On the Impact of Social Comparison on Physician Behavior

How to cite

Mueller, Tobias; Gerfin, Michael; Gestel, Raf Van; Doorslaer, Eddy Van. 2019. "On the Impact of Social Comparison on Physician Behavior." AEA RCT Registry. AEARCTR-0004747. https://doi.org/10.1257/rct.4747-1.2

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

Tobias Mueller* , Michael Gerfin , Raf Van Gestel , Eddy Van Doorslaer

* Primary investigator

Registry ID
AEARCTR-0004747
DOI
https://doi.org/10.1257/rct.4747-1.2
First registered
Last updated
Status
Completed
Trial dates
2019-11-01 – 2021-09-01
Countries
Switzerland (Bern)
JEL codes
I11, I13, C21
Registry record
http://www.socialscienceregistry.org/trials/4747

Abstract

Primed by the ongoing political and academic debate about rising health costs and the over-provision of medical care in the developed world, this research project strives at gaining an in-depth understanding about how the provision of social comparison information influences physician behavior, adherence to clinical guidelines and quality of care. To this end, we are planning to run a pragmatic randomized trial in which general practitioners (GPs) from the universe of all GPs active in the outpatient sector in Switzerland are randomly assigned to either an intervention group receiving social comparison letters or the control group (no letter). The information in the social comparison letters is extracted from routinely collected insurance claims data and contains information about the health care costs or practice style behavior (e.g. vaccination rate, generic drug use) of a GP in comparison to peers with a comparable risk pool of patients. Previous studies have shown that peer comparison information has the potential to significantly influence individual decision-making in various contexts thereby motivating the extension to the proposed physician population (e.g. Meeker et al., 2016; Allcott & Rogers, 2014).

Intervention

We randomly assigned general practitioners (GPs) from the universe of all primary care providers active in the outpatient sector in Switzerland to receive peer comparison letters containing information on low-value care (i.e., vitamin D testing and generic substitution) or medical spending in comparison to GPs with a comparable risk pool of patients. Primary care providers in the control group receive no social comparison letter.

The letter will be sent out once at the beginning of the intervention period and the short-term reaction is tracked after half a year. To capture the long-term response, information on the main outcomes will be gathered again after one year.

Experimental design

Nation-wide pragmatic randomized trial using a full factorial design allocating physicians to either the control or one of the intervention arms receiving peer comparison information.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual level: Primary care providers active in the Swiss outpatient sector
Clusters
3064 primary care providers (n = 766 per arm)
Observations
3064 primary care providers (n = 766 per arm)
Arms
766 GPs in the control group (no letter); 766 GPs in the generic substitution arm; 766 GPs in the vitmamin D testing arm and 766 in the health spending intervention group
Primary outcomes
Primary outcomes of the field experiment are the health care costs (plus cost subcategories) and low-value care services including vitamin D tests and the share of generics prescribed to patients of GPs active in the Swiss outpatient sector.

Cited by

No citation data available for this registration yet.

Keywords