Figueroa 2016.

Methods	Study design (CBA, ITS): CBA Allocation linked to study or natural experiment: natural experiment Selection of hospitals (intervention/control): comparison 1: all conditions targeted by P4P/selected conditions not targeted by P4P; comparison 2: acute care hospitals/hospitals in another region and critical access hospitals Unit of allocation (region, hospital, country): comparison 1: individual level; comparison 2: region and hospital type Nature of desired change: introduction Data source: 100% Medicare inpatient claims data from 2008 through 2013 Unit of analyses: comparison 1: individual; comparison 2: hospital Number of measurements (before, transition, after, unit [e.g. years]): 14/NA/10 (quarter) Statistical analyses: ‐ Method: difference‐in‐difference; random effects linear spline regression ‐ Adjustment factors: comorbidities, seasonal variation
Participants	Country/region/setting: USA/whole country/hospitals Health system characteristics: Medicare Number of hospitals included in the analysis (intervention/control): 2919/1348 Characteristics of hospitals (before [whole population], intervention/control): acute care hospitals/hospitals in another region and critical access hospitals Number of beds: small: 27.7%/92.7% medium: 57.4%/6.8% large: 13.9%/0.6% Number of patients (mean annual Medicare volume): 2671/385 Number of wards: NR Departments: NR Owner: For profit: 20.4%/5.3% Private not for profit: 64.8%/54.5% Public: 14.8%/40.3% Teaching status: major: 9.0%/0.5% minor: 24.0%/6.4% none: 67.1%/93.2% Location: NR Number of patients included in the analysis (intervention/control): 2,252,818/177,800 Characteristics of patients (intervention/control): acute myocardial infarction, congestive heart failure, and pneumonia/stroke, sepsis, gastroenteritis and esophagitis, gastrointestinal bleed, urinary tract infection, metabolic disorder, arrhythmia, renal failure Age (mean): 79.8/81.0 Gender (male): 42.1%/39.5% Casemix: NR Indications: NR Existing/other quality programs: not reported Other relevant context information: none
Interventions	Rewards or penalizes hospitals based on their performance on multiple domains of care, including clinical processes, clinical outcomes (e.g. 30‐day mortality for acute myocardial infarction, pneumonia, and heart failure), patient experience, and, latter, cost efficiency. Performance is determined based on hospitals’ absolute achievement compared with the national average, or improvement compared with their own performance in the baseline period, depending on which is greater. Funding is designed to be budget neutral; Medicare withholds a percentage of inpatient payments to prospectively paid hospitals and then redistributes this money back to hospitals based on their performance. National in scope and obligatory. Control: Comparison 1: No P4P Comparison 2: Probably mixture of basic P4P (incentives of hospitals without penalties) and no P4P
Outcomes	Mortality (30 days)
Notes	Funding/conflict of interest: work received no support from any organization; authors no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work Information for subgroup analysis: low baseline performance Other comments: ‐
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	CBA
Allocation concealment (selection bias)	High risk	CBA
Blinding of participants and personnel (performance bias) All outcomes	High risk	CBA, intervention cannot be blinded
Blinding of outcome assessment (detection bias) Objective outcomes	Low risk	Mortality
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Not sufficiently reported
Selective reporting (reporting bias)	Low risk	No indication for selective reporting
Other bias	Low risk	No evidence of other risk of bias
Baseline outcomes similar All outcomes	High risk	Difference in baseline outcomes
Free of contamination	High risk	Comparison 1: high risk, allocation on individual level and intervention cannot be blinded Comparison 2: low risk, allocation on hospital level
Baseline characteristics similar	High risk	Difference in baseline characteristics