RCT on Information Dosage for Health Insurance Uptake in Guinea-Bissau
How to cite
Costa, Carine Canonici; Catania, Giacomo; Jau, Aliu Bá; Barro, Sié.
2026. "RCT on Information Dosage for Health Insurance Uptake in Guinea-Bissau." AEA RCT Registry. AEARCTR-0018046.
https://doi.org/10.1257/rct.18046-1.0
Download:
BibTeX
·
RIS— for reference managers without a browser
translator. Zotero and most
others pick up this page's citation metadata directly.
Carine Canonici Costa*
, Giacomo Catania
, Aliu Bá Jau
, Sié Barro
This study tests whether short information videos can increase interest in and demand for a new voluntary health insurance scheme in Guinea-Bissau. We will survey about 1,200 adults in urban areas and randomly assign them to watch different versions of a brief insurance information video. Version 1 talks about health insurance in general and the benefits and risks associated with its adoption, version 2 describes a plan with national coverage, and version 3 highlights both national and international coverage, including cross-border care. Immediately after the video, we will measure how interested people would be in enrolling, how much they say they would be willing to pay for different insurance products, and how their beliefs and trust in insurance change. We will also track whether people make follow-up phone calls to a project hotline and re-contact them by phone after around two months to see whether any changes in demand and beliefs persist over time.
Experimental design
This study embeds an individual-level randomized controlled trial in a baseline survey of about 1,200 adults in four urban areas of Guinea-Bissau (Bissau, Canchungo, Bafatá, Gabú). After completing baseline questions, respondents are randomly assigned to one of three short information videos: a control video with general information about health insurance practices, a video that additionally describes a scheme with national coverage, or a video that adds emphasis on international coverage and cross-border care. Immediately after viewing the assigned video, respondents answer a short module on interest in enrolling, willingness to pay for two insurance product modalities (reimbursement vs. direct payment), and perceptions and trust in insurance. Approximately two months later, respondents are re-interviewed by phone to measure persistence of changes in demand and beliefs.
Design details
Randomization method
Randomization is implemented by the survey software at the individual level using programmed randomization logic within the baseline questionnaire (variable rand_treat mapping to treat_group for Control, T1, T2).
Randomization unit
The unit of randomization is the individual respondent.
Clusters
N/A
Observations
The planned sample size is approximately 1,200 individual respondents, stratified across four main urban areas (Bissau; Canchungo; Bafatá; Gabú).
Arms
The study anticipates approximately equal allocation across the three information arms. With a total planned sample of about 1,200 individuals, this implies roughly:
≈ 400 individuals in Control (general information video).
≈ 400 individuals in Treatment 1 (general information + national coverage).
≈ 400 individuals in Treatment 2 (general information + national and international coverage).
Primary outcomes
Midline demand outcomes (immediately post-intervention):
Ordinal interest in enrolling in health insurance (q38_interest, 0–4 scale).
Binary “high interest” indicator (q38_interest ≥ 3).
Willingness to pay (WTP) for Product A and Product B (full coverage), based on the max WTP elicitation for each product.
Midline beliefs and trust:
An insurance perceptions and trust index constructed from items on perceived protection, complexity, waste, and distrust of insurance companies.
Behavioral outcome:
Indicator for at least one inbound call to the project hotline within a two-week window after the interview.
Endline outcomes (≈2 months later), defined as analogues of the midline measures:
Interest in enrolling (ordinal and high-interest indicator).
WTP for Product A and Product B (comparable to midline).
Perceptions and trust index constructed on the same basis as at midline.
Secondary outcomes
Perceived ability to pay (affordability confidence).
Stated main concern about subscribing to health insurance.
Stated product preference between Product A (reimbursement) and Product B (direct payment).
Salience of cross-border care and perceived quality of care, especially for the arm with international coverage.
Interest in being recontacted if the insurance product is implemented.
Selected post-intervention intentions regarding health service use, when explicitly elicited as responses attributable to the information exposure.
Cited by
No citation data available for this registration yet.