Childhood Socioeconomic Circumstances and Cause-specific Mortality in Adulthood: Systematic Review and Interpretation

There is convincing evidence that exposures acting across the life course influence adult health outcomes (1–3). Lifecourse epidemiology examines a range of potential processes through which exposures acting at different stages of life can, singly or in combination, influence disease risk (table 1) (4). In the critical period model, an exposure acting at a specific time has long-lasting effects on the structure or function of the body. The fetal origins hypothesis, in its original formulation, took this approach (5). Other examples of processes where outcomes appear to depend upon the time window during which an exposure acts are limb development (in relation to maternal thalidomide use); infection with hepatitis B and risk of adulthood liver cancer (with very early postnatal infection being most implicated); and environmental lead exposure, which results in serious neurodevelopmental deficits only if occurring in infancy and childhood (3). However, the influence of exposures acting during critical periods of susceptibility may be modified by later life exposures. This is the case for the associations of birth weight with coronary heart disease, high blood pressure, and insulin resistance, where associations are stronger (or only evident) among those who become obese during adolescence or adulthood (6–8). In contrast, the accumulation of risk model suggests that effects accumulate over the life course, although it also allows for developmental periods during which susceptibility may be greater (4). Health damage may increase with the duration and/or number of detrimental exposures. This has been shown in relation to exposure to poor socioeconomic conditions, where additive effects of experiencing low socioeconomic position across different stages of the life course influence the risk of several diseases (9, 10). Accumulation of risk can also be due to clustering of exposures. For example, children from a lower socioeconomic background are more likely to be of low birth weight, to have poorer diets, to be more exposed to passive smoking and some infectious agents, and to have fewer educational opportunities (4). Exposures may also form chains of risk and thus tend to occur together. For example, low educational attainment will increase the probability that a person works in an occupation with a high risk of toxic exposures and lower income (3). Coronary heart disease is a good example of a disease that develops throughout the life course. Although coronary heart disease manifests itself in adulthood, atherosclerosis, an important underlying process leading to the disease, begins at a much earlier age. Fatty streaks in the artery walls are found in children (11, 12), and arterial lesions are evident in young men suffering violent deaths (13–15). Moreover, the presence of risk factors during childhood or adolescence is associated with an increased risk of developing coronary heart disease (16). For example, serum cholesterol in early adult life was strongly associated with coronary heart disease in the Johns Hopkins Precursors Study (17). The magnitude of this association was greater than that reported with respect to adulthood cholesterol measures. Recent studies have suggested that unfavorable preadulthood measures of cholesterol, blood pressure, and adiposity are associated with increased intimal-medial thickness, a presymptomatic indicator of coronary heart disease (18, 19). Socioeconomic circumstances at different stages of the life course can influence specific adulthood health outcomes through the different pathways outlined in table 1. While associations of adulthood socioeconomic position with a wide range of specific health outcomes have been described extensively (20–25), relatively little investigation of how early life social circumstances influence adult health has been conducted (26). This paper therefore presents a systematic review of epidemiologic evidence on the association between preadulthood socioeconomic position and cause-specific mortality. We performed a systematic review of individual-level studies examining childhood socioeconomic circumstances and adult overall and cause-specific mortality. For the electronic search, childhood socioeconomic circumstances were defined in general and via specific indicators such as father/mother/parental education or occupation, housing conditions, overcrowding, number of siblings, maternal marital status, illegitimacy, and residence in an orphanage or similar facility. Full details on the search strategy are available from the authors. The electronic search was supplemented by the authors’ files. We excluded those publications that reported exclusively on infant/childhood mortality or adolescent suicide. However, studies that presented all-age mortality were included. In these populations, most deaths are likely to have occurred in adulthood, except for those due to external causes. We also excluded studies that grouped nonfatal and fatal events to ensure comparability between the outcomes. We have not performed formal meta-analysis as the necessary conditions of comparability of exposures and outcomes, together with homogeneity of association direction and strength (27), are not—and perhaps could not be expected to be—met. We located 29 studies of childhood socioeconomic position and all-cause or cause-specific mortality, two based on different subgroups of the Longitudinal Study in the United Kingdom and two based on different groups of the Harvard Alumni Study. A full description of these studies can be found in Web table 1. (This information is described in the first of five supplementary tables; each is referred to as “Web table” in the text and is posted on the Journal’s website (http://epirev.oupjournals.org/).) All were prospective or cohort designs with the exception of two case-control studies (28, 29) and one cross-sectional survey of reported sibling mortality (30). The populations included in these studies were from the United Kingdom, Sweden, Finland, Norway, Denmark, the Netherlands, the United States, and Russia. The oldest cohorts were born at the beginning of the last century, with only three studies including people born before that (28, 31, 32). The majority included people born during or before the 1940s and 1950s, and the youngest birth cohorts dated from the late-1950s to the 1960s in the United Kingdom (33), Sweden (34), Finland (35, 36), the Netherlands (37), and the United States (38). Thus, with the exception of the youngest cohorts, most participants lived their childhood, or were born, at the beginning of the century and lived through the Second World War at some point in their lives. Sixteen studies included samples representative of the general population. Nineteen of the 29 studies measured the participants’ socioeconomic circumstances during childhood or young adulthood. The remainder obtained data on childhood circumstances by participant recall during adulthood. The father’s occupation was the most common indicator of childhood socioeconomic position. Other indicators used were crowding, housing characteristics (either as a score of several items or evaluating each item separately), home ownership, parental education, number of siblings, farm size, indicators of the mother’s marital status, and presence of both natural parents during childhood. Most studies reported participants’ own occupation (current or longest) and deprivation indices as measures of adult socioeconomic position. Education was also assessed in some studies, but this was treated mainly as a measure of adult socioeconomic position despite that fact that, for most people, education is completed early in life and will also capture aspects of childhood circumstances (39). Eighteen of 22 studies found a higher risk of all-cause mortality associated with worse childhood socioeconomic circumstances (Web table 2) (29–31, 33, 37, 38, 40–51). For example, in the Collaborative Study and in the Boyd Orr cohort, men whose fathers had manual occupations or were unemployed had higher risk of death (40, 43). A similar effect was also found in the Boyd Orr cohort with other childhood circumstances, such as crowding (trend p of (trend p and poor (trend p In a the birth cohort, men and whose fathers had manual occupations and risk of death In cohorts, similar of effects were had a with a manual occupation in higher risk in Sweden higher risk in the Netherlands (37), higher risk in and higher risk in a cohort of men in Finland of of in the Study in the United States who low socioeconomic position during childhood had an increased mortality risk reported a lower probability of to at among whose fathers were farm or In a all-cause mortality was higher for those who some deprivation during childhood (30). The on the only for overall socioeconomic circumstances may have for a period during adulthood, and it the to how the influence of childhood adulthood socioeconomic position could be different in different for adult socioeconomic position and/or adult risk factors the associations in most studies, although the influence of childhood circumstances For example, in the Collaborative men whose fathers in manual occupations had 22 higher risk of mortality for their adult social of and This risk higher for the presence of risk factors in adulthood blood pressure, cholesterol, and effects were for by adult socioeconomic position in the cohort of young men of of and in the (30). the other was of childhood effects in the Boyd Orr cohort the birth cohort and the Study these studies that people experiencing poorer socioeconomic circumstances during childhood a higher risk of overall mortality, of adult socioeconomic position. This was for cohorts that lived in the United Kingdom, the United States, and Russia. increased mortality childhood deprivation was also in some This was the case despite these experiencing conditions during childhood than and despite the in fewer deaths due to disease mortality and a higher due to external causes. The risk associated with lower childhood socioeconomic position not by adult socioeconomic position and adult risk studies examining disease 31, of of childhood socioeconomic conditions were associated with overall disease mortality in five of of studies have been reported in more than one the number of and later the number of studies (table studies that assessed three found a greater risk of disease mortality with childhood social circumstances in a with a higher number of was not associated with increased risk of disease mortality in two studies childhood circumstances associated with disease mortality for adult socioeconomic circumstances and/or adult risk However, in the cohort of young the effect of father’s social on disease mortality was by adult occupation of of Coronary heart disease or heart was to childhood circumstances in (40, of studies (table with parents in manual occupations a higher risk of coronary heart disease adult of their adult socioeconomic for adult socioeconomic position the strength of the associations in some (40, but not born to had a higher risk of coronary heart disease, but only among those who in a Most studies childhood circumstances with father’s social The housing conditions during childhood in were to adult coronary heart disease more was associated with higher risk of coronary heart disease in the Collaborative although this association was by adult and risk factors the other in the Harvard and Alumni of were more likely to have had However, the of was based on the of those to be this could death from other being associated with number of Moreover, the of father’s social was in this of In the cohort, increased crowding was not to mortality from coronary heart disease In the results of these studies that socioeconomic circumstances during both childhood and adulthood to adult coronary heart disease mortality. marital may be a for an of deprivation than that by low socioeconomic position. of studies reported a higher risk of overall among those who childhood socioeconomic circumstances (table In the Boyd Orr and cohorts, socioeconomic circumstances were during childhood through adulthood and the effect of childhood circumstances on overall for adult socioeconomic position This suggests that exposures to poor socioeconomic circumstances during childhood be stronger of than of coronary heart mortality risk from was higher among men whose fathers had manual occupations in the Collaborative but not among those who had more (40, However, the risk of and death from was for men whose fathers had manual occupations with those whose fathers had to for adult socioeconomic position and risk factors For the were and for adult circumstances and risk factors The for deaths and in the was higher with increased crowding in the home for not the association of with crowding The Health Study not an association of adulthood of childhood social circumstances with mortality. and the number of were not associated with higher mortality risk due to heart disease in one although the the of this had to an association There was little evidence that children whose fathers were from lower social of of or who had more had higher risk of overall cancer mortality (Web table The exception was the Collaborative although this effect was by for adult socioeconomic circumstances that cancer mortality an of diseases from very different this is perhaps not In a from Norway, was an increased risk of death due to cancer among men who were exposed to worse housing conditions during childhood, of adult socioeconomic circumstances results were found in the Collaborative Study (40, and the Boyd Orr cohort although in these the association was by adult socioeconomic position. There was association of childhood circumstances with a of other in the Collaborative Study In the Collaborative men whose fathers had manual occupations or those who had more had higher mortality from of adult circumstances (40, The Study found higher risk of and cancer among those who had the housing conditions during childhood but this was There was association between childhood socioeconomic characteristics and later death from in three studies (40, from cancer mortality in Sweden or from in In childhood socioeconomic circumstances have a influence on cancer and are likely to with adult circumstances, to cancer through exposure to effect was found for other or overall studies from the United Kingdom the associations between deaths and childhood circumstances (Web table In the Collaborative men whose fathers had manual occupations had higher mortality risk socioeconomic position and risk factors and score for in of the association with a higher number of more mortality, but this was by adult circumstances not higher mortality with greater crowding or with more There is little evidence on mortality, but it that poor childhood circumstances are to at through adult childhood such as poor childhood poor childhood and housing and passive exposure, and poor are and may to poor health in adult life of five studies found higher risk of and violent death with worse childhood socioeconomic circumstances (Web table (35, of of In the cohort, men whose fathers had manual occupations during childhood had a higher risk of death from external but this was by adult socioeconomic position father’s social was not associated with risk in the Study found that housing conditions during childhood were strongly of death from external among men but not among The Collaborative Study in the United Kingdom not a higher risk of death due to and associated with childhood circumstances (40, but the birth cohort from the United Kingdom reported higher risk among those with fathers of low social during childhood two studies reported on associations of childhood socioeconomic position with risk of in adulthood (Web table Although only deaths were due to in the birth cohort from the United Kingdom, was more likely to occur among those of lower childhood socioeconomic background In the cohort, men whose fathers in manual occupations had a higher risk of this was for by adult socioeconomic position of of parental social was not associated with There were more deaths among men with fathers from a lower social in the cohort but this risk for adult socioeconomic position (Web table of of In the housing conditions during childhood were associated with deaths due to conditions to or of for adult income results were for and other deaths in this Thus, socioeconomic circumstances during childhood some in external of death in but results in the United Kingdom were in the of of death and/or risk processes in these birth cohorts may these different The evidence from individual-level studies suggests that childhood socioeconomic circumstances to a of different of In childhood conditions appear strongly to mortality from cancer and circumstances, together with adulthood socioeconomic to mortality from coronary heart disease, and childhood conditions are not associated with mortality from and but may to external and of in The childhood social circumstances to cause-specific mortality are with data from studies and mortality. is early in is a for an unfavorable of exposures acting during the period the period is is associated with risk of coronary heart disease, and disease and it is not associated or to mortality from such as cancer The association between childhood social circumstances and mortality through a of such as those in the of on adulthood disease outlined by and and presented in table 1. with during infancy and childhood a to the association between poor childhood circumstances and cancer and a critical period model during early life are most to this Moreover, the of associations of cancer and with poor childhood circumstances, together with in their association with and the relation between and cancer risk the that an early life a to early life infection such as due to childhood associated with the other coronary heart disease, and disease appear to be by factors acting across the life course and therefore may more to a risk For example, poor childhood and adult social circumstances could influence cancer risk via increased probability of earlier of lower probability of and higher of smoking an accumulation of exposures over the life course may in higher risk of coronary heart disease In an to the evidence on the potential for early life socioeconomic conditions to influence adult it is to other of epidemiologic studies in to the individual-level studies described In these studies, and maternal at some earlier period have been used as of socioeconomic conditions and of during childhood. these effects of childhood circumstances on and mortality from including but associations with coronary heart disease are and across In an early example, that socioeconomic conditions during childhood be an important of adult used mortality the time of birth as an indicator of the socioeconomic conditions, an approach later in other studies and cause-specific mortality in with mortality in across for mortality as a measure of socioeconomic cancer mortality in was strongly with higher mortality in for for and this for mortality in results were found for and mortality due to The stronger for than for coronary heart disease may the greater influence of childhood deprivation on than with and coronary heart disease being by similar and some This model is also by data on in and coronary heart disease, where have been for the beginning of the century, those of coronary heart disease studies may susceptibility periods during childhood for some diseases may with the disease risk by their in the of For a that the higher of cancer in their of lower cancer from their of that their risk was not by the results with those of born in their risk of and was similar to that of and the of from a lower risk of cancer The of infection during infancy and childhood with respect to cancer in later adulthood is with the of studies a high of cancer among to the United Kingdom for cancer were not perhaps that factors acting in both childhood to of and in adult life to are and studies mortality due to studies were not with both childhood and adulthood of residence to of coronary heart disease and for reported adult of residence as the more important risk indicator for coronary heart disease in disease mortality associated with of birth between and in the United States have also been reported the of disease mortality between born in the and born in the the between and in For coronary heart disease in both men and born in the had similar overall coronary heart disease as with born in the had lower coronary heart disease born in the had higher In contrast, had higher of and heart disease mortality with but the were among born in the the of born in the to born in the of the United States, and the coronary heart disease by born in the may be of early life disease risk in adulthood disease can evidence for childhood of later adult mortality of birth cohort effects has been used as an indicator of the of conditions in childhood on later mortality In a cohort that, the of to with it the mortality throughout adult and that these results are with the that the important from the point of of health of the during life is to the of The of this of evidence have been described and to the in in from and in the of period from cohort effects can a between diseases for which childhood socioeconomic circumstances can have a and those where this is the in socioeconomic circumstances and conditions occurring throughout the last century in most these to for those diseases where these conditions a This may be the case for and effects in cancer have been described and these could a of infection and of conditions that was a in mortality from throughout the century Moreover, the of to increased between the and the with most of the increase between the and the in and blood between the of the 1940s and the in two studies from the United Kingdom and cohort effects across birth from to in an of data adult mortality in the early of the century included from heart disease the of heart and for which is evidence of early life the of the century, mortality was by such as coronary heart disease, and and for which the of deprivation in childhood is much Thus, the of early life socioeconomic will depend on the disease which will across and time There to the in the of an early paper on this in infancy or in adult which is more important for mortality has been that socioeconomic factors in adulthood are to the risk of different diseases a a general association between socioeconomic circumstances and poor health and or of increased of the poor However, this processes of general susceptibility is by evidence that health among the has not been worse that different of socioeconomic health can be found in different and that the magnitude and direction of the social of specific of death are (3). The early studies of and and and who and later mortality, together with cause-specific mortality, to adulthood mortality, specific early life risk The exposure indicator was to and different with adulthood mortality were For deaths from were associated with mortality, and this was as that exposure to maternal factors that occurred before and during was In contrast, deaths from and disease were more to mortality, that exposure to worse conditions, including risk of during the first of life was important This review evidence that the of and adult socioeconomic circumstances across different for and these the specific that socioeconomic circumstances to different health outcomes than a In the of effects of childhood circumstances by adult socioeconomic position as the in this case an increased risk on childhood socioeconomic circumstances be for outcomes with socioeconomic in adulthood. review that this is not the different indicators of childhood socioeconomic position on the that may their with health in adulthood. of early life socioeconomic position may have both general and specific Thus, measures of crowding and number of may poorer conditions and as a for infectious disease for other outcomes such as coronary heart disease may as a general of social associated with overall and cancer mortality may be a of more and/or it may increased exposure to and Other childhood of such as a that different of deprivation than those with more of socioeconomic position and that these can also influence mortality studies measured childhood socioeconomic conditions through recall in adulthood. studies a high in father’s There was of in in father’s social between the and the recall later In the of the in their recall of their father’s occupation with later life early life measures will be more in with measures of adult socioeconomic circumstances, in greater for the childhood measures. for the two indicators will the adulthood Moreover, of the association between early life socioeconomic circumstances and adult mortality will through how childhood factors later life social (4). for diseases where adulthood factors mortality if these adulthood socioeconomic circumstances are at in an of childhood the also be as to the disease approach to the influence of childhood and adulthood social circumstances is to populations with adulthood social This is the case with the the Harvard and the that in the by and In these populations, father’s social was to disease risk in later This review presents evidence of both and of childhood socioeconomic circumstances to mortality from and disease studies a of the of childhood socioeconomic circumstances in adult disease effects are for cancer and socioeconomic conditions during early and adult life accumulate throughout the life course in their influence on the risk of coronary heart disease and The of and adulthood socioeconomic circumstances in the risk of conditions and external of death in different perhaps in how early life factors later processes that influence The results presented can be as evidence for health of early and later life may and developmental processes over the life course of born during the century and in outcomes such as and deprivation in may not be associated with the health as childhood deprivation in in where most of life course on adult health is it is that, in the United States and the United Kingdom, has increased the that of children in those are in are social how children are exposed to deprivation in early For in Sweden, the United States, and the United Kingdom is and and in Sweden this to in the United Kingdom it is only to and in the United States to and that poor children in the United States are worse in of than poor children in of early and later life socioeconomic conditions are important for it is the health of that of children to will be in the and were (in by a in Health from this also to and to and of the studies childhood socioeconomic circumstances and disease mortality of of the studies childhood socioeconomic circumstances and disease mortality of of the studies childhood socioeconomic circumstances and coronary heart disease mortality of the studies childhood socioeconomic circumstances and coronary heart disease mortality of the studies childhood socioeconomic circumstances and mortality of the studies childhood socioeconomic circumstances and mortality