American Journal of Epidemiology Advance Access originally published online on October 13, 2006
American Journal of Epidemiology 2007 165(1):85-93; doi:10.1093/aje/kwj352
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ORIGINAL CONTRIBUTIONS |
Childhood Socioeconomic Position and Cause-specific Mortality in Early Adulthood
1 Division of Epidemiology, Norwegian Institute of Public Health, Oslo, Norway
2 Department of Public Health, Erasmus MC, University Medical Center Rotterdam, Rotterdam, the Netherlands
Correspondence to Dr. B. H. Strand, Division of Epidemiology, Norwegian Institute of Public Health, P.O. Box 4404 Nydalen, NO-0403 Oslo, Norway (e-mail: heine{at}fhi.no).
Received for publication January 24, 2006. Accepted for publication May 25, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
There is growing evidence that childhood socioeconomic position (SEP) influences adult health. The authors' aim was to describe the association between childhood SEP measures (parents' education, occupation, and income) and mortality, for both genders, and to assess to what extent this association was mediated by adult SEP. Registry data for all Norwegians born in 1955–1965 were used. Death records were linked to the cohort, and 6,589 persons died during 1990–2001. Cox's regression was used to calculate relative rates and the relative index of inequality. Low childhood SEP was associated with increased mortality for most causes of death, except for breast cancer, where no association was found. For suicide in women, low childhood SEP was protective. Adult SEP accounted for the associations for total mortality and most causes of death. However, adult SEP accounted for only one half of the association of father's educational level with ischemic heart disease mortality among men. The increased suicide risk among women with high childhood SEP persisted, regardless of adult SEP. In summary, childhood SEP had a direct association with early adult cardiovascular mortality in men and with suicide in women. For other causes of death, childhood SEP was only indirectly associated, mostly through persons' own educational level.
adult; cause of death; child; education; income; mortality; occupations; social class
Abbreviations: ICD-9, International Classification of Diseases, Ninth Revision; ICD-10, International Classification of Diseases, Tenth Revision; SEP, socioeconomic position
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Links between low socioeconomic position (SEP) and poor health are well established, and there is growing evidence that the economic and social positions early in life influence health later in life (1–4). Measures of SEP at different stages of the life course each influence health in later life. Studies that examine the specific causes of death have found considerable heterogeneity, however, in their association with both SEP and the times during life when inequalities are generated. For example, coronary heart disease is a disease that develops throughout the life course, and exposure to risk factors in early childhood has been found to be associated with an increased risk for this disease in adulthood (3). For other causes, such as lung cancer deaths, the relative contribution of child versus adult circumstances may vary in different contexts.
Socioeconomic differences in cause-specific mortality have been investigated in less detail than all-cause mortality or self-rated poor health (4). This partly reflects the limitations of available data sources, which often were confined to small and selected male populations (4). Further, studies of cause-specific mortality usually applied only one or two indicators of childhood SEP simultaneously (3). In addition, there are relatively few studies of recent birth cohorts, where diseases with strong secular declines in mortality, such as stomach cancer, are relatively unimportant and where other causes of death, such as suicide and accidents, are more important. Most deaths in young adulthood are due to external causes, and whether life-course theories are applicable for these causes is uncertain. It has been suggested that inadequate support in early life may give rise to an increased substance use or unnecessary risk taking, leading to an increased risk of accidental death (5).
An example with a focus on recent birth cohorts is the series of studies by Pensola and colleagues (6–9) of younger generations of Finnish people. They found that, compared with childhood SEP, adult SEP had a much larger effect on mortality in early adulthood. However, childhood SEP had an independent association with mortality from cardiovascular diseases and alcohol-related causes. Although similar findings were obtained in the Oslo Mortality Study (10, 11), it is yet uncertain to what extent the Finnish findings represent a new pattern characteristic of younger generations in developed countries.
In this paper, we analyze a large, national cohort of Norwegian men and women to investigate the impact of four different measures of childhood socioeconomic position on total mortality and on cause-specific mortality in young adult life. We have access to Norwegian registry data on the parents' education, occupation, and income, as well as the subjects' own education and income, for all Norwegian men and women born in 1955–1965. The sample, which we followed for mortality between 1990 and 2001, is large and nationally representative, and it has only a very small number of missing observations.
Our key question is how childhood SEP, measured by different socioeconomic indicators, influences mortality levels during early adulthood and to what extent this association can be explained by adult SEP. Is there mainly an indirect association, in which a high SEP of parents sets the pathway for a high SEP by the persons themselves, and thereby lower mortality, or is there also a direct association? In addition, we are interested in comparing four measures of childhood SEP (mother's education, father's education, father's occupation, childhood household income) and their relative importance for mortality in adulthood.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Data were compiled by Statistics Norway and are based on census records from 1970 and 1990. Data include men and women born in 1955–1965 and who were living in Norway at the time of the censuses. Death records for 1990–2001 and information on socioeconomic position in 1970 and 1990 were linked to the census data by means of personal identification numbers. The cohort born in 1955–1965 was chosen for these analyses, because in 1970 these persons were aged 5–15 years and mostly living with their parents. By the time of the 1990 census, the cohort had largely finished their own education. Mother's and father's education, father's occupation, and household income were linked to the cohort members by a unique household code.
Our analysis was restricted to families in 1970 consisting of a married couple and the couple's unmarried children living at home, for children aged 5–15 years. This restriction captured 94 percent of the children aged 5–15 years in private households in 1970, and it did not include lone parents. Single-parent households were excluded, because information on the father's education was not available for most of the households.
Education was classified into four groups: higher education (at least 13 years of education), higher secondary education (12 years), lower secondary education (10–11 years), and basic education (9 years or less). Mother's and father's education in 1970 and the person's own education in 1990 were used.
The three-digit code of the Occupational Classification of Norway was used (12). Occupation was aggregated by use of the Erikson-Goldthorpe scheme known as "EGP-10" (13) and further aggregated into six groups as upper nonmanual (class 1), lower nonmanual (classes 2 and 3), skilled manual (classes 4, 7, and 8), unskilled manual (classes 5, 9, and 10), farmer (class 6), and missing occupation. For occupation, only the father's occupation in 1970 was used in the main analyses. The person's own occupation in 1990 was not measured, because occupational information was available for only a sample of about 20 percent of the 1990 census.
For income, the household equivalent income was measured (14). This was done by summing the income of all household members and adjusting this sum by dividing it by the square root of the household size. This adjustment factor is often applied in the measurement of household income in relation to size (15, 16). In 1990, several people had zero income, and they were placed in a separate group. Having zero income indicates that all household members were outside the workforce. Two variables on income were used: household income in 1970 and household income in 1990. The income measures were slightly different in 1970 and 1990. Income in 1970 was defined as gross income as determined for the tax assessment, including all possible income components (12). The income measure used in the 1990 census was gross income as earned from work, from which returns on capital such as interest received were not included (17).
Mortality follow-up covers the period from November 3, 1990, to December 31, 2001. During this period, 613,807 persons were followed for 6.71 million person-years, and 6,589 died. The causes of death were coded according to the International Classification of Diseases, Ninth Revision (ICD-9), for the years 1990–1995 and according to the International Classification of Diseases, Tenth Revision (ICD-10), for the years 1996–2001. The following groups of deaths were used: all causes, cardiovascular diseases (ICD-9 codes 390–459; ICD-10 codes I00–I99), ischemic heart disease (ICD-9 codes 410–414; ICD-10 codes I20–I25), stroke (ICD-9 codes 430–438; ICD-10 codes I60–I69), cancer (ICD-9 codes 140–239; ICD-10 codes C00–C99), lung cancer (ICD-9 code 162; ICD-10 code C34), breast cancer (ICD-9 code 174; ICD-10 code C50), suicides (ICD-9 codes E950–E959; ICD-10 codes X60–X84, Y87.0), accidents/violence (ICD-9 codes E800–E929, E960–E999; ICD-10 codes V01–X59, Y87.1–Y87.2), and alcohol- and drug dependency-related causes (ICD-9 codes 291–305; ICD-10 codes F10–F16, F18, F19).
With regard to statistical methods, the relations between childhood SEP and adult mortality were analyzed in three steps. First, we calculated age-standardized mortality rates by means of the direct method using the European standard population.
Second, Cox's proportional hazards models were used to calculate relative rates and their 95 percent confidence intervals (18). In the regression models, control was made for age. In additional models, control was also made for adult education and adult income.
Third, the association between childhood SEP and mortality was summarized for each of the childhood SEP indicators by means of the relative index of inequality. This is an often-used summary measure of the size of differences in mortality across all socioeconomic levels (16, 19, 20). The relative indexes of inequality were also fitted by Cox's proportional models. First, each socioeconomic group was assigned a rank score, indicating the proportion of the population having a higher socioeconomic level. The rank score was determined by taking half of the proportion of people with the same level plus the proportion with a higher level (16, 19). Next, this rank score was treated as a continuous variable in the regression models. The exponentiated beta coefficient associated with the rank score yields the relative index of inequality (16, 19). The relative index of inequality can be interpreted as comparing the mortality rate of the hypothetical worst-off with that of the best-off person in the hierarchy. Relative indexes of inequality are presented with 95 percent confidence intervals. STATA, version 8.0, software (StataCorp LP, College Station, Texas) was used for the analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Young adults in 1990 were higher educated than their parents (table 1). Men were more represented in the upper income groups. This male-female difference in household income is possible because of the large number of single persons (28 percent) in this age group in 1990. The age-adjusted mortality rates indicate a stronger association between adult SEP and mortality than between childhood SEP and mortality.
|
In men, the total mortality hazard ratios increased with decreasing childhood SEP for all four SEP indicators, in a stepwise pattern (table 2). In women, the associations were weaker and not so consistent across the four measures of childhood SEP. Regarding parents' education, only women with basic-educated parents had a higher mortality hazard ratio; otherwise, no differences across educational classes were observed. However, for the father's occupation, there was a stepwise pattern also in women. For childhood household income, no association with mortality was seen among women.
|
Most of the observed associations between childhood SEP and mortality disappeared when we adjusted for adult SEP (table 2). The person's own education especially attenuated the association between childhood SEP and mortality, while adult household income did not mediate the association as much. Among men, there was a stepwise association between mortality and parents' educational level, with the highest mortality among sons of basic-educated parents. Among women, only daughters of basic-educated parents were at elevated mortality risk. These associations were fully mediated in both men and women by adult SEP. Sons and daughters whose fathers were outside the workforce or in manual occupations had higher hazard ratios compared with sons and daughters with upper nonmanual class fathers. When adjusted for adult SEP, this association was strongly attenuated and no longer statistically significant. Regarding household income in 1970, men in the lowest group were at increased mortality hazard, but this difference disappeared and was reversed in the fully adjusted model.
In table 3, a cause-specific mortality analysis is presented separately for all four childhood SEP indicators, using the relative index of inequality. For total mortality, the associations with childhood SEP seem stronger than those in table 2, because the relative index of inequality gives much weight to the lowest educational level and its higher mortality, because of the large number of subjects in this level. For cardiovascular mortality in men, childhood SEP had an independent effect, especially on ischemic heart disease mortality. This pattern was robust and seen for three of four childhood SEP indicators, except for income. For women, there was a strong inverse association between cardiovascular mortality and childhood SEP, which was true also for childhood household income. Unlike that in men, this association was fully mediated by adult SEP. For stroke mortality, an association with childhood SEP was seen in men but strongest for mother's education. Among women, no significant associations were found.
|
Among men, a significant association with cancer mortality was seen with father's education and occupation (table 3). Among women, significant associations were found for father's occupation and mother's education. The results suggest only a small independent effect of childhood SEP after adjustment for adult SEP, except for father's occupation, in relation to cancer among women. There were not many lung cancer deaths in this young cohort. Nevertheless, the results showed an inverse, significant association with father's education. The other childhood SEP measures showed a tendency for a similar association, but they failed to be significant. None of the four childhood SEP measures had an effect that was independent from adult SEP. Breast cancer had no strong association with childhood SEP.
For suicide in men, no association with childhood SEP was found in the age-adjusted model. However, when adult SEP was added to the model, there was excess suicide in the childhood high-SEP groups. In women, a very strong positive association between childhood SEP and suicide was found. This association remained highly significant after adjustment for adult SEP, and the pattern was similar for all four SEP measures. For accidental deaths, there was an inverse association with childhood SEP among both men and women. Among women, this association was fully mediated by adult SEP, while in men a substantial part of the association remained in the case of income. For alcohol- and drug-related deaths, an inverse association with childhood SEP was found for all indicators except income. This association was attenuated and reversed when adjusted for adult SEP. For childhood household income, there was a positive association with alcohol- and drug-related deaths among both men and women. This association remained when adjusted for adult SEP.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In this study, a pattern of increasing mortality with decreasing childhood SEP was found. The results were consistent for the different SEP indicators, except for household income, which showed a weaker association with mortality. A similar pattern, with increasing mortality with decreasing childhood SEP, was found for almost all specific causes of death that have been examined. The exceptions were breast cancer, where no association with childhood SEP was found, and suicide in women, where low childhood position was protective.
Adult SEP accounted almost entirely for the association between childhood SEP and total mortality, as well as most specific causes of death. A person's own education was an especially strong mediator, while household income in adulthood hardly contributed to the association between mortality and childhood SEP. Among men, adult SEP accounted for only one half of the association of father's educational level with cardiovascular mortality and ischemic heart disease. The increased suicide risk among daughters of high-SEP parents persisted, regardless of adult SEP. A similar association emerged for men when the effect of adult SEP was controlled for.
Evaluation of data and methods
We used four measures of childhood SEP and compared them with regard to their relative importance for mortality. Father's occupation is the most widely used indicator for childhood SEP (3). Other common indicators are parents' educational levels, while the childhood household income level is not so widely used. The importance of considering parents' educational level as a measure of childhood SEP relevant to childhood health status has been emphasized in previous studies (21, 22). Large-scale studies show that the health and welfare of children are linked to the educational level of their parents (22). Nevertheless, parental education is just one among different socioeconomic indicators, and each of them might need to be studied for its own sake.
Examining and comparing different indicators of childhood SEP may shed light on the mechanisms that may explain their link with health in adulthood (3). In our paper, however, we have not directly compared the different childhood SEP indicators with regard to their effect on mortality, because these indicators may have been measured with different degrees of accuracy in the 1970 census. Instead, we have analyzed the four indicators in parallel ways, in order to determine any effects that childhood SEP may exert on adult mortality.
We observed that the effect of childhood SEP can largely be explained by that of adult SEP. This result could possibly be biased if adult SEP would have been measured more accurately than childhood SEP. We estimate that there is no evidence to support this suspicion. All measures of education are based on educational registries, which are expected to be equally good in 1970 and 1990. Therefore, the precision of father's and mother's education may be as good as the precision of the child's own education. Similarly, adult income may be measured about as accurately as adult education, because all income information was derived from tax registries.
The analyses were not adjusted for occupation at adult age. This is because occupation was registered for only a 20 percent sample of the Norwegian population. This smaller sample would not have sufficient statistical power for cause-specific mortality analyses. An analysis of total mortality (similar to the one reported in table 2) using this 20 percent sample showed that the person's own occupation in 1990 was not a strong mediating factor between the father's education and mortality. The person's own occupation did not add to adult income in explaining this association.
The zero-income group in 1990 suffered seven times the mortality risk of the nonzero group. The zero-income group was better educated than was the nonzero group, and their fathers were as well educated as were the fathers of the nonzero group, but there were far more single persons and lone parents among the zero group (data not presented). Of the male deaths in the zero-income group, 63 percent were due to violent/external causes, compared with 51 percent among males in the nonzero-income group. Alcohol abuse might be an underlying factor for this group, as the relative risk for death from alcohol- and drug-related causes of death among men was 21.7 (95 percent confidence interval: 16.1, 29.3) in a fully adjusted model.
Our age group was still young at the end of follow-up (36–46 years), and external causes of death constituted a large part of the total deaths. It is uncertain whether the same associations with childhood socioeconomic position will be observed when this cohort grows older and other causes of death dominate their mortality profile. Perhaps a stronger direct effect of childhood socioeconomic position will become manifest in all-cause mortality. This might apply especially to women, whose mortality pattern at young ages is dominated by cancer and suicide, which have small or even opposite educational gradients. At older ages, the share of cardiovascular deaths (for which a direct effect of childhood SEP is observed in our study and those of others) increases and so perhaps may the effect of early living conditions on total mortality. Similarly, the relative importance of adult income instead of adult education may be larger for older age groups than in this relatively young cohort. Therefore, the results must be interpreted for this young age group only, and caution must be made when generalizing to other age groups.
Interpretation of key findings
We found that low childhood SEP was associated with increased risk of all-cause mortality among both men and women. This is in accordance with the findings of a recent systematic review of 29 studies of childhood SEP and all-cause and cause-specific mortality from eight countries (3). Our finding that the influence of childhood SEP could be fully explained by adult SEP replicates the findings of a study on Finnish women (9) and in a later study on Finnish men (6). In Finland, most of the mortality differences according to childhood SEP, measured by the occupation of the head of the household, could be attributed to differences in educational pathways in adolescence and early adulthood. Like our study, the Finnish studies used census-based data sets and analyzed similar birth cohorts, which were studied in almost the same period of follow up (1991–1998).
Adult education, which may act as an intermediary factor between childhood SEP and late-adult SEP (19), was strongly associated with mortality regardless of childhood SEP. Mortality among men with a basic education was more than three times higher than that among men with a higher education; in women, it was more than two times higher. Similar results were reported in Finland (6, 9). The pathways linking education and mortality have been discussed previously at some length, and at least three mechanisms have been distinguished. First, unobserved factors such as cognitive ability may lead to both good health and better education. Second, education may act through improving health-related knowledge and thereby healthier lifestyles, which reduce mortality risk. Third, education is an important determinant of occupational career, in which a high and stable position in the labor market constitutes a continuous opportunity for ensuring high income and favorable living conditions (23–27).
We found the strongest direct association between childhood SEP and mortality for cardiovascular diseases. After control for adult SEP, about half of the excess risk among men from low childhood SEP families remained. This finding is in line with most previous studies (6, 10, 28–30). Anders Forsdahl, David Barker, and others have argued that maternal, fetal, and infant nutrition are important early life influences that affect later adult health, especially cardiovascular disease (31–33). More specifically, Barker argued that impaired growth in childhood permanently affects or "programs" the structure and physiology of a range of organs and tissues.
For stroke mortality, the gradient in childhood SEP was fully explained by adult SEP. This result contrasts with the observation of an independent association made in two other studies, that is, in the Boyd Orr cohorts (34) and in the Oslo cohorts (11). There may be two main explanations for this discrepancy. First, our own study has low precision for stroke, but so did these other two studies (the Boyd Orr cohorts had 50 stroke deaths, and the Oslo cohorts had 78 such deaths); the discrepancy may thus be due to chance fluctuations. Second, the other studies did not assess the role of adult education, which in our study was found to be the key adult SEP factor.
There was little evidence for a direct association between childhood SEP and overall cancer mortality, except for the association between cancer mortality among women and their father's occupation. The general lack of associations for cancer mortality is in accordance with the results from most other studies (3). The results for lung cancer indicated a large role for own adult SEP, with no evidence on a residual role of parental SEP. This is in line with a previous study from Norway, which found a larger effect of adult SEP than of childhood SEP on lung cancer (11). However, there was some evidence of a residual role of childhood SEP in their study. In other studies, the association between lung cancer death and childhood SEP was largely explained by adult SEP (3). For breast cancer, the results indicated no association with childhood SEP. Similar results were reported in a previous study that found that early life factors had a modest influence on breast cancer risk in postmenopausal women (35).
The association between childhood SEP and suicide differed between men and women. Among men, there was a weak, inverse association, which was reversed when adjusted for adult SEP. Similar results were found in a study among Finnish men (8). Among women, we found an excess suicide risk in daughters from high-SEP families. This association was highly significant and not attenuated by adult SEP. We identified only one other study of the association with parental SEP and suicide in women, and this Finnish study found no association between parental occupational class and suicide (9). The reasons for this link are not studied, but it might be hypothesized that not meeting the high demands set out from highly educated parents might create stress and low self-esteem.
There were more alcohol- and drug-related deaths among men and women with a lower childhood SEP, but this excess risk diminished after accounting for the person's own education and adult income. In contrast, two other studies (8, 11) found an effect of childhood SEP after accounting for adult SEP. The discrepancy between these results and our own findings might be due to the fact that these studies did not assess the role of adult education. Other studies have also reported a higher risk of deaths from external causes among men whose fathers had a low SEP (8, 11), but they have failed to find an association in women (8, 9, 36).
Conclusions
In our data, childhood socioeconomic position had a direct association with early adult cardiovascular mortality, especially for ischemic heart disease in men. Suicide in women was also linked to childhood SEP, but in the reverse direction: Daughters of high-SEP families suffered more than double the suicide risk compared with daughters from low-SEP families. Total mortality and other causes of death were indirectly associated with childhood SEP, mostly through the person's own educational level. This might imply that improved educational opportunities for children and adolescents and employment opportunities for young adult men may be critical for avoiding the long-term health effect of poor socioeconomic position in childhood.
|
ACKNOWLEDGMENTS |
|---|
Conflict of interest: none declared.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
- Barker DJ and Osmond C. (1986) Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet 1:1077–81.[Web of Science][Medline]
- Forsdahl A. (1973) Points which enlighten the high mortality rate in the county of Finnmark. Can the high mortality rate today be a consequence of bad conditions of life in childhood and adolescence? (In Norwegian). Tidsskr Nor Laegeforen 93:661–7.[Medline]
- Galobardes B, Lynch JW, Davey Smith G. (2004) Childhood socioeconomic circumstances and cause-specific mortality in adulthood: systematic review and interpretation. Epidemiol Rev 26:7–21.
[Free Full Text] - In Davey Smith G (Ed.). Health inequalities: lifecourse approaches (2003) (Policy Press, Bristol, United Kingdom).
- Kalland M, Pensola TH, Merilainen J, et al. (2001) Mortality in children registered in the Finnish child welfare registry: population based study. BMJ 323:207–8.
[Free Full Text] - Pensola T and Martikainen P. (2004) Life-course experiences and mortality by adult social class among young men. Soc Sci Med 58:2149–70.[CrossRef][Web of Science][Medline]
- Pensola TH and Valkonen T. (2002) Effect of parental social class, own education and social class on mortality among young men. Eur J Public Health 12:29–36.
[Abstract/Free Full Text] - Pensola TH and Martikainen P. (2003) Cumulative social class and mortality from various causes of adult men. J Epidemiol Community Health 57:745–51.
[Abstract/Free Full Text] - Pensola TH and Martikainen P. (2003) Effect of living conditions in the parental home and youth paths on the social class differences in mortality among women. Scand J Public Health 31:428–38.[CrossRef][Web of Science][Medline]
- Claussen B, Davey Smith G, Thelle D. (2003) Impact of childhood and adulthood socioeconomic position on cause specific mortality: the Oslo Mortality Study. J Epidemiol Community Health 57:40–5.
[Abstract/Free Full Text] - Naess O, Claussen B, Davey Smith G. (2004) Relative impact of childhood and adulthood socioeconomic conditions on cause specific mortality in men. J Epidemiol Community Health 58:597–8.
[Free Full Text] - Vassenden K. (1987) Censuses 1960, 1970 and 1980. Documentation of the comparable files(Statistics Norway, Oslo, Norway).
- Erikson R and Goldthorpe JH. (1992) The constant flux: a study of class mobility in industrial societies(Clarendon Press, Oxford, United Kingdom).
- Galobardes B, Shaw M, Lawlor DA, et al. (2006) Indicators of socioeconomic position (part 1). J Epidemiol Community Health 60:7–12.
[Abstract/Free Full Text] - Dalstra JA, Kunst AE, Geurts JJ, et al. (2002) Trends in socioeconomic health inequalities in the Netherlands, 1981 –1999. J Epidemiol Community Health 56:927–34.
[Abstract/Free Full Text] - Kunst AE, Bos V, Lahelma E, et al. (2005) Trends in socioeconomic inequalities in self-assessed health in 10 European countries. Int J Epidemiol 34:295–305.
[Abstract/Free Full Text] - Norway Statistics. (1999) Population and housing census 1990. Documentation and main figures(Statistics Norway, Oslo, Norway).
- Clayton D and Hills M. (1993) Statistical models in epidemiology(Oxford University Press, Oxford, United Kingdom).
- Davey Smith G, Hart C, Hole D, et al. (1998) Education and occupational social class: which is the more important indicator of mortality risk? J Epidemiol Community Health 52:153–60.[Abstract]
- Mackenbach JP and Kunst AE. (1997) Measuring the magnitude of socio-economic inequalities in health: an overview of available measures illustrated with two examples from Europe. Soc Sci Med 44:757–71.[CrossRef][Web of Science][Medline]
- Hauser RM. (1994) Measuring socioeconomic status in studies of child development. Child Dev 65:1541–5.[CrossRef][Web of Science][Medline]
- Zill N. (1996) Parental schooling & children's health. Public Health Rep 111:34–43.
- Blane D, White I, Morris J. (1996) Education, social circumstances and mortality. In Blane D, Brunner E, Wilkinson RG (Eds.). Health and social organisation(Routledge, London, United Kingdom) pp. 171–87.
- Fuchs VR. (1979) Economics, health, and post-industrial society. Milbank Mem Fund Q Health Soc 57:153–82.[CrossRef][Web of Science][Medline]
- Leigh JP. (1983) Direct and indirect effects of education on health. Soc Sci Med 17:227–34.[CrossRef][Web of Science][Medline]
- Pincus T and Callahan LF. (1994) Associations of low formal education level and poor health status: behavioral, in addition to demographic and medical, explanations? J Clin Epidemiol 47:355–61.[CrossRef][Web of Science][Medline]
- Weber M. (1968) Economy and society(Bedminster Press, New York, NY).
- Beebe-Dimmer J, Lynch JW, Turrel G, et al. (2004) Childhood and adult socioeconomic conditions and 31-year mortality risk in women. Am J Epidemiol 159:481–90.
[Abstract/Free Full Text] - Osler M, Andersen AM, Due P, et al. (2003) Socioeconomic position in early life, birth weight, childhood cognitive function, and adult mortality. A longitudinal study of Danish men born in 1953. J Epidemiol Community Health 57:681–6.
[Abstract/Free Full Text] - Smith GD, McCarron P, Okasha M, et al. (2001) Social circumstances in childhood and cardiovascular disease mortality: prospective observational study of Glasgow University students. J Epidemiol Community Health 55:340–1.
[Free Full Text] - Barker DJ, Gluckman PD, Godfrey KM, et al. (1993) Fetal nutrition and cardiovascular disease in adult life. Lancet 341:938–41.[CrossRef][Web of Science][Medline]
- Forsdahl A. (1978) Living conditions in childhood and subsequent development of risk factors for arteriosclerotic heart disease. The cardiovascular survey in Finnmark 1974 –75. J Epidemiol Community Health 32:34–7.
[Abstract/Free Full Text] - Osmond C, Barker DJ, Winter PD, et al. (1993) Early growth and death from cardiovascular disease in women. BMJ 307:1519–24.
[Abstract/Free Full Text] - Frankel S, Davey Smith G, Gunnell D. (1999) Childhood socioeconomic position and adult cardiovascular mortality: the Boyd Orr Cohort. Am J Epidemiol 150:1081–4.
[Abstract/Free Full Text] - Titus-Ernstoff L, Egan KM, Newcomb PA, et al. (2002) Early life factors in relation to breast cancer risk in postmenopausal women. Cancer Epidemiol Biomarkers Prev 11:207–10.
[Abstract/Free Full Text] - Neeleman J, Wessely S, Wadsworth M. (1998) Predictors of suicide, accidental death, and premature natural death in a general-population birth cohort. Lancet 351:93–7.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Braveman and C. Barclay Health Disparities Beginning in Childhood: A Life-Course Perspective Pediatrics, November 1, 2009; 124(Supplement_3): S163 - S175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Maty, J. W. Lynch, T. E. Raghunathan, and G. A. Kaplan Childhood Socioeconomic Position, Gender, Adult Body Mass Index, and Incidence of Type 2 Diabetes Mellitus Over 34 Years in the Alameda County Study Am J Public Health, August 1, 2008; 98(8): 1486 - 1494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-M. Alavian, M. E. Motlagh, G. Ardalan, M. Motaghian, A. H. Davarpanah, and R. Kelishadi Hypertriglyceridemic Waist Phenotype and Associated Lifestyle Factors in a National Population of Youths: CASPIAN Study J Trop Pediatr, June 1, 2008; 54(3): 169 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B Galobardes, J W Lynch, and G D. Smith Is the association between childhood socioeconomic circumstances and cause-specific mortality established? Update of a systematic review J Epidemiol Community Health, May 1, 2008; 62(5): 387 - 390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Baker, L. W. Olsen, and T. I.A. Sorensen Childhood Body-Mass Index and the Risk of Coronary Heart Disease in Adulthood N. Engl. J. Med., December 6, 2007; 357(23): 2329 - 2337. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Naess, B. H Strand, and G. D. Smith Childhood and adulthood socioeconomic position across 20 causes of death: a prospective cohort study of 800 000 Norwegian men and women J Epidemiol Community Health, November 1, 2007; 61(11): 1004 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Melchior, T. E. Moffitt, B. J. Milne, R. Poulton, and A. Caspi Why Do Children from Socioeconomically Disadvantaged Families Suffer from Poor Health When They Reach Adulthood? A Life-Course Study Am. J. Epidemiol., October 15, 2007; 166(8): 966 - 974. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Kelishadi Childhood Overweight, Obesity, and the Metabolic Syndrome in Developing Countries Epidemiol. Rev., May 3, 2007; (2007) mxm003v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||