AMI 12.4 11.0 -1.4 11.0 10.7 -0.3 Heart failure 17.9 16.6 -1.3 16.6 16.1 -0.6 Pneumonia 28.0 25.9 -2.2 27.2 26.3 -0.9 Reference conditions 13.3 13.0 -0.3 11.7 11.0 -0.7 Mortality measured in percentage points.
- 8. Difference-in-differences analyses of adjusted mortality Health condition Between-Region Triple Difference Difference in Differences Reference conditions 0.3 (-0.4 to 1.1) - Incentivised conditions -0.9 (-1.4 to -0.4) -1.3 (-2.1 to -0.4) AMI -0.3 (-1.0 to 0.4) -0.6 (-1.7 to 0.4) Heart failure -0.3 (-1.2 to 0.6) -0.6 (-1.8 to 0.6) Pneumonia -1.6 (-2.4 to -0.8) -1.9 (-3.0 to -0.9) Mortality measured in percentage points (95% CI).
- 9. Headline results • There was a larger overall reduction in mortality of 1.3 percentage points in the North West when the P4P was introduced • Relative rate reduction of 6% • Over 18 months, equates to a reduction of 890 deaths (95% CI, 260 to 1500) amongst population of 70,644 patients with these conditions
- 10. Further analyses • No significant differences in proportions of patients discharged to institutions • Trends in mortality were similar in the North West to the rest of England before introduction of the scheme • Results unaffected by controlling for baseline mortality and changes in patient volumes • Similar results when exclude the south of England • Largest reductions in mortality achieved in small Trusts and Trusts rated “excellent” or “good” by CQC • Cost-effectiveness – scheme cost £13M to set-up, administer and provide bonuses – estimated to have generated over 3,000 Quality-Adjusted Life Years – cost-per-QALY well below NICE threshold
- 11. How and why? • Results differ from those found for HQID in the US • Not feasible that the mortality reductions were only due to improvements on the incentivised process measures • Providers adopted range of quality improvement strategies • Identification and targeting of particular patient groups • Principal differences from US scheme – Universal participation – Size of bonus – Probability of bonus – Regional collaboration
- 12. Implications • NB. have only considered first 18 months of scheme • Pay-for-Performance can be associated with substantial mortality reductions • Financial incentive not as high-powered as QOF, BPTs, CQUIN • AQ is a P4P programme: – regional initiative – new data collection and public reporting – bonuses to clinical teams
- 13. Concluding remarks • NW SHA imported a P4P scheme from the US from October 2008 • Translated to NHS context – universal participation, regional collaboration • Associated with a substantial reduction in mortality • Cost-effective use of resources in first 18 months • Not just direct result of improvements in the incentivised measures • A quality improvement programme supported by financial incentives • Differs in some potentially important ways from other P4P initiatives adopted in the NHS
- 14. Reduced mortality with Hospital Pay for Performance in England Matt Sutton For more information: matt.sutton@manchester.ac.uk Work published as: Sutton M et al, NEJM 2012; 367: 1821-8.