Coal, children, and avoidance behavior

Last registered on November 15, 2019


Trial Information

General Information

Coal, children, and avoidance behavior
Initial registration date
November 15, 2019

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
November 15, 2019, 10:02 AM EST

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



Primary Investigator

Department of Economics, Faculty of Law, Charles University

Other Primary Investigator(s)

PI Affiliation
Faculty of Medicine, Kyrgyz-Russian Slavic University, Bishkek
PI Affiliation
New School of Economics, Satbayev University
PI Affiliation
International School of Economics, Kazakh British-Technical University

Additional Trial Information

In development
Start date
End date
Secondary IDs
The adverse effects of air pollution on newborns' health are well established, particularly in developed countries with relatively low ambient pollution levels. The purpose of this pilot study is to evaluate the potentially beneficial effects of avoidance behavior during pregnancy in an abnormally polluted urban area in the city of Bishkek, Kyrgyzstan.

Bishkek is a valley city with a population of 1m that suffers from very high levels of air pollution, particularly during winter when the pollution from coal-based centralized heating system enters the city. The public information on air pollution in Bishkek appears to be limited, as evidenced by the fact that daily measurements and air pollution forecasts, which are a norm in developed countries, are not being provided by public authorities but instead supplied by a private initiative. No public warning system responding to frequent extreme pollution levels, nor mitigation measures, are being utilized.

The exposure to pollution and health effects can be reduced by staying indoors during high pollution periods, avoiding exertion, limiting ventilation, or using air filters. However, little is known about the avoidance behavior and health benefits of avoidance strategies in general (Laumbach, Meng, and Kipen 2015). There are only a few studies on this topic (Graff Zivin and Neidell 2009; Moretti and Neidell 2010; Neidell 2004, 2009), all from developed countries. No research exists on avoidance behavior and its effects in developing or middle-income countries.

The efficacy of avoidance behavior in developed countries is a priori ambiguous. Arceo et al. (2016) hypothesize that avoidance behavior may be costlier and less effective in developing countries, due to worse access to health care and lower quality of housing stock, implying that a marginal decrease in air pollution might have a large overall health impact.

We hypothesize that some avoidance behavior (staying indoors, limiting ventilation, etc.) may be more effective in developing countries as the potential for exposure reduction is greater given the higher baseline pollution levels. This may specifically apply to countries with cold winters, as their housing stock is likely to have better insulation than in countries with a warm climate. At the same time, limited awareness about the pollution-related health risks and ways to minimize them, together with the lack of information about the actual pollution levels, may limit individuals’ ability to respond and lead to avoidance opportunities being left unexploited.

Thus apart from contributing to the general knowledge about avoidance behavior and its effects, this study may contribute to a better understanding of opportunities to limit exposure in highly polluted urban areas.
External Link(s)

Registration Citation

Aldashev, Alisher et al. 2019. "Coal, children, and avoidance behavior." AEA RCT Registry. November 15.
Sponsors & Partners

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

Request Information
Experimental Details


Pregnant mothers will be provided with envelopes containing brochures with a calendar tracking pregnant mothers' outdoor activity during pregnancy and a brief survey and asked to distribute these to their patients.
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Birth weight, preterm birth, head circumference, Apgar score.
Primary Outcomes (explanation)
The outcomes are standard measurements taken upon birth that are known to strongly correlate with the infant health.

Secondary Outcomes

Secondary Outcomes (end points)
- Outdoor activity measured by the outdoor activity calendar provided in the envelope.

- Exposure to ambient air pollution measured by outdoor activity calendar matched with area-specific high-frequency air pollution data from sensors.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The study will be implemented in cooperation with Professor Assymbekova Clinic, a private clinic in Bishkek attended by middle-class mothers where about 1000 children per year are born. The packages will be distributed by prenatal care doctors as part of their routine consultations and will carry a unique ID number, which will later be used to match data from anonymized medical records.

The doctors will be fully briefed about this study, but will not know the treatment status of individual mothers. The clinic frequently participates in medical research and doctors are used to that.

Experimental Design Details
Randomization Method
Each package will be identified by an ID number and the treatment status will be randomly assigned will be randomly selected using R's sample {base} command.
Randomization Unit
In order to be able to test for potential information spillover effects, which would pollute the control group, we will vary treatment saturation across weeks (0, 50, and 80 percent). Thus the average treatment saturation will be about 44 percent.
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
We expect to invite approximately 20 mothers per week. If 80 percent agree to participate we need 69 weeks to achieve the needed sample size of at least 1100.
Sample size: planned number of observations
At least 1100.
Sample size (or number of clusters) by treatment arms
1100 mothers in 70 weeks.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The effect of avoidance behavior will be a product of three variables: (i) the behavioral response, (iii) the effect of (i) on exposure, and (iii) the effect of exposure on the outcome. Only the third can be gauged from existing literature: Two recent meta-studies, Stieb et al. (2012) and Sun et al. (2016) report that a 10 microgram/m3 increment in PM2.5 exposure during pregnancy leads to 16g to 23g reduction in birth weight, respectively. We do not have exact pollution data for Bishkek yet, but take the levels from similarly polluted Almaty (in neighboring Kazakhstan on the other side of the mountains). Given the large levels and variation of pollution in (mean 56 and IQR 55 microgram/m3, based on AirKaz data from March 2017 to March 2018), a reduction in exposure by 10 microgram of PM2.5 per m3 due to avoidance appears plausible. According to the 2015 MICS survey, the average birth weight in Kazakstan between 2013 and 2015 was 3.4 kg with a standard deviation of 0.55 kg. An expectation of a meaningful effect in the magnitude of 0.15 of the standard deviation (82.5 grams) may thus be plausible. Using the standard criterion of statistical significance of 0.05 and power 0.9 and the sampling ratio (control/treated) of 1.5, the implied minimal sample size for testing whether the birth weight of the treated is larger than that of the control (one-sided test) is at least 1100 observations. Apart from the standard t-test, we plan to use the non-parametric exact test of stochastic inequality (Schlag 2015), which tend to be more powerful.
Supporting Documents and Materials

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

Request Information

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

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

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials