# paper.pdf ## Metadata - PDFFormatVersion=1.7 - Language=en-US - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260514123812+05'30' - Creator=Adobe InDesign 19.5 (Windows) - ModDate=D:20260514123819+05'30' - Producer=Adobe PDF Library 17.0 - Title=Early childhood asthma and adiposity until adolescence: A prospective birth cohort study using body fat percentage - Trapped=False - Custom.WPS-ARTICLEDOI=10.1111/pai.70360 - Custom.WPS-JOURNALDOI=10.1111/(ISSN)1399-3038 - Custom.WPS-PROCLEVEL=3 - jav:journal_article_version=VoR - prism:doi=10.1111/pai.70360 - pdfx:wps-articledoi=10.1111/pai.70360 - pdfx:wps-journaldoi=10.1111/(ISSN)1399-3038 - pdfx:wps-proclevel=3 - dc:title=Early childhood asthma and adiposity until adolescence: A prospective birth cohort study using body fat percentage ## Contents ### Page 1 Pediatr Allergy Immunol. 2026;37:e70360. | 1 of 10https://doi.org/10.1111/pai.70360wileyonlinelibrary.com/journal/paiReceived: 11 February 2026 | Revised: 15 April 2026 | Accepted: 17 April 2026DOI: 10.1111/pai.70360O R I G I N A L A R T I C L E Early childhood asthma and adiposity until adolescence: A prospective birth cohort study using body fat percentage Mónica Rodrigues1,2 | Margarida Henriques 3 | Susana Santos1 | Pedro Moreira1,4 |André Moreira1,2,3,4,5This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution inany medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2026 The Author(s). Pediatric Allergy and Immunology published by European Academy of Allergy and Clinical Immunology and John Wiley & Sons Ltd.1EPIUnit ITR – Instituto de Saúde Públicada Universidade do Porto (ISPUP),Universidade do Porto, Porto, Portugal 2 Basic and Clinical Immunology, Department of Pathology, Faculty of Medicine, University of Porto, Porto, Portugal 3 Faculty of Medicine, University of Porto, Porto, Portugal 4Faculty of Nutrition and Food Sciences,University of Porto, Porto, Portugal5 Immuno-Allergology Department, CentroHospitalar São João, Porto, PortugalCorrespondenceMónica Rodrigues, EPIUnit, Institute of Public Health of the University of Porto(ISPUP), Universidade do Porto, Rua dasTaipas, n° 135, 4050-600 Porto, Portugal.Email: monicarodrigues@med.up.ptFunding informationFundação para a Ciência e a Tecnologia,Grant/Award Number: 2023.02362.BD,UID/4750/2025, LA/P/0064/2020 and2024.10106.CEECINDEditor: Ömer KalayciAbstract Background: Asthma and obesity frequently coexist in childhood, yet whether asthmapredisposes to later adiposity remains underexplored, particularly using measuresbeyond body mass index (BMI). Considering this, we aimed to examine the associationbetween asthma in early childhood and subsequent adiposity through adolescence,assessed using BMI and body fat percentage, and to evaluate sex-specific effects.Methods: This prospective study included 4717 participants from the Portuguesepopulation-based Generation XXI birth cohort. Physician-diagnosed asthma at age4 years was the exposure. Incident obesity and overweight/obesity were assessedat ages 7-, 10-, and 13-years using BMI and body fat percentage estimated bybioelectrical impedance analysis and classified according to age- and sex-specificpercentiles. Children who were already overweight or obese at age 4 were excludedfrom the respective analysis. Cox proportional hazards models were used to estimatehazard ratios (HR) and 95% confidence intervals (CI), adjusting for confounders. Effectmodification by sex was tested, and sex-stratified analyses were performed.Results: Asthma was associated with a higher risk of incident obesity defined by body fatpercentage in adjusted analyses (aHR = 1.91, 95% CI: 1.16–3.16), whereas associationswith BMI-defined obesity were not statistically significant. Sex-stratified analysesshowed that these associations were confined to girls, who had increased risk of obesity(aHR = 1.97, 95% CI: 1.19–3.26) and overweight/obesity (aHR = 1.75, 95% CI: 1.02–3.01)defined by body fat percentage. No significant associations were observed among boys.Conclusions: Asthma in early childhood was associated with increased adiposity inadolescence assessed by body fat percentage, particularly among girls. These findingssuggest that children with asthma may develop excess fat accumulation not fullycaptured by BMI and support the integration of body composition and metabolic risk assessment into pediatric asthma care.K E Y W O R D Sadiposity, adolescence, asthma, body fat percentage, childhood, obesity, overweightMónica Rodrigues and Margarida Henriques share first authorship. ### Page 2 2 of 10 |RODRIGUES et al.1 | I NTRO D U C TI O NAsthma and obesity are among the most prevalent chronic conditionsin childhood and adolescence, both showing rising incidence and sub- stantial long-term health consequences.1,2 Their frequent coexistencehas driven increasing interest in shared biological and behavioral path-ways originating early in life. 1, 3–6 While obesity is a well-establishedrisk factor for asthma, the reverse temporal relationship, whether asthma in early childhood predisposes to later obesity, remains com- paratively underexplored. Several mechanisms may explain how asthma could influ- ence subsequent excessive bodyweight or adiposity. Respiratorysymptoms, exercise-induced bronchoconstriction, and fear of ex-acerbations may limit physical activity, particularly when asthma is poorly controlled, promoting sedentary behaviors. 7 Dietarypatterns may also contribute, as energy- dense diets have beenassociated with both asthma and obesity.1,8 Dietary patterns may mediate part of the association if asthma leads to lifestyle changes that influence diet. In addition, asthma treatment has been implicated in weight gain trajectories, although evidence is inconsistent and may reflect disease severity rather than treat-ment effects.3,9Beyond behavioral factors, biological pathways linking asthmato adiposity have been proposed. Asthma is increasingly recog- nized as a condition with systemic inflammatory features, andchronic airway inflammation may contribute to metabolic dys-regulation.10,11 Adipokines such as leptin have been implicated inboth asthma pathophysiology and fat accumulation. Leptin mayincrease airway inflammation and influence T- cell function, withone study suggesting higher leptin levels in children with asthmaindependent of body mass index (BMI).12 Moreover, sharedearly-life determinants—including maternal adiposity, gestationalweight gain, smoking during pregnancy, and other environmentalexposures may influence the development of both asthma and ex-cess adiposity.1,3,4 Longitudinal evidence studying asthma as a determinant of later obesity remains limited.3,5,6,13 Existing cohort studies arerelatively few, often have short follow-up periods, 4,5 and rely predominantly on BMI to define obesity. 3–6, 13 BMI does not dis-tinguish fat from lean mass and may misclassify adiposity during growth and pubertal development.14 Moreover, potential sex-specific associations have not been consistently examined, de- spite known differences in asthma expression, physical activitypatterns, and hormonal influences between boys and girls.In this context, we investigated the prospective association be- tween physician-diagnosed asthma at 4 years of age and subsequentadiposity up to 13 years in a large population-based birth cohort. Weassessed incident obesity and combined overweight/obesity usingboth BMI- and body fat percentage-based definitions. We also ex- plored sex-specific associations to better characterize adiposity riskin children with early-life asthma.2 | M E TH O DS 2.1 | Study design and participantsThis prospective study was conducted within the Generation XXIpopulation-based birth cohort from Portugal.15,16 Between 2005and 2006, all mothers delivering live-born infants in five public ma-ternity units in the metropolitan area of Porto were invited to par- ticipate. A total of 8495 mothers (92%) and 8647 newborns wereenrolled at baseline. Detailed maternal, obstetric, demographic, and lifestyle information was collected through structured interviews, and neonatal data were obtained from clinical records.The entire cohort was invited to the follow-up study waves at fourdifferent time-points considered for the present study: when the chil- dren were 4 years old, 2009–2011 (n = 7459, 86%); 7 years old, 2012– 2014 (n = 6889, 80%); 10 years old, 2015–2016 (n = 6397, 74%); andat 13 years old, 2018–2020 (n = 4640, 54%). Participation was lowerat the 13-year visit because follow-up was stopped early due to theCOVID-19 pandemic. Children without information on asthma statusor adiposity measures (BMI or body fat percentage) at age 4 were ex- cluded. To ensure the assessment of incident outcomes over follow-up,children who already presented obesity or overweight/obesity at age4 were excluded from the respective analyses. Specifically, when obe- sity was the outcome, children with obesity (BMI z-score >2 or ≥95percentile for body fat percentage for age and sex) at age 4 were ex-cluded from the respective analysis; when overweight/obesity was theoutcome, children with overweight (BMI z-score >1 and ≤2 or ≥85thand <95th percentile for body fat) or obesity at age 4 were excluded.Only children with at least one subsequent follow-up assessment withavailable adiposity data were included.Of the 8647 children enrolled at birth, 4717 (54.6%) met the in- clusion criteria and were retained in the analyses (Figure 1). Differences between included and excluded participants wereexamined and, although some characteristics differed statistically,effect sizes were small or negligible, suggesting limited potential forselection bias (Table 1).2.2 | Exposure: Asthma at age 4 yearsThe exposure of interest was physician-diagnosed asthma, which wasobtained based on the International Study of Asthma and Allergies inKey messageAsthma in early childhood seems to be associated with in-creased body fat in adolescence, particularly among girls. Integrating obesity risk awareness into pediatric asthma caremay help reduce cardiometabolic burden in the long-term. ### Page 3 | 3 of 10RODRIGUES et al.Childhood (ISAAC) standardized questionnaire, considering a positiveanswer to the following question “Has your child ever been diagnosedwith asthma by a physician?” as reported at the 4-year- old visit.2.3 | Outcomes: Adiposity measures2.3.1 | Body mass index (BMI) Height and weight were measured at 4, 7, 10, and 13 years of ageby trained researchers using standardized procedures. BMI was cal- culated as weight (kg) divided by height squared (m2) and convertedinto age- and sex-specific BMI z-scores.17,18 At the 4-year follow-up,when children's ages ranged from 44 to 71 months, BMI z-scores werecalculated using the World Health Organization (WHO) growth stand-ards18 or the WHO growth reference for school-aged children,17 asappropriate for each child's age. For the subsequent visits (7, 10, and13 years), BMI z-scores were calculated according to the WHO refer- ence for school-aged children and adolescents.17 BMI z-scores werethen classified as underweight (z-score <−2), normal weight (z-score≥−2 and ≤+1), overweight (z-score >+1 and ≤+2), or obesity (z-score>+2), according to WHO classification.172.3.2 | Body fat percentageBody fat percentage was assessed using tetrapolar bioelectricalimpedance analysis (BIA) with the Akern BIA Anniversary 101 de-vice. The device was calibrated according to manufacturer specifi-cations. Measurements were performed with participants wearing light clothing and no metal accessories, after fasting and bladder emptying.Fat-free mass was estimated using the equation developed bySchaefer et al.19 which incorporates bioelectrical impedance, height, and age and was validated in pediatric populations. Fat mass wascalculated as the difference between body weight and fat-free mass.Body fat percentage categories (underfat defined as <2nd percentile,normal fat as 2nd-85th percentile, overfat as 85th-95th percentile,and obese as ≥95th percentile) were defined using age- and sex-specific reference percentiles proposed by McCarthy et al. 20 Due to the lack of body fat reference curves for the entire age period, forchildren younger than 5 years, the lowest age category available inthe reference was applied. Sensitivity analyses were conducted re- stricting the sample to children aged ≥5 years at the 4-year follow-upfor body fat outcomes to ensure robustness of findings (analysis onTable S7).F I G U R E 1 Flowchart of the study participants. *The analytic sample included 4717 children; however, the number of participantsincluded in each analysis differs because eligibility depended on the availability of outcome-specific data (BMI or body fat percentage)across follow-up assessments. Incident obesity defined by BMI (Inc Ob BMI), incident obesity defined by body fat percentage (Inc Ob Fat%),incident overweight or obesity defined by BMI (Inc OwOb BMI), and incident overweight or obesity defined by body fat percentage (IncOwOb Fat%). ### Page 4 4 of 10 |RODRIGUES et al.TA B L E 1 Maternal and child characteristics according to included and excluded participants in the Generation XXI cohort.Sample size, n Included (n = 4717, 54.6%) Excluded (n = 3930, 45.4%) p-ValueMother age, years, mean ± SD 8641 29.77 ± 5.24 28.06 ± 5.87 <.001 Maternal education, n (%)<=9y 8634 1992 (42.2%) 2255 (57.6%) <.001>=10 e < =12 338 (7.2%) 315 (8.0%)>12 2387 (50.6%) 1347 (34.4%)Parity, multiparous, n (%) 8487 1921 (41.5%) 1730 (44.9%) .001GWG categories, n (%)Insufficient 7904 1134 (25.7%) 898 (25.7%) .078Adequate 1670 (37.9%) 1250 (35.7%) Excessive 1602 (36.4%) 1350 (38.6%)Maternal BMI Underweight weight, n (%) 8348 184 (4.0%) 158 (4.3%) .003Normal weight, n (%) 3089 (66.4%) 2332 (63.1%)Overweight, n (%) 998 (21.5%) 833 (22.5%)Obese, n (%) 380 (8.2%) 374 (10.1%)Maternal smoking in pregnancy, n (%) 8489 931 (20.1%) 1024 (26.6%) <.001 Breastfeeding 7398 4355 (93.2%) 2522 (92.7%) .480Sex, n (%)Female 8647 2291 (48.6%) 1946 (49.5%) .380 Male 2426 (51.4%) 1984 (50.5%)Birth weight<2500 g 8647 462 (9.8%) 343 (8.7%) .089> = 2500 g 4255 (90.2%) 3587 (91.3%) Asthma 4y, n (%) 6988 181 (3.8%) 124 (5.5%) .002Physical Activity 4y, n (%)Yes 7383 3268 (69.6%) 1871 (69.6%) 1.000No 1427 (30.4%) 817 (30.4%) Physical Activity 7y, n (%)Yes 4941 2077 (55.3%) 653 (55.1%) .878No 1678 (44.7%) 533 (44.9%) Physical Activity 10y, n (%)Yes 6389 2741 (62.6%) 1176 (58.4%) .001No 1635 (37.4%) 837 (41.6%) Physical Activity 13y, n (%)Yes 4561 2034 (59.1%) 634 (56.6%) .140No 1410 (40.9%) 487 (43.4%)Dietary Pattern, n (%)Healthy 5831 1989 (42.7%) 442 (37.7%) .006Snacking 691 (14.8%) 178 (15.2%) Energy-dense 1979 (42.5%) 552 (47.1%)Note: Effect sizes for variables showing statistically significant differences were: maternal age (Cohen's d = 0.310), maternal education (Cramér'sV = 0.165), maternal BMI category (Cramér's V = 0.041), parity (φ = 0.035), maternal smoking during pregnancy (φ = 0.078), physician-diagnosedasthma at 4 years (φ = 0.037), dietary patterns at 4 years (Cramér's V = 0.042), and physical activity at 10 years (φ = 0.040). Abbreviations: BMI, body mass index; GWG, gestational weight gain; n, number of participants; SD, standard deviation; y, years. ### Page 5 | 5 of 10RODRIGUES et al.2.4 | Definition of events and time-to-eventoutcomesFour time-to-event outcomes were defined using repeatedassessments at ages 7, 10, and 13 years, with age as the underlyingtime scale and entry at age 4 years. Outcomes included: (1) incidentobesity defined by BMI; (2) incident obesity defined by body fatpercentage; (3) incident overweight or obesity defined by BMI; and(4) incident overweight or obesity defined by body fat percentage.Event time was defined as the age at which the outcome was first observed, and children who did not develop the outcome were censored at their last available assessment.2.5 | CovariatesCovariates were selected according to previous literature if theywere related to asthma diagnosis exposure and obesity outcome. Thisincluded: child's sex (Female/Male), birthweight (<2500 g/≥2500 g),parity (primiparous/multiparous), maternal smoking status dur- ing pregnancy (ever vs. no-smoking), maternal age (years) at child'sbirth, GWG categories (insufficient, adequate, and excessive),21 ma- ternal BMI (underweight, normal weight, overweight, and obese)22and breastfeeding (ever vs. never). Maternal level of education wasassessed through the number of completed schooling years and cat- egorized into ≤9, 10–12, or >12 years.23Children's physical activity status at 4-, 7-, 10- and 13-year-old fol- low-up visits were also considered, based on dichotomous questions. At4 years of age, parents answered the question: “Does the child practiceany type of scheduled or regular sport?” at 7, 10, and 13 years, the ques- tion was: “Does the child/adolescent practice any type of scheduled orregular sport, outside of school?”, categorized as yes vs. no.24Children's dietary intake at 4 years of age was assessed using a 35-item qualitative food frequency questionnaire (FFQ) that capturedusual consumption over the previous 6 months. 25 Dietary patterns at age 4 were previously derived through latent class analysis. Three patterns were identified based on response profiles and mean dailyconsumption frequencies. The first pattern (“Energy dense foods”– “EDF”) reflected higher intakes of energy-dense, micronutrient-poor foods and beverages (e.g., sweets, sugar-sweetened drinks,processed and high-fat foods). The second pattern (“Snacking”) wascharacterized by lower consumption of foods typically eaten at mainmeals and relatively higher intake of snack-type foods. The third(“Healthier”) pattern showed greater consumption of fruits, vegeta-bles, and fish and lower intake of energy dense food items. 26A directed acyclic graph (DAG) was used to confirm the minimalsufficient adjustment set for confounding control (Figure S1).2.6 | Statistical analysisContinuous variables are presented as means and standard devia-tions, and categorical variables as counts and percentages. Group comparisons were performed using independent sample t-tests forcontinuous variables and chi-square tests for categorical variables. Associations between asthma at age 4 and subsequent adiposityoutcomes were estimated using Cox proportional hazards models,reporting hazard ratios (HR) and 95% confidence intervals (CI). Bothcrude and adjusted models were fitted. The proportional hazardsassumption was verified by visual inspection of log-minus-log plots,with no violations detected.Effect modification by sex was evaluated by including an inter- action term between asthma and sex, and, as p for interaction was<.05, sex-stratified analyses were also performed. Missing data oncovariates were handled using multiple imputation by fully condi- tional specification (FCS) in SPSS, assuming missing at random, using10 imputed data sets. Complete-case analyses were performed andpresented on Table S1. Kaplan–Meier curves and correspondinglog-rank test results comparing cumulative incidence of outcomesby asthma status are presented in Figures S2–S13 and Tables S2– S5. Sensitivity analyses additionally adjusting for dietary patternsand physical activity variables at age 4 are presented in Table S6.Descriptive analyses in the full cohort and by sex regarding dif-ferences in body fat percentage, BMI and asthma are presented in Table S8. Descriptive analysis of children with asthma by sex are alsopresented in Table S9.2.7 | EthicsAll phases of the study complied with the Ethical Principles forMedical Research Involving Human Subjects expressed in theDeclaration of Helsinki*. The baseline and follow-up evaluationswere approved by the University of Porto Medical School/S. JoãoHospital Centre Ethics Committee, except the 13-year follow-upthat was approved by the ISPUP Ethics Committee. At baseline andfollow-up evaluations, all procedures were explained to participants,and an informed consent was signed by one of the parents or legalguardians (at 13 years, it was also signed by the participants).The baseline evaluation was additionally approved by the DataProtection National Commission, and the study follows the presentEU General Data Protection Regulation under close supervision ofthe Data Protection Office of ISPUP.273 | R E S U LT SAt 4 years of age, 181 children (3.8%) had a physician diagnosis ofasthma. Among these children, asthma persistence was high acrossfollow-up, with 76.8% reporting asthma at 7 years, 71.9% at 10 years,and 74.0% at 13 years, among those with available data at eachtime point. Baseline maternal and child characteristics according to asthma status are shown in Table 2. Children with asthma were morefrequently boys (63.0% vs. 51.0%, p = .002) and had slightly youngermothers (28.8 vs. 29.8 years, p = .002). Differences were also observedin maternal pre- pregnancy BMI (p = .003) with a lower proportion of ### Page 6 6 of 10 |RODRIGUES et al.mothers of children with asthma having normal weight (61.6% vs.66.6%). Maternal gestational weight gain categories also differed(p = .038), with excessive GWG more common among mothers ofchildren with asthma (45.2% vs. 36.0%). Maternal education, smokingduring pregnancy, breastfeeding, parity, and birth weight did not differ significantly between groups.Asthma at age 4 was positively associated with subsequentobesity defined by body fat percentage (Table 3). In the overallsample, children with asthma had a higher risk of incident obesitybased on body fat percentage in adjusted analyses (aHR = 1.91, 95%CI: 1.16–3.16), whereas the crude estimate did not reach statisticalsignificance (HR = 1.30, 95% CI: 0.96–1.75). Sex-stratified analysesTA B L E 2 Maternal and child characteristics according to asthma status at 4 years the Generation XXI cohort.Sample size, n Asthma (n = 181, 3.8%) Without asthma (n = 4536, 96.2%) p-ValueMother age, years, mean ± SD 4717 28.78 ± 5.96 29.81 ± 5.20 .002 Maternal education, n (%)<=9y 4717 83 (45.9%) 1909 (42.1%) .178> = 10 e < =12 7 (3.9%) 331 (7.3%)>12 91 (50.3%) 2296 (50.6%)Parity, multiparous, n (%) 4634 77 (44.0%) 1844 (41.4%) .486GWG categories, n (%)Insufficient 4406 41 (24.4%) 1093 (25.8%) .038Adequate 51 (30.4%) 1619 (38.2%) Excessive 76 (45.2%) 1526 (36.0%)Maternal BMI Underweight weight, n (%) 4651 11 (6.2%) 173 (3.9%) .003Normal weight, n (%) 109 (61.6%) 2980 (66.6%)Overweight, n (%) 41 (23.2%) 957 (21.4%)Obese, n (%) 16 (9.0%) 364 (8.1%)Maternal smoking in pregnancy, n (%) 4643 44 (24.9%) 887 (19.9%) .103 Breastfeeding 4675 169 (93.4%) 4186 (93.1%) .907 Sex, n (%)Female 4717 67 (37.0%) 2224 (49.0%) .002 Male 114 (63.0%) 2312 (51.0%)Birth weight<2500 4717 21 (11.6%) 441 (9.7%) .404>=2500 160 (88.4%) 4095 (90.3%) Physical Activity 4y, n (%)Yes 4695 121 (66.9%) 3147 (69.7%) .411No 60 (33.1%) 1367 (30.3%) Physical Activity 7y, n (%)Yes 3755 66 (50.4%) 2011 (55.5%) .248No 65 (49.6%) 1613 (44.5%) Physical Activity 10y, n (%)Yes 4376 108 (63.5%) 2633 (62.6%) .806No 62 (36.5%) 1573 (37.4%) Physical Activity 13y, n (%)Yes 3441 68 (51.1%) 1966 (59.4%) .058 No 65 (48.9%) 1345 (40.6%)Diet Pattern, n (%)Healthy 4659 74 (41.3%) 1915 (42.7%) .916 Snacking 28 (15.6%) 663 (14.8%) Energy-dense 77 (43.0%) 1902 (42.5%) Abbreviations: BMI, body mass index; GWG, gestational weight gain; SD, standard deviation; y, years. ### Page 7 | 7 of 10RODRIGUES et al.revealed marked differences. Among girls, asthma was associatedwith a significantly higher risk of incident obesity defined by bodyfat percentage in both crude (HR = 1.91, 95% CI: 1.16–3.15) and ad- justed models (aHR = 1.97, 95% CI: 1.19–3.26). In contrast, no signif- icant associations were observed among boys (aHR = 1.00, 95% CI:0.68–1.47). When overweight or obesity was defined using body fat per-centage, asthma was associated with a significantly increased riskin adjusted analyses (aHR = 1.76, 95% CI: 1.03–3.00). This associa-tion was driven by girls, among whom asthma was associated witha higher hazard of developing excess adiposity (aHR = 1.75, 95%CI: 1.02–3.01). No significant associations were observed amongboys.In line with these findings, when examining differences withinchildren with asthma (Table S9), girls were less likely than boys toreport practicing sports outside of school at 13 years of age (33.9%vs. 61.4%, χ2 , p < .001).4 | D I S CUS S I O NIn this large prospective birth cohort, asthma diagnosed at 4 yearsof age was associated with an increased risk of developing adipos-ity until adolescence when adiposity was assessed using body fat percentage, but not consistently when defined by BMI. These as- sociations were largely confined to girls, highlighting important sex-specific differences in the asthma-adiposity relationship.Our findings suggest that early childhood asthma may predis- pose to increased fat accumulation rather than generalized weightgain. This distinction is clinically relevant, as BMI does not differ- entiate between fat and lean mass and may inadequately captureexcess adiposity during periods of rapid growth and pubertal de- velopment. Several studies have demonstrated that BMI may mis - classifies a substantial proportion of children with excess body fat,thereby underestimating obesity prevalence and related health risks.14,17,20,28Although most previous research has focused on obesity asa risk factor for asthma, a growing body of longitudinal evidencesupports the reverse pathway. Multi-cohort studies have reportedan increased risk of later obesity among children with asthma, witha 23% increased risk of developing obesity (95% CI: 4%–44%) re- ported in the study by Stratakis et al.6 and an adjusted hazard ratioof 1.66 (95% CI: 1.18–2.33) reported in the study by Contreraset al. 7 Evidence from individual cohort studies further supports this as- sociation. In the Southern California Children's Health Study (CHS),it was found that children with asthma had a 51% increased risk ofdeveloping obesity during childhood and adolescence (HR = 1.51;95% CI: 1.08–2.10) compared with children without asthma at base-line. 4 Additional analyses from the CHS using joint transition mod- eling with cross-lagged structures showed that non-obese childrenwith asthma had 37% higher odds of becoming obese at the nextannual visit (OR = 1.38; 95% CI: 1.12–1.71), while obesity at baselinedid not predict subsequent asthma in the short term (OR = 1.25; 95%CI: 0.94–1.62).13Our results are consistent with these findings but extend exist-ing knowledge by demonstrating that associations are stronger and more evident when adiposity is defined by body composition ratherTA B L E 3 Associations between asthma at 4 years and four obesity-related events in adolescence (incident obesity based on BMI/on bodyfat percentage, incident overweight or obesity based on BMI/on body fat percentage) in the Generation XXI cohort.N (Total/Events) Crude HR (CI 95%) p-Value Adjusted HR (CI 95%)a p-valueIncident Obesity Based on BMI 4597/631 1.19 (0.81–1.75) .371 1.63 (0.89–2.98) .116 Girls 2236/265 1.68 (0.92–3.07) .094 1.59 (0.87–2.92) .136 Boys 2361/366 0.94 (0.57–1.54) .793 0.87 (0.52–1.44) .578Incident Obesity Based on Body Fat Percentage3507/1014 1.30 (0.96–1.75) .092 1.91 (1.16–3.16) .011 Girls 1718/460 1.91 (1.16–3.15) .011 1.97 (1.19–3.26) .008 Boys 1789/554 1.06 (0.73–1.55) .761 1.00 (0.68–1.47) .990Incident Overweight or Obesity Based on BMI3526/1053 1.00 (0.72–1.38) .999 1.07 (0.60–1.90) .816 Girls 1691/491 1.06 (0.60–1.89) .831 1.01 (0.57–1.79) .981Boys 1835/562 0.95 (0.64–1.41) .812 0.97 (0.65–1.45) .893Incident Overweight or Obesity Based on Body Fat Percentage3044/1195 1.13 (0.84–1.52) .431 1.76 (1.03–3.00) .039Girls 1478/536 1.60 (0.94–2.73) .082 1.75 (1.02–3.01) .042 Boys 1566/659 0.95 (0.66–1.37) .787 0.88 (0.61–1.27) .493Note: A child sex*asthma interaction term was additionally included when analyses were performed in the whole sample, (ref=girls). Abbreviations: BMI: body mass index; CI: confidence interval; HR: hazard ratio.aAdjusted for maternal age, maternal education, maternal BMI, parity, gestational weight gain (GWG), maternal smoking in pregnancy, breastfeeding,birth weight, and child sex. ### Page 8 8 of 10 |RODRIGUES et al.than BMI alone. This may partly explain inconsistencies across co- horts relying exclusively on BMI-based outcomes. Several mechanisms may underlie the observed association.Children with asthma may limit physical activity due to respiratorysymptoms, fear of exacerbations, or reduced exercise tolerance, lead- ing to lower energy expenditure.4,8 Unhealthy dietary patterns 3 and asthma treatments9,30–33 have also been proposed as contributing factors, although evidence remains inconsistent. In addition, asthmais increasingly recognized as a condition with systemic inflammatoryfeatures.10,11 Pro-inflammatory mediators and adipokines, such asleptin, may link airway inflammation with metabolic dysregulation and fat accumulation, even in the absence of overt obesity.12A key finding of our study is the clear sex-specific pattern, withassociations observed almost exclusively among girls. Importantly,our analyses compared children with and without asthma withineach sex. Therefore, although girls may have a higher body fat per- centage overall, this is unlikely to fully explain the observed associ-ation, which reflects differences between children with and withoutasthma within the same sex. Behavioral factors may partly explainthis difference. In our cohort, girls with asthma were substantially less likely than boys to report participation in sports during adoles-cence, consistent with literature showing greater perceived barriers to physical activity among girls. 33 Reduced physical activity may therefore act as an important mediator linking asthma to later ad-iposity in girls.Biological mechanisms may further contribute to sex differences.Pubertal hormonal changes influence both immune responses and metabolic regulation. Estrogens and progesterone have been shownto enhance Th2-mediated inflammation and airway hyperrespon- siveness, while androgens exert anti-inflammatory and bronchodi-latory effects. 34 Around puberty, asthma prevalence and severitytend to increase among girls, coinciding with hormonal changes that may also favor fat deposition.35From a public health and clinical perspective, our findings un-derscore the importance of early, integrated prevention strategiesof excessive weight gain in children with asthma. Physical activityimproves cardiopulmonary fitness, lung capacity, muscular strength, and endurance, 36 while also reducing airway eosinophilic inflamma-tion. 37 In children and adolescents with asthma, regular physical ac- tivity lowers hospitalization rates, reduces school absenteeism, anddecreases the use of bronchodilators. 38 It also improves daytime andnocturnal asthma control, enhances sleep quality, and supports psy- chological well-being.38,40 Besides these effects, physical activity also contributes to obesity prevention by promoting energy balance, thereby addressing these two interrelated conditions, asthma and obesity.3,15 Given the considerable inter-individual variability amongpatients with asthma, personalized exercise prescriptions may leadto better clinical outcomes and quality of life.37,40 Beyond respira-tory health control, attention should be given to growth patterns, body composition, and lifestyle counseling, particularly among girls. Early identification of children at higher metabolic risk may help mit- igate long-term cardiometabolic consequences.Our study has some limitations that should be acknowledged.Asthma diagnosis was based on parental report of a physician di-agnosis and did not capture disease severity, persistence, or treat-ment factors that may influence adiposity trajectories. However, thequestionnaires used were based on the standardized InternationalStudy of Asthma and Allergies in Childhood (ISAAC),41 a widely used and validated approach in epidemiological studies. 42 Loss tofollow-up may have introduced selection bias; however, differencesbetween included and excluded participants had small effect sizes,suggesting limited impact on the observed associations. Moreover, body fat percentage categories were defined using UK reference curves, which may not perfectly reflect the distribution of adiposity in Portuguese children. However, these reference curves have theadvantage of being age- and sex-specific and being widely used inepidemiological research. In addition, as the study was conducted within a single Portuguese birth cohort, the generalisability of thefindings to other populations may be limited. Additionally, transientwheezing and current asthma were not analyzed separately. Thismay influence the interpretation of findings, potentially attenuating the observed associations. Despite these limitations, the study has several importantstrengths. Its prospective design, large population-based sample,and long follow-up into adolescence allowed assessment of tempo- ral associations between early-life asthma and subsequent adiposity.By focusing on incident outcomes and excluding prevalent cases, wereduced the likelihood of reverse causation. Adiposity was assessedusing repeated, objective measurements, rather than self-reported.The use of body fat percentage derived from validated bioelectri- cal impedance equations, subsequently classified according to theage- and sex-specific body fat reference curves, represents a majormethodological advance over BMI-only approaches, allowing for astandardized assessment of adiposity across developmental stages.Additionally, although body fat percentiles were applied using ref- erence curves available from 5 years onwards, sensitivity analysesrestricted to children aged ≥5 years at 4-year-old follow-up yieldedconsistent results (Table S7). Furthermore, we conducted bothcomplete-case analyses and analyses using multiple imputations.Results were consistent between both analyses, with comparable effect estimates, direction, and statistical significance. Finally, con-founder selection guided by a directed acyclic graph strengthens the robustness of causal inference. In conclusion, asthma in early childhood was associated with anincreased risk of subsequent adiposity when assessed using bodyfat percentage, particularly among girls. These findings indicate thatchildren with asthma may be prone to excess fat accumulation that isnot fully captured by BMI alone. From a clinical perspective, routinefollow-up of children with asthma should extend beyond respiratorycontrol to include careful monitoring of growth patterns and body composition, alongside early counseling on physical activity and healthy lifestyle behaviors. Integrating metabolic risk awareness intopediatric asthma care may help reduce long-term cardiometabolicburden, especially in girls. ### Page 9 | 9 of 10RODRIGUES et al.AU T H O R CO N T R I B U T I O N SMargarida Henriques: Investigation; writing – original draft; visuali- zation; writing – review and editing; data curation; formal analysis.Mónica Rodrigues: Investigation; writing – original draft; meth- odology; visualization; writing – review and editing; formal analy- sis; data curation; supervision; conceptualization. André Moreira:Conceptualization; writing – review and editing; supervision; writing– original draft. Pedro Moreira: Writing – review and editing; su -pervision. Susana Santos: Supervision; writing – review and editing.AC K N OW L E D G M E N T SWe gratefully acknowledge the families enrolled in Generation XXIfor their kindness, the participating hospitals and their staff for their help and support, and all previous and current research and field team members for their enthusiasm and perseverance.F U N D I N G I N FO R M AT I O NG21 was funded by Programa Operacional de Saúde – Saúde XXI,Quadro Comunitário de Apoio III and Administração Regional deSaúde Norte (Regional Department of Ministry of Health). Thiswork was supported by FCT Fundação para a Ciência e Tecnologia,I.P. through the projects with references UID/4750/2025 andLA/P/0064/2020 and DOI identifiers https:// doi.org/10.54499/UID/ 04750/ 2025and https:// doi.org/10.54499/ LA/P/0064/2020.M.R. was supported by the Portuguese Foundation for Science andTechnology – FCT – project reference 2023.02362.BD: https://doi.org/ 10.54499/ 2023.02362.BD and DOI identifier: http://doi.org/10. 54499/ 2023.02362.BD. S.S. was supported by national fundsthrough the Foundation for Science and Technology (FCT), underthe Scientific Employment Stimulus Program – Individual Call (CEECIndividual), reference 2024.10106.CEECIND.CO N FL I C T O F I N T E R E S T S TAT E M E N TNo conflicts of interest to report.P E E R R E V I E WFor transparency, the peer review documents associated with this article are available at https://doi.org/10.1111/pai.70360.DATA AVA I L A B I L I T Y S TAT E M E N TThe data that support the findings of this study are available on re- quest from the corresponding author. The data are not publicly avail-able due to privacy or ethical restrictions.O R C I DMónica Rodrigues https://orcid.org/0000-0002-7932-9460André Moreira https://orcid.org/0000-0002-7294-9296R E F E R E N C E S1. Permaul P, Kanchongkittiphon W, Phipatanakul W. Childhoodasthma and obesity—what is the true link? Ann Allergy Asthma Immunol. 2014;113(3):244-246. 2. Krol KA, Litonjua AA. Beyond obesity: the complex relationshipbetween early growth trajectories and later lung function. J Allergy Clin Immunol. 2021;148(3):713-715. 3. Chen Z, Salam MT, Alderete TL, et al. Effects of childhood asthmaon the development of obesity among school-aged children. Am J Respir Crit Care Med. 2017;195(9):1181-1188. 4. Shan LS, Zhou QL, Shang YX. Bidirectional association betweenasthma and obesity during childhood and adolescence: a system- atic review and meta-analysis. Front Pediatr. 2020;8:576858. 5. Stratakis N, Garcia E, Chandran A, et al. The role of childhoodasthma in obesity development: a Nationwide US multicohortstudy. Epidemiology. 2022;33(1):131-140. 6. Contreras ZA, Chen Z, Roumeliotaki T, et al. Does early onsetasthma increase childhood obesity risk? A pooled analysis of 16European cohorts. Eur Respir J. 2018;52(3):1800504. 7. Panagiotou M, Koulouris NG, Rovina N. Physical activity: a missinglink in asthma care. J Clin Med. 2020;9(3):706. 8. Calcaterra V, Verduci E, Ghezzi M, et al. Pediatric obesity-relatedasthma: the role of nutrition and nutrients in prevention and treat-ment. Nutrients. 2021;13(11):3708. 9. Moitra S, Carsin AE, Abramson MJ, et al. Long-term effect ofasthma on the development of obesity among adults: an interna-tional cohort study. ECRHS Thorax. 2023;78(2):128-135. 10. Tattersall MC, Jarjour NN, Busse PJ. Systemic inflammation inasthma: what are the risks and impacts outside the airway? J Allergy Clin Immunol Pract. 2024;12(4):849-862. 11. Gkrinia EMM, Belancic A. The mechanisms of chronic inflammationin obesity and potential therapeutic strategies: a narrative review.Curr Issues Mol Biol. 2025;47(5):357. 12. Guler N, Kirerleri E, Ones U, Tamay Z, Salmayenli N, DarendelilerF. Leptin: does it have any role in childhood asthma? J Allergy Clin Immunol. 2004;114(2):254-259.13. Zhang Y, Chen Z, Berhane K, et al. The dynamic relationship be-tween asthma and obesity in schoolchildren. Am J Epidemiol.2020;189(6):583-591. 14. Javed A, Jumean M, Murad MH, et al. Diagnostic performance ofbody mass index to identify obesity as defined by body adiposityin children and adolescents: a systematic review and meta-analysis.Pediatr Obes. 2015;10(3):234-244. 15. Alves E, Correia S, Barros H, Azevedo A. Prevalence of self-reportedcardiovascular risk factors in Portuguese women: a survey after de-livery. Int J Public Health. 2012;57(5):837-847. 16. Larsen PS, Kamper-Jørgensen M, Adamson A, et al. Pregnancyand birth cohort resources in europe: a large opportunity for aetiological child health research. Paediatr Perinat Epidemiol.2013;27(4):393-414. 17. de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J.Development of a WHO growth reference for school-aged childrenand adolescents. Bull World Health Organ. 2007;85(9):660-667. 18. WHO. Training Course on Child Growth Assessment. WHO; 2008. 19. Schaefer F, Georgi M, Zieger A, Schärer K. Usefulness of bioelec- tric impedance and skinfold measurements in predicting fat-freemass derived from total body potassium in children. Pediatr Res.1994;35(5):617-624. 20. McCarthy HD, Cole TJ, Fry T, Jebb SA, Prentice AM. Body fat refer-ence curves for children. Int J Obes. 2006;30(4):598-602. 21. Institute of Medicine. In: Rasmussen KM, Yaktine AL, eds. Weight Gain During Pregnancy: Reexamining the Guidelines. The NationalAcademies Press; 2009. 22. World Health Organization. Obesity: Preventing and Managing the Global Epidemic: Report of a WHO Consultation. World HealthOrganization; 2000. 23. UNESCO, S.C. International Standard Classification of Education (ISCED). UNESCO, S.C; 1997. ### Page 10 10 of 10 |RODRIGUES et al.24. Barros L, Lopes C, Oliveira A. Child and family characteristics areassociated with a dietary variety index in 4-year-old children fromthe generation XXI cohort. Nutr Res. 2019;63:76-85. 25. Vilela S, Severo M, Moreira T, Ramos E, Lopes C. Evaluation of ashort food frequency questionnaire for dietary intake assessmentamong children. Eur J Clin Nutr. 2019;73(5):679-691. 26. Durao C, Severo M, Oliveira A, et al. Association between dietarypatterns and adiposity from 4 to 7 years of age. Public Health Nutr.2017;20(11):1973-1982. 27. Association, W.M. World Medical Association Declaration ofHelsinki: Ethical Principles for Medical Research Involving HumanSubjects. 2013 https:// www.wma.net/what- we-do/medic al- eth- ics/ declaration- of-helsinki/28. Zeng Q, Dong SY, Sun XN, Xie J, Cui Y. Percent body fat is a betterpredictor of cardiovascular risk factors than body mass index. Braz J Med Biol Res. 2012;45(7):591-600. 29. Sullivan PW, Ghushchyan VH, Globe G, Schatz M. Oral corticoste- roid exposure and adverse effects in asthmatic patients. J Allergy Clin Immunol. 2018;141(1):110-116.e7. 30. Sweeney J, Patterson CC, Menzies-Gow A, et al. Comorbidity insevere asthma requiring systemic corticosteroid therapy: cross-sectional data from the optimum patient care research data-base and the British thoracic difficult asthma registry. Thorax.2016;71(4):339-346. 31. Cooper V, Metcalf L, Versnel J, Upton J, Walker S, Horne R. Patient-reported side effects, concerns and adherence to corticosteroid treatment for asthma, and comparison with physician estimates ofside-effect prevalence: a UK-wide, cross-sectional study. NPJ Prim Care Respir Med. 2015;25:15026. 32. Umlawska W. Adipose tissue content and distribution in chil-dren and adolescents with bronchial asthma. Respir Med.2015;109(2):200-207. 33. Cowley ES, Watson PM, Foweather L, et al. ‘Girls Aren't meant toexercise’: perceived influences on physical activity among adoles- cent girls-the HERizon project. Children (Basel). 2021;8(1):31. 34. Fuseini H, Newcomb DC. Mechanisms Driving Gender Differencesin Asthma. Curr Allergy Asthma Rep. 2017;17(3):19. 35. Chowdhury NU, Guntur VP, Newcomb DC, Wechsler ME. Sex andgender in asthma. Eur Respir Rev. 2021;30(162):210067. 36. Liu W, Feng Z, Song S, Lei S. The effectiveness of physical activity inasthma management: an overview of systematic reviews. PLoS One.2025;20(7):e0325488. 37. Francisco CO, Bhatawadekar SA, Babineau J, Reid WD, YadollahiA. Effects of physical exercise training on nocturnal symptoms inasthma: systematic review. PLoS One. 2018;13(10):e0204953. 38. Privitera A, Privitera S. Physical exercise in asthma adolescents: aconcept review. Multidiscip Respir Med. 2023;18(1):924. 39. Xu HZ, Lin N, Bai GN, et al. Comprehensive exercise recommenda-tions for pediatric asthma: an evidence synthesis. World J Pediatr.2026;22(2):174-193. 40. Georgoulis M, Grapsa I, Arnaoutis G, et al. Association betweenphysical activity and indicators of overweight/obesity and metabol-ically unhealthy obesity risk in children and adolescents: a system- atic review of prospective epidemiological studies and randomizedcontrolled trials in Western countries. Obes Rev. 2026;27(1):1-28. 41. Asher MI, Keil U, Anderson HR, et al. International study of asthmaand allergies in childhood (ISAAC): rationale and methods. Eur Respir J. 1995;8(3):483-491. 42. Islam MS, Huq S, Ahmed S, et al. Operational definitions of paedi-atric asthma used in epidemiological studies: a systematic review. J Glob Health. 2021;11:04032.S U P P O R T I N G I N FO R M AT I O NAdditional supporting information can be found online in theSupporting Information section at the end of this article.How to cite this article: Rodrigues M, Henriques M, SantosS, Moreira P, Moreira A. Early childhood asthmaand adiposity until adolescence: A prospective birth cohortstudy using body fat percentage. Pediatr Allergy Immunol. 2026;37:e70360. doi:10.1111/pai.70360