Association of active and passive smoking with adolescent asthma: a systematic review and meta-analysis

Abstract

Objective

To synthesize evidence on the association of active or passive smoking with asthma in adolescents.

Data Sources

A comprehensive search of Embase, PubMed, Scopus, and Web of Science were conducted from inception to January 2026. Only human studies published in English were included.

Study Selection

Observational studies evaluating active or passive smoking in adolescents aged 10–19 years were eligible. Studies reporting odds ratio (OR), hazard ratio (HR), relative risk (RR), prevalence ratio (PR), incidence rate ratio (IRR), or prevalence odds ratio (POR) with 95% confidence interval (95% CI) were included. Two reviewers independently screened studies.

Data Extraction

Key study characteristics, including authors, year, setting, age range, design, sample size, exposure type, asthma outcome, adjusted covariates, and effect estimates, were independently extracted and cross-verified by two reviewers, with discrepancies resolved by consensus.

Data Synthesis

Seventy-seven studies met inclusion criteria, with 72 contributing to meta-analysis. Active smoking (adjusted odds ratio [aOR] = 1.16; 95% CI: 1.13–1.20), cigarette use (aOR = 1.19; 95% CI: 1.13–1.26), e-cigarette use (aOR = 1.13; 95% CI: 1.10–1.16), and passive smoking (aOR = 1.23; 95% CI: 1.17–1.29) were all significantly associated with asthma. Parental smoking conferred elevated risk (aOR = 1.85; 95% CI: 1.33–2.36), and exposure from friends showed the strongest association (aOR = 3.70; 95% CI: 1.63–5.78).

Conclusions

Both active and passive smoking significantly increase asthma risk in adolescents, with both cigarette use and e‑cigarette use showing adverse associations. Friend smoking represents the most potent passive exposure. According to the GRADE assessment, the certainty of evidence for these associations was rated as very low, reflecting limitations in study design, confounder adjustment, and exposure measurement. Future studies should incorporate longitudinal designs, validated biomarkers, and standardized outcome measures to better characterize dose–response patterns and underlying mechanisms.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27640-6.

Key messages

What is already known on this topic.

•Active and passive smoking are established contributors to respiratory morbidity in adolescents, yet prior reviews have largely focused on passive exposure and often reported inconsistent associations.

•However, evidence has remained fragmented, with limited assessment of peer-related exposure, differential parental effects, or product-specific risks.

What this study adds

•This study provides the most comprehensive quantitative synthesis to date, demonstrating that both active and passive smoking significantly increase asthma risk in adolescents.

•It identifies friend smoking as the strongest passive exposure source and reveals divergent effects of maternal versus paternal smoking.

•Subgroup analyses further clarify risk gradients by exposure type, timing, study design, and asthma phenotype.

How this study might affect research, practice or policy

•These findings highlight the need for adolescent-focused tobacco control strategies that incorporate peer-network dynamics alongside family-level interventions.

•The strong associations observed reinforce the urgency of stricter smoke-free policies, targeted school-based prevention programs, and surveillance systems that account for emerging tobacco products.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27640-6.

Introduction

Asthma is a prevalent chronic respiratory condition with significant global impact [1]. Despite advances in clinical management, epidemiological data reveal rising asthma incidence across specific geographical regions [2]. In 2022, 11% of adolescents globally reported asthma symptoms in the preceding 12 months, with prevalence exceeding 20% in parts of Europe, South America, and South Africa, whereas remaining lower—between 5% and 10%—across Southeast Asian countries [3]. However, asthma remains a major concern among adolescents, highlighting the need for deeper investigation into its genetic, environmental, and socioecological determinants [4]. Adolescent asthma, a chronic respiratory disorder with considerable global prevalence, affects individuals aged 10–19 who are particularly vulnerable due to ongoing pulmonary development. Its pathophysiology is strongly influenced by various environmental exposures [5]. Among these, active and secondhand tobacco smoke represent the most prominent modifiable risk factor for asthma in this age group [6].

Globally, adolescent smoking remains prevalent (median prevalence 10.3% across 142 countries) [7]. In the USA, 10.1% of high school students use any tobacco product, with 7.8% using e-cigarettes—a behavior distinct from smoking, commonly termed vaping or electronic nicotine delivery systems (ENDS) use [8]; in China, 4.2% smoke cigarettes and 2.4% use e-cigarettes [9]. Given nicotine dependence impairs respiratory development [10], understanding early tobacco use is critical. Epidemiological studies link active smoking to adolescent asthma—e.g., a South African cohort (adjusted odds ratio [aOR]: 1.84; 95% confidence interval [95% CI]: 1.08–3.16) [11] and Brazilian surveys (aOR:1.36; 95% CI [1.30–1.41]; adjusted prevalence ratio [aPR]: 1.82; 95% CI: [1.30–2.56]) [12, 13]—though null findings also exist [14–16]. Passive smoking is widespread: approximately 40% of children under 15 are regularly exposed [17], with higher prevalence in schools (80.5%) than in homes (60.1%) [18]. Prenatal exposure (maternal smoking) increases offspring asthma risk by 21–85% [19], and postnatal exposure worsens asthma control and lung function [20, 21]. However, inconsistent exposure assessment may explain non-significant results [16, 22, 23], highlighting the need for rigorous studies integrating active and passive exposures.

Critically, the persistent fragmentation of evidence has resulted in a concerning body of inconsistent findings. While secondhand smoke has been a focal point of public health campaigns, the escalating prevalence of adolescent smoking and the parallel rise of e-cigarette use remain an under-addressed public health crisis. This oversight is alarming given that adolescence represents a critical window of respiratory vulnerability, during which the combined burden of active initiation and passive co-exposure may precipitate irreversible lung function deficits. Moreover, the absence of unified analytical frameworks that simultaneously account for distinct exposure types, timing, sources, and emerging risks has left key questions unanswered. For instance, whether household smoking restrictions alone are sufficient given high exposure in schools, or whether the influence of maternal, paternal, and peer smoking differs, remains unclear. Without a robust synthesis of the global evidence that rigorously addresses heterogeneity, public health strategies risk being neither adequately targeted nor optimally effective. Therefore, there is an urgent need to move beyond piecemeal findings toward a methodologically rigorous, comprehensive quantification of the associations between active smoking, secondhand smoke exposure, and asthma.

To date, however, no comprehensive systematic review or meta-analysis has systematically synthesized the global evidence on adolescent asthma in a manner that simultaneously accounts for the distinct and combined effects of active and passive smoking while rigorously addressing the key sources of heterogeneity identified above. Existing syntheses have largely been constrained by a narrow focus on either active or passive exposure alone, precluding a holistic understanding of the total tobacco-related burden during this critical developmental window. Furthermore, prior meta-analyses have rarely incorporated stratification by product type (combustible cigarettes versus e-cigarettes), exposure timing (current versus ever), asthma definition (current versus ever asthma), or source of passive exposure (mother, father, versus peers)—factors that are essential for disentangling the heterogeneous findings reported across primary studies. In the absence of such a methodologically rigorous synthesis, the precise magnitude and patterns of association between tobacco exposure and adolescent asthma remain insufficiently characterized. To address this gap, we conducted a systematic review and meta‑analysis to quantify the associations of active smoking (by type) and passive smoking (by source) with adolescent asthma, while rigorously addressing key sources of heterogeneity identified in the literature.

Methods

This systematic review was conducted following the Cochrane Handbook for Systematic Reviews of Interventions and aimed to synthesize high-quality evidence on the association between active and passive smoking and asthma incidence. The protocol for this systematic review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA) guidelines and the Meta-analysis of Observational Studies in Epidemiology (MOOSE) [24, 25] (Supplementary file 1). This systematic review was prospectively registered with International Prospective Register of Systematic Reviews (PROSPERO) on 4 December 2024 under registration number CRD42024622246 (https://www.crd.york.ac.uk/prospero/), and received ethical approval from the Research Ethics Committee, Faculty of Medicine, Chiang Mai University, Thailand (Approval No. Exemption 0391/2025) on 29 May 2025.

Data sources and search strategy

For this study, a systematic literature search was conducted across four electronic databases: Embase, PubMed, Scopus, and Web of Science. The search commenced on December 5, 2024, and concluded on January 1, 2026. Given the substantial content overlap among major biomedical databases, the search was primarily focused on these four sources, which together ensure comprehensive coverage of the relevant literature. The search strategy combined Medical Subject Headings (Mesh) and free-text terms in titles and abstracts, using keywords such as “determinants,” “predictors,” “risk factors,” “tobacco smoke pollution,” “smoking,” “ENDS,” “asthma,” “adolescent,” and “child.” Controlled vocabulary and database-specific subject headings were also applied to ensure comprehensive coverage. To ensure comprehensive coverage, the search was supplemented by manual screening of reference lists from included studies and relevant prior reviews. The complete search strategy for all databases is outlined in Supplementary file 2.

Eligible criteria

Studies were included if they met the following criteria: (1) published up to January 2026; (2) original observational designs, including cross-sectional, case-control, or cohort studies; (3) only full-text articles published in English were included; (4) investigation of postnatal exposure to smoking, including both active and passive (secondhand) smoking; (5) asthma as the primary outcome, defined by either self-reported physician-diagnosed asthma using a questionnaire adapted from the International Study of Asthma and Allergies in Childhood (ISAAC) protocol, or formally documented diagnosis by a licensed medical professional; (6) study population comprising adolescents aged 10–19 years; (7) analysis of the association between smoking exposure and asthma; and (8) application of multivariable regression analysis. Effect estimates were required to be reported as odds ratio (OR), hazard ratio (HR), relative risk (RR), prevalence ratio (PR), incidence rate ratio (IRR), or prevalence odds ratio (POR), with corresponding 95% CI.

Exclusion criteria included conference abstracts, case reports or series, reviews, animal studies, and expert opinions. Studies relying solely on self-reported asthma or asthma-like symptoms without clinical confirmation by either physician diagnosis or formal medical records were excluded to reduce misclassification bias and improve diagnostic validity.

Exposure and of outcome classification

The primary exposures of interest were active smoking and passive smoking. Active smoking was defined as the use of any tobacco or nicotine product, including: (1) cigarettes; (2) electronic cigarettes (e-cigarettes/vapes/ENDS); (3) water pipes/hookahs/shisha; (4) heat-not-burn tobacco products; (5) cigars; and (6) unidentified tobacco products. Passive smoking was defined as the involuntary inhalation of tobacco smoke from others, including parents (mother, father, or both), friends, or unidentified individuals in the environment.

Exposure timing was categorized as: (1) current smoking (use within the past 30 days); (2) past or ever smoking (any history of smoking, even a single instance); (3) both past and current smoking (ongoing use with prior history); and (4) unidentified smoking status (unspecified timing of use).

The outcomes of interest were categorized as current or ever asthma. Current asthma was defined as a physician-diagnosed condition accompanied by prescribed asthma medication use or wheezing episodes within the preceding 12 months. Ever asthma referred to individuals previously diagnosed with asthma by a physician but not treated in the past year. Asthma outcomes were assessed via (1) parent- or self-reported physician diagnosis or (2) documented diagnosis in medical records.

Data extraction

All relevant studies identified from the four electronic databases were imported into EndNote X9 (Thomson Reuters, USA) for systematic management, including duplicate removal and preliminary screening. Two independent reviewers (W.W.S. and S.K.) screened titles, abstracts, and full texts based on predefined inclusion criteria. Discrepancies were resolved through discussion with a third reviewer (R.S.) to achieve consensus.

For each eligible study, the following data were extracted: first author, year of publication, study location, age range of participants, study design, sample size, type of exposure (active or passive smoking), method of exposure assessment (e.g., questionnaire, interview, or biomarker analysis), asthma outcome (current or ever asthma), method of outcome assessment (e.g., self-reported physician diagnosis or medical records), key findings, and covariates adjusted for in the analysis. Data extraction was performed independently by two reviewers and subsequently cross-verified for consistency (W.W.S. and S.K.).

Risk of bias assessment

The methodological quality of included studies was assessed using standardized tools developed by the National Heart, Lung, and Blood Institute (NHLBI) for observational cohort, cross-sectional, and case-control studies [26]. These tools are designed to evaluate key elements of internal validity. For cohort and cross-sectional studies, the NHLBI tool consists of 14 items rated as “yes,” “no,” or “other” (e.g., “cannot determine,” “not applicable,” or “not reported”). Based on the number of affirmative responses, studies were classified as good (11–14), fair (6–10), or poor quality (1–5). For case-control studies, the NHLBI tool includes 12 items and follows a similar rating system, with quality classified as good (9–12), fair (5–8), or poor (1–4). Two reviewers (W.W.S. and W.K) independently evaluated study quality. Discrepancies were resolved through discussion with a third reviewer (R.S.) to achieve consensus.

Certainty of evidence

For the overall certainty of evidence, the same two reviewers independently assessed each pooled outcome using the GRADE approach, implemented via GRADEpro GDT (GRADEpro Guideline Development Tool, McMaster University and Evidence Prime, 2021) [27]. Certainty was rated as high, moderate, low, and very low based on five domains: risk of bias, inconsistency, indirectness, imprecision, and publication bias [28]. Any disagreements were resolved by consensus, with consultation from a third reviewer (R.S.) when necessary.

Data analysis

Effect estimates reported across the included studies comprised OR, HR, RR, PR, and POR. For meta-analytic synthesis, all effect measures were converted to OR, a commonly used metric in meta-analyses. Studies reporting PR or RR [29–31] were converted using the following standard formulas [32, 33]:

Of the three studies mentioned above [29–31], only one reported a low prevalence of physician‑diagnosed asthma (7.4% [31]). Given this prevalence was below 10%, the rare‑outcome assumption applied, under which the OR approximates the RR. To confirm this approximation, we converted the reported RR to an OR using standard equations. As shown in the forest plot, the converted OR closely matched the original RR, supporting the validity of the conversion. We therefore used the OR derived from the RR in the meta‑analysis. For the other two studies [29, 30], where the reported prevalence of asthma exceeded 10%, we retained PR as originally reported.

Heterogeneity was assessed using Cochran’s Q test and the I² statistic, with thresholds interpreted as low (< 25%), moderate (25–50%), or substantial (> 50%). A fixed-effect model was applied when heterogeneity was low, assuming a common true effect. For moderate to high heterogeneity (I² ≥ 25%), a random-effects model using inverse-variance weighting was used to account for between-study variability. Subgroup analyses were conducted based on type (e.g., cigarettes or e-cigarettes), study design (cross-sectional, cohort, or case-control), exposure timing (current, ever, or unspecified), and asthma classification (current or ever asthma). Sensitivity analyses explored the robustness of findings across exposure types, effect measures, and model selection (fixed or random effects).

Due to the limited number of studies (n < 3) examining specific tobacco products (e.g., water pipes, heat-not-burn products, cigars), meta-analysis was not performed for these exposures to avoid low statistical power and unreliable estimates. The meta-analysis was therefore restricted to cigarettes and e-cigarettes, where sufficient data was available. We presented the study results using evidence tables and graphically using forest plots to display the effect sizes from individual studies and the pooled estimates from meta-analyses.

Publication bias was assessed using funnel plots and Egger test, with effect estimates plotted against sample size. Asymmetry was evaluated to detect small-study effects or potential bias. When present, the trim-and-fill method was applied to adjust for potential missing studies. All statistical analyses were conducted using STATA version 17.0 (StataCorp LLC, College Station, TX, USA), with statistical significance set at p value < 0.05 (two-tailed).

Result

Study selection

A total of 35,567 records were initially identified through database searches. After removing duplicates, 21,899 unique records underwent title and abstract screening, resulting in 386 articles deemed potentially eligible. Following full-text review based on predefined inclusion criteria, 77 studies were included in the systematic review, of which 72 were eligible for meta-analysis (Fig. 1, Supplementary file 3).

Fig. 1.

The PRISMA flow diagram of study selection

The study characteristics regarding the association between active smoking and adolescent asthma

This analysis synthesized 38 studies [29–31, 34–68] published between 1994 and 2024, with sample sizes ranging from 347 to 216,056 participants. Of these, six employed cohort designs and 32 used cross-sectional methodologies. Exposure assessed primarily through self-reported questionnaires. Asthma outcomes included current asthma (n = 28) and ever asthma (n = 15). The studies represented diverse geographic regions, including Alaska, Arab countries, Brazil, Canada, the Carolinas, Croatia, Malaysia, the Maltese Islands, the Netherlands, Peru, Iran, China (n = 2), Sweden (n = 2), South Korea (n = 7), and the United States (n = 13). Most studies investigated cigarettes and e-cigarettes; fewer examined cigars (n = 3), heat-not-burn tobacco products (n = 2), and water pipes (n = 1). Exposure timing was categorized as current smoking (n = 16), past/ever smoking (n = 16), or unspecified (n = 5). Additional study characteristics are presented in Table 1.

Table 1.

The studies regarding the association between active smoking and adolescent asthma

Author, year (region)	Age (years)	Study design	Sample size	Health outcome	Type of tobacco	Findings POR/PR/OR/RR/HR (95%CI)	Confounders/covariates
Larsson., 1995 (Sweden) [34]	16–19	CO	2,308	Current asthma c	Cigarette a,1	OR = 1.00 (0.50–1.80)	Gender
				Ever asthma c	Cigarette a,1	OR = 1.90 (1.10–3.40) *
Lam et al., 1998 (China) [35]	13–15	CS	6,304	Ever asthma c	Cigarette a,2	OR = 1.05 (0.82–1.36) for tried smoking cigarette only	Gender, age, area of residence, and type of housing
					Cigarette a,2	OR = 1.43 (0.94–2.19) for ever used cigarette, not now
					Cigarette a,1	OR = 1.09 (0.63–1.88) for < 1/week
					Cigarette a,1	OR = 1.13 (0.57–2.26) for 1–6/week
					Cigarette a,1	OR = 1.18 (0.76–1.83) for > 6/week
Montefort et al., 1998 (Maltese Islands) [36]	13–15	CS	4,184	Ever asthma c	Cigarette a,4	OR = 1.46 (1.12–1.91) *	Gender, age, passive smoking, busy road, pets, atopic relatives, and blankets
Norrman et al.,1998 (Sweden) [37]	13–16	CO	1,112	Current asthma c	Unidentified tobacco a,2	OR = 2.1(0.46–9.8)	Month of birth, sex, parental smoking, own smoking, type of home, and visible mold at home, atopy, furry pets at home, and heredity of asthma or rhinoconjunetivitis
					Unidentified tobacco a,1	OR = 1.0(0.34–2.8)
				Ever asthma c	Unidentified tobacco a,1	OR = 3.3(0.2–57)
Wang et al., 1999 (China) [38]	11–16	CS	155,283	Current asthma c	Cigarette a,4	OR = 1.29 (1.17–1.42) *	Gender, age, residence, parents’ education, exercise habit, Chinese incense, alcohol drinking, and environmental tobacco smoke
Lewis et al., 2004 (Alaska) [31]	Grade 6th to 9th	CS	377	Ever asthma c	Cigarette a,1	RR = 0.60 (0.17–2.16) for ≧ smoked 1 cigarette	Gender, residence duration in a village, tobacco smoke exposure, unknown smoking status, and potential atopy
Sturm et al., 2004 (Carolina) [39]	Grade 7th and 8th	CS	11,378	Current asthma c	Cigarette a,2	OR = 1.02 (0.98–1.07)	Gender, race, use of gas stove, parental asthma history, allergy status, socioeconomic status, and other competing sources of tobacco smoke exposure
					Cigarette a,1	OR = 1.10 (1.04–1.16) *
Jones et al., 2006 (USA) [40]	Grade 9th to 12th	CS	13,222	Current asthma c	Cigarette a,2	OR = 1.10 (0.95–1.30)	Gender, race, grade, tried to quit using cigarettes, lifetime inhalant use, and current inhalant use
					Cigarette a,1	OR = 1.60 (1.30–2.00) *
					Cigar a,1	OR = 1.00 (0.90–1.20)
Van De Ven et al., 2007 (Netherlands) [41]	15–17	CO	7,426	Current asthma c	Unidentified tobacco a,1	OR = 0.96 (0.55–1.66) for experimental smoking OR = 2.08 (0.88–4.90) for regular smoking	Age, education level, and ethnicity
Bae et al., 2011 (Korea) [42]	13–17	CS	75,238	Current asthma c	Cigarette a,1	OR = 1.25 (1.06–1.48) for use ≧ 1 days * OR = 1.15 (0.93–1.42) for use ≧ 20 days OR = 1.93 (1.51–2.46) for use ≧ 10 cigarettes/day *	Gender, grade, self–rated school records, perceived socioeconomic status, current alcohol use, and suicidal ideation
					Cigarette a,2	OR = 1.13 (0.95–1.33) for use before 13 years of age
				Ever asthma c	Cigarette a,1	OR = 1.30 (1.13–1.49) for use ≧ 1 days * OR = 1.45 (1.20–1.74) for use ≧ 20 days * OR = 2.14 (1.62–2.84) for use ≧ 10 cigarettes/day *
					Cigarette a,2	OR = 1.35 (1.14–1.61) for use before 13 years of age *
Hedman et al., 2011 (Sweden) [43]	16–17	CO	2,805	Ever asthma c	Cigarette a,1	OR = 1.36 (0.93–1.98)	Gender, family history of asthma, current place of residence, house dampness and birth weight
Lawson et al., 2011 (Canada) [44]	11–15	CS	4,726	Current asthma c	Unidentified tobacco a,1	OR = 1.10 (0.79–1.53) for smoking tobacco occasionally OR = 1.04 (0.70–1.55) for smoking tobacco daily	Geographic region, obesity status, physical activity, whole milk and vegetable consumption, friend or parental smoking, age, gender, family affluence scale, and ethnicity
Gudelj et al., 2012 (Croatia) [45]	14–16	CS	4,027	Current asthma c	Unidentified tobacco a,4	OR = 2.373 (1.508-3.734) *	Area of residence, gender, allergy household members, each crowdedness type of fuel used for heating, central or classic gas used for cooking, pets, dog or cat ownership over 12 months, and passive smoking
Checkley et al., 2013 (Peru) [46]	13–15	CS	1,002	Current asthma d	Unidentified tobacco a,2	OR = 2.10 (0.70–6.40) for 461 atopic children OR = 1.60 (0.20–13.9) for 541 no atopic child	Total serum IgE, BMI, age maternal education, people per house income per month, concrete floor, and gender
Kim et al., 2013 (Korea) [47]	12–18	CS	3,432	Current asthma c	Cigarette a,1	OR = 1.18 (0.77–1.82) for 1–9 days smoking OR = 1.00 (0.47–2.09) for 10–19 days smoking OR = 1.77 (1.26–2.49) for 20–30 days smoking *	Gender, academic achievements, allowance, duration of diagnosis, asthma treatment, drinking days in the past 30 days, and secondhand smoke at home in the past 7 days
Norbäck et al., 2014 (Malaysia) [48]	Junior high schools	CS	462	Ever asthma c	Unidentified tobacco a,1	OR = 2.34 (0.82–6.72)	Gender, race, parental asthma, and allergy
Hedman et al., 2015 (Sweden) [49]	12–19	CO	2,747	Current asthma c	Unidentified tobacco a,2	HR = 1.49 (0.98–2.26)	Gender, parental history of asthma, number of siblings, ever cat house, dampness, height, BMI, weight, self–reported traffic exposure, maternal environmental tobacco smoke, living in an apartment, allergic sensitization, and respiratory infections
				Ever asthma c	Unidentified tobacco a,2	HR = 1.33 (0.92–1.92)
Cho et al., 2016 (Korea) [50]	High school students	CS	674	Current asthma c	Cigarette a,2	OR = 0.99 (0.75–1.31)	Gender, city size, multi–cultural family status, overweight status, secondhand smoking at home, atopic dermatitis and allergic rhinitis history
					Cigarette a,1	OR = 1.47 (1.05–2.06) *
					E– cigarette a,2	OR = 0.96 (0.42–2.19)
					E– cigarette a,1	OR = 2.77 (1.31–5.85) *
Choi et al., 2016 (USA) [51]	High school students	CS	36,085	Current asthma c	E– cigarette a,2	OR = 3.96 (1.49–10.56) *	Age, race, gender, metropolitan status, and living with someone who smoking
					E– cigarette a,1	OR = 1.78 (1.20–2.64) *	Age, race, gender, metropolitan status, days smoked cigarettes in the past 30 days, positive social norm towards smoking, and secondhand smoke exposure
Kim et al., 2017 (Korea) [52]	12–18	CS	216,056	Current asthma c	E– cigarette a,1	OR = 1.13 (1.01–1.26) *	Age, physical exercise, gender, obesity, region of residence, economic level, educational level of father, education level of mother, and passive smoking
Schweitzer et al., 2017 (USA) [53]	12–18	CS	6,089	Current asthma c	Cigarette a,2	OR = 1.27 (1.05–1.54) *	Stratum, school clustering, gender, age, overweight, educational intentions, and ethnicity
					Cigarette a,1	OR = 1.23 (0.92–1.64)
					E– cigarette a,2	OR = 1.22 (1.01–1.47) *
					E– cigarette a,1	OR = 1.48 (1.24–1.78) *
				Ever asthma c	Cigarette a,2	OR = 1.01 (0.82–1.24)
					Cigarette a,1	OR = 1.17 (0.88–1.54)
					E– cigarette a,2	OR = 1.19 (0.99–1.43)
					E– cigarette a,1	OR = 1.20 (1.00–1.44) *
Marques et al., 2019 (Brazil) [30]	13–15	CS	211,176	Current asthma c	Cigarette a,4	PR = 1.38 (1.30–1.47) for 109,104 child in 2012 *	Gender, age, skin color, maternal educational attainment, economic index, alcohol drinking in previous 30 days, and parents who smoke
					Cigarette a,4	PR = 1.25 (1.16–1.35) for 102,072 child in 2015 *
Bayly et al., 2019 (USA) [54]	11–17	CS	11,830	Current asthma c	Cigarette a,2	OR = 1.23 (0.99–1.52)	Age, gender, race, metropolitan status, and housing type
					Cigarette a,1	OR = 1.92 (1.28–2.68) *
					Cigarette a,4	OR = 1.47 (0.76–2.86)
					E– cigarette a,2	OR = 1.01 (0.81–1.25)
					E– cigarette a,1	OR = 0.90 (0.71–1.15)
					E– cigarette a,4	OR = 0.84 (0.42–1.69)
					Cigar a,2	OR = 0.78 (0.60–1.03)
					Cigar a,1	OR = 0.84 (0.60–1.18)
					Cigar a,4	OR = 2.40 (1.25–4.63) *
					Water pipe a,2	OR = 1.01 (0.81–1.25)
					Water pipe a,1	OR = 0.76 (0.56–1.06)
					Water pipe a,4	OR = 0.81 (0.49–1.31)
Han et al., 2019 (USA) [55]	Grade 9th to 12th	CS	24,612	Current asthma c	Cigarette a,1	OR = 1.12 (0.91–1.38)	Age, gender, race, BMI, average hour of sleep < 8 h., ate fruit or vegetables < 7 times/day, drank regular soda, pop, psychosocial stressors, depressive symptoms, suicidal behavior, violent behavior exposure, and feeling sad or hopeless in the past year
Lee et al., 2019 (Korea) [56]	12–18	CS	14,829	Current asthma c	Cigarette a,2	OR = 1.11 (0.86–1.42)	Age, gender, obesity, residential area, family economic status, and physical activity
					E– cigarette a,2	OR = 0.86 (0.59–1.25)
					Heat–not–burn tobacco a,2	OR = 1.54 (0.95–2.49)
Alnajem et al., 2020 (Arab) [29]	16–19	CS	1,565	Current asthma c	Cigarette a,2	PR = 1.67 (1.08–2.58) *	Gender, age, household secondhand smoke exposure, household secondhand aerosols from electronic cigarettes, and exposure to secondhand smoke, secondhand aerosols from electronic cigarettes in public places
					Cigarette a,1	PR = 1.73 (1.01–3.21) *
					E– cigarette a,2	PR = 1.43 (0.70–2.92)
					E– cigarette a,1	PR = 1.85 (1.03–3.41) *
Chung et al., 2020 (Korea) [57]	13–18	CS	60,040	Current asthma c	Cigarette a,2	OR = 1.20 (1.00–1.50)	Age, gender, BMI, residential area, regular exercise, sedentary time, exposure to secondhand smoke, socioeconomic status, and the presence of allergic rhinitis
					Cigarette a,1	OR = 1.60 (1.10–2.20) *
					E– cigarette a,2	OR = 1.80 (1.10–3.00) *
					Heat–not–burn tobacco a,2	OR = 3.80 (1.50–9.60) *
Han et al., 2020 (USA) [58]	Grade 9th to 12th	CS	21,532	Ever asthma c	Cigarette a,1	OR = 1.01 (0.82–1.26) for use cigarette < 10 days OR = 1.20 (0.93–1.55) for use cigarette ≧ 10 days	Age, gender, race, overweight or obesity, and at least 1 dental office visit in the previous year, frequent marijuana and e–cigarette use, and food allergy
					E– cigarette a,1	OR = 1.13 (0.97–1.33) for use of electronic vapor < 10 days OR = 1.27 (1.05–1.55) for use of electronic vapor ≧ 10 days *
					E– cigarette a,1	OR = 1.16 (0.95–1.41) for use vaping	Age, gender, race, overweight or obese, dental office visit in the previous year
Wills et al., 2020 (USA) [59]	14–18	CS	14,652	Ever asthma c	Cigarette a,2	OR = 1.01 (0.81–1.25)	E– cigarette smoking, marijuana use, gender, obesity, and age
					Cigarette a,1	OR = 1.23 (0.92–1.64)
					E– cigarette a,2	OR = 1.16 (1.01–1.33) *	Cigarette smoking, marijuana use, gender, obesity, and age
					E– cigarette a,1	OR = 1.29 (1.07–1.55) *
Chaffee et al., 2021 (USA) [60]	Grade 9th and 10th	CS	6,298	Current asthma c	E– cigarette a,2	OR = 0.90 (0.72–1.12) for ever use (n = 4,875)	Gender, race, school lunch program participation, personal income (young adults), age, past 30–day combustible tobacco, and past 30–day use of cannabis
					E– cigarette a,1	OR = 1.27 (1.01–1.59) for used 1–5 days (n = 4,875) *
					E– cigarette a,1	OR = 1.85 (1.52–2.26) for used 6–30 days (n = 4,875) *
					E– cigarette a,2	OR = 1.15 (0.73–1.82) for ever use (n = 1,423)
					E– cigarette a,1	OR = 0.76 (0.48–1.17) for used 1–5 days (n = 1,423)
					E– cigarette a,1	OR = 0.98 (0.59–1.64) for used 6–30 days (n = 1,423)
					E– cigarette a,1	OR = 1.14 (0.78–1.65) for vape pens/tanks (n = 4,875)
					E– cigarette a,1	OR = 0.73 (0.37–1.44) for disposables use (n = 4,875)
					E– cigarette a,1	OR = 0.64 (0.29–1.39) for box mods use (n = 4,875)
					E– cigarette a,1	OR = 1.54 (1.10–2.14) for multiple device (n = 4,875) *
					E– cigarette a,1	OR = 1.79 (0.83–3.85) for vape pens/tanks (n = 1,423)
					E– cigarette a,1	OR = 1.51 (0.71–3.24) for disposables use (n = 1,423)
					E– cigarette a,1	OR = 1.78 (0.36–8.70) for box mods use (n = 1,423)
					E– cigarette a,1	OR = 2.10 (1.07–4.15) for multiple device (n = 1,423) *
Cherian et al., 2021 (USA) [61]	12–17	CS	9,769	Current asthma c	E– cigarette a,2	OR = 1.13 (0.85–1.50)	Age, gender, race, parent education, tobacco used by other household members, rules about combustible tobacco product use inside home, lifetime number of cigarettes used, and lifetime number of cigars used
Norbäck et al., 2021 (Malaysia) [62]	14	CS	462	Current asthma c	Unidentified tobacco a,1	OR = 2.84 (0.81–9.83)	Gender, ethnicity, atopy and current smoking
Sabeti et al., 2021 (Iran) [63]	14–19	CS	1,459	Current asthma c	Unidentified tobacco a,1	OR = 1.91 (1.21–3.02) *	Location, gender, BMI, physical activity, parent’s education, parent’s income, residence type, ventilation, pets, fruit, vegetables, fast foods, history of allergic rhinitis, and family history of asthma
				Ever asthma c	Unidentified tobacco a,1	OR = 0.72 (0.32–1.63)
Kim et al., 2022 (Korea) [64]	13–17	CS	57,303	Current asthma c	Unidentified tobacco a,1	OR = 1.607 (1.414–2.264) for current single use one of combustible cigarettes, e–cigarettes, and heated tobacco products * OR = 2.626 (1.626–4.240) for current dual use combustible cigarettes, e–cigarettes, and heated tobacco products *	Age, BMI, gender, living with family, family economic status, secondhand smoking, current drinking, and regular exercise
Miura et al., 2022 (USA) [65]	Grade 9th to 12th	CS	21,622	Current asthma c	Cigarette a,1	OR = 1.40 (1.09–1.81) *	Race or ethnicity, housing, urbanicity, and gender
					E– cigarette a,1	OR = 1.10 (0.96–1.27)
					Cigar a,1	OR = 1.36 (1.07–1.73) *
				Ever asthma c	Cigarette a,1	OR = 1.43 (1.06–1.92) *
					E– cigarette a,1	OR = 1.12 (0.96–1.31)
					Cigar a,1	OR = 1.10 (0.84–1.44)
Roh et al., 2023 (USA) [66]	13– 17	CS	13,399	Ever asthma c	E– cigarette a,2	OR = 1.32 (1.06–1.66) for never use combustible product child (n = 1,765) *	Gender, age, race or ethnicity, BMI, other substance use, depression, and combustible products
					E– cigarette a,2	OR = 1.18 (1.02–1.37) for never use combustible product child (n = 11,634) *
Williams et al., 2023 (USA) [67]	Grade 10th and 12th	CS	150,634	Current asthma c	Cigarette a,4	OR = 0.95 (0.60–1.51) for model 1(n = 133,717)	Age, gender, parental education, race or ethnicity, and 3 types of household cigarette, e–cigarette and cannabis use.
					Cigarette a,4	OR = 0.79 (0.43–1.45) for model 2 (n = 113,922)
					E– cigarette a,4	OR = 1.17 (1.03–1.33) for model 1 (n = 133,717) *
					E– cigarette a,4	OR = 1.12 (0.97–1.28) for model 2 (n = 113,922)
				Ever asthma c	Cigarette a,4	OR = 0.89 (0.70–1.13) for model 1 (n = 133,717)
					Cigarette a,4	OR = 0.80 (0.60–1.05) for model 2 (n = 113,922)
					E– cigarette a,4	OR = 1.11 (1.04–1.19) for model 1(n = 133,717) *
					E– cigarette a,4	OR = 1.10 (1.02–1.18) for model 2 (n = 113,922) *
Yao et al., 2024 (USA) [68]	12–17	CO	9,422	Current asthma c	E– cigarette a,1	OR = 0.37 (0.07–1.87)	Age, gender, race, household income, parents’ education, home tobacco rules, and living with tobacco use

^a Questionnaire or interview; ^c Parent- or child self-reported physician diagnosis; ^d Medical records; ¹ Current smoking; ² Past/ever smoking, ³ Past and current smoking, ⁴ Unidentified; BMI Body mass index, OR Odds ratio, HR Hazard ratio, RR Relative risk, PR Prevalence ratio, IRR Incidence rate ratio, POR Prevalence odds ratio (POR), CS Cross–sectional study, CO Cohort study, CC Case–control study; * p value < 0.05

Study characteristics regarding the association between passive smoking and adolescent asthma

The 57 included studies, published between 1994 and 2024, featured sample sizes ranging from 40 to 216,056 participants [29, 31, 35, 37–39, 41, 43–47, 52, 54, 56, 67–107]. Study designs comprised 16 cohort studies, 2 case-control studies, and 39 cross-sectional studies. Smoking exposure was predominantly assessed through questionnaires. The reported outcomes were current asthma and ever asthma. A detailed summary of study characteristics and findings is presented in Table 2.

Table 2.

The studies regarding the association between passive smoking and adolescent asthma

Author, Yeas, (region)	Age (years)	Study design	Sample size	Health outcome	The type of exposure	PR/OR/RR/HR (95%CI)	Confounders/covariates
Norrman et al., 1994 (Sweden) [69]	14	CS	404	Current asthma c	Father smoking a	OR = 0.83 (0.35–2.00) for ex–smoker OR = 0.79 (0.34–1.85) for current smoker	Month of birth, gender, heredity of asthma or rhino conjunctivitis, parental smoking, type of residence, visible mold at home, and location of school among atopic pupils participating in interview
					Mother smoking a	OR = 2.68 (1.08–6.65) for ex–smoker * OR = 1.63 (0.72–3.67) for current smoker
				Ever asthma c	Father smoking a	OR = 0.88 (0.40–1.96) for ex–smoker OR = 0.86 (0.40–1.85) for current smoker
					Mother smoking a	OR = 2.68 (1.16–6.15) for ex–smoker * OR = 1.55 (0.74–3.26) for current smoker
Lam et al., 1998 (China) [35]	13–15	CS	6,304	Ever asthma c	Father smoking a	OR = 0.92 (0.72–1.17)	Gender, age, area of residence, and type of housing
					Mother smoking a	OR = 1.32 (0.71–2.45)
Norrman et al., 1998 (Sweden) [37]	13–16	CS	1,112	Current asthma c	Mother smoking a	OR = 2.10 (0.80–5.50) for former smoking	Month of birth, sex, atopy, furry pets at home, parental smoking, type of home, heredity of asthma or rhinoconjunetivitis, own smoking, and visible mold at home
						OR = 1.50 (0.80–3.00) for current smoking
				Ever asthma c	Mother smoking a	OR = 0.40 (0.00–21.0) for former smoking
				Current asthma c	Father smoking a	OR = 1.00 (0.60–2.50) for former smoking
						OR = 1.20 (0.40–2.70) for current smoking
				Ever asthma c	Father smoking a	OR = 0.20 (0.00–4.20) for current smoking
Wang et al., 1999 (China) [38]	11–16	CS	165,173	Current asthma c	Unidentified a	OR = 1.08 (1.05–1.12) *	Gender, age, residence, parents’ education, exercise habit, Chinese incense, cigarette smoking, and alcohol drinking
Räsänen et al., 2000 (Finland) [70]	16	CO	4,538	Ever asthma c	Mother smoking a	OR = 1.59 (1.09–2.31) for former smoking	Gender, non–perinatal factors common to both members of the pair parental asthma, parental hay fever, number of older siblings, father’s occupation, hay fever in the adolescent
Shima et al., 2000 (Japan) [71]	11–12	CO	840	Current asthma c	Unidentified a	OR = 0.51 (0.19–1.35)	Gender, history of allergic diseases, respiratory diseases under 2 years of age, breastfeeding in infancy, parental history of allergic diseases, outdoor and indoor NO2 concentration, and use of unvented heater in winter
Lee et al., 2001 (Korea) [72]	12–15	CS	38,955	Current asthma c	Unidentified a	OR = 0.99 (0.87–1.13)	Age, gender, BMI, city position, electricity bill, carpet use, dog, and living environment
Mannino et al., 2001 (USA) [73]	12–16	CS	1642	Current asthma c	Unidentified a, b	OR = 0.7 (0.3–1.7) for medium cotinine exposure 0.60–3.23 nmol/L	Race, ethnicity, sex social economic status, parental history of asthma, and family size
						OR = 1.7 (0.7–7.3) for high exposure cotinine 3.24–113.6 nmol/L
				Ever asthma c	Unidentified a, b	OR = 0.8 (0.4–1.8) for medium cotinine exposure 0.60–3.23 nmol/L
						OR = 1.5 (0.7–3.3) for high exposure cotinine 3.24–113.6 nmol/L
Pokharel et al., 2001 (Haryana) [74]	11–15	CC	Cases (n = 40) Controls (n = 80)	Current asthma c	Unidentified a	OR = 3.33 (1.85–7.65) *	Age, gender, height, weight, family history of asthma, worm infestation, fuel used for cooking, location of kitchen, food allergies, pets, and absence of windows in rooms
Golshan et al., 2002 (Iran) [75]	12–15	CS	1,348	Current asthma c	Mother and father smoking a	OR = 1.91 (1.67–2.19) for boy *	Parents education, history of similar illness in parents, history of similar illness in siblings, history of similar illness in other relatives, cockroaches in the household, chicken pets and cats in the household
Lewis et al., 2004 (Alaska) [31]	Grade 6th to 9th	CS	377	Ever asthma c	Unidentified a	RR = 3.90 (1.38–11.00) for high exposure to tobacco smoke exposure *	Gender, residence duration in a village, tobacco smoke exposure, unknown smoking status, and potential atopy
Sturm et al., 2004 (Carolina) [39]	Grade 7th and 8th	CS	11,378	Current asthma c	Unidentified a	OR = 1.33 (1.22–1.44) for ≤ once per month * OR = 1.48 (1.35–1.62) for once per month * OR = 1.62 (1.49–1.76) for 2–4 times/month * OR = 1.72 (1.60–1.84) for every day *	Gender, race, use of gas stove, parental asthma history, allergy status, socioeconomic status, and other competing sources of tobacco smoke exposure
Arshad et al., 2005 (UK) [76]	10	CO	1,456	Current asthma c	Mother and father smoking a	OR = 1.99 (1.15–3.45) for at child 1 year *	Gender, parental smoking, atopic skin test, recurrent chest infections, food allergy, recurrent chest infections, eczema, food allergy, maternal asthma, sibling asthma, and paternal eczema
Hyvärinen et al., 2005 (USA) [77]	11–13	CO	81	Current asthma c	Unidentified a	OR = 1.354 (0.523–3.509)	Gender and age on entry into study
Magnusson et al., 2005 (Denmark) [78]	14–18	CO	7,844	Ever asthma c	Unidentified a	OR = 1.10 (0.90–1.40)	Socioeconomic group, maternal occupation, maternal age in pregnancy, coffee consumption in pregnancy, parity, breastfeeding, gender, and maternal smoking in late pregnancy
Navon et al., 2005 (USA) [79]	Grade 6th to 12th	CS	3,125	Ever asthma c	Unidentified a	OR = 1.30 (1.10–1.70) for currently live with a smoker *	Race, ethnicity, female guardian education, male guardian education, and gender
Bjerg–Bäcklund et al., 2006 (Sweden) [80]	11–12	CO	1,870	Current asthma c	Mother smoking a	OR = 1.41 (0.95–2.08) for current smoking	Gender, family history of asthma, breast feeding, birth weight, respiratory infections, mother current smoker, cat ever at home, dog ever at home, living in damp house, living in apartment, allergic sensitization, and citrus fruits each week
				Ever asthma c	Mother smoking a	OR = 1.50 (1.06–2.11) for current smoking *
Tsai et al., 2006 (China) [81]	11–12	CS	2,290	Current asthma c	Unidentified a	OR = 0.90 (0.65–1.26)	Residential districts, gender, physician–diagnosed allergy, parental history of respiratory symptom, having pets, having stuffed toys/fur pets, mosquito–repellent fumes, incense–burning fumes, use of dehumidifier, use of air conditioner, having indoor plants chemical vapor, gas leaks, carpeted floor, use of wool blanket, leaky water/water puddle at home, and moldy wall/furniture
				Ever asthma c	Unidentified a	OR = 0.82 (0.55–1.23)
Anthracopoulos et al., 2007 (Greek) [82]	12–16	CO	1,453	Current asthma c	Mother smoking a	OR = 0.93 (0.50–1.71) for 12–14 year of age child at home (n = 695)	Mother allergic, father allergic, premature birth, mother smoked during pregnancy, shares bedroom living in apartment, mother smoking at home, any smoking at home, wheeze ever, wheeze in the last 12 months, asthma diagnosed ever, wheeze during or after exercise in the last 12 months, and hay fever ever hay fever in the last 12 months
						OR = 0.58 (0.31–1.08) for 14–16 year of age child at home (n = 758)
Hublet et al., 2007 (Belgium, Canada, Denmark, Finland, France Netherlands)[83]	15	CS	1,261	Current asthma c	Friend smoking a	OR = 8.39 (5.68–12.39) for ≧ half of friend smoking *	Gender, age and socioeconomic level of the parents
					Mother and father smoking a	OR = 4.98 (3.16–7.85) *
					Unidentified a	OR = 3.11 (2.02–4.78) *
Van De Ven et al., 2007 (Netherlands) [41]	13–14	CO	4,762	Current asthma c	Father smoking a	OR = 1.13 (0.67–1.91)	Age, gender, education, and ethnicity
					Mother smoking a	OR = 0.90 (0.51–1.57)
Tsai et al., 2010 (China) [84]	Grade 7th and 8th	CS	5,019	Current asthma c	Father smoking a	OR = 1.08 (0.76–1.52)	Age, gender, parental education, family history of asthma, family history of atopy, gestational age, and community
					Mother and father smoking a	OR = 1.76 (0.77–4.01)
					Mother smoking a	OR = 1.62 (0.47–5.50)
				Ever asthma c	Father smoking a	OR = 1.03 (0.81–1.32)
					Mother and father smoking a	OR = 1.60 (0.86–2.96)
					Mother smoking a	OR = 1.05 (0.36–3.03)
Chen et al., 2011 (China) [85]	12–14	CC	Cases (n = 193) Controls (n = 386)	Ever asthma c	Unidentified a	OR = 3.15 (1.93–5.15) for ≥ 1 cigarette/day *	Level of education of the parents, family income, family history of asthma, family history of atopy, and in–utero maternal smoking
						OR = 2.63 (1.44–4.78) for 1–2 cigarettes/day *
						OR = 3.87 (2.06–7.27) for ≥ 3 cigarette/day *
Hedman et al., 2011 (Sweden) [43]	16–17	CO	2,805	Ever asthma c	Mother smoking a	OR = 1.42 (1.07–1.87) for 7–8 years child *	Gender, family history of asthma, current place of residence, house dampness and birth weight
						OR = 1.40 (1.04–1.88) for 16–17-year child *
						OR = 1.51 (1.07–2.14) for 7–16 years child *
Lawson et al., 2011 (Canada) [44]	11–15	CS	4,726	Current asthma c	Friend smoking a	OR = 1.37 (1.03–1.82) for smoke occasionally *	Geographic region, obesity status, physical activity, whole milk and vegetable consumption, personal smoking, age group, gender, family affluence scale, and ethnicity
						OR = 1.77 (1.26–2.48) for smoke daily *
					Mother and father smoking a	OR = 1.20 (0.96–1.51)
					Unidentified a	OR = 1.09 (0.89–1.33)
Vlaski et al., 2011 (Macedonia) [86]	13–14	CS	3,026	Ever asthma c	Father smoking a	OR = 1.02 (0.57–1.83)	Gender, current fruit and vegetables and cereals intake, gas/wood cooking and heating exposure at home, current cat and dog, maternal educational level, and siblings
					Mother smoking a	OR = 1.76 (0.96–3.22)
Gudelj et al., 2012 (Croatia) [45]	14–16	CS	4,027	Current asthma c	Unidentified a	OR = 1.406 (0.958–2.062)	Area of residence, gender, allergy household members, each crowded type of fuel used for heating, central/classic gas used for cooking, pets, and dog or cat ownership over 12 months
Banda et al., 2013 (USA) [87]	12–18	CS	246	Current asthma d	Unidentified a	OR = 1.03 (0.30–3.53) for > 2 asthma night awakenings in the past month	Gender and clustering of children in families
Checkley et al., 2013 (Peru) [46]	13–15	CS	1,002	Current asthma c	Unidentified a	OR = 0.50 (0.20–1.30) for atopic asthma child	Total serum IgE, BMI, maternal education, people per house income/per month, concrete floor, gender, and age
						OR = 0.50 (0.10–2.10) for no atopic asthma child
Kim et al., 2013 (Korea) [47]	12–18	CS	3,432	Current asthma c	Unidentified a	OR = 1.11 (0.90–1.37) for 1–6 days exposure at home in the past 7 days	Gender, academic achievements, allowance, duration of diagnosis, asthma treatment, and drinking and smoking days during the past 30 days
						OR = 1.33 (0.99–1.78) for current everyday exposure at home in the past 7 days
Oluwole et al., 2013 (Nigeria) [88]	13–14	CS	1,736	Ever asthma c	Mother and father smoking a	OR = 2.81 (1.46–5.40) *	Urbanicity, home fuel, lorries pass through street, cat in home in past year, active smoking status, and parent smoking (old siblings, young siblings, and all siblings)
Duksal et al., 2014 (Turkey) [89]	13–14	CS	7,082	Ever asthma c	Unidentified a	OR = 1.14 (0.68–1.91) for at home in 2002 (n = 3,004)	Gender, family history of atopy, active smoking, domestic animals at home, stuffed toys, education level of parent, number of people living in home, sharing bedroom, heating system, and bath in sunlight house
						OR = 1.10 (0.89–1.36) for at home in 2008 (n = 4,078)
Akcay et al., 2014 (Turkey) [90]	13–14	CS	9,991	Ever asthma c	Mother smoking a	OR = 1.11 (0.93–1.31)	Gender, history of family atopy, time television watched in a week, use paracetamol in last 12 months, siblings, born in Istanbul, time lived in Istanbul, education level of child’s parents, presence of domestic animals at home, smoking of father, history of tonsillectomy and adenoidectomy
Kanamori et al., 2015 (USA) [91]	12–17	CS	9,658	Ever asthma c	Friend smoking a	OR = 6.23 (4.12–9.41) for female have at least 1 close friend who smokes cigarettes *	Weekly income, education, and self–perceived health status
					Friend smoking a	OR = 4.44 (2.79–7.06) for male have at least 1 close friend who smokes cigarettes *
					Unidentified a	OR = 3.59 (2.17–5.92) for female in the room *
						OR = 2.69 (1.63–4.45) for female in the car *
						OR = 1.29 (0.76–2.18) for male in the room
						OR = 4.67 (2.89–7.54) for male in the car *
Lee et al., 2015 (Korea) [92]	Middle school students	CS	1,395	Ever asthma c	Father smoking a	OR = 1.02 (0.72–1.44) for former smoking OR = 0.99 (0.71–1.37) for current smoking	Age, gender, BMI, a parental history of any allergic diseases, exposure to tobacco smoking, and household income
Dai et al., 2016 (Australia) [93]	18	CO	620	Current asthma c	Father smoking a	OR = 0.77 (0.20–2.93) for boys	Age, height, parental asthma at birth, parental education and adolescent smoking
						OR = 3.45 (1.25–9.54) for girls *
					Mother smoking a	OR = 0.25 (0.03–2.35) for boys
						OR = 1.37 (0.24–7.84) for girls
					Unidentified a	OR = 0.56 (0.15–2.10) for boys
						OR = 3.44 (1.36–8.77) for girls *
Kitsantas et al., 2016 (USA) [94]	13–17	CS	28,807	Current asthma c	Unidentified a	OR = 2.07 (1.15–3.70) for obese child in home*	Gender, race or ethnicity, poverty level, and consistency of health insurance
						OR = 1.23 (0.74–1.97) for obese child not home
Arrais et al., 2017 (Angola) [95]	13–14	CS	3,128	Current asthma c	Father smoking a	OR = 1.25 (0.87–1.78) for at home *	Rhinitis in last 12 months, eczema ever, cooking fuel used, indoor home cooling system, frequency of paracetamol intake, number of siblings, frequency of passage of trucks in front of home, pet, smoking at home, and BMI
					Mother smoking a	OR = 1.51 (0.87–2.62) for at home
Hallit et al., 2017 (Lebanon) [96]	16	CS	527	Ever asthma c	Mother smoking a	OR = 0.295 (0.075–1.157) for mother smoking during infancy	Age, gender, district, education of parent, room number at home, and persons at home
Kim et al., 2017 (Korea) [52]	12–18	CS	216,056	Current asthma c	Unidentified a	OR = 1.11 (1.03–1.21) for 1–2 days a week *	Age, physical exercise, gender, obesity, region of residence, economic level, educational level of father, education level of mother, active, passive smoking, and electronic cigarette
						OR = 1.15 (1.04–1.28) for 3–4 days a week *
						OR = 1.40 (1.28–1.53) for ≧ 5 days a week *
Lalu et al., 2017 (India) [97]	Adolescence	CS	629	Current asthma c	Friend smoking a	OR = 2.16 (1.17–3.97) for smoke cigarette in child presence *	Age, gender, place of residence, type of house, pets at home, usage of wood at home, friends usually smoke in your presence, family history of asthma, ever smoked, and current smoker
					Unidentified a	OR = 1.37 (0.73–2.55) for smoking at home
Park., 2017 (Korea) [98]	14.7	CS	56,840	Current asthma c	Unidentified a	OR = 1.19 (1.01–1.40) for male at home *	Age, family wealth, area of residence, school type, intact family, co–residence with parents, parental education, and perceived academic performance, binge drinking, habitual drug use, and BMI
						OR = 1.04 (0.97–1.11) for in home exposure at least 1 day during the past 7 days among male
						OR = 1.32 (1.10–1.59) for female at home *
						OR = 1.12 (1.05–1.19) for in home exposure at least 1 day during the past 7 days among female *
Szentpetery et al., 2017 (Sweden) [99]	12	CO	2,290	Current asthma c	Unidentified a	OR = 1.40 (0.90–2.10) for in utero or before age 2 years	Parental asthma, gender, obesity at age 4 years, allergic rhinitis at age 4 years, and early–life second–hand smoke
Madani et al., 2018 (North Macedonia and Canada) [100]	13–14	CS	4,226	Ever asthma c	Mother smoking a	OR = 1.23 (0.89–1.72) for current maternal smoking cigarette	Gender, BMI status, paracetamol use, current maternal smoking, gas cooking, electric heating, cat in the home, dog in the home, physical activity, and type of food consumption
Thacher et al., 2018 (Europe) [101]	14–16	CO	7,970	Current asthma c	Unidentified a	OR = 1.15 (1.00–1.31) *	Gender, parental education level, parental allergy, older siblings, breastfeeding, study center, intervention arm, and early day–care attendance
Bayly et al., 2019 (USA) [54]	11–17	CS	11,830	Current asthma c	Unidentified a	OR = 1.19 (1.05–1.35) *	Age, gender, race or ethnicity, metropolitan status, and housing type
						OR = 1.27 (1.11–1.47) for electronic nicotine delivery systems aerosols *
Lee et al., 2019 (Korea) [56]	12–18	CS	14,829	Current asthma c	Unidentified a	OR = 1.63 (1.33–2.00) for ≧ 5 days/ week at home *	Age, gender, obesity, residential area, family economic status, and physical activity
						OR = 1.08 (0.94–1.24) for 1–4 days/week at home
						OR = 1.68 (1.37–2.07) for ≧ 5 days/ week at school *
						OR = 1.42 (1.25–1.62) for 1–4 days/ week at school *
Mallol et al., 2019 (Chile) [102]	13–14	CS	2,736	Current asthma c	Unidentified a	OR = 1.46 (1.09–1.97) for current exposure *	Gender, maternal tertiary education and current tobacco smoking
Skrzypek et al., 2019 (Poland) [103]	13–15	CS	936	Ever asthma c	Unidentified a	OR = 1.71 (0.82–3.58) for living in the vicinity of a main road	Gender, BMI, maternal employment, exposure to environmental tobacco smoke at home, type of heating, traces of moisture or mold in the place of residence, and parental allergy
						OR = 1.59 (0.76–3.33) for traffic intensity near the place of residence
Sordillo et al., 2019 (USA) [104]	11.9–16.6	CO	996	Current asthma c	Mother smoking a	OR = 1.13 (0.69–1.84) for former smoking	Child gender, age at outcome, child’s race/ethnicity, sine and cosine of date of birth, and maternal pre–pregnancy BMI, smoking status, acetaminophen intake, education level, maternal and paternal history of asthma
Alnajem et al., 2020 (Arab) [29]	16–19	CS	1,565	Current asthma c	Unidentified a	PR = 1.49 (1.01–2.23) for e–cigarettes exposure 1–2 days in past 7 days *	Gender, age, cigarette smoking status, electronic cigarette use status, exposure to household secondhand smoke, and exposure to secondhand smoke and/ or secondhand aerosols from electronic cigarettes in public places
						PR = 1.56 (1.13–2.16) for e–cigarettes exposure ≧ 3 days in past 7 days *
Kim et al., 2020 (Korea) [105]	Grade 11th	CO	68,043	Ever asthma c	Unidentified a	OR = 1.05 (0.98–1.13)	Urbanity, perceived economic status, BMI, and environmental tobacco smoke
Vlaski et al., 2020 (Macedonia) [106]	12–15	CS	4,804	Current asthma c	Unidentified a	OR = 2.46 (0.92–6.57) for at home	Year of study, age, gender, current paracetamol use, current cat and dog ownership, frequency of trucks passing, passive smoking at home, TV watching time, mother’s education level, being overweight, and food intake
				Ever asthma c	Unidentified a	OR = 1.63 (0.94–2.84) for at home
Sabeti et al., 2021 (Iran) [63]	14–19	CS	1,459	Current asthma c	Unidentified a	OR = 1.07 (0.68–1.70)	Location, gender, BMI, physical activity, parent’s education and income, residence type, ventilation, pets, fruit, vegetables, fast foods, history of allergic rhinitis, and family history of asthma
				Ever asthma c	Unidentified a	OR = 0.92 (0.47–1.81)
Williams et al., 2023 (USA) [67]	Grade 10th and 12th	CO	113,922	Current asthma c	Unidentified a	OR = 1.16 (1.07–1.27) for current household cigarette smoking *	Age, gender, parental education, race or ethnicity, and 3 types of household cigarette, e–cigarette and cannabis use
						OR = 1.31 (1.19–1.45) for current household vaping (e–cigarette) *
				Ever asthma c	Unidentified a	OR = 1.08 (1.03–1.13) for current household cigarette smoking *
						OR = 1.02 (0.96–1.09) for current household vaping (e–cigarette)
Satybaldiyeva et al., 2024 (USA) [107]	Grade 8th, 10th and 12th	CS	158,937	Current asthma c	Unidentified a	OR = 1.01 (0.92–1.10) for combustible tobacco exposure	Grade, gender, race or ethnicity, urbanicity, and secondhand combustible tobacco smoke
Yao et al., 2024 (USA) [68]	12–17	CO	9,422	Current asthma c	Unidentified a	OR = 1.51 (1.00–2.28) for living with someone who uses tobacco	Age, gender, race, household income, parents’ education, home tobacco rules, and tobacco use
						OR = 1.87 (0.54–6.55) for unknown

^a Questionnaire or interview; ^b Blood; ^c Parent- or child self-reported physician diagnosis; ^d Medical records; BMI Body mass index, OR Odds ratio, HR Hazard ratio, RR Relative risk, PR Prevalence ratio, IRR Incidence rate ratio, POR Prevalence odds ratio (POR), CS Cross–sectional study, CO Cohort study, CC Case–control study; * p value < 0.05

Risk of bias

Quality assessment scores ranged from 4 to 11. Of the included studies, 67 were rated as fair quality (scores 6–10), and five were classified as poor quality (scores 1–5). Only five studies met the criteria for high quality (score = 11), attributed primarily to their cohort design, low loss to follow-up, appropriate adjustment for key confounders, and adequate exposure timing. Common methodological limitations included the absence of sample size justification, reliance on self-reported questionnaires for exposure and outcome assessment, lack of repeated measurements, and suboptimal study design. The overall risk of bias was generally low, although concerns regarding external validity were common across studies. Detailed assessments are provided in Supplementary file 4.

Certainty of evidence

According to the GRADE assessment, the certainty of evidence for the association between both active and passive smoking and asthma in adolescents was rated as very low. Downgrading was chiefly attributed to the inclusion of both cohort and cross-sectional study designs, with cross-sectional studies unable to establish temporal sequence between exposure and outcome. Additionally, several studies lacked adequate adjustment for important confounders, and exposure assessment was predominantly based on self-reported questionnaire data, further limiting the certainty of the evidence (Supplementary file 5).

Association between active smoking and adolescent asthma

Twenty-six studies were eligible for inclusion in qualitative synthesis, comprising 88.46% cross-sectional and 11.54% cohort designs. Findings were mixed, with some studies reporting a significant association between smoking and increased asthma risk, while others observed no statistically significant relationship. For example, Montefort et al. [36] reported a significant association (aOR = 1.46; 95% CI: 1.12–1.91), consistent with Wang et al. [38] (aOR = 1.29; 95% CI: 1.17–1.42) and Sturm et al. [39] (aOR = 1.10; 95% CI: 1.04–1.16). In contrast, Hedman et al. [43], Han et al. [55], and Cherian et al. [61] found no significant association (aOR = 1.36; 95% CI: 0.93–1.98; aOR = 1.12; 95% CI: 0.91–1.38; and aOR = 1.13; 95% CI: 0.85–1.50, respectively). Other studies, including those by Larsson [34], Jones et al. [40], Bae et al. [42], and Schweitzer et al. [53], yielded conflicting results, potentially due to variation in exposure timing or duration (Table 1).

In the meta-analysis, the random-effects model demonstrated that active smoking was significantly associated with increased asthma risk (aOR = 1.16; 95% CI: 1.13–1.20; I² = 51.8%; p value < 0.001) (Fig. 2).

Fig. 2.

Meta-analysis of the association between active smoking and adolescent asthma

Association between passive smoking and adolescent asthma

A total of 56 studies met the eligibility criteria for inclusion in qualitative synthesis. The majority employed cross-sectional designs and reported a significant association between passive smoking exposure and increased asthma risk. For example, the studies were conducted in the United Kingdom [76], Sweden [80], Iran [75], the United States [54, 67, 79, 94], India [97], Chile [102], Nigeria [88], China [38, 85], and multiple European countries [83, 101]. Several studies conducted in Korea [52, 56, 98] and Canada [44] also supported this association. However, also subsets of studies reported no statistically significant relationship between passive smoking and asthma [35, 69, 78, 95] (Table 2).

Meta-analysis using a random-effects model revealed a significant association between passive smoking and asthma risk (aOR = 1.23; 95% CI: 1.17–1.29; I² = 75.6%; p value < 0.001) (Fig. 3).

Fig. 3.

Meta-analysis of the association between passive smoking and adolescent asthma

Subgroup and sensitive analysis

Substantial heterogeneity was observed in studies on active smoking (I² = 51.8%), prompting subgroup analyses by type (cigarette vs. e-cigarette), study design (cohort or cross-sectional), asthma outcome (current or ever asthma), and exposure timing (current, ever, or unspecified) (Table 3; Fig. 2). Similarly, passive smoking studies showed considerable heterogeneity (I² = 75.6%), necessitating stratification by source of exposure (e.g., father, mother, both parents, friends, or unspecified), study design (cohort, case-control, or cross-sectional), and asthma classification (current or ever asthma) (Table 3; Fig. 3).

Table 3.

Subgroup and sensitivity analysis

Group of smoking	Condition for analysis		Studies included	Fixed-effect meta།analysis		Random-effect meta།analysis
				aOR	95%CI	aOR	95%CI
Active smoking	All	All	26	1.12	1.10-1.14	1.16	1.13-1.20
	Type of exposure	Cigarette	20	1.11	1.08-1.14	1.19	1.13-1.26
		E-cigarette	15	1.13	1.10-1.16	1.13	1.10-1.16
	Study design	Cohort study	3	1.15	0.80-1.51	1.13	0.66-1.61
		Cross-sectional study	23	1.12	1.10-1.14	1.16	1.12-1.20
	Outcome	Current asthma	19	1.15	1.12-1.17	1.20	1.14-1.27
		Ever asthma	12	1.14	1.10-1.18	1.16	1.11-1.21
	Time of exposure	Cigarette- current smoking	17	1.18	1.14-1.23	1.30	1.21-1.39
		Cigarette- ever smoking	11	1.06	1.02-1.09	1.11	1.04-1.18
		Cigarette- unidentified	5	1.30	1.25-1.36	1.16	0.96-1.35
		E-cigarette- current smoking	11	1.16	1.11 -1.22	1.20	1.09-1.30
		E-cigarette- ever smoking	11	1.12	1.05-1.19	1.12	1.04-1.21
		E-cigarette- unidentified	2	1.11	1.06-1.16	1.11	1.06-1.16
		Cigarette + E-cigarette- current smoking	20	1.17	1.14-1.21	1.25	1.18-1.32
		Cigarette + E-cigarette- ever smoking	15	1.07	1.04-1.10	1.12	1.06-1.17
		Cigarette + E-cigarette- unidentified	5	1.20	1.17-1.24	1.14	1.03-1.26
Passive smoking	All	All	56	1.14	1.12-1.16	1.23	1.17-1.29
	Type of exposure	Father smoking	9	1.00	0.89-1.11	1.00	0.89-1.11
		Mother smoking	16	1.16	1.05-1.27	1.20	1.02-1.37
		Mother and father smoking	6	1.62	1.44-1.80	1.85	1.33-2.36
		Friend smoking	4	1.72	1.40-2.03	3.70	1.63-5.78
		Unidentified	37	1.14	1.12-1.15	1.23	1.16-1.30
	Study design	Cohort study	16	1.09	1.06-1.12	1.13	1.05-1.21
		Cross-sectional study	38	1.16	1.14-1.18	1.26	1.19-1.33
		Case-control study	2	3.09	2.10-4.09	3.09	2.10-4.09
	Outcome	Current asthma	40	1.17	1.15-1.19	1.25	1.18-1.33
		Ever asthma	26	1.07	1.04-1.10	1.08	1.04-1.12

aOR adjusted odds ratio

Subgroup analyses showed significant associations between active smoking and adolescent asthma: cigarette use (aOR 1.19; 95% CI: 1.13–1.26; I²=59.4%) and e‑cigarette use (aOR 1.13; 95% CI: 1.10–1.16; I²=39.7%). Current smoking (aOR 1.25; 1.18–1.32; I²=52.2%) conferred higher risk than ever smoking (aOR 1.12; 1.06–1.17; I²=34.1%), with current cigarette use strongest (aOR 1.30; 1.21–1.39; I²=52.2%). Risks were elevated for current asthma (aOR 1.20; 1.14–1.27; I²=70.9%) and ever asthma (aOR 1.16; 1.11–1.21; I²=40.5%). Cross‑sectional studies showed significant associations for combined cigarette/e‑cigarette use (aOR 1.16; 1.12–1.20; I²=65.1%).

In passive smoking subgroup analyses, significant associations with adolescent asthma were seen across study designs: cross-sectional (aOR 1.26; 95% CI: 1.19–1.33; I²=79.0%), cohort (aOR 1.13; 95% CI: 1.05–1.21; I²=49.5%), and case-control (aOR 3.09; 95% CI: 2.10–4.09; I²=0.0%). By exposure source, mother smoking (aOR 1.20; 95% CI: 1.02–1.37; I²=31.4%), both parents (aOR 1.85; 95% CI: 1.33–2.36; I²=74.4%), friend smoking (aOR 3.70; 95% CI: 1.63–5.78; I²=86.1%), and unidentified sources (aOR 1.23; 95% CI: 1.16–1.30; I²=82.1%) were significant. By asthma status, associations held for current asthma (aOR 1.25; 95% CI: 1.18–1.33; I²=80.3%) and ever asthma (aOR 1.08; 95% CI: 1.04–1.12; I²=56.9%).

Publication bias and funnel plot

Publication bias was assessed using funnel plots and Egger’s linear regression test. For active smoking, funnel plots were moderately symmetrical. Egger’s test was non‑significant for overall smoking (p = 0.15) and e‑cigarette use (p = 0.14). Cigarette‑specific analyses showed no evidence of small‑study effects (p = 0.573), supporting the robustness of the pooled findings. For passive smoking, Egger’s tests were non‑significant for all identified sources: fathers (p = 0.66), mothers (p = 0.89), parents (p = 0.62), and friends (p = 0.69); funnel plots were symmetrical, indicating no publication bias. For unspecified sources, however, Egger’s test showed significant asymmetry (p < 0.05). Trim‑and‑fill imputed five missing studies, modestly reducing the aOR from 1.36 to 1.30, suggesting limited impact of publication bias on the overall findings (Fig. 4).

Fig. 4.

Funnel plots for active and passive smoking

Discussion

To our knowledge, this is the first systematic review and meta-analysis to concurrently examine differential effects by product type and source of passive tobacco smoke exposure on adolescent asthma. Three key findings emerged: both cigarette and e-cigarette use were associated with asthma; among passive exposures, friend smoking showed the strongest association, followed by maternal smoking, while paternal smoking was not significant.

Meta-analysis findings indicate that both cigarette and e-cigarette use are independently associated with adolescent asthma (cigarette: aOR = 1.19; 95% CI: 1.13–1.26; e-cigarette: aOR = 1.13; 95% CI: 1.10–1.16). These associations are underpinned by distinct toxicological profiles. With more than 7,000 compounds, including particulate matter, volatile organic compounds, and carcinogens such as benzene and formaldehyde [108]—cigarette smoke contributes to asthma pathogenesis via multiple biological pathways. First, cigarette smoke induces pro-inflammatory cytokines (e.g., IL-8, IL-17), driving neutrophil and eosinophil recruitment and airway inflammation [109–111]. Second, it disrupts immune homeostasis by promoting Th2-dominant responses, increasing IgE levels and mast cell [112]. Third, chronic exposure impairs epithelial integrity, weakens airway barrier function, and induces airway hyperresponsiveness [113, 114]. Cigarette smoke also generates high levels of reactive oxygen species, depletes antioxidants such as glutathione, and damages airway mucosa [115], promoting airway remodeling, including smooth muscle hypertrophy and fibrosis [116]. Toxicants such as cyanide and heavy metals enhance the virulence of respiratory pathogens (e.g., Streptococcus pneumoniae, Haemophilus influenzae) by promoting biofilm formation [117, 118], while tar impairs mucus clearance, prolonging pathogen retention [119]. These processes amplify the infection–inflammation cycle, increasing the frequency and severity of asthma exacerbations. Nicotine and its metabolites further activate cholinergic receptors, enhancing myosin phosphorylation via Ca²⁺/RhoA signaling and increasing goblet cell mucus secretion, thereby contributing to airway obstruction and acute exacerbations [120–122].

Similarly, e-cigarette use has been associated with asthma in adolescents, with growing evidence supporting multiple biologically plausible mechanisms. First, e-cigarette aerosols impair airway epithelial integrity, reducing ciliary beat frequency (up to 53%) and disrupting tight junction proteins (occludin-1, ZO-1), thereby increasing permeability to allergens and delaying repair for up to 72 h [123]. Second, metal particles from heating coils, particularly iron, induce oxidative stress via MMP12 pathways; animal models show upregulation of antioxidant response elements and MUC5AC expression, promoting mucus hypersecretion and airway remodeling [124]. Third, constituents such as nicotine and flavorings trigger pro-inflammatory cytokine responses (IL-6, IL-8) and elevate oxidative stress markers [125]. Concurrently, flavorings (e.g., cinnamon, menthol) disrupt oral commensal streptococci, increasing bacterial killing and surface hydrophobicity, thereby altering biofilm formation and microbial homeostasis [126]. Proteomic analyses further reveal sex-dependent alterations in pulmonary metabolic and lipid pathways following adolescent exposure [127]. Consistently, epidemiological studies demonstrate a dose–response relationship, with daily use conferring the highest asthma risk [128].

Heterogeneity was substantial among cigarette studies (I² = 59.4%) and moderate among e-cigarette studies (I² = 39.7%). For cigarettes, variability likely arose from differences in study design, exposure timing, asthma definitions, and dose–response effects. Subgroup analysis by exposure timing reduced I² to 52.2% and 42.7%, indicating that smoking duration contributed substantially to heterogeneity. Additional sources include population differences and residual confounding. Lower heterogeneity in e-cigarette studies may reflect more consistent exposure assessment in recent literature; however, methodological inconsistencies remain, particularly in quantifying aerosol constituents and classifying device types. Variations in exposure metrics (e.g., pack-years vs. vaping frequency) and the lack of standardized biomarkers for nicotine delivery further complicate dose–response comparisons, underscoring the need for harmonized research protocols.

Regarding passive smoking, the systematic review and meta-analysis found that exposure to friend smoking was associated with the highest risk of adolescent asthma (aOR = 3.70), compared with maternal smoking (aOR = 1.20), both parents smoking (aOR = 1.85), and paternal smoking (aOR = 1.00). It is possible that adolescents spend increasing time with peers and less time with parents as they age, which amplifies the influence of friends during this developmental period. Another possibility is that this association probably reflects an indirect behavioral pathway. Peer smoking acts as a social catalyst, increasing adolescents’ own smoking initiation and thus active exposure to tobacco toxicants implicated in asthma pathogenesis. Previous studies demonstrated that adolescents with smoking peers were significantly more likely to initiate smoking, an effect that in some cases exceeded that of parental smoking [129, 130]. Similarly, recent meta-analyses show substantial effect sizes for peer influence across various forms of adolescent substance use, including tobacco, consistent with processes of social conformity and behavioral alignment within peer groups. Perceived peer smoking norms and the number of smoking friends also independently predict smoking onset, underscoring the role of social modeling and normative influence [131, 132]. These findings align with social learning theory, which posits that adolescents adopt behaviors modelled by influential peers to gain social acceptance [133–135]. Notably, individuals were more likely to quit smoking if others in their network also quit, reinforcing the notion that smoking behavior is socially contagious and shaped by network dynamics [136]. Together, these observations suggest that the strong association between friend smoking and adolescent asthma is driven primarily by the behavioral pathway of smoking initiation, rather than by direct passive exposure to second-hand smoke from peers. However, interpretation of the pooled effect for friend smoking should be cautious. Only four moderate-quality studies were included, limiting generalizability. Inconsistent adjustment for confounders, particularly socioeconomic status and parental monitoring, may also influence the observed effect. Substantial heterogeneity (I² = 86.1%) likely reflects cross-cultural differences in peer dynamics, family structure, exposure measurement, and smoking timing. Future studies should use longitudinal designs with repeated measures of peer influence and smoking trajectories, ideally complemented by qualitative assessments of contextual moderators (e.g., school tobacco policies, community smoking prevalence). These findings also suggest that school-based interventions targeting peer networks may be more effective than exclusively family-centered approaches for adolescent tobacco prevention.

Additionally, this study confirmed that both mother and father smoking is a significant risk factor for adolescent asthma, while mother smoking alone demonstrated a strong significant association. In contrast, father smoking alone showed no association (aOR = 1.00; 95% CI: 0.89–1.11). This differential pattern likely reflects distinct exposure pathways and methodological considerations rather than inherent differences in biological effects. Maternal smoking confers risk through multiple exposure windows, including prenatal and postnatal periods. Prenatal exposure has been linked to epigenetic reprogramming affecting immune regulation, including Th1/Th2 imbalance, which can predispose offspring to allergic asthma [19]. Furthermore, maternal smoking may have transgenerational impacts via mitochondrial DNA alterations in oocytes, as supported by animal and human studies [137, 138]. These mechanisms provide a plausible basis for the strong and consistent association between maternal smoking and childhood respiratory outcomes observed across epidemiological studies. In contrast, paternal smoking is primarily a source of postnatal exposure, as it does not directly involve the intrauterine environment. The absence of a statistically significant association should be interpreted with caution. Collectively, the absence of a significant association between paternal smoking alone and adolescent asthma does not preclude an effect but highlights the need for more refined exposure assessment. Future studies should incorporate objective biomarkers (e.g., cotinine) and life-course approaches to disentangle exposure timing and sources, and to assess the independent effects of maternal versus paternal smoking on adolescent asthma.

Several limitations should be considered when interpreting the findings of this systematic review and meta-analysis. First, most included studies relied on self-reported measures of tobacco exposure, with only two studies employing biomarker-based assessments. Self-reported exposure is susceptible to recall bias and misclassification, limiting the ability to establish accurate dose-response relationships [139]. Second, most included studies employed cross-sectional designs, which preclude the establishment of temporal or causal inference due to the simultaneous assessment of exposure and outcome [140, 141]. Third, substantial heterogeneity was observed across several exposure categories, this heterogeneity likely reflects differences in study design, population characteristics, exposure measurement, outcome definition, and statistical modeling approaches across studies. Fourth, the substantial variation in covariate adjustment across studies limits comparability and leaves open the possibility of residual confounding, particularly from unmeasured factors such as socioeconomic status, household environmental exposures, and co-exposure to multiple tobacco products [142–144]. Fifth, although a comprehensive search strategy was employed without geographical restrictions, the analysis may have omitted relevant non-English publications or unpublished studies due to language and publication bias. Moreover, studies reporting significant associations may be more likely to be published than those with null findings [145, 146], potentially inflating the magnitude of pooled estimates.

Reflecting these methodological limitations, the GRADE framework rated the certainty of evidence as low to very low for the associations examined. This rating highlights the need for future prospective studies with biomarker-validated exposure assessment and standardized outcome definitions. Nevertheless, despite this low certainty, the consistent direction of effect across subgroup analyses supports the robustness of these associations and underscores their relevance for informing future research priorities.

Conclusion

This systematic review and meta‑analysis suggest that active and passive tobacco smoke exposure were associated with adolescent asthma risk. Friend smoking showed the strongest association. High heterogeneity, self‑reported exposure, and cross‑sectional designs limit causal inference, with very low-grade certainty. Standardized exposure assessment and biomarker‑validated longitudinal studies are warranted.

Abbreviations

OR

Odds ratio

HR

Hazard ratio

PR

Prevalence ratio

IRR

Incidence rate ratio

RR

Relative risk

POR

Prevalence odds ratio

95%CI

95% confidence intervals

ENDS

Electronic nicotine delivery systems

GRADEpro GDT

Grading of Recommendations, Assessment, Development and Evaluations pro Guideline Development Tool

WHO

World Health Organization

aOR

Adjusted odds ratio

aPR

Adjusted prevalence ratio

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

Protocols MOOSE

Meta-analysis of Observational Studies in Epidemiology

PROSPERO

International Prospective Register of Systematic Reviews

Mesh

Medical Subject Headings

ISAAC

International Study of Asthma and Allergies in Childhood

NHLBI

National Heart, Lung, and Blood Institute

VOCs

Volatile organic compounds

AHR

Airway hyperresponsiveness

ROS

Reactive oxygen species

*

p value < 0.05

CS

Cross-sectional study

CO

Cohort study

CC

Case-control study

SE

Standard error

Authors’ contributions

Conceptualization, R.S, and W.W.S.; methodology, R.S., and W.W.S.; software, W.W.S.; validation, R.S., W.W.S., W.K, S.K.; formal analysis, R.S., W.W.S., W.K, S.K.; investigation. R.S., and W.W.S.; resources, R.S., W.W.S; data curation, R.S.; writing—original draft preparation W.W.S.; writing-review and editing, R.S., and W.W.S.; visualization, R.S., and W.W.S.; supervision, R.S.; project administration, R.S. All authors have read and agreed to the published version of the manuscript.

Funding

This study received no external funding.

Data availability

The data used in the study can be made available from Ratana Sapbamrer (corresponding author) on reasonable request.

Declarations

Ethics approval and consent to participate

The study was registered under PROSPERO (CRD42024622246 on 4 December 2024.). Ethics approval and consent to participate Ethics approval: This study has been approved by the Research Ethics Committee, Faculty of Medicine, Chiang Mai University, Thailand (EXEMPTION 0391/2025).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.