Abstract
Background
Rising temperatures have become one of the most critical public health challenges in the era of global climate change. Frequent exposure to temperature extremes not only threatens the health of pregnant women but also poses significant risks to fetal development. Increasing evidence from global studies indicates that both high and low temperature exposures are associated with adverse birth outcomes such as preterm birth, low birth weight, and stillbirth. However, this issue remains inadequately explored in the Indian context, particularly through longitudinal district-level analyses.
Methods
The present study examines the relationship between ambient temperature and stillbirth rate across districts of Bihar from 2009 to 2019, employing fixed-effects panel regression, distributed lag, and distributed lag non-linear models.
Result
The findings reveal a significant non-linear U-shaped relationship between temperature and stillbirth rate, indicating that exposure to either extreme high or low temperature substantially elevates the risk of stillbirth. Specifically, exposure to high temperatures around 35 °C or low temperatures near 18 °C increases stillbirth risk by approximately 7–14%.
Conclusions
These results highlight the thermal sensitivity of pregnancy outcomes and the dual burden posed by temperature extremes. Despite certain limitations, the study underscores the need to integrate climatic factors into maternal health policy to enhance resilience and safeguard the health of mothers and their developing foetuses.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12884-025-08620-x.
Keywords: Ambient temperature, Climate change, Environmental health risk, Panel data analysis
Introduction
Climate change has emerged as one of the dominant global issues in the 21st century for the globe. Temperatures is rising, resulting in an increased frequency and severity of extreme weather events posing severe risks to both ecosystems and living organisms [1]. The steady rise in global mean temperature, coupled with more frequent heatwaves, drought and erratic precipitation pattern, has disrupted agricultural systems, water availability, and human livelihoods through several pathways. These climatic changes exert multifaceted impacts that extend beyond environmental degradation, influencing food security, economic productivity, and public health outcomes [2, 3]. Even these impacts are distributed disproportionately across different socio-economic class of the society. Vulnerable populations like child, pregnant women, and elderly are at facing risk due to these changes [4, 5].
Among the multifaceted impact of the climate change, adverse health outcomes are one of them, affected directly or indirectly by either by temperature or by precipitation. Rising temperature possess risk to both children and pregnant women by increasing the risk of heat related morbidities like heat stress, stroke or cramp etc [6]. Pregnant women are thermally more sensitive to high temperature and even a moderate temperature fluctuation can influence their physiological condition and fetal development [7]. The evidence from the previous studies suggests that exposure to extreme temperatures may increase the likelihood of adverse pregnancy outcomes, including preterm birth, low birth weight, and stillbirth, through mechanisms such as placental dysfunction, dehydration, or maternal cardiovascular strain [8–10]. These outcomes highlight temperature as a critical yet often overlooked environmental determinant of maternal health and fetal well-being [11].
During last decades, the global scientific community has made considerable progress in quantifying the effects of temperature extremes on pregnancy outcomes across the globe. Likewise, a study from China has found that exposure to extreme temperature events during pregnancy increases the risk of still birth by 36–83% [12]. Similar evidence was observed in a population-based study from Taiwan, which identified a distinct U-shaped association between ambient temperature and stillbirth, with elevated risks at both cold and hot extremes [13]. A multi-country study across 14 nations using Demographic and Health Survey (DHS) data also found that exposure to high temperatures increased the risk of both stillbirth and preterm birth [14]. A study by Auger et al. in Canada using the longitudinal data founds that odd of still birth was 1.16 times at 26 °C compared to 20 °C [15]. Basu et al. carried similar type of study on California state of united states of America and found that 10 0F increase in temperature is positively associated with 10.4% change in still birth incidence [16]. Even studies bases on meta-analyses have further confirmed that heat exposure in the final weeks of gestation increases the odds of stillbirth by 5–10% [17].
In recent years, some studies have increasingly highlighted the role of socio-environmental determinants that exacerbate temperature-related pregnancy risks. For instance, a 2025 cross-country analysis demonstrated that energy poverty significantly elevates stillbirth, neonatal, and infant mortality, with stillbirths increasing by 13.16% in low- and lower-middle-income countries lacking adequate energy access [18]. Similarly, a parallel work from China has concluded that exposure to extreme temperature events raises the risk of stillbirth by 36–83%, reaffirming the sensitivity of fetal development to both heat and cold [19]. Likewise, DLNM-based studies, such as those conducted in Taiwan, report a distinct U-shaped association between ambient temperature and stillbirth risk, with elevated risks at both hot and cold extremes [13]. Meta-analytic evidence further supports this non-linear association across diverse populations globally [17]. Collectively, these study findings emphasize that thermal stress interacts with structural vulnerabilities, such as energy deprivation, poor housing, and socio-economic marginalization, to heighten risks during pregnancy [20], particularly in LMIC settings like India.
In India, a few studies have examined the relationship between temperature exposure and adverse pregnancy outcomes such as preterm birth and low birth weight. For instance, McElroy et al. reported that approximately 3.01% of low-birth-weight cases in India were attributable to exposure to high mean temperatures [14]. Similarly, a cohort study involving 800 pregnant women in Chennai found that exposure to high occupational heat was positively associated with the risk of stillbirth and miscarriage [10]. While study from Taiwan demonstrate a strong temperature–stillbirth relationship using advanced modelling approaches, similar longitudinal evidence from low- and middle-income regions, particularly India, remains limited [8, 13]. This gap highlights the need for district-level analyses that can account for spatial heterogeneity and climatic variability. The present study seeks to address this gap by employing longitudinal district-level data to investigate the temperature–stillbirth relationship and to assess whether a non-linear (U-shaped) association exist between them, as previously reported in global studies. This analysis aims to provide empirical insights that can inform climate-resilient maternal health policies in low- and middle-income settings.
Data and method
Data
The present study uses monthly district-level secondary data for stillbirth counts and total deliveries from two publicly accessible data portal. Stillbirth counts and total deliveries data were acquired from the Health Management Information System (HMIS), Ministry of Health and Family Welfare, Government of India, which compiles routine facility-based maternal and child health indicators from public and accredited private health facilities nationwide. Monthly maximum temperature data for each district were obtained from the India Meteorological Department (IMD), which provides gridded temperature estimates at 1° × 1° spatial resolution for India.
The study sample includes all 38 districts of Bihar from January 2009 to December 2019, yielding 4,900 district-month observations. Missing values in HMIS reporting, primarily due to under-reporting or delayed data entry, were addressed using the seasonal decomposition-based interpolation (SEATS) method, following the exclusion of districts with missing observations at the beginning or end of their time series [21]. This approach allowed the seasonal component of the data to be preserved while minimizing distortion of underlying temporal patterns. Temperature exposure for each district-month was obtained by overlaying district boundaries with gridded maximum temperature data from the India Meteorological Department (IMD). The study adopts a district-level ecological panel design, combining monthly stillbirth data from HMIS with gridded maximum temperature records from IMD for the period January 2009–December 2019 (Figure S1).
The study period was restricted to January 2009–December 2019 to ensure complete and consistent annual coverage for all districts. Although HMIS provides data from April 2008 to March 2020, the first and last year’s contain incomplete monthly records, making them unsuitable for full-year comparisons. Reporting quality also stabilized following national digitization reforms implemented after 2008, whereas data from 2020 onward were affected by COVID-19–related service disruptions that introduced irregularities unrelated to temperature exposure. IMD temperature data were extracted for the same period to align with the availability and reliability of HMIS stillbirth records, thereby ensuring a coherent and comparable time series for longitudinal analysis.
Variable description
Stillbirths is defined as fetal deaths occurring after 28 weeks of gestation, consistent with national reporting guidelines [22]. To assess the relationship between temperature exposure and still birth rate, still birth rate was calculated using the total monthly still birth cases and total delivery at district level in similar month. The formula for the calculation of stillbirth rate is-
![]() |
Monthly maximum temperature for each district was used as the explanatory exposure variable. Extreme cold and heat thresholds were identified empirically by modelling risk across the full temperature range using the DLNM, consistent with prior studies that define thermal extremes based on modelled exposure–response functions rather than fixed percentiles [13, 23].
Statistical analysis
In this study, both descriptive and inferential statistics has been utilized to evaluate the relationship between temperature exposure and still birth rates. Mean still birth rate were calculated for monthly to identify the months of high still birth rate.
A series of panel data regression models were also applied in this study, which includes the fixed-effect panel regression model, distributed lag model, and distributed lag non-linear model.
Fixed-effect (FE) panel regression model
The present study uses a fixed-effects panel regression model, consistent with previous climate–health research. A Hausman test confirmed that this specification was appropriate (χ² = 1129, df = 2, p < 0.001), indicating that district-level differences were correlated with temperature exposure. Controlling for these stable district characteristics allows a more accurate assessment of how temperature variations influence stillbirth rates over time [24, 25].
![]() |
Where:
➢ Yit = stillbirth rate in district i at time t
➢ Tit= monthly maximum temperature
➢ Tit2= quadratic term capturing non-linear effects
➢ γi = district fixed effects
➢ δt = month & year fixed effects
➢ ϵit = error term
A quadratic term for maximum temperature was included in the fixed-effects model to account for the established nonlinear relationship between temperature and adverse pregnancy outcomes, particularly the U-shaped association documented in previous stillbirth studies [13]. Evidence shows that both low and high temperature exposures elevate stillbirth risk, making a linear specification insufficient. The squared term enables the model to capture increasing risk at thermal extremes while maintaining an interpretable structure. To further assess nonlinear effects, we also estimated a distributed lag model (DLM) and distributed lag non-linear model (DLNM) [16, 17].
Distributed Lag Model (DLM)
Distributed lag model was employed to investigate delayed physiological effects of heat exposure on pregnancy [13]. Temperature lag of 0 to 3 month has been used to observed how it affects the still birth.
![]() |
Where lags l = 0,1,2,3 represent exposure up to three months prior. This approach helps to captures the delayed potentially risks of high temperature exposure on fetal development and placental dysfunction.
Distributed Lag Non-Linear Model (DLNM)
Although the DLNM framework was originally developed for single-location time-series analyses, it has since been widely extended to multi-location and panel data settings in environmental epidemiology. This panel-DLNM approach enables the estimation of non-linear and delayed temperature effects while simultaneously accounting for spatial heterogeneity and unobserved district-level characteristics through fixed effects. Multi-city studies from China, Australia, Europe, and North America have successfully adopted this framework to evaluate temperature–health associations in settings with multiple geographic units [26–28]. Following this established methodological practice, the present study applies a DLNM within a district-level panel design to capture both the non-linear exposure–response relationship and the distributed lag effects of temperature on stillbirth risk, consistent with prior epidemiological applications [13].
![]() |
Where:
cb(⋅) = cross-basis matrix using spline functions for exposure and lag
Maximum lag = 3 months
Degrees of freedom (df = 3 for exposure, df = 3 for lag)
Although the outcome variable (stillbirth rate) in the study is continuous in nature, the DLNM framework estimates changes in the log rate, and relative risks (RR) are derived by exponentiating the predicted change in log rate relative to a reference temperature [26, 29].
Result
Descriptive statistics
The study utilizes final dataset of 4,900 district-month observations from all 38 districts of Bihar between January 2009 and December 2019. After imputation and data cleaning, the dataset included a total of 1,98,38,540 live births and 2,94,728 stillbirths. The mean still birth rate was 15.97, with maximum 64.21, minimum 0.15 and standard deviation of ± 7.65 per 1000 birth for the state of Bihar during 2009-19. Mean maximum temperature in that similar duration was 31.19 °C, ranging from minimum 17.42 °C to 41.51 °C with a standard deviation of ± 4.59 °C (Table 1).
Table 1.
Descriptive statistics of still birth and maximum temperature during 2009-19
| Measure | Mean | SD | Min | Max |
|---|---|---|---|---|
| Stillbirth rate (per 1,000 births) | 15.97 | 7.65 | 0.15 | 64.21 |
| Maximum temperature (°C) | 31.19 | 4.89 | 17.42 | 41.59 |
Source: HMIS, MoHFW; IMD
The seasonal trend of the still birth rate was found to be highest during the month of July, which tends to decline in later months, However, a sudden spike is observed in the still birth rate during the month of January which later decrease again in February month (Fig. 1). The seasonal trend of the maximum temperature reveals that minimum temperature was lower during January and exhibits an increasing trend up to the month of May and June. Later that, the maximum temperature tends to decrease and reaching lowest in the month of December (Fig. 2).
Fig. 1.

Seasonal trend in mean stillbirth rate across Bihar (2009–2019)
Fig. 2.

Seasonal trend in maximum temperature rate across Bihar (2009–2019)
Fixed effect panel regression model
The results of the fixed-effects panel regression model reveal that exposure to higher temperatures has a statistically significant positive association with stillbirth rates across the state. Specifically, a 1 °C increase in maximum temperature is associated with a 0.125 per 1,000 live births rise in the stillbirth rate (p < 0.001; Table S2). Later, we incorporate both linear and quadratic terms of maximum temperature in the fixed effect panel regression model to capture potential non-linear effects by controlling for unobserved, time-invariant district-level heterogeneity and temporal seasonality through district, month, and year fixed effects. The estimated coefficients show that the linear term of temperature (β = −1.39, p < 0.001) is negatively associated with stillbirth rate, while the quadratic term (β = 0.020, p < 0.01) is positive. This result indicates a U-shaped relationship between temperature and stillbirth rate across the study region. Monthly coefficients reveal significantly higher stillbirth rates from March to November, with peak values observed during July. The year-specific coefficients exhibit a consistent declining trend from 2010 to 2019, suggesting a gradual reduction in stillbirth rates over time (Table 2). Moreover, the robustness check using the clustered standard error generated similar outcomes, which confirm the consistency and reliability of obtained model estimates. Overall, the result of the model indicates a significant variation in stillbirth rate for both temperature and time by taking account of district-specific characteristics and seasonality.
Table 2.
Effect of monthly maximum temperature on stillbirth rate
| Variable | Estimate | Std. Error | t-value | p-value |
|---|---|---|---|---|
| Max Temperature (°C) | −1.39 | 0.41 | −3.39 | < 0.001 |
| Max Temperature² | 0.020 | 0.01 | 3.20 | < 0.01 |
| February | 1.076 | 0.74 | 1.45 | 0.15 |
| March | 3.119 | 1.104 | 2.82 | < 0.01 |
| April | 3.9 | 1.296 | 3.01 | < 0.01 |
| May | 4.242 | 1.318 | 3.22 | < 0.01 |
| June | 4.558 | 1.3 | 3.51 | < 0.001 |
| July | 5.422 | 1.172 | 4.63 | < 0.001 |
| August | 4.151 | 1.166 | 3.56 | < 0.001 |
| September | 3.349 | 1.153 | 2.9 | < 0.01 |
| October | 3.157 | 1.087 | 2.91 | < 0.01 |
| November | 2.551 | 0.908 | 2.81 | < 0.01 |
| December | −0.022 | 0.5 | −0.04 | 0.96 |
| Year 2010 | 3.62 | 0.439 | 8.24 | < 0.001 |
| Year 2011 | 1.696 | 0.439 | 3.86 | < 0.001 |
| Year 2012 | 0.86 | 0.435 | 1.98 | < 0.05 |
| Year 2013 | −0.520 | 0.434 | −1.20 | 0.23 |
| Year 2014 | −1.384 | 0.441 | −3.14 | < 0.01 |
| Year 2015 | −2.895 | 0.431 | −6.72 | < 0.001 |
| Year 2016 | −4.331 | 0.429 | −10.09 | < 0.001 |
| Year 2017 | −4.748 | 0.43 | −11.04 | < 0.001 |
| Year 2018 | −5.516 | 0.435 | −12.69 | < 0.001 |
| Year 2019 | −6.282 | 0.437 | −14.39 | < 0.001 |
Source: Author
Association between temperature and stillbirth
The results of the Distributed Lag Model (DLM) were consistent with those of the fixed-effects panel regression, even after accounting for temperature exposure up to a three-month lag (Table 3). The linear term of maximum temperature remained negative (β = −1.19, p < 0.01), while the quadratic term was positive (β = 0.0175, p < 0.01), reaffirming the presence of a U-shaped relationship between temperature and stillbirth rate. The lag coefficients for one to three months were negative and statistically insignificant, except for the second-month lag, which was marginally significant (p = 0.053), suggesting a weak delayed effect of temperature on stillbirth outcomes. Overall, these results indicate that the non-linear association between temperature and stillbirth risk persists even when considering short-term lagged exposures, although the model could be further refined to better capture delayed temperature effects.
Table 3.
Distributed lag model (Lag 0–3 months) for temperature–stillbirth association
| Predictor | Coefficient (β) | Std. Error | p-value |
|---|---|---|---|
| Max temperature (°C) | −1.19 | 0.41 | < 0.01 |
| Max temperature² | 0.018 | 0.0063 | < 0.01 |
| Lag 1 month | −0.05 | 0.09 | 0.579 |
| Lag 2 months | −0.16 | 0.08 | 0.053 |
| Lag 3 months | −0.02 | 0.08 | 0.834 |
Source: Author
The exposure–response curve (Fig. 3) validates the U-shaped relationship between temperature and stillbirth rate observed in the fixed-effects and distributed lag models. The relative risk of stillbirth is elevated at both low and high temperature extremes, with 31 °C identified as the minimum-risk threshold. Exposure to lower maximum temperatures shows a notably higher risk, such as at 18 °C, where the risk of stillbirth increases by approximately 14% (RR = 1.149, 95% CI: 1.014–1.302). As temperature rises toward the threshold, the risk decreases and becomes minimal around 31 °C. However, beyond this threshold, the risk increases again, with exposures of 35 °C and 38 °C raising the risk by about 7% and 4%, respectively, relative to the reference (Table 4). The lag–response curves (Fig. 4) further reveal that the risk is highest during exposure to both cold and extreme heat conditions. This risk declines after the first month of exposure, reaches its lowest point around the second month, and rises again by the third month, indicating a possible delayed effect of temperature extremes on fetal health (Fig. 4). Overall, these findings confirm that both cold and heat exposures contribute to higher stillbirth risk with evident short-term and lagged effects.
Fig. 3.

Cumulative exposure–response relationship between maximum temperature and stillbirth risk
Table 4.
Distributed lag non-linear model
| Temperature (°C) | RR | 95% CI |
|---|---|---|
| 31 °C* (Reference) | 1 | – |
| 18 °C | 1.149*** | 1.014–1.302 |
| 35 °C | 1.073*** | 1.043–1.103 |
| 38 °C | 1.041*** | 1.002–1.081 |
Source: Author; (p-value ***<0.001)
Fig. 4.

Lagged effects of cold, high, and extreme heat exposure on stillbirth risk (up to 3-month lag)
Discussion
The findings of the present study indicate a positive association between extreme temperature exposure and stillbirth rate. Exposure to either extreme heat or cold increases the risk of stillbirth by approximately 7–14%, illustrating a U-shaped relationship between temperature and stillbirth incidence [30]. This U-shaped pattern aligns closely with evidence from a previous study conducted in Taiwan [13], which similarly reported elevated stillbirth risks at both cold and hot temperature extremes using DLNM. Such consistency across studies strengthens the biological and epidemiological plausibility of dual thermal vulnerability during pregnancy. Overall, our results support the growing international evidence that fetal development is highly sensitive to temperature variations. With ongoing climate change, this risk is expected to intensify due to the increasing frequency and duration of heatwaves. Women from socio-economically marginalized communities remain particularly vulnerable because of limited access to cooling resources and inadequate healthcare support [31]. These findings underscore the urgent need for proactive policy interventions aimed at enhancing maternal health resilience, improving heat adaptation strategies, and ensuring health equity and reproductive justice for both mothers and their developing foetuses.
The study findings suggest that mean still birth rate for the state of Bihar during 2009–2019 was 16.0, comparable to the estimates of the previous study [32]. Result of different statistical models consistently demonstrate a significant non-linear relationship between temperature and stillbirth, characterized by a distinct U-shaped exposure–response pattern [13]. Both the linear and quadratic terms of maximum temperature were statistically significant, indicating that stillbirth risk increases at both lower and higher temperature extremes while remaining minimal at moderate levels. Consistent with our findings, Yang et al. also reported that temperature exposures during earlier weeks and months of pregnancy exert delayed effects on stillbirth risk, particularly during the third trimester. This supports the DLNM results of the present study, suggesting that temperature-related fetal stress may accumulate over pregnancy, intensifying adverse outcomes at later stages. Similar U-shaped associations between temperature and adverse pregnancy outcomes, including stillbirth and preterm birth, have been documented in other regions also [13, 33]. The association in our study persisted even after incorporating lagged temperature exposures up to three months, suggesting that the influence of temperature on fetal outcomes extends beyond immediate exposure periods. The distributed lag non-linear model (DLNM) further confirmed that exposure to extreme cold (18 °C) and extreme heat (≥ 35 °C) substantially elevates the relative risk of stillbirth, with 31 °C identified as the threshold for minimal risk. Furthermore, the results indicate a steady decline in stillbirth rates after 2010, which may reflect the expansion of maternal healthcare services and improved institutional delivery coverage across districts [34]. The implementation of national maternal health initiatives such as the Janani Suraksha Yojana (JSY) likely contributed to this improvement by enhancing access to skilled care during pregnancy and childbirth [35, 36]. Overall, the findings reinforce the vulnerability of pregnant women to climatic stress while highlighting the positive impact of health system strengthening on reducing stillbirth incidence in recent years.
Despite these robust findings, the study has several limitations. First, the analysis relied on aggregated monthly stillbirth counts reported through the Health Management Information System (HMIS), which may be prone to underreporting or inconsistencies, particularly in rural and resource-constrained districts [37, 38]. The HMIS does not capture detailed maternal or socio-demographic characteristics, such as maternal age, education, parity, nutritional status, and antenatal care utilization, thereby limiting the ability to control for important confounders commonly included in perinatal epidemiological studies. Second, temperature exposure was assessed using district-level ambient measurements, which may not fully reflect individual-level exposure due to variations in housing, occupation, and microclimatic conditions. Such ecological exposure misclassification is likely non-differential and may attenuate the estimated associations. Third, the study focused exclusively on temperature and did not consider other environmental determinants such as humidity, air pollution, or rainfall variability, which may interact with heat or cold to influence fetal health [39, 40]. These limitations were considered when interpreting the results, and future research incorporating individual-level socio-demographic information and multiple environmental exposures is needed to strengthen causal inference and improve spatial precision in evaluating climate-related pregnancy risks.
The findings of this study carry important implications for public health planning and climate adaptation strategies. The identified U-shaped relationship between temperature and stillbirth risk suggests that both heat and cold extremes can adversely affect maternal health, emphasizing the need for temperature-sensitive maternal care interventions. Integrating climate information systems into maternal health programs could help anticipate and mitigate risks during periods of extreme weather. Strengthening local health infrastructure, promoting timely antenatal check-ups, and disseminating early warning advisories during high-temperature months can substantially reduce preventable stillbirths. Additionally, improving the accuracy and completeness of HMIS data through enhanced reporting mechanisms and digital monitoring will be critical for evidence-based decision-making. In the broader context, the results underscore the necessity of incorporating maternal and reproductive health indicators into climate resilience and adaptation frameworks at both state and district levels to safeguard vulnerable populations from the growing impacts of climate variability.
Conclusion
This study finding provide evidence of a non-linear association between ambient temperature and stillbirth rates in Bihar, with both cold and heat extremes contributing to elevated risk. Using fixed-effects panel regression and DLNM, we identified a U-shaped relationship, with the lowest risk around 31 °C and increased risks at approximately 18 °C and ≥ 35 °C. These findings indicate that fetal outcomes are sensitive to temperature variations across both immediate and lagged exposure periods. While the analysis is constrained by the use of aggregated HMIS data and the absence of maternal and socio-demographic covariates, the results contribute important empirical insight from a resource-limited setting. The study underscores the relevance of temperature variability as a determinant of stillbirth risk and provides a foundation for future research incorporating individual-level data and broader environmental exposures.
Supplementary Information
Acknowledgements
The authors acknowledge the continuous support of their PhD supervisor and senior during the study time period.
Authors’ contributions
**AY: ** Conceptualization, data analysis, visualization, writing—original draft preparation, review, and editing. **AR: ** Data curation, review, and editing. All authors have read and approved the final version of the manuscript.
Funding
Not applicable.
Data availability
Data used in this study are publicly accessible. Stillbirth data were obtained from the Health Management Information System (HMIS), Ministry of Health and Family Welfare, Government of India ( [https://hmis.mohfw.gov.in/#!/](https://hmis.mohfw.gov.in/#!/) ), and temperature data were sourced from the India Meteorological Department (IMD) gridded datasets ( [https://www.imdpune.gov.in/cmpg/Griddata/Max_1_Bin.html](https://www.imdpune.gov.in/cmpg/Griddata/Max_1_Bin.html) ). The processed analytical dataset and statistical code used for this study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Abdul-Nabi SS, Al Karaki V, Khalil A, El Zahran T. Climate change and its environmental and health effects from 2015 to 2022: A scoping review. Heliyon. 2025;11(3):e42315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ahmed T, Scholz M, Al-Faraj F, Niaz W. Water-Related impacts of climate change on agriculture and subsequently on public health: A review for generalists with particular reference to Pakistan. Int J Environ Res Public Health. 2016;13(11):1051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Saleem A, Anwar S, Nawaz T, Fahad S, Saud S, Ur Rahman T, et al. Securing a sustainable future: the climate change threat to agriculture, food security, and sustainable development goals. JUmm Al-Qura Univ Appll Sci. 2025;11(3):595–611.
- 4.HK E. Why Marginalised Groups are Disproportionately Affected by Climate Change. Earth.Org. 2022. Available from: https://earth.org/marginalised-groups-are-disproportionately-affected-by-climate-change/ . Cited 10 Nov 2025.
- 5.Zahnow R, Yousefnia AR, Hassankhani M, Cheshmehzangi A. Climate change inequalities: A systematic review of disparities in access to mitigation and adaptation measures. Environ Sci Policy. 2025;165:104021. [Google Scholar]
- 6.Ouambo Talla AW, Ascierto A, Sanvido A, Severi P, Trabuio A, Ceccolini F et al. The heat, the heart and beyond: a narrative review of the many ways climate change impacts human health. Front Clim. 2025;7. Available from: https://www.frontiersin.org/journals/climate/articles/ . Cited 10 Nov 2025.
- 7.Mahmood FM. Global warming and adverse pregnancy outcomes – the need for research and intervention. Int J Health Sci (Qassim). 2024;18(5):1–3. [PMC free article] [PubMed] [Google Scholar]
- 8.Konkel L. Taking the heat: potential fetal health effects of hot temperatures. Environ Health Perspect. 2019;127(10):102002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rashid H, Kagami M, Ferdous F, Ma E, Terao T, Hayashi T, et al. Temperature during pregnancy influences the fetal growth and birth size. Trop Med Health. 2016;45:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rekha S, Nalini SJ, Bhuvana S, Kanmani S, Vidhya V. A comprehensive review on hot ambient temperature and its impacts on adverse pregnancy outcomes. J Mother Child. 2023;27(1):10–20. [DOI] [PMC free article] [PubMed]
- 11.Asamoah B, Kjellstrom T, Östergren PO. Is ambient heat exposure levels associated with miscarriage or stillbirths in hot regions? A cross-sectional study using survey data from the Ghana maternal health survey 2007. Int J Biometeorol. 2018;62(3):319–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Li Z, Li S, Su S, Morawska L, Jalaludin B, Dharmage S, et al. Extreme temperature events exposure and risks of stillbirths: nationwide large birth cohort study in China. Environ Int. 2025;204:109854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yang HY, Lee JKW, Chio CP. Extreme temperature increases the risk of stillbirth in the third trimester of pregnancy. Sci Rep. 2022;12(1):18474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McElroy S, Ilango S, Dimitrova A, Gershunov A, Benmarhnia T. Extreme heat, preterm birth, and stillbirth: A global analysis across 14 lower-middle income countries. Environ Int. 2022;158:106902. [DOI] [PubMed] [Google Scholar]
- 15.Auger N, Fraser WD, Smargiassi A, Bilodeau-Bertrand M, Kosatsky T. Elevated outdoor temperatures and risk of stillbirth. Int J Epidemiol. 2017;46(1):200–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Basu R, Sarovar V, Malig BJ. Association between high ambient temperature and risk of stillbirth in California. Am J Epidemiol. 2016;183(10):894–901. [DOI] [PubMed] [Google Scholar]
- 17.Chersich MF, Pham MD, Areal A, Haghighi MM, Manyuchi A, Swift CP, et al. Associations between high temperatures in pregnancy and risk of preterm birth, low birth weight, and stillbirths: systematic review and meta-analysis. BMJ. 2020;371:m3811. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sen KK, Chapman AJ, Saha BB. Analyzing the influence of energy poverty on newborn mortality through econometric and machine learning approaches. Renewable and Sustainable Energy Reviews. 2025;217(C). Available from: https://ideas.repec.org//a/eee/rensus/v217y2025ics1364032125004587.html . Cited 10 Dec 2025.
- 19.Zhang H, Mao Z, Huang K, Wang X, Cheng L, Zeng L, et al. Multiple exposure pathways and health risk assessment of heavy metal(loid)s for children living in fourth-tier cities in Hubei Province. Environ Int. 2019;129:517–24. [DOI] [PubMed] [Google Scholar]
- 20.Sorensen C, Murray V, Lemery J, Balbus J. Climate change and women’s health: impacts and policy directions. PLOS Med. 2018;15(7):e1002603. [DOI] [PMC free article] [PubMed]
- 21.Moritz S, Bartz-Beielstein T, imputeTS. Time series missing value imputation in R. R J. 2017;9(1):207–18. [Google Scholar]
- 22.Dandona R, George S, Majumder M, Akbar M, Kumar GA. Stillbirth undercount in the sample registration system and National family health survey, India. Bull World Health Organ. 2023;101(3):191–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Brimicombe C, Wieser K, Monthaler T, Jackson D, Bont JD, Chersich MF et al. Effects of ambient heat exposure on risk of all-cause mortality in children younger than 5 years in Africa: a pooled time-series analysis. Lancet Planet Health. 2024;8(9):e640–6. [DOI] [PubMed]
- 24.Deschênes O, Greenstone M. Climate Change, Mortality, and adaptation: evidence from annual fluctuations in weather in the US. Am Economic Journal: Appl Econ. 2011;3(4):152–85. [Google Scholar]
- 25.Burgess R, Deschenes O, Donaldson D, Greenstone M, Weather, death in India. climate change and. 2017; Available from: https://www.lse.ac.uk/economics/Assets/Documents/personal-pages/robin-burgess/weather-climate-change-and-death.pdf
- 26.Guo Y, Gasparrini A, Armstrong BG, Tawatsupa B, Tobias A, Lavigne E, et al. Temperature variability and mortality: A Multi-Country study. Environ Health Perspect. 2016;124(10):1554–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Gasparrini A, Armstrong B, Kovats S, Wilkinson P. The effect of high temperatures on cause-specific mortality in England and Wales. Occup Environ Med. 2012;69(1):56–61. [DOI] [PubMed] [Google Scholar]
- 28.Gasparrini A, Armstrong B, Kenward MG. Distributed lag non-linear models. Stat Med. 2010;29(21):2224–34. [DOI] [PMC free article] [PubMed]
- 29.Gasparrini A, Leone M. Attributable risk from distributed lag models. BMC Med Res Methodol. 2014;14(1):55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ruan T, Yue Y, Lu W, Zhou R, Xiong T, Jiang Y, et al. Association between low ambient temperature during pregnancy and adverse birth outcomes: A systematic review and meta-analysis. Chin Med J (Engl). 2023;136(19):2307–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Richards M, Huang M, Strickland MJ, Newman AJ, Warren JL, D’Souza R, et al. Acute association between heatwaves and stillbirth in six US States. Environ Health. 2022;21(1):59. [DOI] [PMC free article] [PubMed]
- 32.Dandona R, Kumar GA, Akbar Md, Bhattacharya D, Nanda P, Dandona L. Deferred and referred deliveries contribute to stillbirths in the Indian state of bihar: results from a population-based survey of all births. BMC Med. 2019;17(1):28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Sexton J, Andrews C, Carruthers S, Kumar S, Flenady V, Lieske S. Systematic review of ambient temperature exposure during pregnancy and stillbirth: Methods and evidence. Environ Res. 2021;197:111037. [DOI] [PubMed] [Google Scholar]
- 34.Mahapatro SR, Sethy M. Maternal satisfaction in health facilities during childbirth and newborn survival in Bihar state of India. Discov Public Health. 2025;22(1):212. [Google Scholar]
- 35.Lim SS, Dandona L, Hoisington JA, James SL, Hogan MC, Gakidou E. India’s Janani Suraksha Yojana, a conditional cash transfer programme to increase births in health facilities: an impact evaluation. Lancet. 2010;5(9730):2009–23. [DOI] [PubMed]
- 36.Nandeep ER, Mamidi RS, Pagidoju S, Pamidi S, Mummadi MK, Reddy GVR, et al. Comparison of Janani Suraksha Yojana (JSY) and augmented Arogya Laxmi scheme (ALS) in improving maternal and child health outcomes in urban settlements of Hyderabad, South India. BMC Pregnancy Childbirth. 2024;24(1):188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Meghani A, Tripathi AB, Bilal H, Gupta S, Prakash R, Namasivayam V, et al. Optimizing the health management information system in Uttar Pradesh, india: implementation insights and key learnings. Glob Health Sci Pract. 2022;10(4):e2100632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Sakia N, Husain Z, Bora R. Challenges of HMIS in India. IEG. 2013. Available from: https://iegindia.org/policy-brief/challenges-of-hmis-in-india/ . Cited 10 Nov 2025.
- 39.Dehdashti B, Bagheri N, Amin MM, Hajizadeh Y. Impacts of climate changes on pregnancy and birth outcomes: A review. Int J Environ Health Eng. 2020;9(1):24. [Google Scholar]
- 40.Veras MM, Saldiva PHN. Impact of air pollution and climate change on maternal, fetal and postnatal health. J Pediatr (Rio J). 2024;101(Suppl 1):S48–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data used in this study are publicly accessible. Stillbirth data were obtained from the Health Management Information System (HMIS), Ministry of Health and Family Welfare, Government of India ( [https://hmis.mohfw.gov.in/#!/](https://hmis.mohfw.gov.in/#!/) ), and temperature data were sourced from the India Meteorological Department (IMD) gridded datasets ( [https://www.imdpune.gov.in/cmpg/Griddata/Max_1_Bin.html](https://www.imdpune.gov.in/cmpg/Griddata/Max_1_Bin.html) ). The processed analytical dataset and statistical code used for this study are available from the corresponding author upon reasonable request.




