# paper.pdf ## Metadata - PDFFormatVersion=1.4 - Language=en-US - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260617090618+05'30' - Creator=Adobe InDesign 15.1 (Windows) - ModDate=D:20260617090619+05'30' - Producer=Adobe PDF Library 15.0 - Trapped=False - Custom.jobID=c6465d2a-a3f0-4017-af77-afa6e7ecbea3 - dc:identifier=doi:bmjresp-2025-003721 - dc:format=application/pdf - xmp:createdate=2026-06-17T09:06:18+05:30 - xmp:metadatadate=2026-06-17T09:06:19+05:30 - xmp:modifydate=2026-06-17T09:06:19+05:30 - xmp:creatortool=Adobe InDesign 15.1 (Windows) - xmpmm:instanceid=uuid:6bbb46b4-59f0-4693-8d6c-4a3c2322168f - xmpmm:originaldocumentid=xmp.did:B2C59A163194E611BD009A7CB80B8EC7 - xmpmm:documentid=xmp.id:313188fe-94ed-0c43-bd86-8b5682fcfa67 - xmpmm:renditionclass=proof:pdf - xmpmm:derivedfrom=xmp.iid:9b6b2f22-512d-5540-943b-35151a675a45xmp.did:dedefd0f-a4e6-9643-b6a8-a7f3e6f9196bxmp.did:B2C59A163194E611BD009A7CB80B8EC7default - xmpmm:history=convertedfrom application/x-indesign to application/pdfAdobe InDesign 15.1 (Windows)/2026-06-17T09:06:18+05:30 - pdfx:jobid=c6465d2a-a3f0-4017-af77-afa6e7ecbea3 - pdf:producer=Adobe PDF Library 15.0 - pdf:trapped=False ## Contents ### Page 1 1VankerA, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721To cite: Vanker A, Brittain K, Barnett W, et al. Antenatal environmental exposure to indoor air pollution and environmental tobacco smoke: association with birth outcomes in an African birth cohort. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/ bmjresp-2025-003721► Additional supplemental material is published online only. To view, please visit the journal online (https://doi.org/10.1136/bmjresp-2025-003721). Received 2 January 2026 Accepted 30 April 2026 1Department of Paediatrics and Child Health and SAMRC Unit on Child and Adolescent Health, University of Cape Town, Red Cross War Memorial Children’s Hospital, Rondebosch, South Africa 2Department of Public Health, School of Global Health, Meharry Medical College, Meharry Medical College, Nashville, Tennessee, USACorrespondence toDr Aneesa Vanker;aneesa.vanker@uct.ac.zaAntenatal environmental exposure to indoor air pollution and environmental tobacco smoke: association with birth outcomes in an African birth cohortAneesa Vanker ,1 Kirsty Brittain,1 Whitney Barnett,1,2 Heather J Zar1Environmental exposure© Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY. Published by BMJ Group.ABSTRACTIndoor air pollution (IAP) and environmental tobacco smoke exposure (ETS) are important antenatal environmental exposures that may adversely impact birth outcomes.Objectives To investigate the impact of IAP and ETS during pregnancy on birth outcomes including respiratory distress in an African birth cohort, the Drakenstein Child Health Study.Setting Prospective cohort study conducted in a periurban setting in sub-Saharan Africa within a large birth cohort.Participants Consenting pregnant women (n=1137) enrolled between 20 and 28 weeks’ gestation, with 1143 live births included in the analysis. Participants were recruited antenatally, with inclusion based on residency in the study area and availability of exposure and outcome data.Primary and secondary outcome measures Primary exposures included antenatal IAP (particulate matter (PM10), carbon monoxide (CO), nitrogen dioxide, sulphur dioxide, volatile organic compounds) and ETS (maternal urine cotinine). Outcomes included birth anthropometry, prematurity, low birth weight and neonatal respiratory distress.Results Among 1143 live births, 17% were premature, 15% had low birth weight and 7% had respiratory distress. Maternal active smoking was associated with reduced weight- for- age z- scores (WfAz; β −0.54; p<0.001). Babies exposed to PM10 above ambient standards were 1.88 times more likely to experience respiratory distress compared with those with exposure below ambient standards (p=0.058). Similarly, exposure to CO above ambient standards was associated with a trend towards a doubled odds of respiratory distress in babies at birth (OR: 2.19 (0.97 to 4.98); p=0.061).Conclusions Antenatal exposure to IAP and ETS is associated with adverse birth outcomes, including impaired growth and increased risk of respiratory distress. These findings support the need for targeted public health interventions to reduce exposure during pregnancy and highlight the importance of improving air quality to mitigate long-term respiratory health risks.INTRODUCTIONAir pollution exposure is well recognised as a global health emergency,1 and indoor or household air pollution is especially problematic in many African countries as well as other low and middle-income countries (LMIC) globally, where access to clean fuels for household energy is limited.2 Further- more, the burden of tobacco smoke exposure in pregnant women remains a public health concern given the potential adverse impact on the health of both the mother and fetus and is often under-reported.3 The antenatal period is a vulnerable time with the developing fetus at risk from noxious environmental exposures. Increasing evidence describes the susceptibility of preg- nant women to air pollution with transpla- cental spread of pollutants and widespread distribution of black carbon and other pollut- ants in the organs of the unborn fetus.4 5 This is particularly important given that in utero exposure has been shown to impact not only birth outcomes but also longitudinal healthWHAT IS ALREADY KNOWN ON THIS TOPIC⇒ Air pollution and tobacco smoke exposure have lifelong health impacts starting in early life, but measured exposure data, especially from low and middle-income country African settings, are scarce and their effects on birth outcomes remain poorly described.WHAT THIS STUDY ADDS⇒ This study provides measured epidemiological data on household air pollution and tobacco smoke ex- posure in a peri-urban African context and demon- strates their association with key birth outcomes, including respiratory distress, an area with previ- ously limited evidence.HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY⇒ The findings highlight the urgent need for univer- sal access to clean energy and support targeted antenatal public health interventions to reduce ex- posure, with the goal of improving long-term health outcomes. ### Page 2 2VankerA, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721Open accesstrajectories.6 7 While the impact of tobacco smoke expo- sure on foetal development and birth outcomes has been well described,8 9 the impact of both environmental tobacco smoke (ETS) and indoor air pollution (IAP) exposure on birth outcomes remains limited, particu- larly in LMIC and African settings. Recent studies from Africa have focused on particulate matter (PM2.5) or carbon monoxide (CO) exposure and their impact on birth outcomes, in particular, birth weight, prematurity and perinatal mortality.10–13 However, few studies have compared the impact of multiple individual household pollutants, nor included explorations of the effect of tobacco smoking and exposure on birth outcomes.14 Many pollutants contribute to IAP including particu- late matter (PM), CO and volatile organic compounds (VOC), benzene and toluene. Tobacco smoke is an important contributor to IAP, especially in areas where smoking prevalence is high and household ventilation poor.15 16 As epidemiological data on IAP and ETS in Africa are limited with many studies still relying on self-reported exposure,17 comprehensively assessing multiple types and sources of IAP, including ETS, is important to identify key drivers of birth outcomes to appropriately inform public health interventions. Therefore, we aimed to investigate the impact of multiple sources of IAP and ETS on birth outcomes in a South African birth cohort study, the Drakenstein Child Health Study (DCHS).18METHODSA prospective study of antenatal exposure to IAP and ETS in the DCHS was done. Pregnant women in a peri-urban area were enrolled at 20–28 weeks’ gestation at two primary healthcare clinics, Newman (serving a predominantly mixed-ancestry population) and Mbek- weni (predominantly black African population), using convenience sampling from 1 March 2012 to 31 March 2015, to ensure enrolment across different time periods and seasons. Written informed consent, in the partici- pants’ home language (English, Afrikaans or isiXhosa), was obtained at enrolment. Antenatal care was provided free of charge at these public sector clinics; all births or any intercurrent hospitalisations occurred at the single central public hospital, Paarl Hospital. Participant demo- graphics were collected at enrolment, objective measures of IAP and ETS were collected at 28–32 weeks’ gestation, and birth data along with a second measure of ETS were collected at delivery or abstracted from hospital folders. Gestational age (GA) at delivery (in weeks) was calculated from an antenatal ultrasound in the second trimester; if this was not available, then fundal height at enrolment or maternal recall of last menstrual period was used.Assessing environmental exposuresMeasuring IAP exposureAn antenatal home visit, conducted by trained field- workers, between 28 and 32 weeks’ gestation was done to assess the home environment and to measure IAP using devices placed within the home. Data gathered included access to electricity, types of alternate fuels used for household energy, household size and position of dwelling in relation to passing traffic.16 Dwellings were categorised using six factors: type of home (formal vs informal), primary building material (brick or cement vs other materials), water supply (piped into dwelling or yard), toilet facilities (non-communal flush), kitchen type (separate room in house) and ventilation in the kitchen area (pipe or duct to exterior).16 A home lacking two or more of these dimensions was categorised as a poor struc- ture.19 20 To assess IAP, the most common pollutants and by-prod- ucts of combustion were measured. On consultation with air quality specialists (SGSR environmental), personal samplers were used based on available, validated equip- ment. Devices were placed in participants’ homes, with all measurements done in the communal/main living room, away from windows and doors, approximately 1.5 m from the ground, as previously described.16 Devices were left in the home for 24 hours to 2 weeks depending on the measurement.16 Particulate matter (PM10) was measured using a personal air sampling pump [SKC AirChek 52R] and CO with an AltairR carbon monoxide single gas detec- tion unit, left in homes for 24 hours.16 Diffusion tubes placed in homes for 2 weeks measured nitrogen dioxide (NO2)/sulphur dioxide (SO2; RadielloR adsorbent filters in polyethylene diffusive body) and VOC, benzene and toluene (MarkesR thermal desorption tubes).16 An average concentration based on the 2-week duration in the home was obtained for NO2/SO2 and VOC; 24-hour averages were obtained for PM10. CO data were down- loaded to a computer and the frequency of exceedance above the hourly ambient standard was calculated.16 The South African National Ambient Air Quality Standards21 were used to define expected exposure levels for each pollutant based on an averaging period of 1 year for each measure; PM10: 40 μg/m3, NO2: 40 μg/m3, SO2: 50 μg/m3, benzene: 5 μg/m3, toluene: 240 μg/m3; and based on a 1-hour average for CO: >30 mg/m3 (not more than 88 hours).21Measuring ETS exposureAs a metabolite of nicotine and a robust proxy for maternal smoking or exposure to tobacco smoke, maternal urine cotinine was measured at the second antenatal visit (28–32 weeks’ gestation) and again at birth; the highest result was used to assign smoking status.22 Urine cotinine tests were performed using IMMULITER 1000 Nicotine Metabolite Kit (Siemens Medical Solutions Diagnos- ticsR, Glyn Rhonwy, United Kingdom). This provided a quantitative test using a competitive chemilumines- cent immunoassay, which contained solid-phase beads coated with polyclonal rabbit anticotinine antibody. The test had a calibration range of 10–500 μg/mL, with an analytical sensitivity of 2 ng/mL.23 Urine cotinine levels were classified as <10 ng/mL (non-smoker), 10–499 ng/ ### Page 3 Vanker A, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721 3Open accessmL (passive smoker/exposed), or ≥500 ng/mL (active smoker) according to the manufacturer’s directions and the testing laboratory’s limits.22Birth outcomesAll births were attended by a member of the study team. Birth outcomes included birth weight, prematurity (<37 weeks GA) and presence of respiratory distress. birth weight and length were measured in the labour ward by trained clinical staff as part of routine care and obtained from hospital records. Infant weight was measured to the nearest 10 g and length to the nearest completed 0.5 cm according to standard operating procedures.24 GA at birth was estimated based on a second trimester ante- natal ultrasound where available (68%); if ultrasound was unavailable, GA was estimated using fundal height (28%) or maternal recall of the last menstrual period (4%).24 In a subgroup of participants, labour ward anthropometric measurements were compared with study staff meas- urements and no significant differences were observed, supporting measurement reliability. Weight- for- age z-scores (WfAz) at birth were calculated using the WHO standards for full-term infants and the Fenton growth charts for preterm infants.25 26 Respira- tory distress was defined as a documented respiratory problem (grunting, fast breathing, recession, cyanosis or apnoea) or receipt of supplemental oxygen or respira- tory support at birth.Statistical analysisData were analysed using STATA V.14.2 (Stata Corpo- ration, College Station, Texas). In descriptive analyses, maternal, household and birth characteristics were summarised across study site, and IAP and maternal urine cotinine were summarised using the ambient standards and urine cotinine thresholds described above. The effects of IAP and maternal urine cotinine on WfAz at birth were explored in linear regression models, and the effects on preterm birth and respira- tory distress after birth were explored in logistic regres- sion models. Throughout, a two-sided p value of <0.05 was considered to be statistically significant, and results are presented with 95% CIs. Confounders were selected a priori based on directed acyclic graphs (online supple- mental figures S1 and S2) and included maternal educa- tion and household income. In sensitivity analyses, models were additionally adjusted for maternal age at enrolment and maternal HIV infection (online supple- mental table S1). Sensitivity analyses were conducted to assess the robust- ness of the findings. These included models stratified by study site (online supplemental table S2) and analyses examining associations between maternal smoking and birth outcomes using alternative cotinine-based smoking classifications (online supplemental table S3).Patient and public participationWhile patients or the public were not directly involved in the design or conduct of the research, local community members played an important role in its implementa- tion. Fieldworkers were recruited from within the study communities to support data collection, which enhanced trust and cultural relevance. In addition, community engagement activities were carried out to assess the acceptability of home visits and the placement of meas- uring equipment in households.RESULTSOf 1225 mothers enrolled, 67 (5.5%) were lost to follow- up prior to delivery; 9 (0.7%) experienced miscarriages; and 12 (1.0%) experienced stillbirths, resulting in 1143 live births among 1137 mothers (figure 1). Median maternal age was 25.8 (IQR 22.0– 30.8) years and 392 (35%) were primigravid, with 244 (21%) HIV- infected. Almost all HIV- infected women (97%) were taking antiretroviral therapy. Levels of education were low, with 39% of mothers having completed secondary or any tertiary educa- tion (n=445); 51% of households had at least one parent employed (n=578); and most households had an average monthly income of less than ZAR 5000 (US$260). Gestational complications were diagnosed in a minority of women, including hypertension in 52 women (5%), pre- eclampsia in 29 (3%) and asthma in 17 (2%; table 1).Home environment, IAP and ETS exposuresA total of 884 home visits were conducted antena- tally. From the home visits conducted, the majority 66% (n=553) were formal structures, but 278 (33%) had two or fewer of the basic household dimensions. More Mbekweni homes (39%) had two or fewer basic household dimensions compared with Newman (27%). Although most homes had access to elec- tricity (93%), fossil fuels were still used for cooking (21%) or heating (28%), with more Mbekweni than Newman homes using fossil fuels for cooking or heating (table 1). Of the 884 home visits completed antenatally, IAP data were available for PM10 levels in a total of 764 homes, benzene in 735, CO in 714, and NO2 and SO2 in 756 (see figure 1). Incomplete IAP data primarily reflected equipment stock limitations, device loss or interference in households or samples not passing laboratory quality control checks. Almost half of homes had measured PM10 (45%) and benzene (45%) levels above acceptable ambient standards with no significant differences between sites. For CO levels, 11% of homes were above acceptable ambient stan- dards. However, for NO2 and SO2, the proportion of homes with levels above ambient standards was very low (3% and 0.5%, respectively), and these pollutants ### Page 4 4VankerA, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721Open accesswere, therefore, not included in the analyses of the impacts of IAP on birth outcomes (table 2). Overall, 1088 mothers had urine cotinine measure- ments (493 from Newman and 595 from Mbekweni). Cotinine measurements were obtained during in- person maternal visits, and missing data were primarily due to missed study visits rather than labora- tory processing issues. Based on these measurements, maternal active tobacco smoking prevalence was 32% overall, much higher for Newman mothers compared with Mbekweni mothers (53% vs 15%). A further 45% of mothers were classified as passive smokers, with aFigure 1 Study flow diagram. CO, Carbon monoxide; IAP, indoor air pollution; PM10, particulate matter. ### Page 5 Vanker A, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721 5Open accessTable 1 Maternal and household characteristics by study siteCharacteristics Total Mbekweni NewmanNumber of mothers 1137 628 509 Median (IQR) age at enrolment 25.8 (22.0, 30.8) 26.7 (22.3, 31.6) 24.7 (21.4, 29.2) Married/cohabiting 458 (40%) 237 (38%) 221 (43%) Primigravid 392 (35%) 200 (32%) 192 (38%) EducationPrimary/some secondary 692 (61%) 389 (62%) 303 (60%)Completed secondary/any tertiary 445 (39%) 239 (38%) 206 (40%)Parent employed 578 (51%) 300 (48%) 278 (55%) Average household income (ZAR)R5000/month 159 (14%) 68 (11%) 91 (18%) Household density—median (IQR)Household size 4 (3, 6) 4 (3, 6) 5 (4, 7) Persons per room (n=834) 2 (1, 2) 2 (1, 3) 1 (1, 2)Persons per sleeping room (n=830) 3 (2, 4) 3 (2, 4) 3 (2, 5) Household dimensionsType of home: formal (n=836) 553 (66%) 270 (61%) 283 (72%) Material: brick/cement (n=832) 548 (66%) 270 (61%) 278 (71%)Water supply: piped (n=825) 721 (87%) 378 (86%) 343 (89%)Toilet: non- communal flush (n=832) 597 (72%) 327 (74%) 270 (69%)Kitchen type: separate room (n=821) 359 (44%) 87 (20%) 272 (72%)Kitchen ventilation: pipe/duct (n=821) 37 (5%) 37 (8%) 0 (0%) Household dimension categories (n=836)≤2 dimensions 278 (33%) 172 (39%) 106 (27%)>2 dimensions 558 (67%) 271 (61%) 287 (73%)Electricity in the home (n=834) 773 (93%) 393 (89%) 380 (97%) Fossil fuel (coal, wood, paraffin, gas) usedCooking (n=816) 175 (21%) 138 (32%) 37 (10%)Heating (n=524) 148 (28%) 140 (35%) 8 (6%) Type of stove (n=816)Electric 734 (90%) 378 (86%) 356 (94%)Paraffin 127 (16%) 123 (28%) 4 (1%)Gas 55 (7%) 23 (5%) 32 (8%)Wood 3 (0.4%) 2 (0.5%) 1 (0.3%)Coal 2 (0.3%) 2 (0.5%) 0 (0%) Distance of home from continuously passing trucks (n=833) Passing trucks (n=833)<50 m 380 (46%) 287 (65%) 93 (24%)50–100 m 129 (15%) 94 (21%) 35 (9%)100–200 m 91 (11%) 56 (13%) 35 (9%)200–500 m 95 (11%) 3 (0.7%) 92 (23%)Continued ### Page 6 6VankerA, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721Open accesshigher prevalence in Mbekweni mothers compared with Newman mothers (52% vs 36%; table 2).Birth outcomesThere were 1143 live births, 53% male, with median GA 39 (IQR 37, 40) weeks and 17% (189) were preterm (<37 weeks gestation), of which 128 (68%) were late preterm (≥34 weeks gestation). Vaginal deliveries occurred in 906/1136 births (80%). The median birth weight was 3080 (IQR 2710–3420) g, with Mbekweni babies weighing more than Newman babies (median 3150 vs 3000). The median WfAz was −0.25 (IQR −0.93–0.40). RespiratoryCharacteristics Total Mbekweni Newman>500 m 138 (17%) 1 (0.2%) 137 (35%)HIV- infected 244 (21%) 228 (36%) 16 (3%)On antiretroviral therapy (n=244) 237 (97%) 221 (97%) 16 (100%) Maternal health complications diagnosed during pregnancyDiabetes 14 (1%) 7 (1%) 7 (1%)Preeclampsia 29 (3%) 15 (2%) 14 (3%)Hypertension 52 (5%) 30 (5%) 22 (4%)Asthma 17 (2%) 7 (1%) 10 (2%)Anaemia 250 (22%) 161 (26%) 89 (17%) Data presented as n (%) of mothers or median with interquartile range (IQR). ZAR, South African rand.Table 1 ContinuedTable 2 Antenatal indoor air pollution and environmental tobacco smoke (maternal urine cotinine) by study siteMeasure Total Mbekweni NewmanNumber of mothers 1137 628 509 Particulate matter (PM10) g/m3 (n=764) 33.4 (12.4, 64.8) 31.9 (12.3, 62.5) 35.7 (12.8, 66.1) PM10 above ambient standards 344 (45%) 174 (43%) 170 (47%) Nitrogen dioxide (NO2) μg/m3 (n=756) 7.1 (3.3, 12.7) 6.9 (2.5, 14.6) 7.2 (3.8, 11.3) NO2 above ambient standards 19 (3%) 16 (4%) 3 (0.8%) Sulphur dioxide (SO2) μg/m3 (n=756) 0.0 (0.0, 0.3) 0.0 (0.0, 0.3) 0.0 (0.0, 0.2) SO2 above ambient standards 4 (0.5%) 4 (1%) 0 (0%) Carbon monoxide (CO) mg/m3 (n=714) 0(0, 110) 0(0, 65) 0(0, 120) CO above ambient standards 80 (11%) 39 (10%) 41 (13%) Benzene μg/m3 (n=735) 4.3 (1.8, 11.4) 4.5 (1.5, 17.7) 3.9 (1.8, 8.6) Benzene above ambient standards 333 (45%) 184 (47%) 149 (43%) Toluene μg/m3 (n=735) 16.8 (7.1, 44.3) 16.0 (5.9, 42.9) 17.4 (8.2, 46.0) Toluene above ambient standards 67 (9%) 36 (9%) 31 (9%) Maternal urine cotinine (n=1088)Non- smoker 249 (23%) 196 (33%) 53 (11%)Passive smoker/exposed 486 (45%) 308 (52%) 178 (36%)Active smoker 353 (32%) 91 (15%) 262 (53%) Data presented as n (%) of mothers or median with interquartile range (IQR). South African National Ambient Air Quality Standards based on an averaging period of 1 year: PM10: 40 μg/m3, NO2: 40 μg/m3, SO2: 50 μg/m3, benzene: 5 μg/m3, toluene: 240 μg/m3; and based on a 1 hour average for CO: >30 mg/m3 (not more than 88 hours). Maternal urine cotinine as a proxy for environmental tobacco smoke: <10 ng/mL (non-smoker), 10-499 ng/mL (passive smoker/exposed) or ≥500 ng/mL (active smoker). ### Page 7 Vanker A, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721 7Open accessdistress was found in 76/1143 (7%) babies at birth, with no differences across site. Most babies with respira- tory distress (58/76; 76%) required hospital admission (table 3).Associations between IAP, maternal urine cotinine and birth outcomesNo associations were observed between IAP and WfA z- scores at birth in either unadjusted models or models adjusted for maternal education and household income (table 4). However, high maternal urine cotinine levels (indicating active smoking) were significantly associ- ated with a decrease in WfA z-score after adjustment for maternal education and household income (β coeffi- cient: −0.54; 95% CI −0.71 to −0.37; p<0.001). Neither IAP nor maternal urine cotinine were significantly associated with preterm birth in unadjusted analyses or after adjust- ment for maternal education and household income. Exposure to PM10 or CO above ambient standards was associated with an increased odds of respiratory distress at birth, although these associations did not reach statis- tical significance. After adjustment for maternal educa- tion and household income, babies exposed to PM10 above ambient standards were 1.88 times more likely to experience respiratory distress compared with those with exposure below ambient standards (95% CI 0.98 to 3.61; p=0.058). Similarly, exposure to CO above ambient stand- ards more than doubled the odds of respiratory distress in babies at birth (OR: 2.19; 95% CI 0.97 to 4.98; p=0.061). Results were unchanged after additional adjustment for maternal age at enrolment and maternal HIV infection (online supplemental table S1). In sensitivity analyses stratified by study site, associations between environmental exposures and birth outcomes were broadly consistent with those observed in the pooled analyses (online supplemental table S2). Additional sensi- tivity analyses using alternative categorisations of maternal smoking based on urine cotinine also produced similar findings (online supplemental table S3).DISCUSSIONIn this study, in a poor peri-urban South African population, we found high levels of maternal urine cotinine, a proxy for ETS, and of exposure to IAP especially to PM10 and benzene, with almost half of all households being exposed to levels above ambient standards. Maternal urine cotinine above the threshold indicating active smoking was associ- ated with lower weight-for-age in infants at birth, consistent with prior studies.22 27 28 We also found that higher ante- natal exposure to PM10 and CO trended towards higher odds of respiratory distress. Systematic reviews have shown associations between household air pollution and birth outcomes, including birth weight, prematurity and neonatal mortality; however, note that the majority of studies rely on self-reported air pollution exposures or modelled IAP exposure and many studies use maternal recall of birth outcomes.10 29 30 More recently, studies have assessed the association between particulate matter (PM2.5) or CO from household air pollution and the association with birth outcomes and showed a tendency towards prematurity and low birth weight;11 13 however, few studies measure both IAP and ETS exposure directly nor do they include respiratory distress as an outcome.31 In this study, key pollutants, the result of unclean fuel combustion, were measured in over 700 homes and maternal urine cotinine was quantified in more than 1000 pregnant women, providing robust exposureTable 3 Birth outcomes by study siteMeasure Total Mbekweni NewmanNumber of infants 1143 634 509 Sets of twins 4 4 0 Sets of triplets 1 1 0 Sex—male 586 (51%) 309 (49%) 277 (54%) Median (IQR) gestation at delivery 39 (37, 40) 39 (38, 40) 39 (37, 40) Pre-term birth (<37 weeks gestation) 189 (17%) 106 (17%) 83 (16%)Median (IQR) gestation at delivery among pre-term births (n=189) 35 (33, 36) 35 (33, 36) 35 (32, 36)Vaginal delivery (n=1136) 906 (80%) 481 (77%) 425 (84%)Median (IQR) birth weight in grams (n=1138) 3080 (2710, 3420) 3150 (2780, 3450) 3000 (2630, 3350)Median (IQR) weight- for- age z- score (n=1138) −0.25 (−0.93, 0.40) −0.13 (−0.80, 0.49) −0.43 (−1.09, 0.23)Low birth weight (<2500 grams; n=1138) 176 (15%) 81 (13%) 95 (19%) Respiratory problems at birth 76 (7%) 43 (7%) 33 (6%) Hospitalised with respiratory problems 58 (5%) 31 (5%) 27 (5%) Data presented as n (%) of infants or median with interquartile range (IQR). ### Page 8 8VankerA, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721Open accessdata. Furthermore, all births were attended by a member of the study team and birth outcomes were clearly docu- mented at the regional hospital by the healthcare providers ensuring accurate outcome assessment. Therefore, the present study adds to this existing literature by examining associations between objectively measured IAP and ETS and rigorously collected birth outcome data. The association seen between PM10 and CO exposure above acceptable ambient standards and respiratory distress at birth is an important finding. While these asso- ciations did not reach statistical significance, these findings support the growing body of evidence that antenatal envi- ronmental exposures drive in-utero inflammatory and chemotaxic process significantly impacting not only early-life outcomes but also potentially life-long lung health.32–34 This emphasises the urgent need to reduce antenatal air pollution exposure and to prioritise public health interven- tions that may mitigate these risks. The sources of CO exposure may be both from fuel combustion and a by-product of cigarette smoke.35 36 Although we did not find an association between measured household CO levels and maternal smoking, this may beTable 4 Impact of indoor air pollution above ambient standards and environmental tobacco smoke (maternal urine cotinine) on infant birth outcomes (A) Weight- for-age z- score at birth, among 1138 babies with birthweight datan Unadjusted model Adjusted modelUnadjusted β (95% CI) P value Adjusted β (95% CI) P value Particulate matter (PM10) 766 −0.08 (−0.22 to 0.06) 0.270 −0.06 (−0.20 to 0.08) 0.414 Carbon monoxide (CO) 716 −0.07 (−0.30 to 0.16) 0.553 −0.08 (−0.31 to 0.15) 0.491 Benzene 737 −0.05 (−0.19 to 0.10) 0.523 −0.03 (−0.17 to 0.12) 0.715 Toluene 737 −0.13 (−0.38 to 0.12) 0.317 −0.13 (−0.38 to 0.12) 0.323 Maternal urine cotinine (vs non- smoker, n=249)Passive smoker/exposed 490 −0.10 [−0.26, 0.05] 0.196 −0.10 [−0.26, 0.05] 0.204Active smoker 353 −0.55 [−0.72, -0.39] <0.001 −0.54 [−0.71, -0.37] <0.001(B) Preterm vs full-term birth, among 1143 babiesn Unadjusted model Adjusted modelUnadjusted OR (95% CI) P value Adjusted OR (95% CI) P value PM10 767 1.07 (0.70 to 1.64) 0.765 1.05 (0.68 to 1.61) 0.836 CO 716 1.45 (0.76 to 2.76) 0.255 1.41 (0.74 to 2.69) 0.304 Benzene 738 1.18 (0.75 to 1.85) 0.478 1.14 (0.72 to 1.79) 0.578 Toluene 738 1.03 (0.47 to 2.24) 0.939 1.02 (0.47 to 2.22) 0.957 Maternal urine cotinine (vs non- smoker, n=249)Passive smoker/exposed 491 1.09 [0.70, 1.68] 0.713 1.03 (0.66, 1.60) 0.905Active smoker 353 1.33 (0.85, 2.08) 0.216 1.20 (0.75, 1.90) 0.444(C) Respiratory distress vs no respiratory distress at birth, among 1143 babiesn Unadjusted model Adjusted modelUnadjusted OR (95% CI) P value Adjusted OR (95% CI) P value PM10 767 1.90 (0.99 to 3.63) 0.053 1.88 (0.98 to 3.61) 0.058 CO 716 2.10 (0.93 to 4.73) 0.074 2.19 (0.97 to 4.98) 0.061 Benzene 738 1.61 (0.84 to 3.08) 0.154 1.63 (0.85 to 3.14) 0.143 Toluene 738 1.15 (0.40 to 3.35) 0.793 1.14 (0.39 to 3.31) 0.811 Maternal urine cotinine (vs non- smoker, n=249)Passive smoker/exposed 491 0.62 (0.33 to 1.16) 0.135 0.63 (0.34 to 1.19) 0.155Active smoker 353 1.04 (0.57 to 1.91) 0.892 1.04 (0.55 to 1.95) 0.904β: β coefficient from linear regression model. Adjusted models: adjusted for maternal education and household income. South African National Ambient Air Quality Standards based on an averaging period of 1 year: PM10: 40 μg/m3, benzene: 5 μg/m3, toluene: 240 μg/m3; and based on a 1 hour average for CO: >30 mg/m3 (not more than 88 hours). Maternal urine cotinine as a proxy for environmental tobacco smoke: <10 ng/mL (non-smoker), 10–499 ng/mL (passive smoker/exposed) or ≥500 ng/mL (active smoker). ### Page 9 Vanker A, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721 9Open accessthe result of multiple household sources of CO. The impact of cigarette smoking on birth weight is still important with potential far-reaching health consequences, including impaired growth and long-term lung health.37 We have already shown that antenatal IAP and ETS exposures impact lung function38 and are associated with lower respi- ratory tract infections in infancy.39 The findings from this study further highlight that these antenatal exposures are an important determinant of birth outcomes. Understanding the social determinants of health is complex40 and while this was not the primary aim of this study, we found several notable differences in antenatal exposures between the two sites with likely consequences for early-life child health. The distinct sociodemographic and environmental profiles of these two communities provide important context for interpreting exposure patterns in this cohort. There was a higher reliance on alternate fossil fuel sources in the Mbekweni community, potentially a reflection of the slightly lower socioeconomic status of this community. Additionally, there were higher maternal urine cotinine levels in the Newman community, likely due to greater social acceptance of women smoking in this community.22 Interestingly, despite these differences, measured IAP exposures were similar across both sites. Although this study was limited to a single antenatal IAP measurement in the last trimester of pregnancy, the findings remain important given the limited African data on directly measured air pollution exposures.29 Future studies should consider measuring IAP at different timepoints throughout pregnancy, and even preconception to improve understanding of exposure timing and outcomes. Furthermore, IAP and ETS have been shown to be associated with stillbirth and perinatal mortality; in this study, the small numbers of miscar- riages and still births precluded this analysis. Another important limitation is that PM10 and not PM2.5 was measured, due to the availability of personal sampling equipment at the time. While PM2.5 is recognised to penetrate deeper into the respiratory system and systemic circulation, we showed high numbers of households with levels above ambient standards for PM10, which is likely to also reflect exposure to PM2.5.11 The associations (though not reaching statistical signif- icance) seen with PM10 exposure are still important and should be considered in strategies to reduce air pollution exposures. Additionally, we were limited by the types of pollutants that could be measured. Form- aldehyde, released from household cleaning products and building materials, is a major IAP that has also been implicated in adverse birth outcomes and longer term health.41 However, this was not included in the panel of VOCs measured. Other important pollut- ants must be included in future studies. Furthermore, future studies with larger samples and more complete exposure measurements should examine the poten- tial joint or interactive effects of IAP and ETS during pregnancy, as these exposures frequently co-occur in many settings. The small numbers of children born premature or with respiratory distress limit the power to detect differences; however, this reflects good ante- natal care and follow-up in a public health setting. In addition, although GA was primarily estimated using second trimester ultrasound, approximately 28% of participants had GA estimated using fundal height, which may introduce some measurement imprecision. Finally, urine cotinine measurements were classified according to the manufacturer’s directions, but we recognise the potential for misclassification of smoking status, especially in the ‘passive smoking’ category. The strengths of this study include (1) the large number of homes visited to obtain objective IAP measurements antenatally; (2) the recruitment period over 3 years allowing for measurements to be collected across seasons and (3) conducting home visits in participant homes removed bias from self-reporting exposures. In addition, high cohort reten- tion through pregnancy and careful phenotyping of infants at birth are key strengths.CONCLUSIONHigh exposure to IAP and ETS in this study high- lights the need for improvement in environmental conditions and greater use of clean fuels in LMICs. The association between maternal urine cotinine and lower weight- for- age at birth underscores the urgent need for effective programmes to address maternal smoking preconception and during pregnancy and to strengthen education about the dangers of IAP expo- sure to the developing infant. Future work should address potential long- term implications of early- life exposures on developmental trajectories of health.Contributors AV conceived the manuscript and was involved in the study design, data collection, reviewing data and manuscript writing. KB was involved in the study design and was the statistician on the manuscript completing all analyses. WB was involved in the study design, data collection, manuscript preparation and reviewing data. HJZ was involved in the study design, reviewing and manuscript writing. All authors contribute to writing, revising and editing the manuscript. AV is the author guarantor.Funding This study was funded by the Bill and Melinda Gates Foundation (OPP1017641), the Discovery Foundation (19997/1), a South African Thoracic Society AstraZeneca Respiratory Fellowship, the National Research Foundation (South Africa) (1233905), a CIDRI (Wellcome Centre for Infectious Diseases Research in Africa) Clinical Fellowship (084323), the SA Medical Research Council. AV was supported by the SAMRC Clinician Postdoctoral Career Development Award Programme, division of Research Capacity Development. We thank the study staff, the clinical and administrative staff of the Western Cape Government Health Department at Paarl Hospital and at the clinics for support of the study. Thank you to SGS Environmental Services for supporting this project. We thank the participants and their families.Competing interests None declared.Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.Patient consent for publication Not applicable.Ethics approval This study involves human participants and was approved by Faculty of Health Sciences Human Research Ethics Committees of the University of Cape Town (149/2013) and of Stellenbosch University, and by the Western Cape Provincial Health Research Committee. Participants gave informed consent to participate in the study before taking part.Provenance and peer review Part of a topic collection; not commissioned; externally peer- reviewed. ### Page 10 10VankerA, et al. BMJ Open Respir Res 2026;13:e003721. doi:10.1136/bmjresp-2025-003721Open accessData availability statement Data are available upon reasonable request. The de- identified data that support the findings of this study are available from the corresponding author upon reasonable request as per DCHS cohort guidelines.Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.Open access This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.ORCID iDAneesa Vanker https://orcid.org/0000-0001-7396-5506REFERENCES1 World Health Organization. WHO global air quality guidelines: particulate matter (pm2.5 and pm10), ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide. Geneva:World Health Organization; 2021.2 Health Effects institute. The state of air quality and health impacts in africa. a report from the state of global air initiative. Boston, MA Health Effects institute; 2022.3 Öberg M, Jaakkola MS, Woodward A, et al. Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. The Lancet 2011;377:139–46.4 Bongaerts E, Lecante LL, Bové H, et al. Maternal exposure to ambient black carbon particles and their presence in maternal and fetal circulation and organs: an analysis of two independent population-based observational studies. The Lancet Planet Health2022;6:e804–11.5 Bové H, Bongaerts E, Slenders E, et al. Ambient black carbon particles reach the fetal side of human placenta. Nat Commun2019;10:3866.6 Sly PD, Bush A. From the Cradle to the Grave: The Early-Life Origins of Chronic Obstructive Pulmonary Disease. Am J Respir Crit Care Med 2016;193:1–2.7 Rylance S, Masekela R, Banda NPK, et al. Determinants of lung health across the life course in sub-Saharan Africa. Int J Tuberc Lung Dis 2020;24:892–901.8 Fazel N, Kundi M, Kazemzadeh A, et al. Environmental tobacco smoke exposure during pregnancy affects complications and birth outcomes in women with and without asthma. BMC Pregnancy Childbirth 2020;20:314.9 Braun M, Klingelhöfer D, Oremek GM, et al. Influence of Second-Hand Smoke and Prenatal Tobacco Smoke Exposure on Biomarkers, Genetics and Physiological Processes in Children-An Overview in Research Insights of the Last Few Years. Int J Environ Res Public Health 2020;17:3212.10 Flanagan E, Oudin A, Walles J, et al. Ambient and indoor air pollution exposure and adverse birth outcomes in Adama, Ethiopia. Environ Int 2022;164:107251.11 Shezi B, Jafta N, Asharam K, et al. Maternal exposure to indoor PM2.5 and associated adverse birth outcomes in low socio-economic households, Durban, South Africa. Indoor Air 2022;32:e12934.12 Bachwenkizi J, Liu C, Meng X, et al. Maternal exposure to fine particulate matter and preterm birth and low birth weight in Africa.Environ Int 2022;160:107053.13 Quinn AK, Adjei IA, Ae- Ngibise KA, et al. Prenatal household air pollutant exposure is associated with reduced size and gestational age at birth among a cohort of Ghanaian infants. Environ Int2021;155:106659.14 Demelash Enyew H, Bogale BG, Hailu AB, et al. Environmental exposures and adverse pregnancy outcomes in Ethiopia: A systematic review and meta-analysis. PLoS One 2023;18:e0288240.15 Gordon SB, Bruce NG, Grigg J, et al. Respiratory risks from household air pollution in low and middle income countries. Lancet Respir Med 2014;2:823–60.16 Vanker A, Barnett W, Nduru PM, et al. Home environment and indoor air pollution exposure in an African birth cohort study. Sci Total Environ 2015;536:362–7.17 Tonne C. A call for epidemiology where the air pollution is. The Lancet Planet Health 2017;1:e355–6.18 Zar HJ, Barnett W, Myer L, et al. Investigating the early-life determinants of illness in Africa: the Drakenstein Child Health Study.Thorax 2015;70:592–4.19 Alkire S, Foster J. Understandings and misunderstandings of multidimensional poverty measurement. J Econ Inequal2011;9:289–314.20 Alkire S, Foster J. Counting and multidimensional poverty measurement. J Public Econ 2011;95:476–87.21 Government Gazette Republic of South Africa. National ambient air quality standards pretoria 2009. 2009 Available: https://www.environment.gov.za/sites/default/files/legislations/nemaqa_airquality_g32816gon1210.pdf22 Vanker A, Barnett W, Brittain K, et al. Antenatal and early life tobacco smoke exposure in an African birth cohort study. Int J Tuberc Lung Dis 2016;20:729–37.23 Siemens. Immulite 1000 nicotine metabolite. 2009 Available: http:// www.healthcare.siemens.com/immunoassay/systems/immulite-1000-im24 Budree S, Stein DJ, Brittain K, et al. Maternal and infant factors had a significant impact on birthweight and longitudinal growth in a South African birth cohort. Acta Paediatr 2017;106:1793–801.25 Fenton TR, Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr2013;13:59.26 Fenton TR, Nasser R, Eliasziw M, et al. Validating the weight gain of preterm infants between the reference growth curve of the fetus and the term infant. BMC Pediatr 2013;13:92.27 Leonardi- Bee J, Smyth A, Britton J, et al. Environmental tobacco smoke and fetal health: systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed 2008;93:F351–61.28 Salmasi G, Grady R, Jones J, et al. Environmental tobacco smoke exposure and perinatal outcomes: a systematic review and meta-analyses. Acta Obstet Gynecol Scand 2010;89:423–41.29 Younger A, Alkon A, Harknett K, et al. Adverse birth outcomes associated with household air pollution from unclean cooking fuels in low- and middle- income countries: A systematic review. Environ Res 2022;204:112274.30 Kadir MM, McClure EM, Goudar SS, et al. Exposure of pregnant women to indoor air pollution: a study from nine low and middle income countries. Acta Obstet Gynecol Scand 2010;89:540–8.31 Shah PS, Balkhair T. Air pollution and birth outcomes: a systematic review. Environ Int 2011;37:498–516.32 Latzin P, Röösli M, Huss A, et al. Air pollution during pregnancy and lung function in newborns: a birth cohort study. Eur Respir J2009;33:594–603.33 Korten I, Ramsey K, Latzin P. Air pollution during pregnancy and lung development in the child. Paediatr Respir Rev 2017;21:38–46.34 Veras MM, de Oliveira Alves N, Fajersztajn L, et al. Before the first breath: prenatal exposures to air pollution and lung development.Cell Tissue Res 2017;367:445–55.35 Dionisio KL, Howie SRC, Dominici F, et al. The exposure of infants and children to carbon monoxide from biomass fuels in The Gambia: a measurement and modeling study. J Expo Sci Environ Epidemiol2012;22:173–81.36 Reynolds CME, Egan B, Kennedy RA, et al. The implications of high carbon monoxide levels in early pregnancy for neonatal outcomes.European Journal of Obstetrics & Gynecology and Reproductive Biology 2019;233:6–11.37 Holgate S, Grigg J, Agius R, et al. Every breath we take: the lifelong impact of air pollution, report of a working party2016: royal college of physicians.38 Gray D, Willemse L, Visagie A, et al. Determinants of early-life lung function in African infants. Thorax 2017;72:445–50.39 Vanker A, Barnett W, Workman L, et al. Early-life exposure to indoor air pollution or tobacco smoke and lower respiratory tract illness and wheezing in African infants: a longitudinal birth cohort study. The Lancet Planet Health 2017;1:e328–36.40 Solar O, Irwin A. A conceptual framework for action on the social determinants of health. In: Social Determinants of Health Discussion Paper 2 (Policy and Practice). Geneva, Switzerland: World Health Organisation, 2010.41 Franklin P, Tan M, Hemy N, et al. Maternal Exposure to Indoor Air Pollution and Birth Outcomes. Int J Environ Res Public Health2019;16:1364.