Socioeconomic status and weight change in adults: a review

https://doi.org/10.1016/j.socscimed.2004.08.056Get rights and content

Abstract

In developed countries, obesity is inversely associated with socioeconomic status (SES) among women, and less consistently among men; whereas, in developing countries, the association is direct. However, the relationship of SES to weight change over time is unknown. This relationship was the focus of the present literature review. It was hypothesized that, compared with persons of higher SES, persons of low SES would show greater weight gain or risk of weight gain over time. A search of electronic databases identified 34 relevant articles from developed countries reporting on studies that assessed the relationship of various measures of SES with weight change over time in adults (there were too few papers from developing countries (n=1) to include). Results of the methodologically strongest studies (those which obtained objectively measured adiposity data and used a follow-up period of 4 years or more) showed that, among non-black samples, there were relatively consistent inverse associations between occupation and weight gain for men and women. When SES was assessed using education, evidence was slightly less consistent, but still provided some support for the hypothesized relationship. However, when income was used as the indicator of SES, findings were inconsistent, although there were fewer studies available. There was little support for a relationship between SES and weight gain for black samples. In the context of the worldwide epidemic of obesity, these findings suggest that in developed countries, weight gain prevention efforts might best be focused on those who are most socioeconomically disadvantaged, particularly those in lower status occupations.

Introduction

Obesity is one of the most prevalent chronic conditions impacting the health of populations worldwide, estimated to affect more than one billion adults (World Health Organization (WHO), 1998). The prevalence of overweight and obesity has been estimated to be between 50% and 65% in developed countries such as the US, UK, and Australia (Australian Institute of Health and Welfare, 2003; Flegal, Carroll, Ogden, & Johnson, 2002; The Comptroller & Auditor General, 2001). Rapid increases in the prevalence of obesity have also been noted in developing countries; for instance, the proportion of adults overweight or obese in China more than doubled during the 1990s, rising from 8% in 1989 to 18% in 1997 (Bell, Ge, & Popkin, 2001). Given the well-established associations of obesity with a range of chronic health conditions and symptoms, research into the cause and prevention of this epidemic is urgently required.

Obesity is related to socioeconomic status. The seminal review of 144 studies by Sobal and Stunkard (1989) conducted 15 years ago demonstrated a direct relationship between SES and obesity in developing societies. That is, men, women and children of higher SES had a higher likelihood of being obese than those of lower SES. In developed societies, an inverse relationship was found for women, whereas relationships for men and children were inconsistent. Since the publication of that paper, to our knowledge there has been no attempt made to review the substantial number of subsequent studies that have investigated these associations.

Such a review is important and timely for a number of reasons. First, there has been no systematic investigation of whether the associations between socioeconomic status and body weight have changed since the 1989 review. This is particularly important, given that the prevalence of obesity has increased rapidly worldwide since the publication of Sobal and Stunkard's review (WHO, 1998). Developed and developing societies are now facing an epidemic of obesity. Driving this epidemic are population increases in mean body weight. Average body mass indices (BMI) of children, adolescents and adults have increased over the past 20 years (WHO, 1998). A steady pattern of weight gain across the lifespan until older adulthood is now typical. Even highly educated groups are at risk of weight gain (Ball, Crawford, Ireland, & Hodge, 2003a). However, whether weight gain is occurring differentially across SES groups has not been systematically examined. Furthermore, results of studies comparing SES gradients in cross-sectional datasets collected across different times or places have been mixed (Flegal, Harlan, & Landis, 1988; Molarius, Seidell, Sans, Tuomilehto, & Kuulasmaa, 2000).

Secondly, an updated review of SES and weight is important, since Sobal and Stunkard incorporated mostly cross-sectional studies in their original review. Such studies are limited in that they represent only a single snapshot of the population at a particular point in time, and hence the degree of change in weight over time, and how this varies by SES cannot be determined conclusively from cross-sectional studies. For these reasons, examination of longitudinal studies of SES and weight change (i.e., with age) is important. A review of longitudinal studies of different samples may also provide more conclusive evidence as to when in the life course SES gradients in obesity are established. While longitudinal studies of weight change in large samples have been less common in the past, in recent years more studies of this nature have been conducted, making such an investigation feasible.

A final reason why a review of socioeconomic factors and obesity is timely is that understanding socioeconomic inequalities in health has become an important priority on the international health agenda (e.g., Blane, 2001). While health and life expectancy are improving worldwide, evidence suggests that SES differentials in mortality and other health outcomes are still maintained or widening (Blane, 2001). Persons of low SES are already at high risk of poorer health from a range of chronic conditions and diseases. Should increases in obesity be accelerated among persons of low SES, this is likely to compound these already widening health inequalities for obesity-related chronic conditions.

Obesity is differentially related to different indices of SES for men and women (Ball, Mishra, & Crawford, 2002b). In their previous review of largely cross-sectional studies, Sobal and Stunkard (1989) did not distinguish between different indices of SES. However, an investigation of associations of weight gain with different dimensions of SES, and their appropriate measures, may help to identify potential aetiological pathways. Furthermore, previous studies have suggested that patterns of associations of SES with obesity are specific for age and sex groups (Ball et al., 2002b), and for developed and developing societies (Sobal & Stunkard, 1989). These issues were taken into account in this review, which examines and presents evidence separately by sex, age group, and SES index.

The aim of this review is to advance Sobal and Stunkard's (1989) findings by examining studies investigating socioeconomic status and weight change. To do so, the review focuses not on the multitude of cross-sectional studies published since that time (many of which have confirmed the findings of Sobal and Stunkard), but on longitudinal studies that enable an investigation of socioeconomic status and weight change in individuals over time. We hypothesized that SES would be related to weight change longitudinally, in the same direction as the cross-sectional associations of SES and body weight reported by Sobal and Stunkard (1989) (i.e., an inverse association in developed countries; a direct association in developing countries).

Section snippets

Search strategy

A detailed search of the following databases was conducted for articles published between 1980 (or the database starting date if after 1980) and the end of 2002: MEDLINE (via Pub Med), Psychinfo, Health & Society, Rural on Aust. Health, Australian Medical Index, Science Direct, Current Contents, CINAHL, Sociological Abstracts, Social Science Plus, Proquest, Web of Science, Expanded Academic, and APAIS. The authors’ own bibliographic libraries were also searched. The following search terms were

Results

A total of 34 papers were reviewed, representing a total of 26 discrete studies (some papers were based on the same study). These comprised five papers based on men only, six based on women only, and 23 papers including both men and women. The methodologies of all papers reviewed are presented in Table 1.

The vast majority of the studies were conducted during the 1980s and 1990s. Education was the most commonly used indicator of SES, included in 28 papers. Fifteen papers reported on

Discussion

In the context of an increasing worldwide epidemic of obesity, this study asked “Does weight change over time in adulthood vary by socioeconomic status?” In particular, we tested the hypothesis that persons of lower socioeconomic status (SES) are at increased risk of weight gain. The results of this review of longitudinal studies in developed countries, while not entirely consistent, provided reasonable support for this hypothesis. Among largely non-black samples, and based on the

Acknowledgements

Kylie Ball and David Crawford are both supported by Career Development Awards funded by the Australian National Health and Medical Research Council/National Heart Foundation of Australia.

References (50)

  • A. Bell et al.

    Weight gain and its predictors in Chinese adults

    International Journal of Obesity

    (2001)
  • M. Bernstein et al.

    Passive smoking, active smoking, and educationtheir relationship to weight history in women in Geneva

    American Journal of Public Health

    (1996)
  • D. Blane

    Commentarysocioeconomic health differentials

    International Journal of Epidemiology

    (2001)
  • G. Burke et al.

    Differences in weight gain in relation to race, gender, age and education in young adultsthe CARDIA study. Coronary artery risk development in young adults

    Ethnicity and Health

    (1996)
  • B.H. Dennis et al.

    Weight gain and its correlates in Poland between 1983 and 1993

    International Journal of Obesity

    (2000)
  • K.M. Flegal et al.

    Prevalence and trends in obesity among US adults, 1999–2000

    JAMA

    (2002)
  • M. Fogelholm et al.

    Predictors of weight change in middle-aged and old men

    Obesity Research

    (2000)
  • L. Fredman et al.

    Weight changean indicator of caregiver stress

    Journal of Aging and Health

    (1997)
  • S. French et al.

    Predictors of weight change over two years among a population of working adultsthe Healthy Worker Project

    International Journal of Obesity

    (1994)
  • P. Froom et al.

    Smoking cessation and body mass index of occupationally active menthe Israeli CORDIS Study

    American Journal of Public Health

    (1999)
  • K.J. Greenlund et al.

    Body mass index in young adultsassociations with parental body size and education in the CARDIA Study

    American Journal of Public Health

    (1996)
  • N. Haapanen et al.

    Association between leisure time physical activity and 10-year body mass change among working-aged men and women

    International Journal of Obesity

    (1997)
  • R. Hardy et al.

    The influence of childhood weight and socioeconomic status on change in adult body mass index in a British national birth cohort

    International Journal of Obesity

    (2000)
  • R. Jeffery et al.

    Socioeconomic status and weight control practices among 20- to 45-year-old women

    American Journal of Public Health

    (1996)
  • R.W. Jeffery et al.

    Socioeconomic status differences in health behaviours related to obesitythe Healthy Worker Project

    International Journal of Obesity

    (1991)
  • Cited by (418)

    • Body-mass index trajectories from childhood to mid-adulthood and their sociodemographic predictors: Evidence from the International Childhood Cardiovascular Cohort (i3C) Consortium

      2022, eClinicalMedicine
      Citation Excerpt :

      Lower baseline socioeconomic position in childhood (indicated by parental education) was associated with a greater likelihood of belonging to the more (improving from high) and less (progressing to moderate, progressing to moderate/high, and progressing to high) favourable BMI trajectory groups, mostly in the MUSC study and in pooled analysis (improving from high and progressing to very high). This finding may be surprising, as evidence from developed countries on socioeconomic position and obesity suggests that lower socioeconomic position (indicated by occupation and to a less extent education) is associated with a greater likelihood of weight gain amongst adults, although only amongst white individuals.39 However, children from higher socioeconomic groups are less likely to be obese and hence have less chance to improve their BMI from childhood into adulthood, which is a likely explanation for this finding.

    View all citing articles on Scopus
    View full text