Expansion of activity-based hospital payment in Israel: evaluation of effects on inpatient activity

Background
In 2013-14, Israel stepped up the replacement of per-diem payments by Procedure-Related Group (PRG) based hospital payments, a local version of Diagnosis-Related groups (DRGs). PRGs were created for selected procedures in urology, general surgery, gynecology and ophthalmology. We analyzed how did this change affect inpatient activities, measured by the number of discharges, average length of stay (ALoS), and the case-severity Charlson Comorbidity Index (CCI).
Methods
We investigated the impacts of the PRG-payment reform on 15 procedures. Observations covered groups of inpatients, by age and gender, who underwent these procedures in 2005-2016 at all non-profit hospitals. We examined the effect of the payment change on the number of discharges, ALoS and CCI using a multivariable analysis of Ordinary Least Squares controlling for patients, hospital characteristics, and year fixed-effects.
Results
Data on 89,533 patients were examined. During the study period, the ALoS decreased except for one procedure, the number of inpatients increased for most procedures, and case severity remained stable. The multivariable analysis suggests that the transition to PRG-payments contributed to changes in ALoS or case severity for only 3 out of 15 procedures examined. The PRG-reform contributed to changes of 10%-45% in the number of patients, but there was no clear trend: it increased in 9, and decreased in 5. The changes did not follow a clear pattern according to procedures’ price changes after the reform.
Conclusions
Factors that may have hampered the effects of the PRG-reform are conflicting incentives created by other co-existing hospital-payment components, such as revenue caps and retrospective subsidies, and the lack of resources to increase productivity.