Impact of pay-for-performance incentives for stroke unit access on hospital costs and utilisation.

Abstract

Background: Pay-for-performance financial incentives for stroke unit access in Queensland hospitals led to improved access and survival. We sought to assess the impact on hospital length of stay (LOS) 30-day readmission and costs. Method(s): Interrupted time series analysis using linked, patient-level hospital admission and costing datasets for all Queensland public hospital admissions, July 2009 to June 2017. We compared change of level and trends in outcomes for stroke with historical trends, and non-incentivised myocardial infarction (MI) controls, before vs after pay-for-performance introduction. Incentives were paid on achievement of incremental targets for proportion of patients admitted to acute stroke units (average $AUD218,000/hospital/year over first three years), followed by a 10% bonus on Diagnosis Related Group funding for eligible patients admitted to stroke units. Implementation was led by a multidisciplinary quality improvement network. Result(s): 23,572 patients with stroke and 39,511 with MI were included. Pay-for-performance had no impact on acute or total LOS. Trends in LOS were similar in both stroke and non-incentivised MI, with a plateau in historical downward trends in both groups following pay-for-performance introduction. There was no difference in trends of costs between stroke and MI, despite an unexplained decrease in MI costs over the year of stroke incentive introduction (MI: -$4 278/admission, 95%CI -$5 280, -$3 275; stroke: - $1 692/admission, 95%CI -$4 440, $1056). Readmissions continued upward trends, unaffected by pay-forperformance. Conclusion(s): By improving quality of care and survival without impact on hospital LOS, costs, or readmissions, these pay-for-performance incentives achieved improved value for healthcare expenditure.