Population data.
The population's data were obtained from the Demographic and Health Surveys (DHS), which are nationally representative household surveys covering worldwide LMICs. These surveys were conducted approximately at 5-year intervals. From 1984 to now, DHS has completed seven rounds of the survey and all surveys were standardized across countries.
For the surveys, females of reproductive age (15–49 years) were interviewed by well-trained interviewers and their socioeconomic status, fertility, reproductive history, etc. were collected using standard questionnaires. Although the surveys were cross-sectional and did not follow the same women over time, an option for longitudinal analysis has been embedded: the contraceptive calendar. The calendar was developed by collecting each woman’s contraceptive behavior and pregnancy experience over 5 to 7 years before the interview. Thus, it can be used to longitudinally analyze changes in women’s reproductive lives in recent periods.
Further, in recent surveys, the location (longitude and latitude) of survey clusters (surveyed village or residential cluster) were available, so that environmental variables such as air pollutants and meteorological factors could be linked to each surveyed sample. Based on this, we included eight DHS surveys from South Asian countries, including three in Bangladesh (DHS phases 5–7), three in Nepal (DHS phases 5–7), one in Pakistan (phase 7), and one in India (phase 7).
We have been approved to use the data by adhering to the data usage guidelines. The DHS data is publically available and anonymous thus no further ethical approval is required.
The current duration (CD) calculations.
Fecundity is the biological ability to reproduce, which is generally indicated by TTP and infertility. Infertility is defined as “a disease of the reproductive system defined by the failure to achieve a clinical pregnancy after 12 months or more of regular unprotected sexual intercourse” by the World Health Organization (WHO) 32. However, measuring the TTP can be challenging regardless of epidemiological study design 33. Pregnancy-based retrospective TTP measurement could miss the women who never get pregnant, while prospective cohorts could miss unplanned pregnancies as the women without pregnancy attempts would not join the cohort. The CD approach was recently developed to estimate the TTP and infertility prevalence among women trying to conceive. The CD refers to a self-reported time of trying to conceive at the time of the interview. It can include couples without pregnancy attempts and couples that will never get pregnant. The efficiency of the CD approach has been validated when compared with the retrospective and prospective designs 33. Previous studies 22,23 have applied the CD approach in the DHS data to estimate the TTP and reported that it was a cost-effective method for measuring infertility in LMICs.
In this study, by referring to Polis’s work 23, we included women ‘at risk’ of pregnancy and collected their information based on the contraceptive calendar. The inclusion criteria were: (1) at the age of 18–44 years, (2) married or cohabitating, (3) sexually active within the past 4 weeks, and (4) not using any birth control method (and had not been sterilized). Women who (1) were currently pregnant, (2) had given birth in the past 3 months or were postpartum amenorrheic, (3) were menopausal or had a hysterectomy or had never menstruated, and (4) had no reproductive calendar data were excluded (Specific inclusion and exclusion of study population are available in Table S11).
We calculated the CD for each woman included in the study. First, we identified reproductive events including pregnancy, live birth, termination, and also the history of contraceptive use and the date of these events. Then women were divided into four groups according to their reproductive characteristics and the CD was calculated: (1) for women who have never used any birth control method but have never conceived, a CD was calculated as the date of interview minus the date of the first cohabitation with a current partner; (2) for women with the most recent event was live birth and did not use contraception since then, a CD was calculated as the date of interview minus duration of postpartum abstinence or duration of postpartum menorrhea whichever was the maximum; (3) for women where the most recent event was termination and did not use contraception since then, a CD was calculated as the date of interview minus date of termination; (4) for women the most recent event was any method of conception but not currently using contraception, a CD was calculated as the date of interview minus date of last contraception used.
Exposure assessment.
The daily mean 2-m air temperature and daily/annual mean surface concentrations of PM1, PM2.5, and PM10 from 2000 to 2018 were obtained using the Modern-Era Retrospective Analysis for Research and Application, version 2 (MERRA-2) reanalysis product with a 0.5°×0.625° horizontal resolution 34–36. The PM was estimated as:
![](data:image/png;base64,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)
where SO4, BCPHOBIC, BCPHILIC, OCPHOBIC, OCPHILIC, DU001, DU002, DU003, DU004, SS001, SS002, SS003, and SS004 are the surface mixing ratios (kg kg-1) of sulfate, hydrophobic black carbon, hydrophilic black carbon, hydrophobic organic carbon, hydrophilic organic carbon, dust (bin 001, 0.1-1.0 µm), dust (bin 002, 1.0-1.8 µm), dust (bin 003, 1.8-3.0 µm), dust (bin 004, 3.0–6.0 µm), sea salt (bin 001, 0.03–0.1 µm), sea salt (bin 002, 0.1–0.5 µm), sea salt (bin 003, 0.5–1.5 µm), and sea salt (bin 004, 1.5-5.0 µm), respectively. AIRDENS represents surface air density (kg m-3). SO4SMASS, BCSMASS, OCSMASS, DUSMASS2.5, and SSSMASS2.5 are surface mass concentrations (kg m-3) of sulfate, black carbon, organic carbon, dust (PM2.5), and sea salt (PM2.5), respectively. MERRA-2 is the latest and first long-term global atmospheric reanalysis to assimilate aerosol observations and represent aerosol-climate interactions from the National Aeronautics and Space Administration (NASA), which was well evaluated and adopted in studies of premature mortality 37–39. The 2-m air temperature and surface concentrations of PM1, PM2.5, and PM10 were bilinearly interpolated into the DHS geocoded residential address.
We assigned daily PM1, PM2.5, and PM10 exposure for each woman from the date they were trying to become pregnant to the date of the interview, by spatially matching the DHS geocoded locations (longitude and latitude) with the gridded PM concentration. In this study, the date trying to become pregnant was calculated as the date of the interview minus the CD period. Given that the TTP is the number of menstrual cycles, we further averaged the daily PM concentration as monthly exposure for statistical analysis. To adjust for the potential confounding effect of ambient temperature, we also collected daily mean temperatures and assigned monthly exposure for each woman as we did for PM exposure.
Estimation of the TTP and infertility prevalence.
Commonly, the CD is regarded as backward recurrence times 40,41. Based on the assumptions that the pregnancy attempts happens at a constant rate (stationarity), and the distribution of TTP is independent of calendar time, the CD can be used to infer an underlying distribution of TTP by applying survival methods that based on the theory of backward recurrence time 42.
We estimated the survival function of TTP from the CD using parametric survival methods relying on the Yamaguchi distribution 42. The corresponding 95%CIs were calculated using bootstrap methods with 500 samples 23,42. The CD was censored after 36 months as in previous studies 23,41. From the estimated survival function, we extracted the estimations and 95% CIs for the median TTP and prevalence of infertility (i.e., the proportion of women not yet pregnant by 12 months of trying according to the WHO definition 32). We further estimated the prevalence of primary (among nulliparous women) and secondary infertility (among parous women), as well as the infertility prevalence in women with and without ever terminated pregnancy separately, by demographic characteristics, age (18–34 vs. 35–44 years), education (no education, primary, secondary or higher education), residence (urban vs. rural), wealth index (poorest, poorer, middle, richer, richest), fertility preference (“does not want another birth soon” vs. “wants birth soon, now or within next nine months”), and knowledge of any modern conceptive method (yes vs. no).
Examination of the effects of air pollution on fecundity.
The estimation of underlying TTP distribution from CD favors the accelerated-failure-time (AFT) model, therefore we used AFT to estimate the effects of air pollution exposure on fecundity. Since the observed CD follow the same structure as TTP 42, the effects of exposure on CD is also an effective estimation of the exposure on TTP 43. The AFT model has been previously applied to CD data 43,44.
As the PM exposure was time-varying, we treated the data as a counting process format and applied the advanced AFT model for effect estimation by risk sets, which enabled us to estimate the effect of PM exposure on the time of achieving pregnancy for each cycle conditioned on pregnancy failure in the previous cycle. Parametric AFT models were used for effect estimation 45, with the CD censored after 36 months 23,41. The AFT model is constructed as follows:
$$log\left({CD}_{i}\right)=\alpha +{\beta }_{PM}{x}_{PM}+{\beta }_{1}{x}_{1}+\dots +{\beta }_{p}{x}_{p}+{\epsilon }_{i}$$
in which i represents the individuals and 𝑙𝑜𝑔(CD𝑖) represents their log-transformed survival time. 𝑥1 to 𝑥𝑝 are the covariables with the coefficients 𝛽1 to 𝛽𝑝; 𝜀𝑖 is the residual. 𝑥PM is the PM exposure (including PM1, PM2.5, and PM10, respectively) and 𝛽PM is the coefficient of each PM exposure per 10µg/m3 increments in the PM concentration. To interpret the effect estimations easier, we further transformed 𝛽PM to fertility time ratio (FTR, the exponential of 𝛽PM). In the model, we set the coefficients as the expected time for achieving pregnancy, thus an FTR > 1 (i.e. β > 0) indicates that females tend to have longer times to conceive, i.e., reduced fecundity.
We ran both unadjusted and adjusted models. In the unadjusted models, we only included the PM exposure and residential cluster (residential cluster defined by the DHS as a sampling unit), which was included as the random effect. In the adjusted models, we additionally included age, education, residence, body mass index, wealth index, parity, and mean temperature. The mean temperature was also treated as a time-varying covariate and included as a monthly temperature quintile in the adjusted models. We applied the above AFT models for each country and also the whole study sample. When the study population was from multiple survey waves (for Bangladesh, Nepal, and the whole studied sample), we additionally included survey waves (as a category variable) in the models.
We also conducted stratified analyses by age (18–34 vs. 35–44 years), parity (nulliparous vs. parous), education (no education, primary, secondary, or higher education), residence (urban vs. rural), and wealth index (poorest, poorer, middle, richer, richest). Cochran Q tests were performed to examine the heterogeneity among different subgroups.
Examination of the TTP extension burden attributable to particulate matters.
We further estimated the extended TTP attributable to particulate matters for the included countries in the survey years. We used a standard method of attributable risk assessment:
$${AF}_{i,t}=1-\frac{1}{{{FTR}_{i}}^{\left({C}_{i}-{C}_{0}\right)/10}}$$
in which i is the country, FTRi is the estimate of the association between PM exposure and TTP, Ci is the annual averaged PM level of each country in the survey years, and C0 is the recommended average annual level of PM based on the global air quality guideline released by WHO in 2021, which is 15 µg/m3 for PM10, and 5 µg/m3 for PM2.5, respectively. As there is no WHO guideline for PM1, we still used the 5 µg/m3 recommended average annual level. Therefore, AFi,t can be explained as the attributable fraction of TTP extension since the PM exposure is beyond the WHO recommended exposure level.
Then the attributable number of TTP extensions was estimated as:
$${AN}_{i,t}={N}_{i,t}\times {Median TTP}_{i,t}\times {AF}_{i,t}$$
here, Ni,t is the average number of females of reproductive age (15–49 years) of each country in the survey years, which was obtained from World Population Prospects 2022, United Nations. Median TTPi,t is our estimated TTP for each country. Therefore, ANi,t can be explained as the extended months of TTP attributable to PM exposure beyond the WHO recommended exposure level.
Sensitivity analyses.
Multiple sensitivity analyses were performed to examine the robustness of TTP estimation and the association between air pollution and fecundity. First, to account for potential delayed pregnancy recognition, we performed a simulation study to account for possible pregnancy recognition bias. We simulated a random month of delay across 0 to 3 months before the inclusion interview with a normal distribution (mean = 1.5, sd = 0.5). The simulated delay months were subtracted from the originally observed CD. Then, we used newly calculated CD to estimate the TTP and infertility, as well as the association of air pollution with fecundity. Second, couples’ attempts to pregnant in a population may vary with time, e.g., a seasonal pattern of pregnancy planning was observed in previous studies 46,47. In this case, the stationarity assumption under which CD can be used to estimate the TTP would be violated, leading to bias. To test and address this issue, we repeated the main analyses examining the effects of air pollution on fecundity by adding the season of initiation in the models. Third, although the questionnaire of the DHS survey has been evaluated and validated, potential recall bias may also exist and bias our estimation. In addition, we estimated TTP relying on the assumptions of stationarity, which tends to hold for shorter intervals. Thus, we restricted our samples to a CD of less than 24 and 12 months to re-estimate the associations. Fourth, we used multiple imputations to deal with missing values of covariates. We used fully conditional specification methods with 400 iterations to generate 5 imputed datasets. Then we repeated the main analysis using each complete dataset and summarized the results based on Rubin's rules 48.
We cleaned the DHS data and calculated CD using Stata 15.0 (StataCorp LP, College Station, TX, USA). All statistical analyses were performed with R (version 4.1.1; R Development Core Team), and packages “survival”, and “eha” were used. All tests were two-sided, and a P-value of < 0.05 was considered to be statistically significant.