Discrete Choice Experiment on Access, Use, and Preferences for Contraceptive Methods in Colombia

How to cite

Lesmes, Paul Andres Rodriguez; Serna, Natalia; Florez, Jorge; Ortiz, Santiago. 2026. "Discrete Choice Experiment on Access, Use, and Preferences for Contraceptive Methods in Colombia." AEA RCT Registry. AEARCTR-0018692. https://doi.org/10.1257/rct.18692-1.0

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

Paul Andres Rodriguez Lesmes* , Natalia Serna , Jorge Florez , Santiago Ortiz

* Primary investigator

Registry ID
AEARCTR-0018692
DOI
https://doi.org/10.1257/rct.18692-1.0
First registered
Last updated
Status
In development
Trial dates
2026-05-25 – 2026-06-30
Countries
Colombia ()
JEL codes
I11, I12, I18, D12, D83, C91, C93
Registry record
http://www.socialscienceregistry.org/trials/18692

Abstract

This study implements a laboratory-based discrete choice experiment to measure women’s preferences and access barriers related to contraceptive methods in Colombia. The study focuses on a dual market in which contraceptive methods can be obtained through the institutional health insurance channel, often with no direct monetary cost but with administrative and waiting-time frictions, or through commercial retail channels, usually with faster access but higher out-of-pocket prices.

Eligible participants are women aged 18 to 49, residing in Colombia, who are not pregnant and have not undergone permanent sterilization. The main experimental sample will include approximately 300 women recruited through the participant pool of the Rosario Experimental and Behavioral Economics Lab (REBEL), with a preliminary focus group or pilot involving 10 to 15 women to validate the instrument. Each participant will complete one session lasting approximately one hour.

The intervention consists of a discrete choice experiment in which participants choose among hypothetical contraceptive alternatives that vary randomly across five attributes: monthly price, waiting time, frequency of administration, place of obtainment, and availability through the health insurance provider. Each choice task presents three hypothetical contraceptive alternatives and an opt-out option. The choice experiment is complemented with a structured survey on knowledge of contraceptive methods and brands, prior use, perceived risks and side effects, access barriers, and health care experiences.

The primary outcomes are stated choices of contraceptive alternatives and estimated preference weights for each experimental attribute. Secondary outcomes include willingness-to-pay or willingness-to-wait measures, perceived ease of access, perceived likelihood of adverse effects, knowledge of contraceptive markets, and reported barriers to access. The study will estimate how economic, administrative, informational, and service-delivery attributes shape contraceptive preferences and how these preferences vary by sociodemographic characteristics, prior use, and health care experiences.

The study involves minimal risk. The survey does not collect personally identifiable information in the analytical dataset, and all research data will be anonymized and stored securely. Participants in the laboratory sessions will receive a fixed compensation for their time.

Intervention

The intervention is a discrete choice experiment on contraceptive access and preferences. Participants will be presented with repeated hypothetical choice tasks in which they choose among three contraceptive alternatives and an opt-out option. The alternatives vary experimentally across five attributes: monthly out-of-pocket price, waiting time, frequency of administration, place of obtainment, and availability through the participant’s health insurance provider.

The randomized variation in these attributes is designed to estimate the relative importance of monetary costs, time costs, administrative frictions, and institutional availability in contraceptive choices. The experiment does not provide contraceptives, medical advice, prescriptions, or health services. It only elicits stated preferences under hypothetical scenarios.

The discrete choice experiment is complemented by a survey measuring participants’ knowledge of contraceptive products and brands, past and current use, perceived risks and side effects, barriers to access, experiences with health providers and pharmacies, and sociodemographic characteristics.

Experimental design

This is a cross-sectional laboratory-based discrete choice experiment accompanied by a structured survey. The study will be conducted primarily at the Rosario Experimental and Behavioral Economics Lab (REBEL) at Universidad del Rosario. Participants will complete the study once, in individual computer stations, under controlled conditions of privacy and silence.

The target population is women aged 18 to 49 residing in Colombia who are not pregnant, have not undergone permanent sterilization, and are able to provide informed consent. Participants will be recruited through the REBEL participant pool. If needed, additional data collection may be implemented online through REBEL or through a specialized panel provider.

Before the main experiment, the instrument will be validated through at least one pilot focus group with 10 to 15 women. This phase will be used to assess comprehension of the attributes, wording, instructions, and sensitivity of the questions.

In the main experiment, approximately 300 women will complete a discrete choice experiment. Each choice task will present three hypothetical contraceptive alternatives and an opt-out option. Alternatives will vary across five attributes: monthly price, waiting time, frequency of administration, place of obtainment, and EPS availability. Attribute levels will be randomized across choice tasks according to a prespecified experimental design. After each choice, participants will answer follow-up questions on perceived adverse effects and ease of access.

The experiment does not assign participants to real contraceptive products or health services. The randomized component is the presentation of hypothetical contraceptive profiles and their attributes within the discrete choice experiment.

Design details

Randomization method
Randomization will be implemented by computer through the survey platform. Attribute levels and hypothetical profiles will be assigned to choice tasks according to a prespecified discrete choice experimental design. Participants will be assigned to blocks or versions of the choice experiment by the survey software.
Randomization unit
The primary randomization unit is the choice profile within participant-level choice tasks. If the final design uses blocks, participants will be randomly assigned to a block of choice tasks. The study does not randomize participants into treatment and control arms that receive real-world interventions.
Clusters
Not applicable. The study is not clustered. Randomization occurs at the level of hypothetical choice profiles and, if applicable, participant-level blocks of the discrete choice experiment.
Observations
Main experimental sample: 300 women. Pilot/focus group validation: 10 to 15 women. Each participant in the main sample will complete multiple choice tasks, generating repeated choice observations at the participant-task-alternative level. The final number of analytical observations will depend on the final number of choice tasks and alternatives per task.
Arms
This study does not have conventional treatment and control arms. All main-study participants complete the discrete choice experiment and structured survey. Variation comes from randomized hypothetical contraceptive profiles within the choice tasks. Planned sample: approximately 300 women in the main discrete choice experiment. If the survey is blocked into versions, participants will be allocated across blocks as evenly as possible.
Primary outcomes
Choice of contraceptive alternative in each discrete choice task. Selection of the opt-out option, defined as choosing not to use any of the hypothetical methods presented. Estimated preference weights for the attributes included in the discrete choice experiment: monthly price, waiting time, frequency of administration, place of obtainment, and availability through the health insurance provider. Marginal rates of substitution across attributes, including willingness to pay for shorter waiting times, preferred place of obtainment, lower administration frequency, and EPS availability.
Secondary outcomes
Perceived likelihood of adverse effects for the alternatives shown in each choice task. Perceived ease or speed of obtaining the alternatives shown in each choice task. Knowledge of contraceptive brands, modes of administration, prescription requirements, EPS coverage, and access channels. Current and past use of contraceptive methods. Reported barriers to obtaining or using the preferred contraceptive method. Reported time required to obtain contraceptives through pharmacies, EPS dispensing points, or medical consultations. Heterogeneity in preferences by age, education, socioeconomic status, relationship status, prior contraceptive use, EPS access, and health-related characteristics.
Sponsors
Name: Stanford Health Policy Url: https://healthpolicy.fsi.stanford.edu/

Cited by

No citation data available for this registration yet.

Keywords