Since its emergence in the late 20th century, the Developmental Origins of Health and Disease (DOHaD) framework has shaped thinking about how early-life exposures influence health across the life course. Historically, DOHaD research and policy have focused predominantly on maternal pregnancy behaviors, reinforcing the assumption that developmental influences operate primarily via mothers. Recent multicohort evidence challenges that emphasis, arguing that broader social and familial contexts—especially socioeconomic position—may exert larger and more consistent influences on child health than many individual prenatal behaviors.
Sharp and colleagues conducted the Exploring Prenatal influences on Childhood Health (EPoCH) study across four birth cohorts spanning two decades. They examined maternal and co-parent smoking, alcohol use, caffeine intake and socioeconomic position in relation to over 70 child health outcomes in early childhood. The study combined conventional epidemiological meta-analyses with complementary causal inference approaches.
Findings reported in the commentary indicate that most parental prenatal behaviors did not show large causal effects on childhood outcomes. Maternal smoking was the main exception: it remained associated with small-for-gestational-age birth, later overweight and obesity, and six psychosocial outcomes. However, associations between maternal smoking and child body mass index were attenuated after adjustment for genetic risk scores, suggesting shared genetic factors may partly explain observed links.
By contrast, socioeconomic position showed the most consistent associations across the measured outcomes, suggesting social circumstances around pregnancy may have greater influence on early-life health than many modifiable prenatal behaviors.
The EPoCH analysis triangulated evidence using multiple methods: conventional meta-analyses across cohorts, mendelian randomization, negative control analyses and the use of genetic risk scores. This combination aims to move beyond the observational associations that have predominated in DOHaD literature and toward stronger causal inference about intrauterine effects versus shared familial or socioeconomic influences.
The commentary highlights that the methodological breadth—applying several complementary approaches across multiple cohorts—constitutes an important contribution to DOHaD research, increasing confidence in the relative size of effects attributed to parental behaviors versus socioeconomic context.
The multicohort results indicate limited causal effects for most examined parental behaviors. Maternal smoking was associated with specific adverse outcomes (small-for-gestational-age birth, later overweight/obesity, and multiple psychosocial outcomes), but the strength of some associations diminished after accounting for genetic risk scores. For other behaviors—maternal and co-parent alcohol consumption and caffeine intake—the study did not find large causal effects across the broad set of child outcomes analyzed.
These findings question the predominant focus in DOHaD on changing individual maternal behaviors during pregnancy as the primary route to improving child health at the population level.
Across the more than 70 outcomes examined, socioeconomic position emerged as the most consistently associated domain. The commentary explains how socioeconomic disadvantage shapes maternal and child health through multiple pathways: nutrition, housing quality, financial security, chronic stress, environmental exposures, access to healthcare and opportunities for healthy behaviors.
The authors draw attention to the 3Cs framework—Clock, Cost and Capacity—to conceptualize barriers families face when attempting to act on health advice. The commentary also cites evidence that unconditional cash transfers during pregnancy can improve infant outcomes, underscoring how upstream social interventions can have measurable early-life benefits.
A key limitation noted is the predominance of white participants across the cohorts, which constrains assessment of how socioeconomic disadvantage intersects with ethnicity. The commentary underlines that persistent ethnic inequalities in pregnancy outcomes are an important consideration and that the study’s demographic composition limits generalisability on that axis.
The authors also note that while the EPoCH platform and analytical code are publicly available—strengthening transparency and reproducibility—further work is needed to explore interactions between socioeconomic position, ethnicity and other contextual factors.
If the goal is to improve child health and reduce inequalities, the commentary argues that interventions narrowly focused on modifying maternal behaviors during pregnancy are unlikely to deliver substantial population-level gains. Instead, research, policy and public health should attend more to the social and economic conditions surrounding pregnancies.
The article recommends that epidemiological research continue to adopt rigorous causal methods and maintain open, reproducible approaches, as Sharp and colleagues have done, but with an explicit orientation toward informing interventions that address structural determinants rather than placing responsibility primarily on individual mothers.
The commentary highlights that many of the largest improvements in child health have resulted from structural policies rather than individual behavior change programs. Examples cited include smoke-free legislation, which reduced children’s exposure to secondhand smoke and improved perinatal outcomes, and the UK Soft Drinks Industry Levy, which has been associated with reductions in childhood asthma admissions.
Economic and social policies are also discussed: an earlier study linking persistent childhood poverty with increased early-adult mortality and evidence that unconditional cash transfers can improve infant health outcomes are used to illustrate how addressing upstream socioeconomic drivers can yield measurable benefits.
The principal contribution of the commentary is to reframe where responsibility for child health should lie. The EPoCH findings suggest that for many outcomes the impact of maternal behaviors is modest compared with the pervasive influence of socioeconomic disadvantage. This challenges the long-standing emphasis in DOHaD and maternity care on changing maternal behaviors during pregnancy and supports a shift toward interventions that modify the structural and societal contexts shaping families’ choices.
By making analytical frameworks openly available and combining rigorous causal methods with transparent reporting, the authors provide a model for epidemiological research designed to inform meaningful action on the social determinants of child health.