Pay-for-performance in health care: Evidence from a nationwide randomized experiment in the home health sector

How to cite

li, jun. 2019. "Pay-for-performance in health care: Evidence from a nationwide randomized experiment in the home health sector." AEA RCT Registry. AEARCTR-0004159. https://doi.org/10.1257/rct.4159-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

jun li*

* Primary investigator

Registry ID
AEARCTR-0004159
DOI
https://doi.org/10.1257/rct.4159-1.0
First registered
Last updated
Status
On going
Trial dates
2016-01-01 – 2020-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/4159

Abstract

This pre-specified analysis plan describes the planned approach and main analyses of a nationwide stratified cluster-randomized trial of a Medicare value-based purchasing program for home health care. The Home Health Value-Based Purchasing program was designed by the Centers for Medicare & Medicaid Services and implemented in January 2016. Randomization was conducted at the state level within nine regions. In the first year, 1,626 home health agencies from 9 states were assigned to the treatment group and 6,981 agencies from 41 states were assigned to the control group. This study will examine the effect of the intervention on health care quality in the first performance year (i.e., from January through December 2016).

Intervention

The Home Health Value-Based Purchasing (HHVBP) program is a Medicare pay-for- performance program structured in a similar way as the Hospital Value-Based Purchasing Program and Skilled Nursing Facility Value-Based Purchasing Program. The program financially rewards (or penalizes) Medicare-certified home health agencies based on their performance on pre-specified measures and for reporting new quality information to CMS.

Experimental design

In November 2015, CMS randomly selected 9 states from 9 regions to participate in the HHVBP. The program started on January 2016 and lasts through December 2020 for Massachusetts, Maryland, North Carolina, Florida, Washington, Arizona, Iowa, Nebraska, and Tennessee.
To select the participants, CMS first divided states into 9 regions based on a variety of characteristics, including geographic proximity, home health utilization rates, ownership model of agencies, the proportion of home health users that are Medicare-Medicaid dually eligible beneficiaries, and average episodes of care per agency. Next, CMS used a random number generator to select one state within each of the 9 regions to be treated.
Agencies that meet the program’s inclusion criteria within the 9 treatment states face mandatory participation. There are two inclusion criteria. First, only agencies that were Medicare certified in the baseline year (2015) and in the performance year (2016) are included. Second, agencies must have at least 5 measures that meet minimum denominator case requirements (e.g., 20 episodes for quality measures and 40 HHCAHPS surveys) in both the baseline and performance periods. Agencies that do not meet these inclusion criteria do not compete for incentive payments.

Design details

Randomization method
Random number generator from computer program (conducted by the Centers for Medicare & Medicaid Services).
Randomization unit
Unit of randomization is state and within each treatment state, agencies that meet the eligibility criteria are included in the program.
Clusters
50 states
Observations
8611 home health agencies
Arms
50 states (9 treatment with 1626 home health agencies, 41 control with 6981 home health agencies)
Primary outcomes
Self-reported outcomes measures: Improvement in Ambulation-Locomotion Improvement in Bed Transferring Improvement in Bathing Improvement in Dyspnea Improvement in Pain Interfering with Activity Improvement in Management of Oral Medications Self-reported process measures: Pneumococcal Polysaccharide Vaccine Ever Received Influenza Immunization Received for Current Flu Season Drug Education on All Medications Provided to Patient/Caregiver Patient Survey: How often the home health team gave care in a professional way How well did the home health team communicate with patients Did the home health team discuss medicines, pain, and home safety with patients How do patients rate the overall care from the home health agency Would patients recommend the home health agency to friends and family Administrative-claims measures: No unplanned Hospitalization No Emergency Department Use without Hospitalization

Cited by

No citation data available for this registration yet.

Keywords