General practitioners’ claiming of fees: An experimental study on the effect of feedback

How to cite

Grøsland, Mari; Haatvedt, Jens-Christian; Reme, Bjørn-Atle. 2024. "General practitioners’ claiming of fees: An experimental study on the effect of feedback." AEA RCT Registry. AEARCTR-0013545. https://doi.org/10.1257/rct.13545-1.0

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

Mari Grøsland* , Jens-Christian Haatvedt , Bjørn-Atle Reme

* Primary investigator

Registry ID
AEARCTR-0013545
DOI
https://doi.org/10.1257/rct.13545-1.0
First registered
Last updated
Status
On going
Trial dates
2024-05-07 – 2024-11-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/13545

Abstract

Each year, Norway disburses reimbursements totaling around 665 millions dollar to its general practitioners (GPs). The system operates on a fee-for-service basis, predominately relying on trust. Under this system, GPs receive compensation determined by their fee utilization which ideally reflects their volume of health services. However, there is considerable variability in fee utilization, leading to diverse amounts of reimbursement being issued to GPs.

This study aims to investigate the extent to which GPs respond to information indicating that they are among the GPs receiving the highest reimbursements from HELFO. More specifically, we randomly assign municipalities into a treatment group and a comparison group, with approximately equal numbers of GPs in each group.

It is expected that the results of this study will serve as a resourceful guidance for policymakers, to how responsive GPs are to information on their reimbursement claims compared to others in a cost-efficient channel. In addition, it will provide a reference regarding which GPs are the most responsive.

Experimental design

We randomize GPs to the treatment and comparison group based on the municipality (district in Oslo, Bergen, Trondheim, and Stavanger) where they work. To enhance statistical power, we implemented a blocked experiment pairing neighboring municipalities. Within each pair one municipality (district) will be randomized to treatment (letter with notification) and comparison (letter without notification). This blocking strategy ensures that the treatment is not randomly concentrated in specific areas of the country, mitigating potential challenges during the analysis phase.

Design details

Randomization method
Randomization is done in office by a computer
Randomization unit
Practice municipality/district
Clusters
391
Observations
1170
Arms
195 municipalities/districts in control, 196 municipalities/districts in treatment
Primary outcomes
Amount of reimbursement paid out to GPs, Number of support calls to HELFO

Cited by

No citation data available for this registration yet.

Keywords