Provider identity and patient demand

How to cite

Stacey, Nicholas; Lagarde, Mylene; Papanicolas, Irene. 2026. "Provider identity and patient demand." AEA RCT Registry. AEARCTR-0018944. https://doi.org/10.1257/rct.18944-1.0

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

Nicholas Stacey* , Mylene Lagarde , Irene Papanicolas

* Primary investigator

Registry ID
AEARCTR-0018944
DOI
https://doi.org/10.1257/rct.18944-1.0
First registered
Last updated
Status
Completed
Trial dates
2021-10-25 – 2021-11-05
Countries
South Africa ()
JEL codes
I11, I14, J71
Registry record
http://www.socialscienceregistry.org/trials/18944

Abstract

This study investigates patients' preferences over their medical providers' characteristics, and in particular their preference for providers of the same race. We analyse data from one arm of a previously fielded survey experiment (AEARCTR-0008353) on the sensitivity of demand to choice effort. In that arm, participants rated a series of hypothetical provider profiles with randomly assigned traits; we used these ratings to build each participant a personalised ranking of real providers with whom they could redeem a consultation voucher they stood a chance of winning for taking part. Because the ratings determined which real providers would be recommended, they were consequential and respondents are incentivized to reveal genuine preferences over provider characteristics. For each profile, participants rated not only their interest in consulting the provider but also how good they expected the provider's care to be. This lets us separate two explanations for any same-race preference: a direct taste for a provider of one's own race, and a belief that same-race providers deliver better care.

Intervention

Participants are presented with a series of provider profiles with randomly assigned characteristics including race, where provider race is implicitly signaled by name. We investigate participants’ ratings of own-race provider profiles vs other-race provider profiles.

Experimental design

Each participant evaluated a series of hypothetical general-practitioner profiles (ten per participant). Every profile displayed the same set of attributes, whose values were independently randomised across profiles and participants: the provider's name (which signalled the provider's race and gender implicitly, through distinctive names rather than explicit labels), gender, region of practice, years of experience, and a patient-satisfaction rating; medical qualification was held constant.

Design details

Randomization method
Randomization by computer.
Randomization unit
Individual-profile level.
Clusters
N=302
Observations
N=302
Arms
N=302
Primary outcomes
Primary outcome is interest in seeking a consultation with the provider: "On a scale of 1 to 10 (where '1' is extremely unlikely and '10' is extremely likely), tell us how likely you would be to choose this doctor for a virtual consultation?"
Secondary outcomes
Secondary outcome is perceived quality of provider's care: "On a scale of 1 to 10 (where '1' is extremely unlikely and '10' is extremely likely), tell us how likely you think this doctor would be able to meet the healthcare needs of someone like you?"

Cited by

No citation data available for this registration yet.

Keywords