Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE ( (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 21 October 2019

Developmental Origins of Health Inequality

Summary and Keywords

Building on early animal studies, 20th-century researchers increasingly explored the fact that early events—ranging from conception to childhood—affect a child’s health trajectory in the long-term. By the 21st century, a wide body of research had emerged, incorporating the original fetal origins hypothesis into the developmental origins of health and disease. Evidence from Organization for Economic Cooperation and Development (OECD) countries suggests that health inequalities are strongly correlated with many dimensions of socioeconomic status, such as educational attainment, and that they tend to increase with age and carry stark intergenerational implications. Different economic theories have been developed to rationalize this evidence, with an overarching comprehensive framework still lacking. Existing models widely rely on human capital theory, which has given rise to separate dynamic models of adult and child health capital within a production function framework. A large body of empirical evidence has also found support for the developmental origins of inequalities in health. On the one hand, studies exploiting quasi-random exposure to adverse events have shown long-term physical and mental health impacts of exposure to early shocks, including pandemics or maternal illness, famine, malnutrition, stress, vitamin deficiencies, maltreatment, pollution, and economic recessions. On the other hand, studies from the 20th century have shown that early interventions of various content and delivery formats improve life course health. Further, given that the most socioeconomically disadvantaged groups show the greatest gains, such measures can potentially reduce health inequalities. However, studies of long-term impacts as well as the mechanisms via which shocks or policies affect health, and the dynamic interaction among them, are still lacking. Mapping the complexities of those early event dynamics is an important avenue for future research.

Keywords: developmental origins, health inequalities, early interventions, health production function, health economics


The study of early development began at the end of the 19th century with the German biologist Wilhelm Roux, the founder of a new discipline that he called “developmental mechanics.” Roux’s experiments with frogs’ embryos, and later the experiments by Hans Driesch with sea urchin eggs, were the first studies on the interactions of an embryo with its environment and how developmental trajectories could be guided by information from its surroundings.

In the first part of the 20th century, infancy and childhood were considered critical phases in determining an individual’s lifespan. The actuary Derrick (1927) showed that age-specific mortality rates plotted against year of birth were parallel; he interpreted this finding as indicating that the secular decline in mortality in England and Wales could be better explained as each succeeding generation displaying a lower mortality risk at all ages (“a ‘generation’ influence,” p. 144). The landmark Lancet paper by Kermack, McKendrick, and McKinlay (1934) was the first to point to the importance of early life environment (and of maternal health): “[t]hese results are consistent with the hypothesis that the important factor from the point of view of the health of the individual during his whole life is his environment up to the age of say 15 years, and that improved conditions at later ages have little direct effect” (Kermack et al., 1934, p. 703). However, interest in early life environment fell after World War II when cohort-based mortality predictions were not fulfilled, since in the 1930s mortality rates for adults stopped declining as a result of higher deaths from heart disease. Kuh and Davey Smith (1993) describe how the “generational approach” to forecasting mortality was rejected by the Statistics Committee of the Royal Commission on Population in the early 1950s.

After a few decades, the idea of early life influences on health gained momentum once more in the 1970s. In 1973, Forsdahl (1973)—reprinted as Forsdahl (2002)—analyzed Norwegian data and posited an association between poor living conditions in childhood and adolescence and adult mortality. Forsdahl’s (1973) analyses were the first of a series of epidemiological association studies which investigated the long-term effects of early conditions, and which had David Barker as one of the most visible supporters. In a first paper, Barker and Osmond (1986) found a large positive correlation between standardized infant mortality rates in 1921−1925 and mortality among adults for ischemic heart disease in 1968−1978. Subsequently, Barker and Osmond (1987) used maternal mortality in 1911−1914 as an indicator of early environment and found a strong correlation with death rates from stroke for the generation born at that time. Barker, Osmond, and Law (1989a) is considered the first paper that gave rise to the “fetal origins hypothesis.” The authors showed that the mortality rates from stroke in local authorities in England for the years 1968−1978 were more closely related to neonatal than postneonatal mortality during 1911−1925, emphasizing the importance of the uterine (rather than postnatal) environment. In a series of papers based on the Hertfordshire Cohort Study, starting with Barker, Osmond, Winter, Margetts, and Simmonds (1989b), they showed that men with the lowest weights at birth and at age 1 had the highest death rates from ischemic heart disease. Subsequently, the fetal origins hypothesis evolved into the “developmental origins of health and disease” (DOHaD; Gluckman & Hanson, 2006) to recognize the fact that the influence of early life environment began at conception and extended well into childhood.

Building on the epidemiology tradition, economics in recent years (the early 21st century) has seen a burgeoning of research investigating the long-term effects of early conditions, and two main surveys of the literature have already been carried out (Currie & Almond, 2011; Almond, Currie, & Duque, 2017). Economists have provided important contributions to the field: They range from developing theoretical frameworks, to establishing causality, understanding mechanisms, and also computing costs and benefits of early interventions. Due to space constraints, this article is limited to reviewing the evidence on long-term health impacts of early events in developed countries. To understand the scope of this article, it may be useful to consider the type of studies that are not included. First, studies that only look at short-term outcomes, such as health at birth or in childhood, are omitted. Second, the article excludes studies that examine exclusively economic outcomes such as employment and earnings; in cases where both health and human capital outcomes are studied in one contribution, it will focus on the former. Third, studies on the impacts of shocks and investments occurring in adulthood are not reviewed. Throughout, the contents largely relies on published work. The fact that, even within these limits, the reviewed body of work is substantial testifies to the tremendous interest generated by this field in the early 21st century, which is why this article also includes some particularly relevant working papers from that time.

This article is organized as follows. The next section, “Evidence of Health Inequalities,” presents a brief overview of the historical evolution of inequalities in health. It is followed by a review of pertinent theoretical frameworks (see “Theory”). Economic evidence on the long-term health effects of early shocks, policies, and their interactions is then examined (see “Shocks,” “Policies,” and “Interactions”). Finally, the Conclusion discusses unanswered questions and future research directions. A set of Appendix tables at the end of this article contains key features of each article in more detail (Figure 1a and 1b).

Developmental Origins of Health InequalityClick to view larger

Figure 1 (a): Mean BMI age profiles by gender.

Developmental Origins of Health InequalityClick to view larger

Figure 1 (b): Mean BMI age profiles by gender. NSHD (1946)High SESLow. NCDS (1958)High SESLow. BCS (1970)High SESLow. MCS (2000/1)High SESLow

Adapted from mean BMI age profiles, by gender, for four birth cohort studies in the United Kingdom: the National Survey of Health and Development (NSHD), the National Child Development Study (NCDS), the British Cohort Study 1970 (BCS), and the Millennium Cohort Study (Cohort and Longitudinal Studies Enhancement Resources, 2018a, 2018b, 2018c). Socioeconomic status is based on a dichotomization of parental occupational social class at birth, according to the Registrar General’s 1990 classification. High SES corresponds to classes I (Professional), II (Managerial), IIINM (Skilled nonmanual). Low SES corresponds to IIIM (Skilled manual), IV (Partly skilled), V (Unskilled). Children whose parents are in the armed forces or are unemployed, long-term sick, or retired are excluded. Observations for which social class was coded as unclassifiable or insufficient information was available in the data are also excluded. For BCS and NCDS, the indicator is based exclusively on the father’s occupation. For MCS, the indicator is based on the highest between father’s and mother’s occupational social class. Both the BMI and social class data are harmonized across cohorts, as part of the CLOSER initiative.

Evidence of Health Inequalities

Income inequality has been rising steadily in Organization for Economic Cooperation and Development (OECD) countries. Evidence from the years 2015 and 2016 shows those at the top of income distribution becoming even wealthier at the expense of a shrinking middle class (Lakner & Milanovic, 2016). Inequality in life expectancy and mortality has increased over the past 20 years in the United States among the older-aged population (Chetty et al., 2016; Case & Deaton, 2015). Other evidence shows that mortality gaps between those most well-off and those least well-off are actually decreasing in individuals of younger ages (Currie & Schwandt, 2016).

Enduring health inequalities are related to educational attainment. For example, education gradients in mortality appear to be increasing both in the United States (Pappas, Queen, Hadden, & Fisher, 1993) and Europe (Mackenbach et al., 2003; Kunst et al., 2004). In the United States, decreasing disability rates are, in fact, mainly experienced by the most educated (Schoeni, Freedman, & Martin, 2008). In essence, each additional year of schooling correlates with better health (Lleras-Muney, 2005), although there is no clear evidence of a causal relationship (Galama, Lleras-Muney, & van Kippersluis, 2018).

Socioeconomic status (SES) relates to health in multiple ways. First, higher SES children enjoy a healthier and more nurturing environment since the prenatal period. For example, low SES mothers tend to behave detrimentally to their fetus in a way that mothers from higher SES do not—such as smoking and drinking during pregnancy (Currie & Moretti, 2003; McCrary & Royer, 2011) or gaining excessive pregnancy weight (Ludwig, Rouse, & Currie, 2013a). Second, low SES children are both more likely to develop health conditions and less able to recover from them (Condliffe & Link, 2008). The effect of SES is then prolonged, as more unhealthy children from poor mothers may not be able to compensate for their accumulated disadvantages. Hence, initial disparities between low and high SES groups increase over the life cycle (Case, Lubotsky, & Paxson, 2002), although they tend to narrow at older ages (Case & Deaton, 2005).1 This “long reach” of childhood socioeconomic disadvantage is documented in various contexts. Figure 1 shows average life cycle profiles of BMI for individuals born into different cohorts in the United Kingdom, separately by parental social class at birth. All cohorts exhibit widening gaps between high and low social class children well into adulthood, with gaps emerging earlier and becoming more pronounced for later cohorts. Further work, which has decomposed the sources of adult health inequality by education, has shown that a substantial part of the gradient can be explained by child development and the family environment in the first 10 years of life (Conti, Heckman, & Urzua, 2010).

Health inequalities also have stark intergenerational implications. Not only do children from disadvantaged backgrounds carry on their health conditions into adulthood, but also it becomes apparent that their own children face increased health risks. For example, Currie and Moretti (2007) find children of mothers born with low birth weight to have a 50% higher likelihood of being born with low birth weight themselves. This effect is particularly strong in high-poverty areas. Moreover, evidence from the United States shows that intergenerational health mobility is much greater than income mobility, with important heterogeneities across gender, race, education, and location (Halliday, Mazumder, & Wong, 2018).


This section provides a brief overview of economic theoretical approaches that can rationalize the mounting empirical evidence on the developmental origins of adult health. Most of these approaches are rooted in human capital theory. Human capital can be defined as the intangible stock of knowledge, skills, personality, and other attributes—including health—that produce economic value in the life of an individual.

Human capital theory, pioneered in the 1960s (Becker, 1962, 1965, 1994; Ben-Porath, 1967), has long been employed as a guiding economic framework for understanding individual decisions across the life course. Human capital theory considers the individual (or the household) as a productive unit, which employs market goods and services in conjunction with its own time and transforms them into commodities that generate utility.2

Stemming from this literature, the seminal work by Michael Grossman (1972a, 1972b, 2000) recognized the peculiarity of health among the dimensions of human capital. The model revolves around the concept of health capital: At any point in time, individuals are endowed with health that provides direct utility but—unlike the commodities in Becker’s approach—is a stock. Health depreciates through time due to the aging process but can be replenished by health investments (traditionally, medical care). In the remainder of this section, this and the related frameworks are referred as the “adult health capital approach.”3

In keeping with the human capital approach, the process of transformation of investments into additional health stock is conceptualized as a “health production function.” In its simplest form, adapted from Grossman (2000):


where H is the health stock, δ‎ is the deterioration rate of health, and I(-) is health investment as a function of medical care M and time spent on medical care T. Basically, individuals are viewed as producers of their own health: They cannot augment their health stock directly but they invest Mt and Tt out of their available monetary and time resources in each period to offset the deterioration of their health capital. The production process is also influenced by the individual’s education E, which is assumed to improve productivity. An individual’s health capital stock in turn influences her utility directly (“consumption motive”) and her labor market productivity and income (“investment motive”). Thus, the demand for medical care is only derived from the demand of health, which has utility and income consequences. Here, health investment is viewed primarily as curative care. An important aspect of this basic formulation is linearity: The beginning-of-period health stock enters separately from the investment function, thus assuming that effectiveness of investment is the same at all levels of initial health.

By explicitly recognizing the dynamic nature of health, Grossman’s health capital model has since become the “workhorse” model in health economics. It has been used as the reference framework for understanding the evolution of individual health during adulthood and late life and has spurred a vast legacy of related literature and theoretical extensions.4 However, it has also attracted substantial criticism due to a number of apparent shortcomings, both in its theoretical implications and its ability to replicate empirical patterns.5 Subsequent work has sought to address such flaws by equipping the standard Grossman model to explain a wider range of empirical phenomena. Galama (2015) shows that assuming decreasing returns to scale (DRTS) in the health production process and correctly interpreting equilibrium conditions derived from the health capital model can address much of the criticisms levied in the past from the theoretical side. Furthermore, Galama and Van Kippersluis (2015) propose an extended Grossman-type model that, thanks to a number of additional features, is able to replicate salient features of adult health trajectories, such as the persistence of socioeconomic health disparities across the life course, and in particular the initial widening of SES gradients and their subsequent shrinking around retirement.6

The Grossman approach is mostly focused on adulthood and older age. It is very plausible that the nature of the human capital production process is very different in earlier stages of life; this seems to be true both in terms of the relevant dimensions of human capital involved and which forms of investments are effective in their accumulation. In the past decade or so, renewed attention has been devoted to understanding the process of human capital development in early life, encompassing the period from conception to late childhood. This increasingly popular strand of literature is henceforth referred to as the “child human capital approach.”

In accordance with human capital theory, the process of development in children is also modeled via a production function (Heckman, 2007):


where θt represents child human capital, It are investments, θtB summarizes background and environment factors, and εt are independent random shocks.7

The theoretical constructs of human capital and investments in this framework are different from the adult human capital approach. It has long been recognized that many aspects of children’s development have important consequences across the life course. Cognitive ability has historically been the main focus, with a large body of work examining its determinants in the early years (Todd & Wolpin, 2003). At the same time, a separate literature studies physical development—often proxied by height and BMI (Mandal & Powell, 2014; Puentes et al., 2016). Finally, research in the early 21st century has seen an increased focus on socioemotional skills (Carneiro, Crawford, Goodman, & Centre for the Economics of Education [Great Britain], 2007). An important feature is that child human capital (θt) is recognized as a multidimensional object, with health, cognitive ability, socioemotional skills, and personality traits among the various dimensions. At the same time, investments also reflect this multidimensionality, ranging from healthy behaviors (smoking in pregnancy, breastfeeding) to time and money spent in stimulating activities.8

The very general formulation of Equation 2 allows for a rich characterization of the development process if a suitably flexible empirical specification (e.g., a constant elasticity of substitution) is chosen. Salient properties include self-productivity (θt+1θt>) and dynamic complementarity (2θt+1θtIt>), whereby current levels of human capital affect both future levels and the productivity of current investment (Cunha & Heckman, 2007).9

Furthermore, this approach recognizes and emphasizes the role of shocks (see “Shocks”). The extent to which a shock εt in period t affects human capital in the long term not only depends on the “biological” effect of the shock on the development process, but also on any investment reaction. Flexible formulations of Equation 2 can inform about whether parental investments act to reinforce or compensate earlier shocks (Currie & Almond, 2011).

Both the adult and child human capital approaches have proven capable of generating consistent predictions that find confirmation in empirical analysis. They build on the same basis—the human capital model—which arguably suggests that a unified theory is possible. However, a framework that synthesizes the two approaches and encompasses the whole life cycle is still fundamentally lacking.

The childhood and adulthood periods have so far been treated and modeled separately. This is perhaps due to the inherent complexity in modeling human capital investment decisions along the life course. Starting from childhood, decisions are made by the parents. Through the secondary and higher-education periods, the individual herself becomes the decision-maker: Adult labor market and health outcomes are determined by the interaction of the human capital stock accumulated during childhood with schooling, labor supply, fertility, and other choices. Existing approaches to model the entire life cycle have required dynastic or overlapping-generation models (Halliday, He, Ning, & Zhang, 2016; Gayle, Golan, & Soytas, 2015). Furthermore, as previously mentioned, the evolution of health has been studied somewhat separately from the other dimensions of human development—a limitation which only work produced after the turn of the millennium has started to overcome (Becker, 2007).

While human capital is the predominant theoretical approach in this field, other paradigms are available. For example, a theory of aging based on the accumulation of health deficits has been advanced by Dalgaard and Strulik (2014). They draw on insights from the medical literature on aging and propose that individuals are characterized by a stock of health deficits Dt, which evolve through time in a fashion similar to health capital in the Grossman model:


where µ is the “natural” rate of aging and E are environmental influences, including deliberate investments such as health behaviors and curative care.

This model generates radically different predictions from the human capital model regarding developmental origins of adult health. The assumption that depreciation is proportional to human capital stock, which underlies the human capital model, means that the effect of early disparities is dampened through time. There is still a possibility that such disparities are persistent, but only via their effect on investments It. In the health deficits framework instead, individuals who are initially less healthy have a faster accumulation of health deficits, allowing for persistent—or widening—disparities in later life. Dalgaard, Hansen, and Strulik (2017) argue that such features make the health deficits approach more suitable for modeling developmental origins of health.


As highlighted in “Theory,” contributions in human capital theory have increasingly explicitly incorporated the role of shocks in the process of human development. Indeed, a substantial number of studies has documented the long reach of early shocks, exploiting quasi-random exposure to adverse events of various types. In this section, they are briefly reviewed by type of shock, broadly defined.

One of the first studies in economics to provide long-term evidence for the “fetal origins hypothesis” investigates the effect of in utero exposure to the 1918 influenza pandemic in the United States and finds that the exposed cohort was more likely to suffer from physical disability in late adulthood (Almond, 2006). In a related contribution, Almond and Mazumder (2005) document a higher likelihood of reporting poor health, a range of functional limitations (hearing, speaking, lifting, and walking), and diabetes or stroke for the affected cohort.10. A contribution by Fletcher (2019) instead suggests no effect of the influenza epidemic on overall mortality or death by specific causes. Turning to Europe, Bengtsson and Helgertz (2015) find significant effects on hospitalizations and mortality from prenatal exposure to the flu in Sweden. While the hospitalization effect varies by gender and time of exposure (for boys it is concentrated in the second trimester, for girls in the third), the mortality effect is found only for boys.11

Another widely investigated event is exposure to famine as a source of quasi-random variation in prenatal nutritional intake. In the theoretical framework in “Theory,” this can be conceptualized as a shock to maternal prenatal investments. A commonly exploited historical setting in the literature is the rationing of food supply, which occurred at the time of World War II. Large effects of famine exposure are found in the Netherlands on the likelihood of hospitalization (Scholte, Van den Berg, & Lindeboom, 2015) and on height (Portrait, van Wingerden, & Deeg, 2017; Van den Berg, Pinger, & Schoch, 2016, the latter including Germany and Greece), and in Germany on obesity and metabolic syndromes (Kesternich, Siflinger, Smith, & Winter, 2015). Furthermore, similar effects have been estimated for adult mortality for the Dutch potato famine, which occurred a century before World War II (Lindeboom, Portrait, & Van den Berg, 2010). In all these studies, men appear more affected by famines than women, consistently with the idea that male fetuses are more vulnerable than female fetuses. Moreover, families living in urban areas appear more affected than those living in rural areas, and especially farmers who could rely on home production to protect themselves against food shortages. However, famine studies have been criticized on the grounds that those particular circumstances must have also caused severe stress (and when occurring during a war, loss of material resources), making it difficult to separately identify the effect of undernutrition. Indeed, studies that have examined more broadly the impact of experiencing a war in early life (Börsch-Supan & Jürges, 2011; Kesternich, Siflinger, Smith, & Winter, 2014; Schiman, Kaestner, & Lo Sasso, 2017; Akbulut-Yuksel, 2017; Havari & Peracchi, 2017) have also found similar adverse long-term consequences.12 Nonetheless, the negative impacts of wars appear to affect both men and women to the same extent as well as individuals belonging to different socioeconomic strata.

Researchers have devised alternative identification strategies to study the effects of prenatal malnutrition, such as exposure to Ramadan for children born to Muslim women. Almond and Mazumder (2011) use variation in the day’s duration during Ramadan observance in a U.S. state (Michigan) and two developing countries (Uganda and Iraq) to identify the reduced-form effect of prenatal Ramadan exposure. They find evidence of significant effects on birth weight (for Michigan) and adult disability of any kind (for Uganda and Iraq), with the strongest effects observed on mental disability.

Another popular strategy to study prenatal nutrition exploits within-twin pairs birth weight differences, determined by random variation in nutrients intake due to different positioning of the twin fetuses in the womb. A first important finding from this strand of work is the strong intergenerational persistence of birth weight (Black, Devereux, & Salvanes, 2007; Royer, 2009). Second, the evidence emerging from these studies on the role of birth weight in determining long-term health is mixed. For example, Behrman and Rosenzweig (2004) find effect of birth weight on adult height but not on obesity using U.S. data, while Black et al. (2007) find that an increase in birth weight is associated with an increase in BMI and height using data from Norway. Moreover, they find substantial distributional effects: For example, the positive association with height only exists for birth weights up to 1.5 kg. Using twin birth data from California, Royer (2009) finds an effect of birth weight on adult health outcomes (such as hypertension) only below the low birth weight threshold. In a contribution based on Swedish data, Bharadwaj et al. (2018) find that birth weight has a significant and economically meaningful impact on sickness benefits take-up, hospitalizations, and mortality (the latter only for males). Lastly, Wernerfelt, Slusky, and Zeckhauser (2017) studied the effects of vitamin D deficiency, common among pregnant women, by exploiting variation in sunlight by county together with time use data in a U.S. sample. They found that children whose mothers had increased sunlight exposure, particularly during the second trimester, had a reduced likelihood of developing asthma.13

In addition to nutrition, maternal stress can be viewed as another input in the production of fetal health. Persson and Rossin-Slater (2018a) use the timing of death of a mother’s family member around birth—whether it occurred in utero or postnatally. Families in both groups are equally likely to be affected by the death of a relative and its consequences, with the only difference being in the timing. They are thus able to distinguish the biological transmission of stress from environmental (e.g., income) effects. They find that prenatal exposure to stress is negatively linked to mental health later in life and equally affects mothers of all education levels—an important result given the strong intergenerational persistence of mental health (Johnston, Schurer, & Shields, 2013).14

Since the seminal contribution of Bowlby (1982), the lack of a secure attachment relationship is a form of early shock that has been widely studied in the biomedical literature. The few existing papers in the economics literature show that child maltreatment has significant adverse long-term consequences, especially for mental health (see, e.g., Conti, Morris, Melnychuk, & Pizzo, 2017; Fletcher, 2009); however, as noted in Doyle Jr. and Aizer (2018), more research is needed on the health effects of child abuse and neglect.15

Environmental shocks due to exposure to various sources of pollution have also received increasing scrutiny. One strand of the literature has studied the long-term impacts of exposure to radioactive fallout. The first studies exploited the 1986 Chernobyl accident and found that Ukrainian cohorts exposed in utero were more likely to be in poor health and to suffer from anemia or gastrointestinal illness 20 years later (Lehmann & Wadsworth, 2011). They also had a higher likelihood of mental disorders and reduced expected survival (Danzer & Danzer, 2016).16 Following this work, Black, Bütikofer, Devereux, and Salvanes (2017) found that radioactive fallout at much lower dosages had a negative impact on cognitive ability (as proxied by IQ tests and education), concentrated among those exposed in the second trimester of pregnancy, but not on physical development.17 Little is known about the long-term health impacts of air pollution from particulates, since most studies focus on education or earnings or on short-term health effects.18 One exception is Bharadwaj, Zivin, Mullins, and Neidell (2016), who studied the effects of prenatal and early-life exposure to the 1952 Great Smog of London on the lifelong development of asthma. While they found a significant effect in childhood asthma prevalence for those being exposed in the first year of life, the effect in adulthood was not significant.

Lastly, a budding literature has studied the long-term consequences of growing up in recessions. Common findings are negative health effects persisting into late adulthood. See Cutler, Miller, and Norton (2007) in the United States, Van den Berg, Lindeboom, and Portrait (2006) and Scholte, van den Berg, Lindeboom, and Deeg (2017) in the Netherlands, and Banerjee, Duflo, Postel-Vinay, and Watts (2010) for France.

A potential limitation of the majority of the studies in this section is that they rely on comparisons between exposed and unexposed cohorts to extreme events, which may make the estimated impacts biased due to selection effects. While the effect of interest is that of prenatal “programming”—in other words, the adverse effect of shocks in utero that children carry over their lives (scarring)—the exposure to such events can have an immediate effect on prenatal mortality for the most vulnerable. This means that the children who survive are the most physically strong and have better outcomes (survival, or culling effect).19 In dealing with the selection problem, most of these studies examine selective mortality by looking at the gender ratio at birth, motivated by the biological theory according to which male fetuses have less chances to survive than females, whether due to vulnerability not favoring males or natural selection in favor of females (Trivers & Willard, 1973).

While a consolidated body of evidence shows that early shocks have long-term consequences, little is known about the underlying mechanisms and particularly about the relative importance of biological pathways versus behavioral responses.20 Answering this question requires moving away from reduced-form analysis to estimate more structural relationships. One of the few exceptions is Yi, Heckman, Zhang, and Conti (2015), who studied how an early health shock affected adolescent human capital. The authors estimated a linear production function using twins data from China and showed that parents made compensatory investments in health and reinforcing investments in education, hence acting as net equalizers. Although biological effects outweighed the intrahousehold allocation effect so that the reduced-form estimate substantially underestimated the full impact of a shock in the case of compensatory responses, their results also showed that some remediation was possible. More work along these lines would be particularly welcome in understanding how different early shocks may be operating in determining subsequent health.


As shown in “Shocks,” children are highly sensitive to adverse circumstances in early life. At the same time, pregnancy and early childhood are viewed as a critical window of opportunity for interventions that promote development. This section reviews evidence on the long-term health effects of a range of early interventions. Notably, only a subset of them refers to the provision of health insurance and health care: The vast majority focus on the so-called social determinants of health, whose importance has been increasingly recognized (see, e.g., Marmot, 2005).

Within each type of policy, a useful distinction is between targeted and universal programs. Targeted programs, like income supplementation, usually are delivered to the neediest, so they are designed to help reduce inequalities. For universal programs, the result is not so clear: Needier families may be less capable to fully reap the benefits of a program, potentially leading to an increase in inequality. This can be contextualized in the theoretical approach presented in “Theory”: Different public investments may exhibit different degrees of complementarity with initial endowments and private investments. Whether the overall effect reduces or promotes inequality hinges on the direction of these complementarities.

Provision of child care has become increasingly salient in the policy debate given that most children are placed in some form of care during their preschool period. Campbell et al. (2014) is the first study to show long-term health effects of an early intervention program (the Abecedarian) providing stimulation, nutrition, and health care to disadvantaged children in the first 5 years of life. The authors find that the children randomized to the treatment group exhibit a lower prevalence of risk factors for cardiovascular and metabolic conditions (such as hypertension) in their mid-30s. Along the same lines, using variation in municipalities’ time of implementation of child care in Denmark, Rossin-Slater and Wüst (2016) show that access to a high-quality preschool targeted to poor children has persistent effects on health outcomes through age 65. However, in many countries, the diffusion of targeted child care programs has only been widely implemented with the beginning of the 21st century so that studies of long-term impacts are still unavailable. Nonetheless, medium-term improvements in several health measures have been found for Head Start, a large-scale program targeting disadvantaged children in the United States (Ludwig & Miller, 2007; Carneiro & Ginja, 2014). Turning to universal preschool, a contribution by Herbst (2017) (which exploits random variation in preschool implementation in the United States during World War II) finds a decrease in work-related disability by middle age.

Home-visiting programs are being increasingly advocated as a promising tool to alleviate health inequalities, although there is a limited number of studies investigating their sustained effectiveness. The longest term evidence available comes from public programs implemented in the Scandinavian countries at the inception of the welfare state. Hjort, Sølvsten, and Wüst (2017) find that a universal home visiting program in Denmark reduced mortality at ages 45−57 as well as the incidence of cardiovascular diseases and hospital admissions. Bhalotra, Karlsson, and Nilsson (2017) examine an early intervention in Sweden, providing information and support to mothers via home visits and clinics, with an emphasis on nutrition and sanitation. They find similar positive effects in terms of reduced mortality by age 50 and increased survival to age 75, especially due to reductions in cancer and cardiovascular mortality. While the authors find no differential effects by gender, they document greater gains for the most vulnerable groups (children outside marriage and children of younger mothers), thus pointing to the potential of reducing health inequalities. In the United States, despite a plethora of studies looking at the effects of various home visiting programs, there has been no study yet showing long-term effects on adult health, as the exposed cohorts are still too young.21

Another important set of policies targeting the immediate postnatal period is family leave policies. As noticed in a review by Datta and Gupta (2018), most of the evidence points to weak or zero effects on children’s medium-term outcomes, especially when the counterfactual is high-quality care. However, only scant evidence exists on children’s long-term health and physical development; this is an important avenue to pursue in future research.

Another line of research focuses on the long-term effects of early childhood access to medical care: a critical input, since early health problems are significant predictors of later health (Currie, Stabile, Manivong, & Roos, 2010). Most research exploits the implementation (1960−1970s) or the expansion (1980−1990s) of Medicaid in the United States, a federal- and state-funded program helping disadvantaged families with children cover medical expenditures. The program has been found to lead to a reduction in health limitations and chronic conditions at ages 18−21 (Thompson, 2017), mortality at age 28 (Brown, Kowalski, & Lurie, 2015), and mortality and disability in the 40s (Goodman-Bacon, 2016). Overall, access to Medicaid is found to be beneficial to participants’ long-term health. In relation to the timing of exposure, Miller and Wherry (2018) have shown that those exposed in utero have larger health improvements in their mid-20s than those who were exposed postnatally. This result is corroborated by Boudreaux, Golberstein, and McAlpine (2016), who also show that health gains are largest among those exposed to Medicaid during the prenatal period. Furthermore, exploiting a discontinuity around a birthday cut-off, Wherry, Miller, Kaestner, and Meyer (2018) find that longer childhood eligibility translates into fewer hospitalizations and reduced disease-related mortality at ages 15−18 (Wherry & Meyer, 2016). The effects are mostly limited to Black cohorts, with Whites having either smaller or no gains. This heterogeneity underpins the importance of health care in childhood on reducing health inequalities by race over the life cycle.

Complementary evidence from the United Kingdom shows that early life medical coverage due to the introduction of the National Health Service (NHS) in 1948 resulted in better health and fewer hospitalizations later in life, especially for women and children in lower social classes (Lührmann & Wilson, 2018). Finally, long-term evidence comes from Norway, where Bütikofer, Løken, and Salvanes (2015) exploit the variation across municipalities in the implementation of child health care centers during the 1930s. This universal intervention was aimed at providing information on infant nutrition and, in general, to promote child well-being. The authors find that the program reduced the incidence of diabetes, hypertension, and heart attacks at age 40 and was also found to increase overall adult height.22

Another strand of work has evaluated specific public health policies, such as universal vaccination. Evidence from Norway, using cohort and regional variation, shows that the 1948 tuberculosis testing and vaccination campaign significantly reduced mortality before the age of 66 (Bütikofer & Salvanes, 2018). Gains were higher for those from low socioeconomic backgrounds, showing again the potential for early childhood interventions to reduce adult health inequalities. Additionally, given that the intervention was administered by the school system, this suggests a potential mechanism behind the widely documented positive effect of schooling on health for early 20th-century cohorts (Mazumder, 2007). In related work, Bhalotra and Venkataramani (2015) exploit the introduction of antibiotic therapies in 1937 in the United States to examine the effects of infant pneumonia on adult outcomes. Although there were large significant effects on all outcomes studied (among which was disability), African Americans living in more racially segregated states gained less from the program compared to those in less segregated states. This suggests the importance of institutions in delivering such programs to the most disadvantaged and, as a consequence, in tackling health inequality. Finally, improvements in sanitation have been shown to be one of the most important and successful interventions delivered in the 20th century in developed countries, especially in larger cities, due to rapid economic growth and urbanization. For example, Kesztenbaum and Rosenthal (2017) show how the expansion of Paris’s sewage system led to an increase in Parisians’ life expectancy by several years. Along the same lines, Cutler and Miller (2005) show how access to clean water decreased mortality rates in the United States, with greater benefits in cities with higher illiteracy rates. However, Anderson, Charles, and Rees (2018) have challenged these findings, showing that none of the several public interventions studied contributed substantially to reductions in mortality. Importantly, Alsan and Goldin (2015) show that the introduction of drinking water and sewerage systems are complementary interventions, displaying higher effectiveness in decreasing infant mortality when combined. While such public health interventions are established in developed countries, disadvantaged families may still face problems in accessing clean water.23 Additionally, new problems are emerging, such as that of hydraulic fracturing (fracking).

Lastly, a separate strand of targeted programs do not provide specific inputs (such as health or child care), but rather monetary resources or improved environments. On the one hand, cash transfers to underprivileged families have been shown to impact children’s long-term health. For example, Aizer, Eli, Ferrie, and Lleras-Muney (2016) have studied the Mothers’ Pension Program in the United States (1911−1931) and found that the children of poor mothers who were accepted into the program were less likely to be underweight and more likely to live longer than the children of those mothers who were rejected from it. Moreover, the program successfully targeted inequality, as the effects were larger for the poorest families. Also in the United States, Hoynes, Schanzenbach, and Almond (2016) find that access to food stamps in the 1960s, in the period between conception and age 5, had a positive effect on adult weight, height, likelihood of disability, and metabolic syndrome. As a program targeted at those living in poverty, it also helped reduce health inequalities. On the other hand, the evidence is more mixed in relation to the health impacts of changes in the neighborhood environment.24

The evidence reviewed in this section shows that early years interventions of various content, especially those targeting poor families, have the potential to improve life course health. While there is a robust body of evidence on causal impacts, particular knowledge about mechanisms is scarce. Understanding why and for whom early interventions improve health across the life course is an important avenue of future work. It requires going beyond reduced-form evidence to estimate more structural relationships, in particular adult health production functions incorporating early life inputs. This would build on a body of literature from the early 21st century, which has shown the role of early childhood development in the production of adult health in all its dimensions: cognitive, socioemotional, and health development (Conti et al., 2010, Conti, Heckman, & Urzua, 2011; Conti & Hansman, 2013; Kaestner & Callison, 2011; Mendolia & Walker, 2014; Bijwaard, van Kippersluis, & Veenman, 2015; Case & Paxson, 2010). A first example of this type of analysis is provided in Conti, Heckman, and Pinto (2016), who develop a novel dynamic mediation analysis to study the mechanisms through which the Perry Preschool and the Abecedarian programs improved adult health. They show a significant role played by improved childhood development above and beyond the effects of experimentally enhanced adult socioeconomic status. In particular, reduced externalizing behavior at ages 7−9 years and improved task orientation in the first 2 years of life were the key drivers of reductions in smoking and hypertension in the Perry and Abecedarian programmes, respectively.


The previous sections documented the available empirical evidence on the effects of shocks and interventions during the prenatal and childhood period on adult health. An unequivocal finding is that children’s early environment can have lasting effects. One particular feature is that the evidence discussed mostly considers one shock or one intervention, at a specific point in time. Yet, multiple shocks of different types (ranging from environmental to socioeconomic) can affect human development since the prenatal period, and very little is known about their consequences. Along the same lines, pregnant women, fetuses, and children may be exposed to different policies at different stages of development. In the theoretical framework, this amounts to studying the role of dynamic complementarities between shocks (ε‎) and investments (I) at different time points in the development process.

Research in these areas is rather limited, partly because of the demanding requirements both in terms of data and research design. Data are very scarce on cohorts who experienced both shocks and interventions in a quasi-random fashion, as well as spans long enough through their lives in order to observe long-term effects on their health. Most of the existing empirical applications use data from developing countries and focus on short-term effects.25 An exception is a historical study by Arthi (2018), who explores whether in utero exposure to a negative shock, the 1930s Dust Bowl in the United States, was remediated by the subsequent implementation of the New Deal. Looking at physical disability and vision and hearing difficulties, she provides evidence supporting this remediating mechanism.

Campbell et al. (2014) is the only study reporting evidence from a sequential randomization in the Abecedarian intervention, where both the treatment and the control group in the preschool stage (ages 0−5) were re-randomized to the school stage (ages 6−8). The authors show that the children who were in the school-age treatment (irrespective of the preschool treatment allocation) did not have significant benefits for their adult health. Another study is that by Rossin-Slater and Wüst (2016), who study the interaction between a nurse home visiting program (targeting children at birth) and a childcare program (targeting children at 3 years of age) that occurred between the 1930s and the 1950s, using administrative data from Denmark. Although they find significant positive effects, overall gains from the childcare program were greater for those unexposed to the nurse home visiting intervention. This seems to run counter to the concept of dynamic complementarity, since the productivity of the later intervention was smaller for those with higher initial stock due to the birth intervention.

Hence, work on the interaction of shocks and policies in different periods within a life course model of health production is a particularly important avenue for future research.


This article reviews the literature in economics on the developmental origins of health. An overview of the available theoretical frameworks through which the long reach of early circumstances can be understood, and through which empirical findings can be contextualized, has been provided. It has also been shown how the frameworks employed to study health production in adulthood and child development have been developed somewhat separately, and an overarching comprehensive theory of health capital production across the entire life course is still fundamentally lacking.

A review of the evidence on the long-term health effects of early-life health shocks of various types, ranging from environmental to nutritional to economic shocks, proposed that these shocks have persistent effects on different measures of health (ranging from height to mortality), with males usually more affected than females whenever gender differences were analyzed.26 At the same time, various policies and interventions of different content and delivery format (ranging from health insurance coverage to home visiting programs) have been effective at improving long-term health, with greater gains for the most socioeconomically disadvantaged groups and for earlier exposures.

The literature has provided robust causal evidence on long-term effects. However, this is mostly limited to reduced-form impacts of single shocks or policies. Several unknowns remain and represent important avenues of future research. First, researchers’ understanding of the mechanisms through which shocks, investments, and policies interact is just in its infancy. This requires estimating more structural relationships via health production functions, which have more demanding data requirements. Second, the public health literature has suggested the importance of the preconception period for future health (Stephenson et al., 2018). Studies in economics usually don’t examine exposures occurring earlier than the pregnancy period. Third, more rigorous life-cycle cost-benefits analyses of early interventions would be beneficial.27 Fourth, more sensitive measures of health phenotypes and underlying biological processes (including genetic and epigenetic markers) would be beneficial in disentangling the various mechanisms through which early shocks and interventions affect life course health.28


Gabriella Conti and Stavros Poupakis have received funding from the European Union’s Horizon 2020 research and innovation program (grant agreement No. 633595 DynaHEALTH). Gabriella Conti has also received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No. 819752).

Further Reading

Almond, D., Currie, J., & Duque, V. (2018). Childhood circumstances and adult outcomes: Act II. Journal of Economic Literature, 56(4), 1360–1446.Find this resource:

Braveman, P., Egerter, S., & Williams, D. R. (2011). The social determinants of health: Coming of age. Annual Review of Public Health, 32, 381–398.Find this resource:

Conti, G., & Heckman, J. J. (2014). Economics of child well-being. In A. Ben-Arieh, F. Casas, I. Frønes, & J. E. Korin (Eds.), Handbook of child well-being (pp. 363–401). Dordrecht, The Netherlands: Springer.Find this resource:

Currie, J., & Almond, D. (2011). Human capital development before age five. In O. Ashenfelter & D. Card (Eds.), Handbook of labor economics (Vol. 4, pp. 1315–1486). Amsterdam, The Netherlands: Elsevier.Find this resource:

Deaton, A. (2003). Health, inequality, and economic development. Journal of Economic Literature, 41(1), 113–158.Find this resource:

Elango, S., García, J. L., Heckman, J. J., & Hojman, A. (2015). Early childhood education (NBER Working Paper No. 21766). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Fogel, R. W. (2012). The escape from hunger and premature death, 1700–2100: Europe, America, and the Third World. Cambridge, U.K.: Cambridge University Press.Find this resource:

Halfon, N., Forrest, C. B., Lerner, R. M., & Faustman, E. M. (Eds.). (2018). Handbook of life course health development. Dordrecht, The Netherlands: Springer.Find this resource:

Heckman, J. J., & Mosso, S. (2014). The economics of human development and social mobility. Annual Review of Economics, 6(1), 689–733.Find this resource:

Hertzman, C., & Boyce, T. (2010). How experience gets under the skin to create gradients in developmental health. Annual Review of Public Health, 31, 329–347.Find this resource:


Adhvaryu, A., Nyshadham, A., Molina, T., & Tamayo, J. (2018). Helping children catch up: Early life shocks and the PROGRESA experiment (NBER Working Paper No. 24848). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Aguilar, A., & Vicarelli, M. (2015). El Niño and Mexican children: Medium-term effects of early-life weather shocks on cognitive and health outcomes. Mimeo. Cambridge, MA: Harvard University.Find this resource:

Aizer, A., Eli, S., Ferrie, J., & Lleras-Muney, A. (2016). The long-run impact of cash transfers to poor families. American Economic Review, 106(4), 935–971.Find this resource:

Akbulut-Yuksel, M. (2017). War during childhood: The long run effects of warfare on health. Journal of Health Economics, 53, 117–130.Find this resource:

Almond, D. (2006). Is the 1918 influenza pandemic over? Long-term effects of in utero influenza exposure in the post-1940 US population. Journal of Political Economy, 114(4), 672–712.Find this resource:

Almond, D., Currie, J., & Duque, V. (2017). Childhood circumstances and adult outcomes: Act II (NBER Working Paper No. 23017). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Almond, D., Edlund, L., & Palme, M. (2009). Chernobyl’s subclinical legacy: Prenatal exposure to radioactive fallout and school outcomes in Sweden. Quarterly Journal of Economics, 124(4), 1729–1772.Find this resource:

Almond, D., & Mazumder, B. (2005). The 1918 influenza pandemic and subsequent health outcomes: An analysis of SIPP data. American Economic Review, 95(2), 258–262.Find this resource:

Almond, D., & Mazumder, B. (2011). Health capital and the prenatal environment: The effect of Ramadan observance during pregnancy. American Economic Journal: Applied Economics, 3(4), 56–85.Find this resource:

Almond, D., & Mazumder, B. (2013). Fetal origins and parental responses. Annual Review of Economics, 5(1), 37–56.Find this resource:

Alsan, M., & Goldin, C. (2015). Watersheds in infant mortality: The role of effective water and sewerage infrastructure, 1880 to 1915 (NBER Working Paper No. 21263). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Anderson, D. M., Charles, K. K., & Rees, D. I. (2018). Public health efforts and the decline in urban mortality (NBER Working Paper No. 25027). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Arthi, V. (2018). “The dust was long in settling”: Human capital and the lasting impact of the American Dust Bowl. Journal of Economic History, 78(1), 196–230.Find this resource:

Banerjee, A., Duflo, E., Postel-Vinay, G., & Watts, T. (2010). Long-run health impacts of income shocks: Wine and phylloxera in nineteenth-century France. Review of Economics and Statistics, 92(4), 714–728.Find this resource:

Barker, D. J., & Osmond, C. (1986). Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. The Lancet, 327(8489), 1077–1081.Find this resource:

Barker, D., & Osmond, C. (1987). Death rates from stroke in England and Wales predicted from past maternal mortality. British Medical Journal (Clinical Research Education), 295(6590), 83–86.Find this resource:

Barker, D. J., Osmond, C., & Law, C. (1989a). The intrauterine and early postnatal origins of cardiovascular disease and chronic bronchitis. Journal of Epidemiology & Community Health, 43(3), 237–240.Find this resource:

Barker, D. J., Osmond, C., Winter, P., Margetts, B., & Simmonds, S. J. (1989b). Weight in infancy and death from ischaemic heart disease. The Lancet, 334(8663), 577–580.Find this resource:

Beach, B., Ferrie, J. P., & Saavedra, M. H. (2018). Fetal shock or selection? The 1918 influenza pandemic and human capital development (NBER Working Paper No. 24725). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Becker, G. S. (1962). Investment in human capital: A theoretical analysis. Journal of Political Economy, 70(5), 9–49.Find this resource:

Becker, G. S. (1994). Human capital: A theoretical and empirical analysis, with special reference to education. 3rd Edition. New York, NY: National Bureau of Economic Research.Find this resource:

Becker, G. S. (1965). A theory of the allocation of time. Economic Journal, 75(299), 493–517.Find this resource:

Becker, G. S. (2007). Health as human capital: Synthesis and extensions. Oxford Economic Papers, 59(3), 379–410.Find this resource:

Becker, G. S., & Murphy, K. M. (1988). A theory of rational addiction. Journal of Political Economy, 96(4), 675–700.Find this resource:

Behrman, J. R., & Rosenzweig, M. R. (2004). Returns to birthweight. Review of Economics and Statistics, 86(2), 586–601.Find this resource:

Bengtsson, T., & Helgertz, J. (2015). The long-lasting influenza: The impact of fetal stress during the 1918 influenza pandemic on socioeconomic attainment and health in Sweden 19682012 (Discussion Paper 9327). Bonn, Germany: IZA Institute of Labor Economics.Find this resource:

Ben-Porath, Y. (1967). The production of human capital and the life cycle of earnings. Journal of Political Economy, 75(4, Part 1), 352–365.Find this resource:

Bhalotra, S., Karlsson, M., & Nilsson, T. (2017). Infant health and longevity: Evidence from a historical intervention in Sweden. Journal of the European Economic Association, 15(5), 1101–1157.Find this resource:

Bhalotra, S., & Venkataramani, A. S. (2015). Shadows of the captain of the men of death: Health innovation, human capital investment, and institutions. Mimeo.Find this resource:

Bharadwaj, P., Lundborg, P., & Rooth, D.-O. (2018). Birth weight in the long run. Journal of Human Resources, 53(1), 189–231.Find this resource:

Bharadwaj, P., Zivin, J. G., Mullins, J. T., & Neidell, M. (2016). Early-life exposure to the great smog of 1952 and the development of asthma. American Journal of Respiratory and Critical Care Medicine, 194(12), 1475–1482.Find this resource:

Bijwaard, G. E., van Kippersluis, H., & Veenman, J. (2015). Education and health: The role of cognitive ability. Journal of Health Economics, 42, 29–43.Find this resource:

Black, S. E., Bütikofer, A., Devereux, P. J., & Salvanes, K. G. (2017). This is only a test? Long-run impacts of prenatal Exposure to radioactive fallout (NBER Working Paper No. 18987), Cambridge, MA: National Bureau of Economic Research.Find this resource:

Black, S. E., Devereux, P. J., & Salvanes, K. G. (2007). From the cradle to the labor market? The effect of birth weight on adult outcomes. Quarterly Journal of Economics, 122(1), 409–439.Find this resource:

Börsch-Supan, A. H., & Jürges, H. (2011). Disability, pension reform and early retirement in Germany (NBER Working Paper No. 17079). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Boudreaux, M. H., Golberstein, E., & McAlpine, D. D. (2016). The long-term impacts of Medicaid exposure in early childhood: Evidence from the program’s origin. Journal of Health Economics, 45, 161–175.Find this resource:

Bowlby, J. (1982). Attachment and loss—Volume I. 2nd Edition. New York, NY: Basic Books.Find this resource:

Bozzoli, C., Deaton, A., & Quintana-Domeque, C. (2009). Adult height and childhood disease. Demography, 46(4), 647–669.Find this resource:

Brown, D. W., Kowalski, A. E., & Lurie, I. Z. (2015). Medicaid as an investment in children: What is the long-term impact on tax receipts? (NBER Working Paper No. 20835). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Brown, R., & Thomas, D. (2016). On the long-term effects of the 1918 US influenza pandemic. Mimeo. Durham, NC: Duke University.Find this resource:

Bütikofer, A., Løken, K. V., & Salvanes, K. G. (2015). Long-term consequences of access to well-child visits (Discussion Paper 29/2015). Bergen, Norway: NHH Department of Economics.Find this resource:

Bütikofer, A., & Salvanes, K. G. (2018). Disease control and inequality reduction: Evidence from a tuberculosis testing and vaccination campaign (Discussion Paper 11709). Bonn, Germany: IZA Institute of Labor Economics.Find this resource:

Campbell, F., Conti, G., Heckman, J. J., Moon, S. H., Pinto, R., Pungello, E., & Pan, Y. (2014). Early childhood investments substantially boost adult health. Science, 343(6178), 1478–1485.Find this resource:

Carneiro, P., Crawford, C., Goodman, A., & Centre for the Economics of Education (Great Britain). (2007). The impact of early cognitive and non-cognitive skills on later outcomes (Technical Report, Centre for the Economics of Education). London, U.K.: London School of Economics. OCLC: 183819510Find this resource:

Carneiro, P., & Ginja, R. (2014). Long-term impacts of compensatory preschool on health and behavior: Evidence from head start. American Economic Journal: Economic Policy, 6(4), 135–173.Find this resource:

Case, A., & Deaton, A. S. (2005). Broken down by work and sex: How our health declines. In D. A. Wise (Ed.), Analyses in the economics of aging (pp. 185–212). Chicago, IL: University of Chicago Press.Find this resource:

Case, A., & Deaton, A. (2015). Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. Proceedings of the National Academy of Sciences, 112(49), 15078–15083.Find this resource:

Case, A., Lubotsky, D., & Paxson, C. (2002). Economic status and health in childhood: The origins of the gradient. American Economic Review, 92(5), 1308–1334.Find this resource:

Case, A., & Paxson, C. (2010). Causes and consequences of early-life health. Demography, 47(1), S65–S85.Find this resource:

Chetty, R., Stepner, M., Abraham, S., Lin, S., Scuderi, B., Turner, N., . . . Cutler, D. (2016). The association between income and life expectancy in the United States, 2001–2014. JAMA, 315(16), 1750–1766.Find this resource:

Cohort and Longitudinal Studies Enhancement Resources. (2018a). Harmonised socio-economic measures in four longitudinal cohort studies: 1970 British cohort study [data collection]. UK Data Service.Find this resource:

Cohort and Longitudinal Studies Enhancement Resources. (2018b). Harmonised socio-economic measures in four longitudinal cohort studies: Millennium cohort study [data collection]. UK Data Service.Find this resource:

Cohort and Longitudinal Studies Enhancement Resources. (2018c). Harmonised socio-economic measures in four longitudinal cohort studies: National child development study [data collection]. UK Data Service.Find this resource:

Condliffe, S., & Link, C. R. (2008). The relationship between economic status and child health: Evidence from the United States. American Economic Review, 98(4), 1605–1618.Find this resource:

Conti, G., Frühwirth-Schnatter, S., Heckman, J. J., & Piatek, R. (2014). Bayesian exploratory factor analysis. Journal of Econometrics, 183(1), 31–57.Find this resource:

Conti, G., & Hansman, C. (2013). Personality and the education–health gradient: A note on “Understanding differences in health behaviors by education.” Journal of Health Economics, 32(2), 480–485.Find this resource:

Conti, G., Hanson, M., Inskip, H., Crozier, S., Cooper, C., & Godfrey, K. (2018). Beyond birth weight: The origins of human capital (IFS Working Paper W18/30). London, U.K.: Institute for Fiscal Studies.Find this resource:

Conti, G., Heckman, J. J., & Pinto, R. (2016). The effects of two influential early childhood interventions on health and healthy behavior. Economic Journal, 126(596), F28–F65.Find this resource:

Conti, G., Heckman, J. J., & Urzua, S. (2010). The education-health gradient. American Economic Review: Papers & Proceedings, 100(2), 234–238.Find this resource:

Conti, G., Heckman, J. J., & Urzua, S. (2011). Early endowments, education, and health (Working Paper 2011–001). Human Capital and Economic Opportunity: A Global Working Group. Chicago, IL: University of Chicago.Find this resource:

Conti, G., Morris, S., Melnychuk, M., & Pizzo, E. (2017). The economic costs of child maltreatment in the United Kingdom: A preliminary study. London, UK: NSPCC Learning.Find this resource:

Cropper, M. L. (1977). Health, investment in health, and occupational choice. Journal of Political Economy, 85(6), 1273–1294.Find this resource:

Cunha, F., & Heckman, J. J. (2007). The technology of skill formation. American Economic Review, 97(2), 31–47.Find this resource:

Cunha, F., Heckman, J. J., & Schennach, S. (2010). Estimating the technology of cognitive and noncognitive skill formation. Econometrica, 78(3), 883–931.Find this resource:

Currie, J., & Almond, D. (2011). Human capital development before age five. In D. Card & O. Ashenfelter (Eds.), Handbook of labor economics (Vol, 4, pp. 1315–1486). Amsterdam, The Netherlands: Elsevier.Find this resource:

Currie, J., Graff Zivin, J., Meckel, K., Neidell, M., & Schlenker, W. (2013). Something in the water: Contaminated drinking water and infant health. Canadian Journal of Economics/Revue canadienne d’´economique, 46(3), 791–810.Find this resource:

Currie, J., Greenstone, M., & Meckel, K. (2017). Hydraulic fracturing and infant health: New evidence from Pennsylvania. Science Advances, 3(12), e1603021.Find this resource:

Currie, J., & Moretti, E. (2003). Mother’s education and the intergenerational transmission of human capital: Evidence from college openings. Quarterly Journal of Economics, 118(4), 1495–1532.Find this resource:

Currie, J., & Moretti, E. (2007). Biology as destiny? Short-and long-run determinants of intergenerational transmission of birth weight. Journal of Labor Economics, 25(2), 231–264.Find this resource:

Currie, J., & Schwandt, H. (2016). Mortality inequality: The good news from a county-level approach. Journal of Economic Perspectives, 30(2), 29–52.Find this resource:

Currie, J., Stabile, M., Manivong, P., & Roos, L. L. (2010). Child health and young adult outcomes. Journal of Human Resources, 45(3), 517–548.Find this resource:

Currie, J., Zivin, J. G., Mullins, J., & Neidell, M. (2014). What do we know about short-and long-term effects of early-life exposure to pollution? Annual Review of Resource Economics, 6(1), 217–247.Find this resource:

Cutler, D., & Miller, G. (2005). The role of public health improvements in health advances: The twentieth-century United States. Demography, 42(1), 1–22.Find this resource:

Cutler, D. M., Miller, G., & Norton, D. M. (2007). Evidence on early-life income and late-life health from America’s Dust Bowl era. Proceedings of the National Academy of Sciences, 104(33), 13244–13249.Find this resource:

Dalgaard, C.-J., & Strulik, H. (2014). Optimal aging and death: Understanding the Preston curve. Journal of the European Economic Association, 12(3), 672–701.Find this resource:

Dalgaard, C.-J. L., Hansen, C. W., & Strulik, H. (2017). Accounting for fetal origins: Health capital vs. health deficits. Technical Report 17–11. Costerton Biofilm Center. Copenhagen, Denmark: University of Copenhagen.Find this resource:

Danzer, A. M., & Danzer, N. (2016). The long-run consequences of Chernobyl: Evidence on subjective well-being, mental health and welfare. Journal of Public Economics, 135, 47–60.Find this resource:

Dardanoni, V., & Wagstaff, A. (1987). Uncertainty, inequalities in health and the demand for health. Journal of Health Economics, 6(4), 283–290.Find this resource:

Dardanoni, V., & Wagstaff, A. (1990). Uncertainty and the demand for medical care. Journal of Health Economics, 9(1), 23–38.Find this resource:

Datta Gupta, N. (2018). Maternity leave versus early childcare: What are the long-term consequences for children? IZA World of Labor.Find this resource:

Derrick, V. (1927). Observations on (1) errors of age in the population statistics of England and Wales, and (2) the changes in mortality indicated by the national records. Journal of the Institute of Actuaries, 58(2), 117–159.Find this resource:

Doyle Jr., J. J., & Aizer, A. (2018). Economics of child protection: Maltreatment, foster care, and intimate partner violence. Annual Review of Economics, 10, 87–108.Find this resource:

Duque, V., Rosales-Rueda, M., & Sanchez, F. (2018). How do early-life shocks interact with subsequent human-capital investments? Evidence from administrative data. IZA World of Labor Conference.Find this resource:

Ehrlich, I., & Chuma, H. (1990). A model of the demand for longevity and the value of life extension. Journal of Political Economy, 98(4), 761–782.Find this resource:

Fletcher, J. M. (2009). Childhood mistreatment and adolescent and young adult depression. Social Science & Medicine, 68(5), 799–806.Find this resource:

Fletcher, J. M. (2019). Examining the long-term mortality effects of early health shocks. Applied Economics Letters, 26(11), 902−908.Find this resource:

Fogel, R. W. (1997). New findings on secular trends in nutrition and mortality: Some implications for population theory. In M. R. Rosenzweig & O. Stark (Eds.), Handbook of Population and Family Economics (Vol. 1, Part A, pp. 433–481). Amsterdam, The Netherlands: Elsevier.Find this resource:

Forsdahl, A. (1973). Momenter til belysning av den høye dødelighet i finnmark fylke. kan den høye dødelighet i dag være en senfølge av meget dårlige levevilkår i barne-og ungdomsalderen? Tidsskrift for den norske lægeforening, 93(10), 661–667.Find this resource:

Forsdahl, A. (2002). Observations throwing light on the high mortality in the county of Finnmark: Is the high mortality today a late effect of very poor living conditions in childhood and adolescence? International Journal of Epidemiology, 31(2), 302–308.Find this resource:

French, E. (2005). The effects of health, wealth, and wages on labor supply and retirement behavior. Review of Economic Studies, 72(2), 395–427.Find this resource:

Galama, T. J. (2015). A contribution to health-capital theory (Technical Report 2015–008). Human Capital and Economic Opportunity: A Global Working Group. Chicago, IL: University of Chicago.Find this resource:

Galama, T. J., Lleras-Muney, A., & van Kippersluis, H. (2018). The effect of education on health and mortality: A review of experimental and quasi-experimental evidence (NBER Working Paper No. 24225). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Galama, T. J., & van Kippersluis, H. (2015). A theory of socioeconomic disparities in health over the life cycle . Economic Journal.Find this resource:

García, J. L., Heckman, J. J., Leaf, D. E., & Prados, M. J. (2016). The life-cycle benefits of an influential early childhood program (NBER Working Paper No. 22993). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Gayle, G.-L., Golan, L., & Soytas, M. A. (2015). What is the source of the intergenerational correlation in earnings? (Working Paper 2015-019A). St. Louis, MO: Federal Reserve Bank of St. Louis.Find this resource:

Gilleskie, D. (2008). Health capital: Theory and empirical evidence. In F. A. Sloan & H. Kasper (Eds.), Incentives and choice in health care (pp. 51–80). Cambridge, MA: MIT Press.Find this resource:

Gluckman, P. D., & Hanson, M. A. (2006). The developmental origins of health and disease. New York, NY: Springer.Find this resource:

Goodman-Bacon, A. (2016). The long-run effects of childhood insurance coverage: Medicaid implementation, adult health, and labor market outcomes (NBER Working Paper No. 22899). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Gould, E. D., Lavy, V., & Paserman, M. D. (2011). Sixty years after the magic carpet ride: The long-run effect of the early childhood environment on social and economic outcomes. Review of Economic Studies, 78(3), 938–973.Find this resource:

Grossman, M. (1972a). The demand for health: A theoretical and empirical investigation. New York, NY: Columbia University Press.Find this resource:

Grossman, M. (1972b). On the concept of health capital and the demand for health. Journal of Political Economy, 80(2), 223–255.Find this resource:

Grossman, M. (2000). The human capital model. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1, pp. 347–408). Amsterdam, The Netherlands: Elsevier.Find this resource:

Halliday, T., Mazumder, B., & Wong, A. (2018). Intergenerational health mobility in the US (Working Paper No. 11304). Bonn, Germany: IZA Institute of Labor Economics.Find this resource:

Halliday, T. J., He, H., Ning, L., & Zhang, H. (2016). Health investment over the life cycle. Technical Report.Find this resource:

Havari, E., & Mazzonna, F. (2015). Can we trust older people’s statements on their childhood circumstances? Evidence from SHARELIFE. European Journal of Population, 31(3), 233–257.Find this resource:

Havari, E., & Peracchi, F. (2017). Growing up in wartime: Evidence from the era of two world wars. Economics & Human Biology, 25, 9–32.Find this resource:

Heckman, J. J. (2007). The economics, technology, and neuroscience of human capability formation. Proceedings of the National Academy of Sciences, 104(33), 13250–13255.Find this resource:

Herbst, C. M. (2017). Universal child care, maternal employment, and children’s long-run outcomes: Evidence from the U.S. Lanham Act of 1940. Journal of Labor Economics, 35(2), 519–564.Find this resource:

Hjort, J., Sølvsten, M., & Wüst, M. (2017). Universal investment in infants and long-run health: Evidence from Denmark’s 1937 home visiting program. American Economic Journal: Applied Economics, 9(4), 78–104.Find this resource:

House, J. S., Lepkowski, J. M., Kinney, A. M., Mero, R. P., Kessler, R. C., & Herzog, A. R. (1994). The social stratification of aging and health. Journal of Health and Social Behavior, 35(3), 213.Find this resource:

Hoynes, H., Schanzenbach, D. W., & Almond, D. (2016). Long-run impacts of childhood access to the safety net. American Economic Review, 106(4), 903–934.Find this resource:

Johnston, D. W., Schurer, S., & Shields, M. A. (2013). Exploring the intergenerational persistence of mental health: Evidence from three generations. Journal of Health Economics, 32(6), 1077–1089.Find this resource:

Kaestner, R., & Callison, K. (2011). Adolescent cognitive and noncognitive correlates of adult health. Journal of Human Capital, 5(1), 29–69.Find this resource:

Kenkel, D. S. (2000). Prevention. In A. J. Culyer & J. P. Newhouse (Eds.), Handbook of health economics (Vol. 1, Part B, pp. 1675−1720). Amsterdam, The Netherlands: Elsevier.Find this resource:

Kermack, W. O., McKendrick, A. G., & McKinlay, P. L. (1934). Death rates in Great Britain and Sweden: Some general regularities and their significance. Lancet, 226, 698–703.Find this resource:

Kesternich, I., Siflinger, B., Smith, J. P., & Winter, J. K. (2014). The effects of World War II on economic and health outcomes across Europe. Review of Economics and Statistics, 96(1), 103–118.Find this resource:

Kesternich, I., Siflinger, B., Smith, J. P., & Winter, J. K. (2015). Individual behavior as a pathway between early-life shocks and adult health: Evidence from hunger episodes in post-war Germany. Economic Journal, 125(588), F372–F393.Find this resource:

Kesztenbaum, L., & Rosenthal, J.-L. (2017). Sewers’ diffusion and the decline of mortality: The case of Paris, 1880–1914. Journal of Urban Economics, 98, 174–186.Find this resource:

Kuh, D., & Smith, G. D. (1993). When is mortality risk determined? Historical insights into a current debate. Social History of Medicine, 6(1), 101–123.Find this resource:

Kunst, A. E., Bos, V., Andersen, O., Cardano, M., Costa, G., Harding, S., . . . Reid, A. (2004). Monitoring of trends in socioeconomic inequalities in mortality: Experiences from a European project. Demographic Research, 2, 229–254.Find this resource:

Lakner, C., & Milanovic, B. (2016). Global income distribution: From the fall of the Berlin Wall to the Great Recession. World Bank Economic Review, 30(2), 203–232.Find this resource:

Lehmann, H., & Wadsworth, J. (2011). The impact of Chernobyl on health and labor market performance. Journal of Health Economics, 30(5), 843–857.Find this resource:

Lindeboom, M., Portrait, F., & Van den Berg, G. J. (2010). Long-run effects on longevity of a nutritional shock early in life: The Dutch potato famine of 1846–1847. Journal of Health Economics, 29(5), 617–629.Find this resource:

Lleras-Muney, A. (2005). The relationship between education and adult mortality in the United States. Review of Economic Studies, 72(1), 189–221.Find this resource:

Ludwig, D. S., Rouse, H. L., & Currie, J. (2013a). Pregnancy weight gain and childhood body weight: A within-family comparison. PLoS Medicine, 10(10), e1001521.Find this resource:

Ludwig, J., Duncan, G. J., Gennetian, L. A., Katz, L. F., Kessler, R. C., Kling, J. R., & Sanbonmatsu, L. (2013b). Long-term neighborhood effects on low-income families: Evidence from moving to opportunity. American Economic Review, 103(3), 226–231.Find this resource:

Ludwig, J., & Miller, D. L. (2007). Does Head Start improve children’s life chances? Evidence from a regression discontinuity design. Quarterly Journal of Economics, 122(1), 159–208.Find this resource:

Lührmann, M., & Wilson, T. (2018). Long-run health and mortality effects of exposure to universal health care at birth. Mimeo. London, UK: Institute for Fiscal Studies.Find this resource:

Mackenbach, J. P., Bos, V., Andersen, O., Cardano, M., Costa, G., Harding, S., . . . Kunst, A. E. (2003). Widening socioeconomic inequalities in mortality in six Western European countries. International Journal of Epidemiology, 32(5), 830–837.Find this resource:

Malamud, O., Pop-Eleches, C., & Urquiola, M. (2016). Interactions between family and school environments: Evidence on dynamic complementarities? (NBER Working Paper No. 22112). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Mandal, B., & Powell, L. M. (2014). Child care choices, food intake, and children’s obesity status in the United States. Economics & Human Biology, 14, 50–61.Find this resource:

Marmot, M. (2005). Social determinants of health inequalities. The Lancet, 365(9464), 1099–1104.Find this resource:

Marmot, M., Ryff, C. D., Bumpass, L. L., Shipley, M., & Marks, N. F. (1997). Social inequalities in health: Next questions and converging evidence. Social Science & Medicine, 44(6), 901–910.Find this resource:

Matsumoto, B. (2018). Family ruptures, stress, and the mental health of the next generation: Comment. American Economic Review, 108(4−5), 1253–1255.Find this resource:

Mazumder, B. (2007). How did schooling laws improve long-term health and lower mortality? Working Paper No. 2006–23. Chicago, IL: Federal Reserve Bank of Chicago.Find this resource:

McCrary, J., & Royer, H. (2011). The effect of female education on fertility and infant health: Evidence from school entry policies using exact date of birth. American Economic Review, 101(1), 158–195.Find this resource:

Mendolia, S., & Walker, I. (2014). The effect of noncognitive traits on health behaviors in adolescence. Health Economics, 23(9), 1146–1158.Find this resource:

Miller, S., & Wherry, L. R. (2018). The long-term effects of early life Medicaid coverage. Journal of Human Resources.Find this resource:

Muurinen, J.-M. (1982). Demand for health: A generalized Grossman model. Journal of Health Economics, 1(1), 5–28.Find this resource:

Oreopoulos, P., Stabile, M., Walld, R., & Roos, L. L. (2008). Short-, medium-, and long-term consequences of poor infant health an analysis using siblings and twins. Journal of Human Resources, 43(1), 88–138.Find this resource:

Pappas, G., Queen, S., Hadden, W., & Fisher, G. (1993). The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. New England Journal of Medicine, 329(2), 103–109.Find this resource:

Persson, P., & Rossin-Slater, M. (2018a). Family ruptures, stress, and the mental health of the next generation. American Economic Review, 108(4–5), 1214–1252.Find this resource:

Persson, P., & Rossin-Slater, M. (2018b). Family ruptures, stress, and the mental health of the next generation: Reply. American Economic Review, 108(4–5), 1256–1263.Find this resource:

Portrait, F., van Wingerden, T., & Deeg, D. (2017). Early life undernutrition and adult height: The Dutch famine of 1944–1945. Economics & Human Biology, 27, 339–348.Find this resource:

Puentes, E., Wang, F., Behrman, J. R., Cunha, F., Hoddinott, J., Maluccio, J. A., . . . Stein, A. D. (2016). Early life height and weight production functions with endogenous energy and protein inputs. Economics & Human Biology, 22, 65–81.Find this resource:

Rossin-Slater, M., & Wüst, M. (2016). What is the added value of preschool? Long-term impacts and interactions with a health intervention (NBER Working Paper No. 22700). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Royer, H. (2009). Separated at girth: US twin estimates of the effects of birth weight. American Economic Journal: Applied Economics, 1(1), 49–85.Find this resource:

Sama-Miller, E., Akers, L., Mraz-Esposito, A., Zukiewicz, M., Avellar, S., Paulsell, D., & Del Grosso, P. (2016). Home visiting evidence of effectiveness review: Executive summary (Technical Report, Office of Planning, Research and Evaluation, Administration for Children and Families). Washington, DC: U.S. Department of Health and Human Services.Find this resource:

Santavirta, N., & Santavirta, T. (2014). Child protection and adult depression: Evaluating the long-term consequences of evacuating children to foster care during World War II. Health Economics, 23(3), 253–267.Find this resource:

Schiman, J. C., Kaestner, R., & Lo Sasso, A. T. (2017). Early childhood health shocks and adult well-being: Evidence from wartime Britain (NBER Working Paper No. 23763). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Schoeni, R. F., Freedman, V. A., & Martin, L. G. (2008). Why is late-life disability declining? The Milbank Quarterly, 86(1), 47–89.Find this resource:

Scholte, R., van den Berg, G. J., Lindeboom, M., & Deeg, D. J. (2017). Does the size of the effect of adverse events at high ages on daily-life physical functioning depend on the economic conditions around birth? Health Economics, 26(1), 86–103.Find this resource:

Scholte, R. S., Van den Berg, G. J., & Lindeboom, M. (2015). Long-run effects of gestation during the Dutch hunger winter famine on labor market and hospitalization outcomes. Journal of Health Economics, 39, 17–30.Find this resource:

Stephenson, J., Heslehurst, N., Hall, J., Schoenaker, D. A., Hutchinson, J., Cade, J. E., . . . Barker, M. (2018). Before the beginning: Nutrition and lifestyle in the preconception period and its importance for future health. The Lancet, 391(10132), 1830−1841.Find this resource:

Thompson, O. (2017). The long-term health impacts of Medicaid and CHIP. Journal of Health Economics, 51, 26–40.Find this resource:

Todd, P. E., & Wolpin, K. I. (2003). On the specification and estimation of the production function for cognitive achievement. Economic Journal, 113(485), F3–F33.Find this resource:

Trivers, R. L., & Willard, D. E. (1973). Natural selection of parental ability to vary the sex ratio of offspring. Science, 179(4068), 90–92.Find this resource:

Van den Berg, G. J., Lindeboom, M., & Portrait, F. (2006). Economic conditions early in life and individual mortality. American Economic Review, 96(1), 290–302.Find this resource:

Van den Berg, G. J., Pinger, P. R., & Schoch, J. (2016). Instrumental variable estimation of the causal effect of hunger early in life on health later in life. Economic Journal, 126(591), 465–506.Find this resource:

Vollmer, S., & Wójcik, J. (2017). The long-term consequences of the global 1918 influenza pandemic: A systematic analysis of 117 IPUMS international census data sets (Discussion Papers, No. 242). Göttingen, Germany: Courant Research Centre: Poverty, Equity and Growth.Find this resource:

Wagstaff, A. (1993). The demand for health: An empirical reformulation of the Grossman model. Health Economics, 2(2), 189–198.Find this resource:

Wernerfelt, N., Slusky, D. J., & Zeckhauser, R. (2017). Second trimester sunlight and asthma: Evidence from two independent studies. American Journal of Health Economics, 3(2), 227–253.Find this resource:

Wherry, L. R., & Meyer, B. D. (2016). Saving teens: Using a policy discontinuity to estimate the effects of Medicaid eligibility. Journal of Human Resources, 51(3), 556–588.Find this resource:

Wherry, L. R., Miller, S., Kaestner, R., & Meyer, B. D. (2018). Childhood Medicaid coverage and later life health care utilization. Review of Economics and Statistics, 100(2), 287–302.Find this resource:

Yemelyanau, M., Amialchuk, A., & Ali, M. M. (2012). Evidence from the Chernobyl nuclear accident: The effect on health, education, and labor market outcomes in Belarus. Journal of Labor Research, 33(1), 1–20.Find this resource:

Yi, J., Heckman, J. J., Zhang, J., & Conti, G. (2015). Early health shocks, intra-household resource allocation and child outcomes. Economic Journal, 125(588), F347–F371.Find this resource:

Zweifel, P., Breyer, F. & Kifmann, F. (2009). Health economics. Dordrecht, The Netherlands: Springer.Find this resource:


Table 1. Summary of Articles Discussed—Shocks

Author (Year), Country

Outcomes and Exposure




Akbulut-Yuksel (2017), Germany

Outcomes: BMI, obesity, metabolic syndrome in adulthood

Exposure: Intensity of World War II destruction (war damage) when in utero or young child during World War II

Diff-in-diff using city and birth cohorts

War children exposed during prenatal or early postnatal periods had about 1 point higher BMI and 16 p.p. higher probability to be obese as adults, and more likely to suffer from stroke, hypertension, diabetes, and cardiovascular disorders (0.23 SD increase in a combined metabolic syndrome index) during adulthood.

Effects larger for females and not significant for males; also, larger effects for urban population. Children of mothers with less than basic education had larger effects, but not for father’s education.

Almond (2006), United States

Outcomes: Physical disability preventing work

Exposure: Influenza exposure in utero

Influenza pandemic and cohort and geographic variation in influenza exposure

The authors find that males in the cohort who experience in utero exposure to the pandemic had increased rates (1% more likely) in physical disability.


Almond & Mazumder (2005), United States

Outcomes: Self-reported health, troubles with hearing, speaking, lifting, walking, diabetes, stroke

Exposure: Influenza exposure in utero

Influenza pandemic and cohort and geographic variation in influenza exposure

The authors find the 1919 cohort to be more likely to self-report fair or poor health (10%) or to have certain functional limitations, such as troubles hearing (19%), speaking (35%), lifting (13%), and walking (17%). Stroke and diabetes were found more likely to occur for this cohort as well.


Almond & Mazumder (2011), United States

Outcomes: Birthweight in Michigan (long-term outcomes for Uganda and Iraq)

Exposure: Mother fasting during pregnancy

Observance of Ramadan during pregnancy for Muslim and non-Muslim mothers, using a diff-in-diff design

The authors use the overlap of Ramadan with pregnancy (and particular trimesters) and a measure of exposure hours using the variation in daylight hours. They find a significant negative effect on birthweight for the first two trimesters and a negative insignificant effect for the third trimester. Although significant, the magnitudes are very low, around 1% decrease of mean birth weight


Arthi (2018), United States

Outcomes: Physical disability and vision and hearing problems

Exposure: Exposure to Dust Bowl when in utero and during childhood

Diff-in-diff using state and birth cohorts

Although the Dust Bowl has generally strong negative effects, especially for those exposed in utero, on physical disability (about 1 p.p.), but not on vision and hearing problems, the implementation of the New Deal is able to recover a large proportion of the damage. This suggests that subsequent interventions can remediate the effects of a negative early-in-life shock.

The negative effect is larger in boys than in girls and significant only for the boys.

Banerjee et al. (2010), France

Outcomes: Adult height, life expectancy, and other health outcomes (exempt from military due to health, myopia, goiter, hernia, spinal problem, epilepsy, low IQ, feeble, blind, deaf)

Exposure: Income shock at year of birth

Diff-in-diff using regional variation in timing of phylloxera that destroyed French vineyards

The authors find an effect on height (1.8 mm), but no effect on any other health outcome or life expectancy. The income shock and corresponding decline in nutrition affected body development, but health status may have been protected by other factors (e.g., public health).


Beach et al. (2018), United States

Outcomes: Height, weight, BMI

Exposure: Influenza exposure in utero

Influenza pandemic and cohort and geographic variation in influenza exposure

The authors revisit the study of the 1918 influenza in the United States by enhancing the analysis using linked data. This provides better geographical variation in influenza exposure as it uses city-level information. Moreover, the authors correct for age heaping (fathers of lower cognitive ability more likely to report ages ending in 0 or 5), which biases the results in the previous literature. While the authors find an effect on education, they find no effects on height, weight, or BMI.


Behram & Rosenzweig (2004), United States

Outcomes: Adult physical characteristics

Exogenous increase in intrauterine nutrient consumption, twin fixed effects

The authors find that increasing birthweight has no effect on the child’s obesity later in life. However, they find that each additional pound increases adult height by 0.6 in.


Bengtsson & Helgertz (2015), Sweden

Outcomes: Hospitalizations and its duration (ages between 54 and 90), and mortality

Exposure: Influenza exposure in utero

Cohort and geographic variation in influenza exposure

Males born in the first quarter of 1919 (exposed in the fetal stage) to the pandemic were 3.6% more likely to be hospitalized and 3.8% for mortality. For females, there was only a significant 2.9% higher risk for hospitalizations.

Hospitalization overall effect was similar between men and women. However, for boys, the effect concentrated in the second trimester, while for girls, it concentrated in the third. Mortality effects only for men.

Bharadwaj et al. (2016), United Kingdom

Outcomes: Adult asthma development

Exposure: Prenatal or during the first year of life to air pollution

Exogenous exposure to air pollution due to the 1952 Great Smog of London

The authors examine the increase in likelihood of asthma development during childhood and adulthood as a result of unexpected exposure to air pollution around birth. They use a sample of children born in London (who suffered because of the smog) and outside London, and various cohorts exposed prenatally and in subsequent years of life. Their main finding is that exposure during the first year or life has a significant effect on the likelihood of childhood asthma (19 p.p.), and an effect on adulthood asthma, which is not significant. The effects of in utero exposure are also not significant and much smaller in magnitude.


Black et al. (2017), Norway

Outcomes: Adult height

Exposure: Nuclear radiation exposure in utero

Regional fallout by wind, rainfall, and topography across and over time; in addition, sibling fixed effects

Although significant effects are found for IQ, education, and earning, the effect on height is much weaker and not consistent across specifications, suggesting that the effect of radiation is concentrated on cognitive ability and not on development.

No heterogeneous effects for education of child’s mother.

Börsch-Supan & Jürges (2011), Germany

Outcomes: Early retirement (before age 55)

Exposure: Born during World War I and exposed to hunger

Variation in birth cohorts born during World War I

The authors find a 30% increase in early retirement below age 55 due to disability for the cohorts born during World War I, which is interpreted as a long-term effect of wartime and hunger on health and the ability to work.

The effect is similar for both men and women.

Brown & Thomas (2016), United States

Outcomes: Physical disability preventing work

Exposure: Influenza exposure in utero

After controlling for parental background, there are no effects.

The authors argue that in utero exposure to the influenza pandemic is not randomly assigned. In fact, exposed families happened to be different than unexposed ones on several observables. For example, fathers of “treated” individuals were found more likely to have a lower socioeconomic status, less likely to be White, more likely to be older or have lower incomes, less likely to be World War II veterans than fathers of unexposed individuals. When controlling for these observables, they find the effect of the pandemic exposure to be statistically insignificant.


Cutler et al. (2007), United States

Outcomes: Health in late life, diabetes, heart condition, high blood pressure, chronic lung disease, psychiatric condition, stroke, arthritis, disability, height, BMI

Exposure: Great Depression and Dust Bowl exposure in utero

Exposure to economic conditions among cohorts

The authors use crop yield, income, and unemployment to capture the effects of the economic conditions controlling for birth cohort effects for the children of farmers. No significant effect is found in any of the health outcomes examined.

No effects found for nonfarmers or farmers.

Danzer & Danzer (2016), Ukraine

Outcomes: Depression and anxiety, subjective survival probabilities

Exposure: Subclinical radiation doses

Regional variation in exposure due to 1986 Chernobyl accident; region, settlement, year and month fixed effects

Twenty years after the accident, the authors find that a one unit increase of natural background radiation increases the likelihood of mental disorder diagnostics (by 1.7 p.p. in the low-dose population), and reduces expected survival probability (by 7.5 p.p.).


Fletcher (2019), United States

Outcomes: Mortality

Exposure: Influenza exposure in utero

Cohort variation in influenza exposure

The author explores whether differential exposure to the pandemic (postnatal vs. at birth and third trimester vs. first and second trimester) has any effect on mortality. The results do not show enough evidence of differences across these groups.


Havari & Peracchi (2017), Europe

Outcomes: Self-reported health, chronic conditions, mental health problems

Exposure: Exposure to war, hunger, and parent absence during childhood

Variation in birth cohorts born during World War I

The authors find that exposure to war and hunger are associated with worse health (3-7 p.p. decrease in likelihood of good self-reported health).

War matters more for females, whereas hunger more for males.

Kesternich et al. (2015), Germany

Outcomes: Obesity in adulthood

Exposure: Hunger episodes during childhood

Food supply (calories per capita per month) varied over time and occupation zone, as additional to hunger recall.

The authors find the treated individuals more likely to spend a higher share of their income on food, leading to a 3-6 p.p. increase in the likelihood of obesity.

The effect on likelihood of obesity is larger for individuals in low-income households. Moreover, the effect on food spending share is also larger for low-income individuals.

Kesternich et al. (2014), Europe

Outcomes: Diabetes, heart disease, height, depression, and self-reported health

Exposure: Living during World War II and living in a region that experienced war combat during childhood

Timing and location of World War II controlling for country and birth year

The authors find significant effects for diabetes (3 p.p.), depression (6 p.p.), and self-reported health (9 p.p.), but not for heart disease and height. Moreover, they identify the following channels of war: hunger, dispossession, persecution, father absence, and childhood immunizations.

The negative health effects do not vary by childhood SES (however, self-reported health effects concentrated on middle class, and heart disease effects concentrated in the middle and lower class).

Lehmann & Wadsworth (2011), Ukraine

Outcomes: Self-reported health, height, BMI, smoking, drinking, anemia, gastrointestinal illness, heart disease at 23−59 years old;

Exposure: Radioactive fallout

Regional variation in exposure due to 1986 Chernobyl accident

The authors find those exposed are more likely to be in poor health (by 13 p.p.), more likely to suffer from anemia or gastrointestinal illness, but not more likely to be underweight or obese, to drink or smoke, all 20 years after.


Lindeboom et al. (2010), The Netherlands

Outcomes: Adult mortality

Exposure: Malnutrition in utero or first month of life

Exposure to Dutch potato famine 1846−1847

The authors find a negative effect (4 years for boys and 2.5 years for girls) of the famine early in life on mortality after age 50. Moreover, they find that higher food prices at birth reduce this effect for children of farmers, suggesting a positive income effect.

The effect is stronger for those in lower social classes. Moreover, men are affected more than women.

Persson & Rossin-Slater (2018a, 2018b), Sweden

Outcomes: Anti-anxiety and depression medications

Exposure: Maternal stress during pregnancy

Variation in timing of death of family member

Exposure to the maternal stress caused by a relative’s death when in utero increases the use of medication for anxiety and depression around age 35 (813% increase in likelihood of using, and 12−19% increase in daily use), when comparing with those experiencing the death right after birth, to control for income shocks.

Effects do not vary by maternal education.

Portrait et al. (2017), The Netherlands

Outcomes: Adult height

Exposure: Childhood undernutrition

Famine during pregnancy due to Dutch Hunger Winter (1944−1945); variation in the moment and place of birth

The authors examine the famine effects at various ages using the unaffected areas as control group. They find significant effects (about 4−4.5 cm) of experiencing the famine for ages under 2, but no significant effects for any other age group until 15 years old.

Effect is stronger for girls than for boys when looking at exposure under age 1 (when exposure under age 2 effects are same for both males and females).

Schiman et al. (2017), United Kingdom and Wales

Outcomes: Self-reported health, incidence of health problems, disability status at ages 40 to 73

Exposure: Infant during World War II

Variation in region and cohort being infant during World War II

The authors exploit variation in infant mortality caused by food rationing and harsh conditions during World War II and find its strong association with later-in-life disability (1 SD increase in infant mortality associated with 39% increase in disability). In particular, the effect was stronger later in life (above age 60). However, the effects on self-reported health and incidence of health problems were not significant.

No heterogeneous effects found for SES.

Scholte et al. (2015), The Netherlands

Outcomes: Hospitalization later in life

Exposure: Malnutrition during pregnancy

Famine during pregnancy due to Dutch Hunger Winter (1944−1945); variation in the moment and place of birth

The authors find exposure to the Dutch Hunger Winter to have higher hospitalization rates (2−7%) in the years before retirement if exposure occurred in middle or late gestation stage.

The effects are stronger for males than for females.

Scholte et al. (2017), The Netherlands

Outcomes: Physical functional limitations in daily life activities

Exposure: Economic conditions early in life

Business cycle around birth

Those born during a recession are more volatile to adverse life events, such as chronic diseases, as the impact (0.09 effect on number of difficulties) on function limitation in later life is much higher (an additional 0.11) for those who experienced such an adverse early-life condition.

The effects are stronger for males for chronic diseases, but not for any other adverse life event.

Van den Berg et al. (2016), Germany, The Netherlands, Greece

Outcomes: Adult height

Exposure: Malnutrition; nutritional shortage due to famine in childhood (ages 0−16)

Instrumental variables using self-reported hunger episodes and variation in birth cohorts exposed to famine

The authors instrument retrospective reporting of hunger episodes during childhood using the exposure to famines induced by World War II.

For males, a negative effect of 3 cm on adult height for undernutrition between in utero and age 4, but no significant effect for females.

Van den Berg et al. (2006), The Netherlands

Outcomes: Mortality: average lifetime

Exposure: Economic conditions early in life

Born in 1872−1876 boom and born in 1877−1881 recession

The authors find a negative effect of a few years in life expectancy for those born in economic boom compared to those born in recession.

Adverse economic conditions at birth affect more men than women, while parents’ social class effect is stronger for women.

Wernerfelt et al. (2017), United States

Outcomes: Incidence of asthma in adulthood

Exposure: Sunlight exposure in utero

Variation in sunlight within-location across counties (and time use data for spending more time outdoors when there is more sunlight)

The authors examine the link between maternal vitamin D levels during pregnancy and likelihood of asthma for the offspring. They achieve this by using variation in sunlight (a source of vitamin D) and find a significant effect for sunlight during the second trimester (a double amount of sunshine is associated with 1 p.p. decrease in probability of asthma), which highlights the critical period of the second trimester of pregnancy.


Yemelyanau et al. (2012), Belarus

Outcomes: Medical visits, being hospitalized, self-reported being in good health (up to age 29)

Exposure: Radioactive fallout in early childhood (ages 1-16)

Regional and age variation in exposure due to 1986 Chernobyl accident

The authors find that individuals exposed in the most contaminated areas have more medical visits (by about 1 visit), less likely to be in good health (by 7 p.p.), though radiation exposure has no effect on the probability of hospitalization. The effects are stronger for younger individuals in the more contaminated.

Effect on medical visits significant only for men (and only for the most affected areas), while effect on self-reported health significant and same for both men and women.

Table 2. Summary of Articles Discussed—Interventions

Author (Year), Country

Outcome and Exposure




Aizer et al. (2016), United States

Outcomes: Longevity and weight in adulthood

Exposure: Cash transfer to poor families

Accepted and rejected applicants to the Mothers’ Pension program

The authors find male children of accepted program participants to be less likely to be underweight (by half) and more likely to live longer lives by 1 year, compared to the rejected ones. However, rejected mothers were on average slightly better off based on observable characteristics. Hence, the effects of the program are likely to understate the benefits of the program.

Effect larger for poorest families. No differential effects for urban and rural areas.

Anderson et al. (2018), United States

Outcomes: Mortality

Exposure: Water filtration and chlorination, sewage, milk bacteriological standards

Variation in the timing of water filtration, chlorination, sewage, and milk standards across cities

The authors examined the interventions of water filtration, chlorination, sewage, and setting milk standards and find that none of them had any effect on infant and total mortality.


Bhalotra & Venkatara-Mani (2017), United States

Outcomes: Work limiting or preventing disability, physical disability

Exposure: Pneumonia exposure

Introduction of antibiotic therapies in 1937 interaction, with variation across race and state in institutionalized segregation

The authors show that men exposed to 1 SD sulfa-driven decline in pneumonia during infancy experienced improvements in work-limiting disability (1 p.p.), but no effect on physical disability.

The gains were smaller for Blacks in segregated states compared to Blacks in the rest of the states.

Bhalotra et al. (2017), Sweden

Outcomes: Mortality: death by age 75, and death by age 50

Exposure: Infant care programs: mother support (nutrition and sanitation) and infant care (home visits and clinics) during infancy

Variation in time and medical districts

The intervention had short- (1.5 p.p.) and long-term (2.5 p.p.) effects on mortality, especially for mortality caused by cancer and cardiovascular diseases.

The effects were stronger for children born outside marriage, and some evidence that they were also stronger for younger mothers. No heterogeneous effects by gender or local economic conditions.

Boudreaux et al. (2016), United States

Outcomes: Self-reported health, blood pressure, heart disease, diabetes, obesity (ages 25−34)

Exposure: Medicaid in early childhood (ages 0−5)

Variation in implementation across the states in combination with individual-level variation in probability to participate

The authors find no significant effect on self-reported health. Regarding the other outcomes, they construct a chronic condition index from these four outcomes and find a significant effect on this index (from no to full exposure, 0.35 SD decrease). However, examining each outcome separately, the effect is concentrated on high blood pressure. The effect becomes much larger (3.5 SD) when interacted with predicted participation probability merged using census information based on state, age, and race.


Brown et al. (2015), United States

Outcomes: Mortality at age 28

Exposure: Medicaid in childhood (ages 0−18)

Variation in implementation across the states in combination with individual-level variation in probability to participate

The authors find a decrease in mortality through age 28, conditional on survival at age 18. While the effect of 1 SD increase in Medicaid eligibility on females is smaller and not significant at 28 (3%), but larger at 26 and 27 (11%), the effect for males is larger and more robust (5%) across all ages. This is probably because male mortality is, in general, higher at these ages.

The effect for men is greater, and more robust, than for women.

Bütikofer et al. (2015), Norway

Outcomes: BMI, likelihood of obesity, blood pressure, hypertension, cardiac risk, cholesterol risk, height at age 40

Exposure: Child health care centers aiming at information on infant nutrition

Diff-in-diff of implementation across municipalities and time

The authors find treated individuals experience a reduction in BMI (0.3 points), likelihood of obesity (2 p.p.), blood pressure (1 mm), hypertension (1 p.p.), cholesterol risk (3 p.p.), and a combined index for bad health (0.24 SD). Effects on the separate outcomes were significant only for men, whereas the effect on index was significant for both. In addition, they show that better nutrition in early-life has positive effects on height (2 cm) at age 40 for both.

Effects are stronger (and significant) for men than women.

Bütikofer & Salvanes (2018), Norway

Outcomes: Adult height

(men only) and mortality (dying before age 66)

Exposure: Tuberculosis testing and vaccination program while at school

Cohort (1948 launch) and regional (infection rates) variation

The authors find that cohorts who were in school during and after the implementation in areas with higher tuberculosis rates gained more longevity as the likelihood of dying before age 66 decreased by 1 p.p. The effect on men’s adult height was a 1.5 cm increase.

The effects on mortality are concentrated in women.

Campbell et al. (2014), United States

Outcomes: Cardiovascular and metabolic diseases in mid-30s; health care utilization at age 30 and obesity

Exposure: Carolina Abecedarian Project (ABC) stimulating early childhood environment

Randomized control trial

The authors find that disadvantaged children randomly exposed to the treatment (ABC intervention) have significantly lower prevalence of risk factors for cardiovascular and metabolic diseases in their mid-30s (diastolic bp: 13.5 mm and systolic bp: 17.5 mm, hypertension: 33 p.p., and vitamin D deficiency: 38 p.p. for males; abdominal obesity: 20 p.p. for women). In addition, treated males are found more likely to have health insurance at age 30 (23 p.p.).

The effects are stronger for men than for women.

Carneiro & Ginja (2014), United States

Outcomes: Self-reported health and overweight at ages 20−21

Exposure: Preschool program Head Start

Fuzzy regression discontinuity design around income eligibility cut-off (multiple points) and instrumental variables regression for eligibility

The authors examine the medium-term effects of Head Start, a preschool program targeting disadvantaged children in the United States. Although they find some effects for chronic conditions and obesity at ages 12−13 and for depression and obesity at ages 16−17, at ages 20−21 there is no significant effect for any health outcome.

The authors attempt to estimate heterogeneous effects, but they report them as imprecise.

Conti et al. (2016), United States

Outcomes: Height, weight, obesity, blood pressure, hypertension, hospitalization

Exposure: Perry Preschool Project and Carolina Abecedarian Project—early childhood

Randomized control trial

The authors find the treated males to have lower BMI (4 points), to be lighter (7 kg) and taller (5 cm), to have lower values of systolic and diastolic blood pressure (13.5 mm and 17.5 mm), less likely to fall into the Stage I hypertension category (34 p.p.) or to be hospitalized (35 p.p.) than the control.

The effects are stronger for men than for women.

Cutler & Miller (2005), United States

Outcomes: Adult mortality

Exposure: Filtration and chlorination

Diff-in-diff exploiting variation in the timing of water filtration and chlorination across cities

Clean water access decreased by almost half the mortality rates in major cities, by 75% infant mortality and by two thirds in child mortality rates. Their analysis suggests that these public health interventions acted as complements. In a cost-benefit analysis, they predict that the return rate for clean water technologies was about 23 to 1 and that the cost per person-year saved was about $500USD (2003).

The impact was larger for cities with higher illiteracy rate.

Goodman & Bacon (2016), United States

Outcomes: Mortality and disability (hearing, vision, mobility, self-care, work limitation) in adulthood up to 60s.

Exposure: Medicaid in childhood (ages 0−18)

Diff-in-diff and cohort eligibility instrumental variables

This study provides an assessment of Medicaid using its original introduction, thus enabling the study of long-term effects for the participants up to their 60s. The author finds strong effects for mortality and disability (both about 20%). Moreover, younger age eligibility is associated with higher gains. In addition, the authors provide an estimate for government savings of 3−11% of the program’s cost

White children have larger gains compared to non-White children.

Gould et al. (2011), Israel

Outcomes: Self-reported health problems, disability income support at ages around 50−60

Exposure: Moving to a more modern environment in childhood

Quasi-random allocation of Jewish Yemenite immigrant families in Israel, after airlift rescue operation in 1949

The authors found no effect of the quality of early childhood environment on late adulthood health outcomes.


Herbst (2017), United States

Outcomes: Work-related dis-ability

Exposure: Universal child

care for children at ages 0−12

Diff-in-diff using variation in eligible age participation and variation across states on spending intensity on preschool program

The program resulted in preschool centers (for children ages 0−5) and after-school services (for children ages 6−12) through the Lanham Act of 1940 due to World War II. The treated cohorts are those born between 1931 and 1946 and the control between 1947 and 1951, along with the interaction of program spending. The author finds that for a $100USD increase in spending, there is a 0.3 p.p. decrease in the probability of work-related disability at ages 24−39.


Hjort et al. (2017), Denmark

Outcomes: Mortality, hospital nights, and cardiovascular disease

Exposure: 1937 Danish home visiting program

Diff-in-diff using variation across municipalities implementing the program, for individuals born between 1935 and 1949

Treated individuals are 5−8% less likely to die middle aged (45−57), less likely to be diagnosed with or die from cardiovascular diseases, and spend about one-half night less in hospital.

Effects on mortality similar for men and women.

Hoynes et al. (2016), United States

Outcomes: Obesity, high blood pressure, heart disease, diabetes, general health status, disability

Exposure: Share of time between conception and age 5 that the Food Stamp Program was available

Diff-in-diff over county and cohort of birth. Also, diff-in-diff-in-diff using variation across subgroups with varying propensities to use the program

The authors find treated individuals (defined as those who have additional exposure to the Food Stamp Program earlier in childhood compared to having it later in childhood) more likely to have a reduction in the incidence of “metabolic syndrome,” constructed from obesity, high blood pressure, heart disease and attack, and diabetes (0.3 SD reduction for full participation compared to none), and more likely to report they are in good health (11%). Access to the program is also found to decrease stunting for treated individuals. The effect on work-limited disability is in the expected direction (i.e., an improvement arisen from access to the program), though its effect is statistically insignificant.

The effect on metabolic syndrome is larger for males than for females; however, the opposite was found for self-reported health.

Kesztenbaum & Rosenthal (2017), France

Outcomes: Total mortality (life expectancy at age 1)

Exposure: Sanitation

Diff-in-diff using time and neighborhood fixed effects

The authors find that sanitation increased the life expectancy of Parisians by 2 years.

The gains are similar for men and women.

Ludwig et al. (2013b), United States

Outcomes: Self-rated health, sleeping hours, mobility troubles, asthma, hypertension, obesity, diabetes and psychological distress, calmness and peacefulness

Exposure: Moving to a less-disadvantaged neighborhood in childhood

Randomized control trial; Moving to Opportunity (MTO) randomized housing mobility experiment

The authors find that a move to a better neighborhood, as a result from the randomization, reduces the likelihood of adult physical health and mental health problems (both standardized indices constructed using the surveyed outcomes) 10−15 years later.

Effects significant only for girls.

Lührmann & Wilson (2018), United Kingdom

Outcomes: Mortality, self-reported, and hospitalizations due to cardiovascular disease at ages 50−60

Exposure: Access to healthcare through NHS

Regression discontinuity design around date of introduction of NHS (July 1948)

The authors find significant effects of the introduction of NHS at birth on mortality (12% decrease) at age 57 and a significant decrease (7%) in the likelihood of cardiovascular diseases among those from the lower class at their 50s. In addition, they also find immediate effects (decrease 17% on infant mortality).

Mortality gains are larger for lower-class individuals; however, there are no differential effects for cardiovascular diseases.

Miller & Wherry (2018), United States

Outcomes: Diabetes, high blood pressure, obesity, heart disease, self-reported health and psychological distress, and hospitalizations, up to mid-30s.

Exposure: Medicaid in utero and infant (first year after birth)

Variation in state and cohort level in the timing and generosity of Medicaid expansions (instrumental variables)

Using both individual survey and administrative hospital data, the authors find that Medicaid prenatal eligibility reduced the likelihood of all chronic conditions, separately and combined in one index (0.03 SD decrease for every 10 p.p. increase in early life eligibility), but there were no effects on self-reported health and psychological distress. In addition, there were lower hospital admissions related to diabetes and obesity (8−10%).

The effects were larger for Black cohorts than for Whites.

Rossin-Slater & Wu¨st (2016), Denmark

Outcomes: Mortality: death by 65 (among others)

Exposure: Preschool childcare and nurse home visiting program

Variation in municipalities that introduced childcare before or after home visiting program

The authors find lower mortality rates for treated individuals through the age of 65 (0.5 p.p.). For women, their results suggest that preschool access lowers the likelihood of heart disease diagnosis (0.03 p.p.), while for men it reduces the cancer diagnosis (0.04 p.p.). Moreover, additional access to the Nurse Home Visiting program reduces the positive effects of the preschool program for most of the studied outcomes.

The mortality effects are larger for women than for men.

Thomson (2017), United States

Outcomes: Self-reported health, health limitation, chronic condition, asthma at ages 18−21

Exposure: Medicaid or CHIP in childhood (ages 0−18)

Instrumental variables using eligibility

The author finds strong effects for all four outcomes and a health index combining all together (0.08 SD for one additional year of eligibility). Furthermore, the gains are larger for those with eligibility at ages 0−5 than for those at ages 6−11 or 12−18.


Wherry & Meyer (2016), United States

Outcomes: Disease-related mortality at 19−23 years old

Exposure: Childhood Medicaid eligibility and coverage at ages 8−14

Regression discontinuity design around birth date cut-off (September 30, 1983) that expands eligibility

Although the authors find that Medicaid eligibility resulted in a decline in mortality at ages 15−18 for Black cohorts (19%), there was no effect of mortality improvement at ages 19−23. For White children there was no mortality improvement.

Significant effects for Black children, while no effects for White children.

Wherry et al. (2018), United States

Outcomes: Hospitalization and emergency department utilization at age 25

Exposure: Childhood Medicaid eligibility and coverage at ages 8 to 14

Regression discontinuity design around birth date cut-off (September 30, 1983) that expands eligibility

Medicaid eligibility is associated with a decrease in hospitalization (7−15%) and emergency department visits (2−5%) at age 25 for Black cohorts (but not at 15, suggesting long-term effects). There were no effects for Whites. Finally, the authors calculate that Black cohorts’ lower utilization at age 25 offsets 3−5% of the cost of expanding Medicaid for Black and non-Black cohorts. This suggests that gains should be even greater for later life.

Significant effects for Black children, while no effects for White children.

Table 3. Summary of Articles Discussed—Birth Weight

Author (Year), Country

Outcome and Exposure




Behram & Rosenzweig (2004), United States

Outcomes: Adult physical characteristics

Exogeneous increase in intrauterine nutrient consumption, twin fixed effects

The authors find that increasing birth weight has no effect on a child’s obesity later in life. However, they find that each additional 1 pound increases adult height by 0.6 in.


Bharadwaji et al. (2018), Sweden

Outcomes: Infant mortality, APGAR (newborn physical condition) score

Sibling and twin pair fixed effects

Infant mortality is reduced by 6 deaths per 1,000 births for a 10% increase in birth weight, where the effect on the APGAR score is also significant (0.47 points increase). Moreover, the effect of birth weight on infant mortality has been declining over time (1920s to 1950s).

No consistent evidence of heterogeneity by maternal age, education, or family composition.

Black et al. (2007), Norway

Outcomes: Infant mortality, height, BMI, birth weight of first offspring

Twin fixed effects, year and month-of birth dummies

The authors differentiate their findings between short-run and long-run effects of birth weight. They show that a 10% increase in birth weight reduced infant mortality by 4 deaths per 1,000 births (short-term outcome) and increases height by 0.57 cm and BMI by 0.11 (long-term outcome for men aged 18−20). Among the female twin subsample who have children, the authors find that a 10% increase in the mother’s birth weight leads to a 1.5% increase in the birth of her first child.


Currie & Moretti (2007), CA, United States

Outcomes: Offspring’s birth weight

Grandmother fixed effects

The authors find that if the mother had a low birth weight, her child is 3.9 p.p. more likely to have a low birth weight.

The effects are stronger for lower SES mothers. Also, the effects are stronger for Blacks than for Whites.

Oreopoulos et al. (2008), Canada

Outcomes: Infant mortality and age 17 survival

Twin study with family and birth year fixed effects

The authors find infants born with an APGAR (a newborn physical condition measure) score below 7 about 10 p.p. more likely to die within 1 year compared to those with scores of 10. Using birth weight rather than Apgar score, the authors find a similar relationship (9 p.p. for those born between 1,000 and 1,500 grams). Infant health (APGAR scores and low birth weight) has no effect on survival at age 17 or on physician utilization between ages 12 and 17.


Royer (2009), United States

Outcomes: Diabetes, pregnancy complications, and offspring birth weight.

Twin-pair fixed effects

The authors find a significant effect (7 grams for every 100 grams in mother’s birth weight) of birth weight on the child’s birth weight, but not for gestational length. Moreover, they find a significant effect on the likelihood of pregnancy complications (5% increase for every 100 grams), but no significant effects on hypertension, diabetes, or anemia.



(1.) The narrowing may in part be a consequence of selective mortality or culling effects.

(2.) As an example, groceries and time spent cooking can be used to produce meals, or a bicycle and time spent pedaling can produce physical activity.

(3.) For more in-depth treatment of the health capital approach and detailed surveys of its theoretical extensions and empirical applications, see, among others, Grossman (2000) and Gilleskie (2008).

(4.) For example, Muurinen (1982) has a generalized version with both consumption and investment motives; Cropper (1977) and Dardanoni and Wagstaff (1987, 1990) study the implications of introducing uncertainty in the model.

(5.) First, Ehrlich and Chuma (1990) claim that the constant returns to scale (CRTS) shape of the investment function adopted by Grossman generates a so-called bang-bang equilibrium, where individuals can immediately adjust their health capital stock to the optimal level; furthermore, due to the same CRTS assumption, individuals would be able to choose to live for an infinite amount of time (Case & Deaton, 2005). Second, it has been noted that model solutions are somewhat “memoryless” in that the optimal choice of health investment does not depend on the histories of health investment and aging (Wagstaff, 1993). Finally, the model has been shown to predict—at least in its most basic formulation—a positive association between health status and health expenditure in the cross section. This feature is inconsistent with empirical evidence showing that healthier people exhibit lower health expenditure (Zweifel, Breyer & Kifmannm, 2009).

(6.) The additional features that are absent in the standard Grossman approach are (mainly): the explicit modeling of labor market returns and unhealthy labor; a shift from a narrow focus on medical care inputs to a wider range of healthy and unhealthy consumption; and an endogenous retirement decision. While this is the first time these features are integrated together in a Grossman-like model, investigation of such aspects is not completely novel; see, for example, Becker and Murphy (1988), Kenkel (2000) for unhealthy behaviors, House et al. (1994), Marmot, Ryff, Bumpass, Shipley, and Marks (1997) for unhealthy work environments, and French (2005) for health and retirement decisions.

(7.) Notice that the Grossman formulation of the health production function in Equation 1 can be viewed as a special case of Equation 2.

(8.) Both human capital and investments are often operationalized as latent variables, using factor-analytic techniques to infer the dimensionality of the latent constructs and extract information from error-ridden measures. See, for example, Cunha, Heckman, and Schennach (2010) or Conti, Frühwirth-Schnatter, Heckman, and Piatek (2014).

(9.) Notice that when human capital and investment are multidimensional, cross-productivities and cross-complementarities are also possible. For example, better health can increase the effectiveness of investments in cognitive skills; different investments (e.g., nutrition and psychosocial stimulation or schooling in different periods) are allowed to exhibit an arbitrary degree of contemporaneous and intertemporal substitutability. Note that investments can include early childhood interventions, as described in “Policies.”

(10.) Brown and Thomas (2016) raised selection concerns about the original analysis, but Beach, Ferrie, and Saavedra (2018) have shown that they are unfounded and have confirmed the original results.

(11.) For a systematic review of studies looking at the 1918 influenza pandemic around the world, see Vollmer and Wójcik (2017).

(12.) Although some of these studies rely on retrospective assessments, it has been shown that older people’s recall of childhood events are trustworthy. For example, Havari and Mazzonna (2015), using survey data from Europe, show that respondents’ recall data are consistent with external aggregate data such as GDP per capita, schooling, and war episodes.

(13.) The importance of nutrition for health and prosperity plays a central role in the work of Fogel; see, for example, Fogel (1997).

(14.) In response to a comment by Matsumoto (2018), Persson and Rossin-Slater (2018b) show that their results are robust to redefining the control group using the expected rather than actual date of birth.

(15.) A related study by Santavirta and Santavirta (2014) examines the effect of being placed in foster care in Sweden on children evacuated from Finland at the time of World War II, and finds no effect of such a change in the family environment on adult depression.

(16.) Negative health effects are also found by Yemelyanau, Amialchuk, and Ali (2012) for Belarus.

(17.) Similar effects (on education but not on children’s health) have been found by Almond, Edlund, and Palme (2009) using exposure to radioactive fallout from Chernobyl in Sweden.

(18.) For a detailed review on effects of early-life exposure to pollution, see Currie et al. (2014).

(19.) See Bozzoli, Deaton, and Quintana-Domeque (2009) for a didactic discussion of scarring and culling effects.

(20.) See Almond and Mazumder (2013) for an excellent review on parental responses.

(21.) See Sama-Miller et al. (2016) for a review of home visiting research literature.

(22.) The Swedish intervention evaluated by Bhalotra et al. (2017), which included both home visits and infant care clinics, also showed long-term health improvements, as reported.

(23.) For example, Currie, Graff Zivin, Meckel, Neidell, and Schlenker (2013) find in utero exposure of these families to contaminated water in New Jersey to negatively affect birth outcomes. Currie, Greenstone, and Meckel (2017) find in utero exposure to fracking to negatively affect birth outcomes. Due to the recency of these events, long-term effects are not yet known.

(24.) While Ludwig et al. (2013b) find long-term improvements in physical and mental health (for females only) from changes to better neighborhoods for low-income families in the United States, Gould, Lavy, and Paserman (2011) find no long-term health impacts from changes in the early environment in Israel.

(25.) See, for example, a handful of working papers: Adhvaryu, Nyshadham, Molina, and Tamayo (2018); Aguilar and Vicarelli (2015); Duque, Rosales-Rueda, and Sanchez (2018); and Malamud, Pop-Eleches, and Urquiola (2016).

(26.) Studying on a more systematic basis the presence of and the possible reasons behind gender differences is also an important avenue of future research.

(27.) A prototype is García, Heckman, Leaf, and Prados (2016). Some papers perform simple cost-benefits analyses; see, for example, Bütikofer et al. (2015).

(28.) See, for example, Conti et al. (2018), who use unique data from ultrasound scans to study fetal development.