The Adoption and Impact of Air Purifiers: Evidence from a Field Experiment in Mexico City

Last registered on February 03, 2026

Pre-Trial

Trial Information

General Information

Title
The Adoption and Impact of Air Purifiers: Evidence from a Field Experiment in Mexico City
RCT ID
AEARCTR-0012227
Initial registration date
October 03, 2023

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
October 04, 2023, 5:04 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
February 03, 2026, 1:42 PM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Affiliation
University of Warwick

Other Primary Investigator(s)

PI Affiliation
IDB
PI Affiliation
World Bank

Additional Trial Information

Status
In development
Start date
2026-02-16
End date
2027-08-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
According to the World Health Organization, air pollution—both ambient and residential-- is the top environmental risk to health. Accordingly, air pollution is a growing topic of study. However, the bulk of the literature is focused on two disparate issues, outdoor air quality in cities and residential air quality related to cookstoves in rural areas. In contrast, very few studies focus on residential air quality in cities of the developing world. Yet, indoor air quality, and especially residential air quality, is likely to be critical for individuals’ outcomes, as people spend 60% of their time indoors. For example, residential air quality could affect sleep duration and quality, which in turn could affect health, wellbeing, and performance. Because outdoor air pollution levels are often higher in poor neighborhoods than in rich neighborhoods and pollutants, such as particulate matter, better permeate older and poorly constructed dwellings, residential air quality may exacerbate health and economic inequalities. Documenting the extent, determinants, and effects of air quality in urban homes in Mexico City will provide an understanding of the scale of the problem. Evidence on the barriers to improving residential air quality and the policy interventions that can overcome them is key to mitigating the impacts of pollution and addressing poverty for millions of households in fast-developing and highly polluted cities of countries like Mexico, India, and China.
This project addresses the following research questions:
1. What is the level of air pollution in a home in a below median income neighborhood in highly polluted city of a middle-income country?
2. What barriers prevent adoption of mitigating technology (air purifiers) among households in a below median income neighborhood in this context and what policies can overcome them?
3. What are the health and wellbeing effects of improved residential air quality for households in a below median income neighborhood in this context?
External Link(s)

Registration Citation

Citation
Behrer, Patrick, Ludovica Gazze and Bridget Hoffmann. 2026. "The Adoption and Impact of Air Purifiers: Evidence from a Field Experiment in Mexico City ." AEA RCT Registry. February 03. https://doi.org/10.1257/rct.12227-2.0
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2026-02-16
Intervention End Date
2026-11-30

Primary Outcomes

Primary Outcomes (end points)
- Air quality: Hourly PM 2.5 at the household level.
- Sleep quality: Standardized index at the participant-day level.
- Wellbeing: Standardized index at the participant level as measured at endline.
- Willingness-to-pay: Incentivized stated price at the participant level as measured at endline.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
- Attrition: share of days with usable wearable data and endline responses for primary outcomes
- Physical health, as measured by index pf self-reports and metrics provided by the wearables (resting heart rate, daily heart rate variability in healthy range)
- Health behaviors: number of steps and time spent in sedentary activities (wearables)
- Self-reported sleep quality index
- Self-reported index of awareness of air quality issues and measures to reduce exposure
- Behavior: ventilation, smoking, and cooking habits as measured by self-reports as well as CO2
- Purifier usage
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Households will be randomized into one of the following three mutually exclusive intervention groups prior to baseline:
• 10-day Trial (200 households): Households will receive an offer for a free 10-day purifier trial (T-10).
• 90-day Trial (200 households): Households will receive an offer for a 90-day purifier trial (T-90).
• Control (C) (200 households): Households will not receive the purifier.

In addition, at endline, we will cross-randomize households into one of three information treatments:
• General Info (200 households): Households will receive information on how purifiers improved air quality and sleep during our pilot in Mexico City.
• Personalized Info (200 households): Households will receive information on how purifiers improved air quality and sleep among households with similar observable characteristics to theirs during our pilot.
• Control (C) (200 households): Households will not receive additional information.
Experimental Design Details
Not available
Randomization Method
Prior to the baseline survey, we will randomize the order of enrolled households into the three intervention treatment groups, generating a list for each enumerator (e.g., a sequence of T10, C, C, T90, T90, T10, ....) that enumerators will follow as they recruit households. In addition, prior to endline, we will cross-randomize households into one of three information treatments.
Randomization Unit
Household.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
600 households for full study.
Sample size: planned number of observations
600 households for full study.
Sample size (or number of clusters) by treatment arms
Control: 200 HHs (roughly 1/3 no info, 1/3 general info, 1/3 personalized)
T10: 200 HHs (roughly 1/3 no info, 1/3 general info, 1/3 personalized)
T90: 200 HHS (roughly 1/3 no info, 1/3 general info, 1/3 personalized)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Human Subjects Committee for Innovations for Poverty Action IRB-USA
IRB Approval Date
2023-07-20
IRB Approval Number
16708