• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of amjphAmerican Journal of Public Health Web SiteAmerican Public Health Association Web SiteSubmissionsSubscriptionsAbout Us
Am J Public Health. 2008 March; 98(3): 468–477.
PMCID: PMC2253566

Emergence of Socioeconomic Inequalities in Smoking and Overweight and Obesity in Early Adulthood: The National Longitudinal Study of Adolescent Health

Abstract

Objectives. We examined whether socioeconomic inequalities in smoking and overweight and obesity emerged in early adulthood and the contribution of family background, adolescent smoking, and body mass index to socioeconomic inequalities.

Methods. Using data from the National Longitudinal Study of Adolescent Health we employed multinomial regression analyses to estimate relative odds of heavy or light-to-moderate smoking to nonsmoking and of overweight or obesity to normal weight.

Results. For smoking, we found inequalities by young adult socioeconomic position in both genders after controlling for family background and smoking during adolescence. However, family socioeconomic position was not strongly associated with smoking in early adulthood. For overweight and obesity, we found socioeconomic inequalities only among women both by young adult and family socioeconomic position after adjusting for birthweight, other family background, and body mass index during adolescence.

Conclusions. Socioeconomic inequalities in smoking emerged in early adulthood according to socioeconomic position. Among women, inequalities in overweight or obesity were already evident by family socioeconomic position and strengthened by their own socioeconomic position. The relative importance of family background and current socioeconomic circumstances varied between smoking and overweight or obesity.

Smoking and obesity among adults are major determinants of population health1 and are associated with adult socioeconomic position (SEP).2,3 Understanding developmental processes of smoking and obesity and their socioeconomic patterns across the life course is crucial to determine optimal times in life for interventions to better limit the population health burden and reduce socioeconomic inequalities in health.4 Smoking and its population health burden are determined at relatively young ages: initiation of smoking occurs most frequently at ages 12 to 14 years and less than 10% of adult smokers initiated smoking after age 19.5,6

On the other hand, despite relatively strong tracking of childhood overweight into adulthood, only one third of obese adults are estimated to have been overweight as a child,7 although this may vary by birth cohort. Smoking by adults is strongly patterned by adult SEP,8 but inconsistent patterns are found during adolescence when most smoking is established.912 This may be because of the blurring of socioeconomic distinctions because of more broad-based patterns of experimentation with smoking among adolescents. Similarly, there is an unclear relationship between SEP and overweight in childhood,13 but both family and adult SEP are predictive of adult obesity.1416 Thus, socioeconomic inequalities in smoking and obesity seem to be crystallized at some time after adolescence, yet our knowledge on when in adulthood such inequalities emerge is limited.

We focused on individuals aged in their early 20s for 2 reasons. First, early adulthood is a pivotal time in the life course that represents a major transitional period in which changes such as residential and financial independence from parents, entry into the labor market, seeking of further education, marriage, and parenthood occur.17,18 Such transitions in early adulthood may also influence behaviors such as smoking and physical activity.19,20 Second, the process of sorting into different socioeconomic paths during early adulthood is of central importance in determining adult SEP. Young adults who pursue tertiary education tend to achieve a higher adult SEP than those who enter the labor market directly after high school. Therefore, young adult SEP is both a marker of cumulative socioeconomic processes transmitted via family background and of future life trajectory that would affect SEP and health in later life. To untangle the effect of young adults’ own SEP on smoking and overweight or obesity in early adulthood from that of family background, we accounted for family background and adolescent factors including smoking and body mass index (BMI; weight in kilograms divided by height in meters squared) during adolescence.

We had 2 specific objectives. First, using the National Longitudinal Study of Adolescent Health (Add Health), we examined whether socioeconomic inequalities in smoking and in overweight and obesity emerged in early adulthood by young adults’ first achieved SEP, that is, their labor market position or further education after high school. Second, we assessed to what extent family background such as family SEP, smoking, and obesity, and adolescent smoking and BMI contributed to the socioeconomic pattern of smoking and overweight or obesity in early adulthood.

METHODS

Study Participants

Add Health is a longitudinal study with a nationally representative sample of US adolescents. Wave 1 interviews were conducted on a stratified random sample of 20 745 adolescents who were in grades 7 through 12 in 1995. Wave 2 interviews were completed in 1996 by 14 736 adolescents who were in grades 7 through 11 at wave 1. Wave 3 interviews, on which our analyses were based, targeted 19 962 wave 1 respondents who were aged at least 18 years and had wave 1 sampling weights. A total of 15 197 respondents aged 18 to 26 years in 2001 to 2002 were interviewed yielding an overall 76% response rate.21

Wave 3 participants who did not have the stratification (region), cluster (school) variable, or wave 3 sampling weight were excluded from the study (n = 875). We further excluded those who, at wave 3, were still attending high school (n = 90), were currently serving in the full-time active-duty military (n = 107), had missing information on education or occupation (n = 376), or were currently pregnant (n = 288). This yielded a total of 13 461 participants, of which 2924 participants had missing values on smoking variables and 486 on overweight and obesity at wave 3. Additionally, there were 13% to 18% of participants whose parents were not interviewed or who had missing values on parent-report information and 0.2% to 4% of participants with missing information on other study variables from their interview.

The final samples of complete case analyses thus were based on 8230 individuals for smoking and 9542 individuals for overweight and obesity. Forty-eight percent of the participants were women, and the mean age was 21.9 years for men and 21.7 years for women at wave 3. There were 67% non-Hispanic White, 15.7% non-Hispanic Black, 11.8% Hispanic, and 5% other racial/ethnic minorities.

Measures of Smoking and of Overweight and Obesity in Early Adulthood (Wave 3)

Smoking status was obtained by self-report on whether respondents had smoked in the past 30 days, the number of days they had smoked, and the number of cigarettes smoked per day during the past 30 days. Heavy smoking was defined as smoking 300 or more cigarettes in the past 30 days (i.e., smoking a half pack or more every day). Those who smoked fewer than 300 cigarettes in the past 30 days were categorized as light-to-moderate smokers. Body mass index, calculated from measured height and weight, was used to determine overweight (BMI = 25.0–29.9 kg/m2) and obesity (BMI ≥ 30.0 kg/m2).

Measures of Young Adult Socioeconomic Position (Wave 3)

Young adult SEP was based on self-report of current employment and educational history after high school (or equivalent) to indicate the first achieved SEP in adulthood. The highest SEP was assigned to those who were attending a 4-year college or graduate school or in the labor market after obtaining an undergraduate degree or higher at wave 3. They were categorized as “going into further education.” For others not in further education, their current occupation, recorded according to the 1998 Standard Occupational Classification list from the Bureau of Labor Statistics, was initially classified into 5 categories22: management or professional, service, sales or office, natural resources or construction or maintenance, and production or transportation or material moving. Because of similar patterns of smoking and of overweight and obesity across categories and small numbers in some occupations (e.g., production or transport, construction or maintenance occupation among women), we used 2 collapsed categories: blue-collar occupation (production or transportation, construction or maintenance, and service) and white-collar occupation (sales or office and management or professional). Individuals who were neither in the labor market nor in further education comprised the lowest SEP category (“no further education and economically inactive”).

Measures of Covariates (Wave 1 or 2)

Covariates were obtained primarily from wave 1 when participants were in grades 7 through 12. Wave 2 data were used in cases when information was missing at wave 1. Family SEP was based on the occupation of the head of household (unemployed = 1; unskilled manual = 2; skilled manual = 3; non-manual = 4), household income (quartile distributions scored 1 to 4), and maternal education (less than high school = 1; high school or equivalent = 2; some college = 3; college graduate or higher = 4). These 3 variables were summed to create a composite variable (range = 3–12) and divided into ter-tiles to indicate low (3–7), middle (8–9), and high (10–12) family SEP. Family structure was coded as 2-parent, single-parent, and other. Family connectedness was measured by 13 items about closeness, perceived caring, and feeling loved and wanted in family (1 = not at all; 5 = very much). A sum of 13 items (range = 13–65) was used; a higher score meant higher family connectedness (Cronbach α = 0.83).23

Family smoking was measured by the presence of a smoker in household (yes or no) and easy access to cigarettes at home during adolescence (yes or no). Parental obesity was categorized into at least 1 parent or no parents being obese, by parental report. Measures of smoking and BMI during adolescence represented the participants’ smoking (yes or no) and BMI (continuous) during the high school years. Self-reported grades were used to calculate grade point average (GPA, range = 1.0–4.0). Depressive symptoms were measured by the Center for Epidemiological Studies Depression Scale.23,24 Birthweight was included in our analysis of overweight and obesity. Of these covariates, household income, maternal education, presence of a smoker in the household, parental obesity, and birthweight of study participants were obtained by parental interviews completed mostly by mothers at wave 1. All other covariates were obtained from participants.

Statistical Analysis

Gender-specific multinomial regression analyses were conducted to estimate the relative odds of heavy or light-to-moderate smoking to nonsmoking and the relative odds of overweight or obesity to normal weight. We initially estimated age- and race/ethnicity–adjusted associations of young adult SEP with smoking and overweight or obesity. This association was then assessed with sequential adjustment for family SEP (model 1), other family characteristics (model 2), and individual factors during adolescence (model 3). Finally, model 4 further adjusted for smoking and BMI during adolescence. We also carried out age-stratified analyses for those aged 11 to 14 years and 15 years and older at wave 1 and found no differences in the results across age groups. Therefore, we presented the results from age-adjusted analyses.

RESULTS

Prevalence of heavy and light-to-moderate smoking by young adult SEP and covariates and their multivariate associations among men and among women are shown in Tables 1 [triangle] and 2 [triangle], respectively. In the fully adjusted model, young adult SEP was not strongly associated with light-to-moderate smoking except among men who had no further education and were economically inactive (odds ratio [OR] = 1.60; 95% confidence interval [CI] = 1.01, 2.52). Men with low family SEP showed lower odds of light-to-moderate smoking compared with those with high family SEP after we adjusted for all covariates, but the estimated odds ratio was somewhat imprecise (OR = 0.80; 95% CI = 0.57, 1.11). Smoking in adolescence was a strong predictor of light-to-moderate smoking after we adjusted for all covariates (OR = 5.29; 95% CI = 4.01, 6.99).

TABLE 1
Prevalence of Smoking and Multivariate Association Between Young Adult Socioeconomic Position and Heavy and Light-to-Moderate Smoking Among Men Aged 18 to 26 Years: National Longitudinal Study of Adolescent Health, 2001–2002
TABLE 2
Prevalence of Smoking and Multivariate Association Between Young Adult Socioeconomic Position and Heavy and Light-to-Moderate Smoking Among Women Aged 18 to 26 Years in the United States: National Longitudinal Study of Adolescent Health, 2001–2002 ...

Prevalence of heavy smoking decreased with young adult SEP from 34% among men who had no further education and were economically inactive to 15% among men in further education. When age and race/ethnicity were adjusted, the odds of heavy smoking were more than 4 times greater among the men who had no further education and were economically inactive (OR = 4.50; 95% CI = 2.74, 7.40), followed by men in blue-collar occupations (OR=2.90; 95% CI=2.10, 3.99) and white-collar occupations (OR=1.54; 95% CI = 1.07, 2.21), compared with men in further education.

With additional adjustment for other family characteristics and individual factors during adolescence in models 2 and 3, the increased odds were attenuated but still found among men in the lowest SEP and those in blue-collar occupations. When we further adjusted for adolescent smoking status in model 4, the increased odds of heavy smoking among men who had no further education and were economically inactive (OR=2.68; 95% CI=1.51, 4.75) and men in blue-collar occupations (OR = 1.65; 95% CI = 1.13, 2.41) remained present.

After we controlled for all covariates, family socioeconomic background was not associated with heavy smoking, but the presence of a smoker in the home (OR = 1.43; 95% CI = 1.05, 1.95) and having easy access to cigarettes at home (OR = 1.34; 95% CI = 0.99, 1.82) were associated with heavy smoking. Young adult men who had higher high school GPAs were less likely to be heavy smokers compared with those with lower GPAs (OR = 0.81; 95% CI = 0.67, 0.99). Smoking during adolescence was a strong predictor of heavy smoking in early adulthood (OR = 7.71; 95% CI = 5.76, 10.31).

Among women, young adult SEP was also inversely associated with light-to-moderate smoking. The inverse associations were attenuated with adjustment for covariates but remained present such that women in blue-collar occupations had 88% increased odds of light-to-moderate smoking (OR = 1.88; 95% CI = 1.28, 2.78) followed by those who had no further education and were economically inactive (OR = 1.45; 95% CI = 0.98, 2.17) and those in white-collar occupations (OR = 1.38; 95% CI = 1.04, 1.84). Among family background factors, presence of a smoker at home seemed to have a persistent effect on light-to-moderate smoking (OR = 1.40; 95% CI = 1.03, 1.90). Smoking during adolescence was also a strong determinant of light-to-moderate smoking in early adulthood after we controlled for all covariates (OR = 7.79; 95% CI = 5.67, 10.71).

Prevalence of heavy smoking among women also decreased with young adult SEP (Table 2 [triangle]). Odds of heavy smoking were highest among those who had no further education and were economically inactive (OR = 5.25; 95% CI = 3.49, 7.78) followed in graded fashion among women in the labor market (blue-collar occupation: OR = 4.62; 95% CI = 3.46, 6.16 and white-collar occupation: OR = 2.38; 95% CI = 1.75, 3.24) compared with those in further education, after we adjusted for age and race/ethnicity. The graded associations were attenuated but persisted after we controlled for family background (models 1 and 2). Smoking during adolescence further attenuated the associations in model 4, but the odds remained higher among those who had no further education and were economically inactive (OR = 3.56; 95% CI = 2.10, 6.02) and those in the labor market (blue-collar occupation: OR = 3.12; 95% CI = 2.10, 4.63 and white-collar occupation: OR = 1.82; 95% CI = 1.28, 2.60).

Although heavy smoking among women was strongly patterned by young adult SEP, it was not associated with family SEP. However, growing up in a single-parented household (OR = 1.37; 95% CI = 1.05, 1.78), having a smoker in the family (OR = 1.96; 95% CI = 1.42, 2.70), and easy access to cigarettes at home during adolescence (OR = 1.34; 95% CI = 1.00, 1.79) increased the odds of heavy smoking, after we adjusted for all covariates. Smoking during adolescence was also strongly associated with the odds of heavy smoking in early adulthood among women (OR = 11.13; 95% CI = 8.10, 15.29).

Prevalence of overweight and obesity by young adult SEP and covariates and their multivariate associations are presented in Tables 3 [triangle] and 4 [triangle] for men and women, respectively. There was no clear socioeconomic patterning of overweight or obesity prevalence by young adult SEP among men. Men in the labor market showed increased odds of obesity compared with those in further education, but the association disappeared when family background factors were accounted for (models 1 and 2). After we controlled for other covariates, family background including parental SEP and parental obesity were not related to overweight or obesity among men. The only strong determinant of overweight or obesity among men was adolescent BMI.

TABLE 3
Prevalence of Overweight and Obesity and Multivariate Association Between Young Adult Socioeconomic Position and Overweight and Obesity Among Men Aged 18 to 26 Years in the United States: National Longitudinal Study of Adolescent Health, 2001–2002 ...
TABLE 4
Prevalence of Overweight and Obesity and Multivariate Association (Adjusted Odds Ratios) Between Young Adult Socioeconomic Position and Overweight and Obesity Among Women Aged 18 to 26 Years in the United States: National Longitudinal Study of Adolescent ...

Table 4 [triangle] shows that women who had no further education and were economically inactive and women in white-collar occupations were more likely to be overweight, but the association was explained by family background including SEP and parental obesity (models 1 and 2). For obesity, there was an inverse graded association with young adult SEP. The inverse association was largely explained by family background, particularly family SEP and parental obesity (models 1 and 2). However, women who had no further education and were economically inactive were more than twice as likely to be obese (OR = 2.61; 95% CI = 0.91, 7.47) as those in further education in the fully adjusted model (model 4). Among women, family SEP was clearly associated with overweight and obesity after adjustment for all covariates and showed a 2-to 4-times increase in the odds of overweight or obesity among those with lower family SEP. Parental obesity and adolescent BMI were also associated with higher odds of overweight and obesity in early adulthood.

DISCUSSION

Consistent with the aims of the study, we found clear socioeconomic inequalities in smoking according to one’s own SEP among both men and women in their early 20s irrespective of family background and smoking during adolescence. Second, family socioeconomic background was not clearly associated with smoking in early adulthood, but family influences were observed through family smoking patterns among both men and women as also found in other studies.25 Family SEP was also not strongly associated with smoking during adolescent years in our data (data not shown, but tables available from the authors upon request). The finding that young adults’ own SEP was a more important contributor to smoking among adolescents and young adults than family SEP is also consistent with other studies.26,27 This suggests that despite tracking of smoking from adolescence into young adulthood28 and their potential influences on social trajectory through young adulthood,29 the socioeconomic patterning of smoking emerged and crystallized as individuals’ own socioeconomic position crystallized in their early 20s.

For overweight and obesity, socioeconomic patterning was found both by family SEP and young adult SEP among women only, suggesting persistent effects of socioeconomic background and strengthening effects of young adults’ own SEP. Family socioeconomic background showed a persistent effect irrespective of parental obesity and young adults’ own SEP among women. Although women with larger BMI as adolescents were more likely to be obese as young adults and to achieve lower SEP, consistent with other studies,30,31 the socioeconomic patterning of obesity evident by family background was further strengthened by the women’s own SEP when they were in their early 20s.

Notably, we found a persistent effect of high school GPA on heavy smoking in early adulthood as observed in other studies,26 in particular among young men. Among women, the association of high school GPA with heavy smoking disappeared with adjustment for adolescent smoking. Koivusilta et al.32 suggested that school grades are a more sensitive indicator of social stratification than traditional indicators such as parental SEP to reveal health inequalities among adolescents. Although our results may lend support to their conclusions, high school GPA was not related to overweight or obesity in our study. It is also worth noting that young adult SEP was associated with obesity only among young women. Studies have found that the association between SEP and obesity is stronger and more consistent among adult women than among men.14,33 A recent study by Wardle et al.34 showed that the association varied by measures of SEP in men such that education was associated with obesity in both men and women whereas occupation-based classification was associated with obesity only in women, which might partly explain the absence of association among men in our study.

Study Limitations

To assess the potential selection bias caused by loss to follow-up, we compared our analytic sample to the original wave 1 sample. The 2 samples were not different on major variables including parental SEP, smoking, and BMI during adolescence. However, those lost to follow-up had lower GPA and were more likely to be from a single-parent family. Because we found that individuals with lower grades and from a single-parent family were more likely to smoke heavily and to have lower SEP in early adulthood, the associations reported here could be underestimates. We also carried out sensitivity analysis by assigning those lost to follow-up as non-smokers and normal-weight individuals and our conclusions remained unchanged (results not shown). To examine the robustness of our results in relation to missing variables, we reanalyzed our data with a category to indicate missing information for our study variables, and our conclusions remained identical.

Although we found clear socioeconomic patterning of smoking and obesity (among women) in young adulthood by their first achieved SEP, there could be other plausible explanations for the observed socioeconomic inequalities. For example, because there is a 4- or 5-year gap between measurements of smoking during adolescence (wave 1 or 2) and early adulthood (wave 3), it is possible that participants would have started smoking after the wave 1 or 2 interview and those initiators may have dropped out of high school or entered blue-collar occupations before wave 3.

Alternatively, the observed associations between young adult SEP and smoking may have been caused by differential cessation rates across SEP.35 A total of 20% had stopped smoking by wave 3 in our data. Prevalence of cessation in the no further education and economically inactive, blue-collar occupation, white-collar occupation, and further education groups was 17%, 19%, 23%, and 22%, respectively, among men. The corresponding figures for women were 21%, 17%, 23%, and 23%. Therefore, it seems plausible that the observed socioeconomic inequalities in smoking by young adult SEP in our study can be partly explained by differential cessation rates by early adulthood. Similarly, socioeconomic inactivity among young women may reflect their marital status and childbearing which would be in turn related to their smoking behavior and BMI. However, further adjustment for marital status, parenthood, and parity did not change our conclusions (results not shown).

Other limitations pertained to the measures used in our study. Our definitions of heavy and light-to-moderate smoking are somewhat arbitrary because there is no consensus in defining heavy smoking in the literature. However, our sensitivity analysis that used daily smoking as an outcome showed essentially unchanged results. Second, our measure of young adult SEP is not conventional. We combined measures of educational history and occupation at the labor market entry rather than using 1 of them or both separately in our multivariate analysis. Our rationale was that because 21.5% of the sample were attending a university or graduate school at the time of the wave 3 interview and they were the most likely to obtain upper–white-collar occupations at their labor market entry, a measure that combined education and occupation would likely yield a better measure of social position in young adulthood than education or occupation taken separately.

Conclusions

Our results showed that smoking was crystallized in a socioeconomically graded fashion as individuals’ own socioeconomic trajectory also became crystallized, such as through pursuing tertiary education or getting a job, irrespective of their family background. By contrast, family socioeconomic background had persistent influences for overweight and obesity and its socioeconomic patterning was strengthened by the individuals’ own SEP, at least among women. Importantly, our study also demonstrated that the relative importance of family background and current SEP as determinants of health and behavior varied by outcomes. Smoking was more strongly related to an individual’s own social position as a young adult than to family SEP. However, overweight and obesity in early adulthood were related to both family SEP and young adult SEP.

This suggests that, although both smoking and overweight or obesity were patterned by adult SEP, the developmental origins of these inequalities lie in different periods across the life course. The observed differences between smoking and overweight or obesity highlighted the importance of considering developmentally specific processes related to socioeconomic disadvantage that link socioeconomic trajectories to health and behavioral trajectories over the life course. Therefore, beneficial effects of intervention efforts to reduce the overall population burden and socioeconomic inequalities in smoking and obesity would be maximized for smoking by targeting young adults who are in a socioeconomically transitional period, and for overweight and obesity by targeting individuals in an earlier stage of life, particularly women.

Acknowledgments

S. Yang was funded in part by the Canadian Institute of Health Research (grant HOA-80072).

Human Participant Protection
The study was approved by the health sciences institutional review board at the University of Michigan.

Notes

Peer Reviewed

Contributors
S. Yang and J. Lynch originated the study and completed the analyses. All authors helped to interpret findings and contributed to writing the article.

References

1. Yusuf S, Hawken S, Ounpuu S, et al. Effect of potentially modifiable risk factors associated with myocar-dial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004;364: 937–952. [PubMed]
2. Lynch JW, Kaplan GA, Salonen JT. Why do poor people behave poorly? Variation in adult health behaviours and psychosocial characteristics by stages of the socioeconomic lifecourse. Soc Sci Med. 1997;44: 809–819. [PubMed]
3. Marmot M, Wilkinson RG. Social Determinants of Health. New York, NY: Oxford University Press; 1999.
4. Lynch J, Davey Smith G. A life course approach to chronic disease epidemiology. Annu Rev Public Health. 2005;26:1–35. [PubMed]
5. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future national survey results on drug use, 1975–2003: Volume I, Secondary school students. Bethesda, Md: National Institute on Drug Abuse; 2003.
6. Preventing Tobacco Use Among Young People: A Report of the Surgeon General. Atlanta, Ga: National Center for Chronic Disease Prevention and Health Promotion; 1994.
7. Freedman DS, Khan LK, Dietz WH, Srinivasan SR, Berenson GS. Relationship of childhood obesity to coronary heart disease risk factors in adulthood: the Bogalusa Heart Study. Pediatrics. 2001;108:712–718. [PubMed]
8. Jarvis MJ, Wardle J. Social patterning of individual health behaviours: the case of cigarette smoking. In: Marmot M, Wilkinson RG, eds. Social Determinants of Health. Oxford, England: Oxford University Press; 1999:240–255.
9. Lowry R, Kann L, Collins JL, Kolbe LJ. The effect of socioeconomic status on chronic disease risk behaviors among US adolescents. JAMA. 1996;276: 792–797. [PubMed]
10. Faulkner DL, Escobedo LG, Zhu BP, Chrismon JH, Merritt RK. Race and the incidence of cigarette smoking among adolescents in the United States. J Natl Cancer Inst. 1996;88:1158–1160. [PubMed]
11. Glendinning A, Shucksmith J, Hendry L. Social class and adolescent smoking behaviour. Soc Sci Med. 1994;38:1449–1460. [PubMed]
12. Soteriades ES, DiFranza JR. Parent’s socioeconomic status, adolescents’ disposable income, and adolescents’ smoking status in Massachusetts. Am J Public Health. 2003;93:1155–1160. [PMC free article] [PubMed]
13. Parsons TJ, Power C, Logan S, Summerbell CD. Childhood predictors of adult obesity: a systematic review. Int J Obes Relat Metab Disord. 1999;23(suppl 8): S1–S107. [PubMed]
14. Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of the literature. Psych Bull. 1989; 105:260–275. [PubMed]
15. Gillman MW. A life course approach to obesity. In: Kuh D, Ben-Shlomo Y, eds. A Life Course Approach to Chronic Disease Epidemiology. 2nd ed. Oxford, England: Oxford University Press; 2004:189–217.
16. Power C, Matthews S. Origins of health inequalities in a national population sample. Lancet. 1997;350: 1584–1589. [PubMed]
17. Schulenberg JE, Maggs JL. A developmental perspective on alcohol use and heavy drinking during adolescence and the transition to young adulthood. J Stud Alcohol Suppl. 2002;(14):54–70. [PubMed]
18. Arnett JJ. Emerging Adulthood: The Winding Road From the Late Teens Through the Twenties. New York, NY: Oxford University Press; 2004.
19. Baranowski T, Cullen KW, Basen-Engquist K, et al. Transitions out of high school: time of increased cancer risk? Prev Med. 1997;26:694–703. [PubMed]
20. Cullen KW, Koehly LM, Anderson C, et al. Gender differences in chronic disease risk behaviors through the transition out of high school. Am J Prev Med. 1999;17:1–7. [PubMed]
21. Chantala K, Kalsbeek WD, Andraca E. Non-response in Wave III of the Add Health Study, 2004. Available at: http://www.cpc.unc.edu/projects/addhealth/files/W3nonres.pdf. Accessed March 9, 2007.
22. Salmon L. 1998 standard occupational classification. Ind Relations. 1999;38:231–235.
23. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401.
24. Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from harm. Findings from the National Longitudinal Study on Adolescent Health. JAMA. 1997;278:823–832. [PubMed]
25. Kestila L, Koskinen S, Martelin T, et al. Influence of parental education, childhood adversities, and current living conditions on daily smoking in early adulthood. Eur J Public Health. 2006;16:617–626. [PubMed]
26. Karvonen S, Rimpela AH, Rimpela MK. Social mobility and health related behaviours in young people. J Epidemiol Community Health. 1999;53:211–217. [PMC free article] [PubMed]
27. Paavola M, Vartiainen E, Haukkala A. Smoking from adolescence to adulthood: the effects of parental and own socioeconomic status. Eur J Public Health. 2004;14:417–421. [PubMed]
28. Kandel DB, Chen K, Warner LA, Kessler RC, Grant B. Prevalence and demographic correlates of symptoms of last year dependence on alcohol, nicotine, marijuana and cocaine in the US population. Drug Alcohol Depend. 1997;44:11–29. [PubMed]
29. Koivusilta L, Rimpela A, Vikat A. Health behaviours and health in adolescence as predictors of educational level in adulthood: a follow-up study from Finland. Soc Sci Med. 2003;57:577–593. [PubMed]
30. Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med. 1997; 337:869–873. [PubMed]
31. Gortmaker SL, Must A, Perrin JM, Sobol AM, Dietz WH. Social and economic consequences of overweight in adolescence and young adulthood. N Engl J Med. 1993;329:1008–1012. [PubMed]
32. Koivusilta LK, Rimpela AH, Kautiainen SM. Health inequality in adolescence. Does stratification occur by familial social background, family affluence, or personal social position? BMC Public Health. 2006; 6:110. [PMC free article] [PubMed]
33. Leigh JP, Fries JF, Hubert HB. Gender and race differences in the correlation between body mass and education in the 1971–1975 NHANES I. J Epidemiol Community Health. 1992;46:191–196. [PMC free article] [PubMed]
34. Wardle J, Waller J, Jarvis MJ. Sex differences in the association of socioeconomic status with obesity. Am J Public Health. 2002;92:1299–1304. [PMC free article] [PubMed]
35. Power C, Graham H, Due P, et al. The contribution of childhood and adult socioeconomic position to adult obesity and smoking behaviour: an international comparison. Int J Epidemiol. 2005;34:335–344. [PubMed]

Articles from American Journal of Public Health are provided here courtesy of American Public Health Association
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles