Indoor air pollution and health outcomes in women in developing Asian countries: a systematic review

Type of study

This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [20]. The PRISMA statement provides a checklist to ensure the transparent and complete reporting of systematic reviews.

Search strategy

A comprehensive electronic search of the literature was performed across five databases: Cochrane Library (Wiley), PubMed, Web of Science, ScienceDirect, and Scopus. The search objective was to identify peer-reviewed studies published in English from 2000 to 2025, that evaluated the association between indoor air pollution and health outcomes in women residing in developing Asian countries. The search strategy combined keywords and MeSH terms related to exposure (e.g., "Air Pollution, Indoor", "household air pollution", "solid fuel", "biomass") and population/outcomes (e.g., "Women's Health", "Asia", "Developing Countries", "Pregnancy Outcome", "Respiratory Tract Diseases"). The full search string adapted for each database is provided in Supplementary Material 4. The final search was conducted on May 31, 2025.

Study selection

Studies were selected if they met the following inclusion criteria: (a) investigated the relationship between indoor air pollution and health outcomes; (b) the study population consisted of women aged ≥18 years residing in developing Asian countries; (c) were observational studies (cross-sectional, cohort, or case-control); and (d) published in English between 2000 and 2025. We excluded literature reviews, commentaries, case reports, editorials, conference abstracts, animal studies, and grey literature. Furthermore, studies that did not explicitly state that they had obtained ethical approval from a relevant ethics committee (e.g., Institutional Review Board/IRB) and/or informed consent from participants were also excluded from this review.

All articles identified were managed using Mendeley v1.19.8. After removing duplicates, titles and abstracts were independently screened by two reviewers (F.R. & W.A.P.). Subsequently, the full texts of potentially eligible articles were retrieved and assessed for final inclusion by the same two reviewers. Any disagreements during the screening and selection process were resolved through discussion with a third reviewer (D.F.).

Methodological quality and bias assessment

The methodological quality and internal validity of the included studies were systematically evaluated to ensure the robustness of the synthesized evidence. This multi-faceted assessment was conducted using specific tools tailored to each study design. The entire process was performed independently by two authors (F.R. and W.A.P.), with any discrepancies resolved through discussion and consensus involving a third author (D.F.).

For cohort and case-control studies, quality was assessed using the Newcastle-Ottawa Scale (NOS) [21]. This tool evaluates three key domains: the selection of study groups, the comparability of the groups, and the ascertainment of either the exposure or outcome. Studies were awarded a maximum of nine stars based on their performance across these domains. In accordance with established criteria, studies with a total score of 6–9 stars were classified as having high quality, while those scoring 0–5 stars were considered to be of low quality.

The quality of cross-sectional studies was evaluated using the Appraisal Tool for Cross-Sectional Studies (AXIS) [22]. This tool consists of a 20-item checklist designed to assess methodological rigor in areas such as study objectives, sample size justification, participant selection, and the validity and reliability of measurement methods. Each item was scored as 'yes' (1 point) or 'no'/'don't know' (0 points). Based on the total score, studies were categorized as high quality (>15 points), moderate quality (10–15 points), or low quality (<10 points).

In addition to quality scoring, the risk of bias for each study was comprehensively assessed using the Office of Health Assessment and Translation (OHAT) tool [23]. This framework is specifically designed for environmental health literature and evaluates the internal validity across critical domains, including exposure assessment, outcome assessment, confounding, selection bias, attrition/exclusion bias, and selective reporting. Each domain for every study was rated as 'low risk', 'probably low risk', 'probably high risk', or 'high risk' of bias. In line with our protocol, studies that were rated as having a 'high' or 'probably high' risk of bias across the key domains (exposure assessment, outcome assessment, and confounding adjustment) were excluded from the final narrative synthesis to ensure the integrity of our findings. Furthermore, studies classified as 'low quality' by the NOS or AXIS tools were not automatically excluded, but their methodological limitations were taken into consideration during the narrative synthesis of the findings.

Data extraction and analysis

A standardized form was used to extract data from each included study, covering: author, publication year, study design, location, number of participants, exposure characteristics, health outcomes, adjusted variables, and main findings. The data extraction was performed independently by two reviewers (F.R. & W.A.P.). Given the significant heterogeneity in exposure assessment methods, specific health outcomes measured, and statistical approaches across the included studies, a quantitative meta-analysis was not feasible. Therefore, the findings were synthesized and are presented as a narrative review.

Study Selection and Characteristics

The initial database search yielded 3,011 articles. After removing 88 duplicates, 2,923 unique articles were screened by title and abstract. Following this, 2,518 articles underwent full-text review, from which 2,401 were excluded, primarily for having an ineligible study location (n=984) or setting (n=660). Of the 117 remaining articles, a final 89 were excluded, predominantly due to inappropriate methodology as per the quality assessment (n=88), such as having a high risk of bias across key domains. Ultimately, this rigorous process yielded 28 studies that met all inclusion criteria for this review. The complete study selection process is illustrated in the PRISMA flow diagram (Figure 1).

Figure 1.

PRISMA flow diagram of literature selection

The 28 included studies were published between 2010 and 2024. The evidence base was geographically concentrated, with the majority of studies originating from China (n=9) and India (n=9). The most common study design was cross-sectional (n=19). Exposure assessment methods varied significantly, ranging from proxy indicators like self-reported fuel type (n=17) to objective personal pollutant measurements (n=11). The health outcomes investigated were diverse, spanning respiratory, cardiovascular, reproductive, cellular, and cognitive endpoints. A detailed summary of each included study is presented in Table 1.

Table 1.

Characteristics of included studies on indoor air pollution and health outcomes in women.

Respiratory Health Outcomes

Respiratory diseases are the most well-documented health effect of household air pollution (HAP) [24, 25]. The combustion of solid fuels releases a complex mixture of pollutants, including particulate matter (PM₂.₅) and polycyclic aromatic hydrocarbons (PAHs), many of which are also found in tobacco smoke [25]. The synthesized evidence from the 28 studies included in this review consistently demonstrates that HAP exposure is a major risk factor for adverse respiratory outcomes in women. This conclusion is robustly supported by a convergence of evidence across different methodological approaches, from self-reported symptoms to objective physiological measurements and mechanistic biomarkers.

Evidence from studies based on self-reported symptoms consistently shows a strong association between HAP and respiratory morbidity. For instance, Tanwar et al. (2024) [26] , in a methodologically robust, cross-sectional study, found that women using unclean fuels had a three-fold higher risk of experiencing respiratory symptoms compared to clean fuel users (aOR: 3.0; 95% CI: 1.5–6.0). This finding aligns with numerous previous reviews which have established self-reported cough, phlegm, and wheezing as primary indicators of HAP-related harm [27, 28]. While subjective, these symptom-based studies are critical as they reflect the perceived disease burden in the population and provide a foundational link between exposure and health.

This association was further corroborated by studies utilizing objective physiological measurements of lung function. Two separate studies, Kaur-Sidhu et al. (2019) [29] and Mitra et al. (2022) [30], both employed spirometry and independently found that women who primarily used solid biomass fuel had significantly lower Forced Expiratory Volume in 1 second (FEV₁) values compared to those using clean LPG. This reduction in FEV₁, a key indicator of airflow limitation, is a primary diagnostic criterion for obstructive lung diseases such as COPD [31]. The convergence of findings from both subjective symptom reports and objective lung function tests strengthens the conclusion that HAP directly impairs pulmonary mechanics.

Furthermore, the inflammatory nature of this physiological impairment was demonstrated through biomarker evidence. A study by Parikh et al. (2021) [32] on pregnant women found that biomass users had significantly elevated levels of fractional exhaled nitric oxide (FeNO), an objective biomarker for airway inflammation. Elevated FeNO is a well-established marker of eosinophilic inflammation, a key feature of Type 2 inflammatory pathways often implicated in atopic asthma [33, 34]. This finding provides a direct molecular link between HAP exposure and a specific type of inflammatory response in the airways.

The collective evidence from these different methodological approaches points towards a coherent biological mechanism: chronic inflammation. The inhalation of pollutants from HAP is understood to trigger a persistent inflammatory cascade within the airways [35, 36]. As evidenced by the elevated FeNO levels, this response involves the recruitment of inflammatory cells that, over time, can lead to airway remodeling, increased mucus production, and structural damage to the lung parenchyma [37, 38]. This chronic process provides a strong biological explanation for both the reported symptoms (e.g., cough and phlegm) and the observed decline in lung function (lower FEV₁), which are hallmarks of chronic obstructive pulmonary disease (COPD) [39].

While these biological mechanisms are universal, the burden of HAP-related respiratory disease falls disproportionately on women due to a critical intersection of social roles and physiological susceptibility. Traditional gender roles in many developing Asian societies designate women as the primary household cooks, leading to higher-intensity and longer-duration exposures over their lifetime, thereby increasing their cumulative risk of lung damage [40]. Furthermore, some evidence suggests that women may be more physiologically susceptible to inhaled pollutants due to generally smaller airway caliber compared to men [41, 42]. This vulnerability is acutely heightened during pregnancy, a unique life stage for women. As demonstrated by the findings of Parikh et al. (2021) [32], significant airway inflammation occurs in pregnant women exposed to HAP. Such respiratory impairment during gestation is not only a direct threat to the mother's health but also poses an indirect risk to the fetus, as maternal hypoxia can compromise oxygen delivery and potentially contribute to the adverse birth outcomes discussed later in this review [43].

Beyond the consistency of the main findings, the included studies also suggest important sources of heterogeneity that may modify the risk of respiratory disease. For instance, cooking duration was identified as a critical factor, with Tanwar et al. (2024) [26] showing that longer daily cooking times significantly increased the risk of impaired lung function, independent of fuel type. This implies that even among biomass users, behavioral factors that reduce exposure duration can be protective. Additionally, housing characteristics, such as the presence of adequate ventilation, were highlighted by Maharana et al. (2018) [44] as significant modifiers of respiratory symptoms. These findings suggest that the respiratory risk from HAP is not uniform but is instead a complex interplay between the type of fuel used, exposure duration, and the household environment.

While the overall findings consistently point towards a detrimental effect, a deeper analysis of the methodologies of the included respiratory studies reveals important nuances. The strength of the observed association appears to be influenced by the rigor of the exposure assessment. Studies that employed objective PM₂.₅ monitoring, such as Mitra et al. (2022) [30], were able to directly link higher pollutant concentrations with poorer lung function. It is plausible that studies relying on crude proxies may underestimate the true magnitude of the risk due to non-differential exposure misclassification [45]. Furthermore, the control of key confounding variables, particularly tobacco smoke, was critical. The findings from studies that methodologically controlled for this by exclusively recruiting non-smokers, such as Kaur-Sidhu et al. (2019) [29], provide some of the cleanest evidence isolating the independent effect of HAP on pulmonary mechanics. This highlights that while the conclusion is robust, the precision of the risk estimates is highly dependent on study design.

These findings strongly reaffirm the conclusions of previous global systematic reviews and meta-analyses, which have robustly established the link between HAP and respiratory morbidity [15]. However, this review contributes to the literature in several critical ways. Firstly, by focusing specifically on developing Asian countries, it highlights that this burden remains severe in a region undergoing rapid but uneven economic transition. Secondly, it synthesizes more recent evidence, including studies from 2022-2024 that were not included in prior global assessments. Finally, the consolidation of evidence across subjective symptoms, objective spirometry, and mechanistic biomarkers like FeNO provides a more comprehensive and layered confirmation of this association within the specific context of Asian women.

Cardiovascular and Metabolic Health Outcomes

Beyond localized effects on the lungs, the evidence synthesized in this review consistently demonstrates that HAP-induced inflammation extends to become a systemic issue, critically impacting cardiovascular and metabolic health. In addition to its well-established impact on respiratory health, emerging evidence increasingly links HAP to adverse cardiovascular and metabolic outcomes [46]. The main finding of this review is that there are significant associations between HAP exposure and increased blood pressure, a higher risk of Metabolic Syndrome (MetS), and Gestational Diabetes Mellitus (GDM). This conclusion is supported by evidence from studies utilizing proxy-based exposure assessments, objective physiological measurements, and relevant clinical biomarkers.

Several studies relying on proxy measures of exposure provided foundational evidence for this association. For instance, a case-control study by Wang et al. (2018) [47] used a composite IAP exposure index based on fuel type, heating methods, and ventilation, finding that women with high exposure had a significantly increased risk of hypertension (aOR: 2.52; 95% CI: 1.46–4.36). Similarly, Du et al. (2023) [48], in a large cross-sectional study in Beijing, found that high HAP exposure, proxied by cooking behavior (≥14 times/week without a range hood), was associated with an increased risk of Metabolic Syndrome (OR: 1.298; 95% CI: 1.028–1.639). While limited by their indirect nature, these studies are crucial as they establish a consistent pattern of risk across large populations.

This association was further strengthened by studies that employed objective clinical measurements for health outcomes. The high-quality multi-country study by Nicolaou et al. (2022) [49], which utilized personal PM₂.₅ monitoring, found that high exposure was linked to clinically significant increases in objectively measured systolic blood pressure in older women. Another large-scale study by Liu et al. (2022) [50] corroborated the link to hypertension, finding that women who cooked frequently had a 2.27-fold higher prevalence ratio for the condition. In the metabolic realm, the high-quality prospective cohort study by Dai et al. (2024) [51] used standardized oral glucose tolerance tests to objectively diagnose GDM, demonstrating a clear link between solid fuel use for heating and an increased risk of the disease (aOR: 1.60; 95% CI: 1.18–2.18).

The evidence extends to the biomarker level, providing insights into the specific metabolic pathways affected by HAP. The study by Liu et al. (2022) [50] found that increased cooking exposure was significantly correlated with an adverse lipid profile, including elevated levels of triglycerides and LDL-cholesterol, and higher levels of homocysteine—all of which are established risk factors for cardiovascular disease [52, 53]. Conversely, the study by Shahriar et al. (2021) [54], which specifically measured endothelial dysfunction via log-transformed reactive hyperemia index (LnRHI), did not find a significant association with HAP exposure. This null finding is highly informative and suggests several possibilities. As the authors noted, it may reflect the young age and relative health of the study population, implying that clinically manifest vascular damage might require longer cumulative exposure or co-existing vulnerabilities. Alternatively, it could suggest that HAP's primary cardiovascular impact may occur through other pathophysiological pathways not captured by RH-PAT, such as direct effects on arterial stiffness or autonomic dysfunction, rather than microvascular reactivity [55].

The most plausible biological explanation for these findings is a multi-step process initiated by an "inflammatory spill-over" from the lungs into the systemic circulation [56]. The process begins when inhaled pollutants, particularly ultrafine particulate matter (PM₂.₅), cross the alveolar-capillary barrier and enter the bloodstream directly, while also triggering a local inflammatory response in the lungs [57]. This leads to the release of pro-inflammatory cytokines (such as TNF-α and IL-6) into circulation, creating a state of low-grade systemic inflammation [58]. This systemic inflammation then promotes endothelial dysfunction by reducing the bioavailability of nitric oxide (NO), a critical vasodilator, leading to increased vascular tone and contributing directly to hypertension [59]. Simultaneously, these inflammatory mediators are known to interfere with insulin signaling pathways in peripheral tissues like muscle and adipose tissue, thereby inducing insulin resistance, which is the core pathophysiological mechanism underlying both Gestational Diabetes Mellitus (GDM) and Metabolic Syndrome (MetS) [60].

The included studies reveal that the cardiovascular and metabolic risk from HAP is not uniform but is significantly modified by several factors. Physiological susceptibility is a key modifier; Nicolaou et al. (2022) [49] found the effect of PM₂.₅ on blood pressure was strongest in older women and those with a higher BMI. Lifestyle factors were also shown to be critical. Dai et al. (2024) [51] demonstrated that a healthy lifestyle conferred a strong protective effect against GDM, even among women exposed to HAP. Furthermore, exposure characteristics are paramount. The finding by Liu et al. (2022) [50] that hypertension risk persisted even among clean fuel users with high cooking frequency underscores that the hazard stems not just from fuel type but also from the cooking process itself (e.g., cooking oil fumes) and highlights the indispensable role of non-fuel interventions like adequate ventilation.

A deeper analysis of the methodologies highlights important nuances in the evidence base. The most compelling evidence comes from studies like Nicolaou et al. (2022) [49], which combined objective personal exposure monitoring with objective clinical outcomes. In contrast, studies relying on crude proxy measures, such as the exposure index used by Wang et al. (2018) [47], are more susceptible to non-differential exposure misclassification, which likely biases the results toward the null and underestimates the true magnitude of the risk. The strength of the evidence is also enhanced by the inclusion of two prospective cohort studies [51, 61], which provide stronger evidence for temporality compared to the more common cross-sectional designs.

The findings of this review, linking HAP to cardiovascular and metabolic diseases, are consistent with the vast body of literature on outdoor air pollution [62, 63]. However, this review provides a unique contribution by synthesizing diverse evidence specifically for HAP exposure among women in developing Asian countries. It moves beyond single-outcome studies to demonstrate a consistent pattern of risk across a spectrum of related conditions—hypertension, Metabolic Syndrome, and GDM. Crucially, by including recent high-quality studies, this review highlights that significant risks can persist even in urban settings with clean fuel use, shifting the focus towards the importance of cooking behaviors and ventilation as critical avenues for intervention.

Taken together, the findings on cardiovascular and metabolic health have profound implications that extend beyond current intervention paradigms. The critical evidence from studies like Liu et al. (2022) [50], demonstrating persistent hypertension risk even among clean fuel users, suggests that a simple transition to cleaner fuels, while essential, is not a panacea. It highlights that pollutants from the cooking process itself, such as cooking oil fumes, are an independent source of risk, thereby elevating the importance of non-fuel interventions like improved ventilation as a primary public health strategy [64]. Furthermore, the association with Gestational Diabetes Mellitus is particularly alarming as it represents a sentinel health event with long-term consequences, increasing the future risk of type 2 diabetes for the mother and metabolic dysfunction for the child [65]. Ultimately, the evidence synthesized here suggests that hypertension, Metabolic Syndrome, and GDM should not be viewed as disparate outcomes, but rather as interconnected manifestations of a single underlying pathway of cardiometabolic disruption driven by HAP-induced systemic inflammation.

Reproductive and Pregnancy Outcomes

One of the most significant contributions of this systematic review is the consolidation of strong evidence linking exposure to Household Air Pollution (HAP) with a range of adverse reproductive and pregnancy outcomes. The synthesis of the 28 included studies consistently demonstrates that maternal HAP exposure is a critical risk factor for failure at every stage of gestation, from early pregnancy loss and structural birth defects to impaired fetal growth at term.

The synthesized evidence indicates a persistent threat from HAP to fetal viability. This finding powerfully reinforces conclusions from previous systematic reviews [11] that have established a link between HAP and adverse pregnancy outcomes. The current review adds granular detail by demonstrating this threat across the entire gestational timeline within Asian populations. At later stages of gestation, the study by Hammad et al. (2024) [66], analyzing large-scale national survey data in India, found that the use of unclean solid fuels significantly increased the risk of stillbirth by 34%. This threat is also evident in early pregnancy, where a case-control study by Zheng et al. (2022) [67] found that high HAP exposure more than doubled the risk of Recurrent Spontaneous Abortion (RSA). This consistency of findings across different stages of pregnancy loss underscores the pervasive impact of HAP.

Beyond pregnancy loss, the most frequently reported impact of HAP is impaired fetal growth, a finding consistent with the well-documented effects of outdoor air pollution on birth outcomes. The studies included in this review provide strong, region-specific evidence for this link. Several studies, including those by Vakalopoulos et al. (2021) [68] in Sri Lanka and Pan et al. (2022) [69] in China , independently found that the use of biomass or solid fuels significantly increases the risk of delivering infants with Low Birthweight (LBW) or who are Small for Gestational Age (SGA). Specifically, Vakalopoulos et al. (2021) [68] reported that the risk of LBW more than doubled among biomass users (aOR: 2.74). While the majority of evidence supports this link, the null finding for wood fuel use from the study by Haider et al. (2016) [70] in Bangladesh deserves critical analysis. The study's reliance on subjective maternal recall for both exposure and outcome likely led to non-differential misclassification, a common limitation in proxy-based exposure studies, which tends to bias results toward the null and thus does not negate the weight of evidence from other studies with more objective measures.

While the link between HAP and respiratory disease is well-established, perhaps the most alarming evidence from this review is the association between maternal HAP exposure and severe, irreversible structural birth defects. This represents a critical extension of the known harms of HAP. A high-quality, population-based case-control study by Chen et al. (2023) [71] found that each unit increase in HAP exposure doubled the risk of Neural Tube Defects (NTDs). This link is strengthened by another robust case-control study by Sun et al. (2022) [72], which utilized objective in-home air monitoring. This study found that high exposure to Total Volatile Organic Compounds (TVOCs) was associated with a more than seven-fold increased risk of Congenital Heart Defects (CHD) (aOR: 7.09). The magnitude of the observed risk, particularly in a study with objective exposure measurement, provides strong evidence that HAP acts as a potent teratogen during the sensitive period of organogenesis in early pregnancy.

The diversity of observed reproductive outcomes can be explained biologically through several interconnected mechanisms. The following paragraphs detail how specific pathways link HAP exposure to the clinical outcomes identified in this review. First, the evidence strongly points to genotoxicity as the direct mechanism for structural birth defects. Inhaled pollutants like polycyclic aromatic hydrocarbons (PAHs) are known to be genotoxic [73]. They can enter the maternal circulation, cross the placenta, and directly cause DNA damage or disrupt critical cellular signaling pathways during the sensitive window of organogenesis. This pathway is empirically supported by the findings from Mondal et al. (2010) [74], which directly demonstrated increased DNA damage and chromosomal instability in women exposed to HAP. This cellular-level damage provides a compelling biological foundation for the observed increases in Neural Tube Defects and Congenital Heart Defects [71, 72].

Second, the consistent findings of impaired fetal growth and pregnancy loss point toward a primary mechanism of placental dysfunction [75]. The same pathway of systemic inflammation discussed previously in the context of cardiovascular outcomes is highly relevant here. This inflammation, triggered by pollutant inhalation, can cause vascular injury within the placenta itself [76]. This compromises its function, reducing uteroplacental blood flow and disrupting the transport of essential nutrients and oxygen to the fetus. This state of placental insufficiency is a well-understood cause of outcomes like LBW, SGA, and stillbirth, directly explaining the findings from studies like Vakalopoulos et al. (2021) [68], Pan et al. (2022) [69], and Hammad et al. (2024) [66].

Finally, these effects are likely exacerbated by chronic fetal hypoxia, creating a "dual-hit" scenario. Pollutants like carbon monoxide (CO), a major component of biomass smoke, have a high affinity for hemoglobin, reducing the oxygen-carrying capacity of the mother's blood [77]. This leads to chronic oxygen deprivation for the fetus, compounding the malnutrition from placental dysfunction with asphyxiation [78]. This dual assault on the fetus further impairs its growth and compromises its viability.

The risk of adverse reproductive outcomes is not uniform and is significantly modified by exposure characteristics and maternal factors, highlighting actionable targets for intervention. Ventilation was identified as a critical modifier; Vakalopoulos et al. (2021) [68] showed the risk of LBW was highest in homes without a chimney, while Zheng et al. (2022) [67] found that the use of a range hood was protective. These findings reinforce public health recommendations that interventions must go beyond just fuel type to include structural improvements like ventilation, which can substantially reduce exposure levels [79]. Furthermore, the study by Zheng et al. (2022) [67] introduced a novel concept of psychological mediation, where HAP-induced depression and anxiety partially explained the risk of RSA. This is a particularly innovative finding that expands our understanding beyond purely physiological pathways and suggests that mental health support could be a novel component of future HAP mitigation strategies.

The quality of evidence for reproductive outcomes is generally high. However, a key weakness is the reliance on proxy measures for exposure. The stark difference in risk estimates between studies using proxies versus objective monitoring [72] suggests that many studies may be underestimating the true magnitude of the risk. This has significant public health implications, as it suggests the current burden of disease attributed to HAP-related reproductive failure could be substantially higher than officially reported [80], leading to an under allocation of resources for mitigation.

These findings are consistent with the broader literature linking outdoor air pollution to adverse birth outcomes. For example, various global meta-analyses have consistently shown that exposure to PM₂.₅ from traffic and industry poses a significant risk for Low Birthweight (LBW), preterm birth, and even autism in children [81–83]. However, this review provides a unique contribution by focusing specifically on Asian women and synthesizing evidence across the entire spectrum of reproductive failure due to HAP exposure, where pollutant concentrations are often much higher than outdoors. By including recent, high-quality regional studies, this review provides a more granular understanding of the risks faced by this demographic.

Overall, the profound link between HAP and adverse pregnancy outcomes has urgent clinical and public health implications. For clinicians, the household cooking environment should be considered an essential component of antenatal risk assessment. For policymakers, these findings demonstrate that investing in clean cooking solutions is a direct and powerful intervention for maternal and child health. The evidence linking HAP to irreversible outcomes like birth defects and conditions that impact future health trajectories (LBW and SGA) highlights an intergenerational pathway of harm, making HAP a critical target for health equity.

Cellular Damage, Epigenetic Aging, and Cognitive Decline

Beyond organ-level impacts, this review identified a compelling cluster of studies providing strong evidence that HAP exposure inflicts damage at the cellular and molecular level. These sub-clinical effects are critical as they represent the biological foundation for the chronic diseases observed. The synthesized evidence demonstrates that HAP is a potent inducer of oxidative stress, which in turn triggers downstream consequences including genotoxicity, accelerated epigenetic aging, and ultimately, cognitive decline. This conclusion is supported by a range of evidence from methodologically diverse studies, including those with robust laboratory-based biomarkers and large-scale population surveys.

The foundational mechanism of cellular damage appears to be oxidative stress. This was demonstrated directly by Rabha et al. (2018) [84], who found that fuelwood users had significantly higher blood levels of the oxidative stress marker malondialdehyde (MDA) and depleted antioxidant defenses compared to LPG users. This state of oxidative stress was shown to directly translate into genotoxicity in a robust laboratory-based study by Mondal et al. (2010) [74]. They found that women exposed to biomass smoke had a significantly higher frequency of micronuclei—a marker of chromosomal damage—and increased DNA strand breaks in their airway and buccal cells. These findings provide direct evidence that HAP is genotoxic, capable of causing damage to the genetic material.

The most direct consequence of HAP-induced oxidative stress is genotoxicity, where cellular genetic material is fundamentally damaged. This process is initiated by the overproduction of Reactive Oxygen Species (ROS)—highly unstable molecules generated from the incomplete combustion of biomass. These ROS directly attack the molecular structure of DNA, leading to adducts and base oxidation, such as the formation of 8-hydroxy-2'-deoxyguinosine (8-OHdG), a classic marker of oxidative DNA damage [85]. Furthermore, they can cause structural lesions, including both single- and double-strand breaks, which compromise the integrity of the chromosome. When the cell's DNA repair mechanisms are overwhelmed or fail, this damage can become fixed, leading to mutations or chromosomal instability. This accumulated damage is visibly manifested as an increased frequency of micronuclei—small, secondary nuclei containing chromosomal fragments—which serve as a robust and well-validated biomarker for genotoxicity [86].

In addition to direct DNA damage, this review found evidence that chronic HAP exposure leaves a "molecular scar" on the genome through epigenetic alterations. A key study by Blechter et al. (2023) [87], conducted in a unique high-risk population in Xuanwei, China, utilized advanced molecular techniques to assess biological aging. They found that exposure to HAP from smoky coal was associated with a significant acceleration of epigenetic aging, with the strongest effects linked to childhood exposure. Accelerated epigenetic aging is a powerful biomarker that reflects the cumulative lifetime impact of environmental damage and is a strong predictor of all-cause mortality and age-related chronic diseases [88].

The tandem effects of oxidative stress and chronic inflammation converge to disrupt the sophisticated epigenetic machinery that regulates gene expression, thereby accelerating biological aging. The epigenome, which comprises chemical marks like DNA methylation and histone modifications, is dynamically maintained by a suite of specific enzymes. Both oxidative stress and inflammatory signaling can directly interfere with the function of these critical regulators, including DNA methyltransferases (DNMTs), which add methyl groups, and histone deacetylases (HDACs), which modify histone proteins [89]. This dysregulation leads to global changes in the DNA methylation landscape, creating an epigenetic pattern that resembles that of a chronologically older individual. This drift is precisely what is captured by epigenetic clocks, which measure these patterns to calculate an individual's biological age. The observed acceleration of epigenetic aging in HAP-exposed individuals is, therefore, a direct molecular readout of the cumulative cellular damage and systemic stress induced by chronic exposure.

The evidence for HAP's systemic impact extends to neurological health. A large, nationally representative study by Dakua et al. (2022) [90], analyzing data from over 22,000 rural women in India, found a significant association between HAP exposure and lower cognitive function scores in middle-aged and elderly women (β=−0.72). This finding is particularly important as it moves beyond cellular markers to a clinically relevant functional outcome, suggesting that the systemic effects of HAP can manifest as neurological impairment later in life.

Beyond direct genotoxicity, oxidative stress from HAP exposure instigates a more insidious, indirect pathway to neurological impairment through the induction of chronic systemic inflammation. The initial oxidative damage in the lungs triggers a persistent inflammatory response, causing the release of pro-inflammatory cytokines, such as TNF-α and IL-6, into the systemic circulation [91]. This state of chronic, low-grade inflammation is particularly detrimental to the central nervous system, as these circulating cytokines are known to compromise the integrity of the blood-brain barrier [92]. This increased permeability allows inflammatory agents to enter the brain parenchyma, activating resident immune cells like microglia and astrocytes and promoting a state of neuroinflammation. This neuroinflammatory environment disrupts normal neuron function, impairs synaptic plasticity, and ultimately promotes neurodegenerative processes that are hallmarks of age-related cognitive decline [93]. Thus, the cognitive deficits observed in HAP-exposed populations are the clinical manifestation of a long-term, inflammation-driven assault on the brain.

The magnitude of cellular and neurological damage is not uniform and appears to be modified by several key factors. The timing of exposure is critical, as demonstrated by Blechter et al. (2023) [87], who found that childhood exposure to HAP had a more profound impact on epigenetic aging in adulthood than recent exposure. The specific type of biomass may also play a role; Mondal et al. (2010) [74] noted that the highest levels of genotoxic damage were found among women who used dung cake as their primary fuel, likely due to its unique chemical composition and combustion properties. These findings suggest that both life-stage vulnerability and specific exposure characteristics are important determinants of the overall risk.

The evidence for these outcomes is strengthened by the use of highly objective and validated laboratory-based measures (DNA damage assays, methylation arrays). The primary methodological limitation, however, lies in linking these sophisticated outcome measures to exposure. For instance, while the study by Dakua et al. (2022) [90] had a very large and representative sample, its exposure assessment was based on a crude proxy, which may have led to misclassification and an underestimation of the true effect. In contrast, studies like Mondal et al. (2010) [74] combined objective outcome measures with direct in-home air monitoring, providing stronger evidence for a direct causal link between particulate matter concentrations and genotoxicity.

While the link between outdoor air pollution and neurological or cellular effects is an emerging field [94], this review makes a critical contribution by synthesizing the specific evidence for HAP among women in developing Asia. Crucially, it highlights the importance of sub-clinical endpoints. Biomarkers like DNA damage and epigenetic aging are not diseases in themselves, but they are powerful indicators of underlying biological processes that precede the onset of clinical disease. By demonstrating that HAP impacts health at this fundamental level, this review provides a strong rationale for preventative action, arguing that interventions are needed long before irreversible clinical outcomes like cancer or dementia manifest.

The Socio-Behavioral Drivers of HAP-Induced Disease

While the preceding sections have detailed the distinct biological pathways, such as chronic inflammation, placental dysfunction, and genotoxicity, that link HAP exposure to specific organ damage, these mechanisms do not operate in a vacuum. They are fundamentally driven by a set of upstream socio-economic and behavioral factors that determine the initial exposure. Understanding these upstream pathways is critical for designing effective interventions.

The most fundamental driver of HAP exposure is the socio-economic mechanism centered on poverty [95]. Economic constraints directly compel households in developing countries to rely on cheap and accessible solid fuels (such as firewood and biomass) and to live in dwellings with inadequate ventilation quality. Many studies in this review statistically control for variables such as income, wealth index, and education level. However, these factors are more than mere confounding variables; they are core components of the causal pathway itself. This cycle is bidirectional: HAP exposure leads to adverse health outcomes, which in turn can reduce labor productivity and increase healthcare expenditures, thereby perpetuating poverty and reliance on unclean fuels [96]. Thus, HAP is a manifestation of a "poverty trap" where socio-economic conditions create the exposure that causes disease, which then exacerbates those same socio-economic conditions.

Layered on top of the socio-economic foundation is a behavioral and cultural mechanism that directly modulates an individual's level of exposure. Traditional gender roles in many Asian societies designate women as the primary household cooks, which systematically leads them to experience exposure of higher intensity and longer duration compared to men [97]. This review found strong evidence that specific behaviors significantly influence risk. For instance, Tanwar et al. (2024) [26] showed that longer daily cooking durations significantly increased the risk of impaired lung function, independent of the fuel type used. Conversely, behaviors that reduce exposure, such as the use of adequate ventilation (e.g., a chimney or range hood), were shown to be protective, reducing the risk of both Low Birthweight (LBW) and Recurrent Spontaneous Abortion (RSA) [67, 68] . This behavioral pathway is important because it highlights that interventions focused on habit changes and simple structural improvements can be effective mitigation strategies.

An innovative and often overlooked mechanistic pathway highlighted by this review is the psychological pathway. A constantly smoky and unhealthy home environment can be a significant source of chronic stress [98]. This review found direct evidence for this pathway from the study by Zheng et al. (2022) [67], which uniquely demonstrated that psychological status (depression and anxiety) acts as a mediator between HAP exposure and the risk of recurrent spontaneous abortion (RSA). This indicates a complex mind-body connection, where a physical environmental stressor (HAP) triggers psychological distress, which then manifests as an adverse physiological health outcome. The recognition of this pathway is critical as it expands the intervention paradigm beyond technical solutions (like clean cookstoves) to also include potential mental health support as part of a holistic approach to mitigating HAP's impact.

Overall, a comprehensive understanding of HAP's health impacts requires a multi-layered framework. These socio-economic, behavioral, and psychological pathways act as the universal precursors that initiate and sustain the conditions of HAP exposure. This exposure, in turn, triggers the cascade of biological mechanisms, inflammation, oxidative stress, and genotoxicity, that ultimately leads to the broad spectrum of damage observed across the respiratory, reproductive, cardiovascular, and cellular systems discussed in this review.

Quality and bias assessment

The methodological quality and risk of bias for the 28 included studies were systematically evaluated, with detailed assessments presented in Supplementary Tables S1-S3. A visual summary of the risk of bias assessment across all domains is presented as a heatmap in Figure 2. This assessment process itself was a critical finding; a substantial number of potentially relevant studies (n=88) that were eligible after full-text screening had to be excluded due to high risk of bias or critical methodological flaws. This high exclusion rate underscores the variable quality of the existing literature and validates the importance of a rigorous, quality-filtered synthesis.

Figure 2.

Risk of bias assessment

Following this filtration, the overall quality of the final included evidence base was heterogeneous. A notable strength across many of the high-quality studies was a "Low" risk of bias in outcome assessment and confounding adjustment, as shown by the predominantly light coloring in these domains on the heatmap. This was attributed to the use of objective and validated methods, such as the use of spirometry to measure lung function (e.g., Kaur-Sidhu et al., 2019 [29]; Mitra et al., 2022) [30], standardized oral glucose tolerance tests to diagnose Gestational Diabetes Mellitus (e.g., Dai et al., 2024 [51]), and robust laboratory-based assays to quantify genotoxicity and cellular damage (e.g., Mondal et al., 2010 [74]).

Conversely, the most critical and pervasive weakness was a "High" risk of bias in exposure assessment. The majority of studies (n=19) relied on self-reported questionnaires (proxy measures), a method highly susceptible to misclassification bias which may underestimate the true magnitude of the risk [45]. This stands in stark contrast to the handful of studies that demonstrated high rigor by using objective personal exposure measurements (e.g., Nicolaou et al., 2022 [99]). Furthermore, selection bias and other potential sources of bias were also frequently identified as areas of concern. Finally, the evidence base is constrained by the predominance of cross-sectional designs (19 (19 of 28 studies), which inherently preclude the establishment of causal inference [100].

Limitations

This systematic review has several limitations that should be considered, stemming from both the inherent constraints of the primary studies included and the review process itself. First and foremost, the conclusions are constrained by the methodological characteristics of the available evidence base. The most significant constraint is the predominance of cross-sectional designs (19 of 28 studies), which, while useful for generating hypotheses, inherently preclude the establishment of causality and temporality between HAP exposure and the observed health outcomes.

Furthermore, a critical weakness across the majority of the synthesized literature (17 of 28 studies) lies in exposure assessment. Most studies relied on crude proxy measures, such as self-reported fuel type, which cannot capture the variability in actual pollutant concentrations driven by factors like ventilation, cooking duration, and housing characteristics. This is susceptible to non-differential exposure misclassification, which typically biases risk estimates toward the null, suggesting that the true magnitude of HAP's health risks may be underestimated by the current body of evidence. While this review includes several large-scale surveys providing high precision in their estimates (reflected in narrow confidence intervals), the evidence base is also constrained by a number of smaller studies whose findings, while valuable, have wider confidence intervals and thus less certainty. Additionally, the findings are heavily concentrated in China and India, which may limit their generalizability to other regions in developing Asia with different cultural practices, fuel types, and environmental contexts.

Second, the review process itself has limitations. The search was restricted to articles published in English, which may have introduced a language bias. As with any review, the potential for publication bias, where studies with statistically significant findings are more likely to be published, cannot be entirely ruled out. Finally, due to the significant methodological and clinical heterogeneity identified—particularly in exposure metrics and the wide array of specific health outcomes measured—a quantitative meta-analysis was not performed. Consequently, the findings are presented as a narrative synthesis, which relies on a qualitative interpretation of the evidence rather than a single, pooled statistical estimate of effect.