**Background:** Low and middle-income countries (LMICs) have experienced increasing prevalence of cardiometabolic risk factors such as diabetes, hypertension, and obesity. The Developmental Origins of Health and Disease hypothesis proposes that early-life adaptations have long-lasting effects on adult cardiometabolic disease. While body composition (fat vs. lean mass) is more relevant to cardiometabolic risk than BMI alone, there is a paucity of studies examining the long-term effects of growth across the entire life span on adult body composition, particularly in LMICs. This study aimed to examine associations among serial measures of linear growth and relative weight with adult body composition in a pooled analysis of the Consortium of Health-Oriented Research in Transitional Societies (COHORTS).
**Methods:** This was a secondary data analysis of six prospective birth cohorts from five LMICs: Brazil (1982 and 1993 Pelotas Birth Cohorts), Guatemala (INCAP Nutrition Trial and Longitudinal Study), India (New Delhi Birth Cohort), the Philippines (Cebu Longitudinal Health and Nutrition Survey), and South Africa (Birth to Twenty Plus Cohort). The analytical sample included 4173 participants (2155 men, 1982 women) with complete and valid weight and height at birth, infancy, childhood, and adolescence, and body composition in adult life (ages 22–46 years). Exposures were birth weight and conditional size measures (standardized residuals) representing linear growth (height) and relative weight (weight increments independent of linear size) in four age intervals: prenatal (birth), infancy (birth to 12–24 months), childhood (12–48 months or 24–102 months depending on cohort), and adolescence (48–180 months or 102–180 months). Outcomes were adult BMI, fat mass index (FMI), fat-free mass index (FFMI), FM/FFM ratio, and waist circumference, all transformed to Z-scores. Site-specific multiple linear regressions stratified by sex were conducted, and effect sizes were pooled using random-effects meta-analysis with restricted maximum likelihood estimation and Knapp-Hartung adjustments. Multiple imputation was used for missing covariates.
**Key Results:** In pooled analyses, birth weight was positively associated with adult FMI (men: β=0.07, 95% CI 0.01 to 0.14; women: β=0.09, 95% CI 0.01 to 0.17) and FFMI (men: β=0.11, 95% CI 0.05 to 0.18; women: β=0.12, 95% CI 0.05 to 0.19), but not with FM/FFM ratio. Conditional relative weight gains in infancy, childhood, and adolescence were positively associated with all adult outcomes. The strongest predictors of adult body composition were relative weight gains in childhood and adolescence. For example, among men, each 1 SD increment in conditional relative weight in childhood was associated with increases of 0.41 SD (95% CI 0.26 to 0.55) in FMI, 0.50 SD (95% CI 0.34 to 0.66) in FFMI, and 0.31 SD (95% CI 0.16 to 0.47) in FM/FFM ratio. Among women, similar patterns were observed, but effect sizes in adolescence were slightly stronger than in childhood for FMI (adolescence: β=0.53, 95% CI 0.44 to 0.61; childhood: β=0.46, 95% CI 0.21 to 0.72) and FM/FFM ratio (adolescence: β=0.43, 95% CI 0.39 to 0.48; childhood: β=0.37, 95% CI 0.07 to 0.66). Conditional height in infancy was positively but weakly associated with FMI (men: β=0.08, 95% CI 0.03 to 0.14; women: β=0.11, 95% CI 0.07 to 0.16) and FM/FFM ratio (men: β=0.07, 95% CI 0.02 to 0.12; women: β=0.10, 95% CI 0.03 to 0.16). Conditional height in childhood was positively but weakly associated with women's adiposity (FMI: β=0.14, 95% CI 0.04 to 0.24; FM/FFM: β=0.16, 95% CI 0.08 to 0.25; waist circumference: β=0.14, 95% CI 0.06 to 0.23). Between-study heterogeneity was low to moderate for most associations, though substantial heterogeneity (I² ≥75%) was observed for some conditional relative weight estimates in childhood and adolescence.
**Clinical Implications:** These findings suggest that prenatal and postnatal relative weight gains are positive predictors of larger body size and increased adiposity in adulthood, with childhood and adolescence being the periods of strongest association. While growth in the first 1000 days of life has a small but long-lasting association with adult body composition, the strongest effects are observed for weight gains in later childhood and adolescence. Faster linear growth in infancy was a significant but weak predictor of higher adult adiposity. Public health interventions in LMICs should focus on promoting adequate linear growth in early life while preventing excessive relative weight gain throughout childhood and adolescence. The consistency of findings across six diverse cohorts from five LMICs suggests these associations are robust and may apply broadly to similar settings. However, the authors caution against generalizability given the selected nature of the cohorts and differences in contexts and time periods.