Elsevier

Preventive Medicine

Volume 36, Issue 6, June 2003, Pages 721-730
Preventive Medicine

Regular article
Socioeconomic disparities in cancer-risk behaviors in adolescence: baseline results from the Health and Behaviour in Teenagers Study (HABITS)

https://doi.org/10.1016/S0091-7435(03)00047-1Get rights and content

Abstract

Background

This study explores the association between socioeconomic deprivation and five factors associated with long-term risk of cancer, in adolescents.

Methods

BMI, fat intake, fruit and vegetable intake, smoking, and exercise were assessed in 4320 students ages 11 to 12, from 36 schools, in the first year of a 5-year longitudinal study of the development of health behaviors (HABITS study). Neighborhood socioeconomic deprivation for each student’s area of residence was matched to their postcode (zip code). We used multiple logistic regression analyses to investigate the relationship between risky behaviors and socioeconomic circumstances.

Results

Univariate analyses showed boys and girls from more deprived neighborhoods were more likely to have tried smoking, to eat a high fat diet, and to be overweight. Girls living in more deprived areas were also less likely to eat five servings of fruit and vegetables or to exercise at the weekend. Most differences persisted after controlling for ethnicity. A clear deprivation gradient emerged for each risk factor, indicating the linear nature of the relationship.

Conclusions

This study demonstrates the influence of deprivation on engaging in cancer-risk health behaviors. These patterns may set young people from more socioeconomically deprived social environments on a trajectory leading to increased cancer mortality in adult life.

Introduction

The burden of cancer is distributed unequally in almost all developed countries, with cancer deaths being higher in groups from deprived socioeconomic backgrounds [1]. There is increasing interest in understanding the roots of these inequalities, with a view to finding better ways to minimize the loss of life to cancer across all sectors of the community. Many processes are likely to contribute to disparities in cancer mortality, including differences in medical care, environmental exposures, reproductive patterns, chronic infections, health behaviors, and utilization of screening. In terms of attributable risk in the population as a whole, tobacco smoking, diet, body weight, and alcohol consumption are estimated to account for the largest fractions of the avoidable risks for cancer [2].

Socioeconomic differentials have been established in many of the contributory behavioral risk factors, supporting the idea that health behavior differences are important influences on the gradient in cancer mortality [3]. Since the hazardous effects of smoking were publicized in the 1950s, dramatic differences in smoking rates between higher and lower socioeconomic status (SES) groups have emerged in most industrialized countries. In the United Kingdom, 15% of men with professional occupations are smokers rising to 44% in unskilled manual men [4]. Similar patterns can be seen in the United States [5] and other European countries [6]. Fresh fruit and vegetables are consumed less among the poor than the rich in many parts of the world. In Britain, intake of fruit and vegetables is almost one and one-half times as high in the highest compared with the lowest income groups [7], and similar associations can be seen with income and education in the United States [8]. Obesity, once confined mainly to the wealthy, has become one of the hallmarks of poverty in industrialized countries. Adult obesity prevalence in Britain is less than 15% in women from professional groups, but rises to 28.1% in women in unskilled jobs [9]. SES patterning in overall physical activity is complex, reflecting lower levels of occupational physical activity, but higher levels of leisure-time physical activity in more affluent groups [10]. However, with progressively fewer jobs involving significant physical activity, an active lifestyle will increasingly depend on leisure-time physical activity throughout the population [11].

These observations highlight the importance of understanding the development of socioeconomic gradients in the behaviors that affect cancer risk, many of which emerge in adolescence. Most smokers take up the habit in their teenage years [4]. Dietary choices move from being primarily determined by family eating patterns in childhood, to being increasingly peer-led or independently determined behaviors in the teenage years [12]. Physical activity, which is part of active play in childhood, becomes one of many competing leisure options in adolescence. Particularly for girls, this is often a time when active pastimes are dropped in favor of more sedentary pursuits [13], [14]. Adolescence is also one of the key times for weight gain, with recent data suggesting that over 17% of 15-year-old girls have a BMI that puts them at risk of obesity in adulthood [15].

Adolescence therefore appears to be a life stage at which important behavioral choices are emerging and trajectories for adult life may be set, and so it may well be a time when socioeconomic differentials in behavior appear. Adolescents become increasingly independent in decision-making, and take progressively more responsibility for social activities, leisure activities, and lifestyle. They also begin to assume responsibility for their own health-related choices, and must develop an understanding of how to promote and protect health.

Research into socioeconomic inequalities in health behaviors during adolescence has produced a mixed picture, and few studies have examined the full range of behaviors under scrutiny in this study. In addition, some studies have used ethnicity as a proxy for socioeconomic deprivation, which has added to the uncertainty of what effect, if any, socioeconomic deprivation has at this formative age. Adolescent smoking has been the focus of many investigations, and although some studies have found no relationship [16], [17], [18], many other studies in the United States, Europe, and Australia observe SES differences [19], [20], [21], [22], [23]. Studies of diet have provided a fairly consistent picture of higher fat intake in more deprived young people [24], [25], [26]. A similar picture has emerged in relation to exercise studies [17], [27], [28], [29]. The findings, however, are again mixed when examining fruit and vegetable intake [30], [31] and obesity rates [32].

The Health and Behaviour in Teenagers Study (HABITS) was set up to produce some more definitive answers with respect to the adoption, development, and change in health behaviors which are related to adult cancer risk. The cohort of adolescents is being assessed annually over 5 years to investigate tracking and clustering of health behaviors as well as psychosocial predictors. The present report examines SES differences in health behaviors in the baseline year of the study (1999). The sample was drawn to be socioeconomically diverse, in order to maximize the power to examine SES differences in the development of health behaviors. We aimed to examine the extent of the SES differences in health behavior engagement, looking for a gradient of effect and not just comparing affluent and deprived young people. We also included a range of health behaviors, to allow us to determine whether socioeconomic deprivation has a similar association with each behavior.

Section snippets

Methods

The HABITS study is a 5-year longitudinal study that began in 1999 and will be completed in 2003. It is a school-based survey, encompassing 36 secondary (high) schools that are visited annually by a team of researchers. The target sample in the baseline year was all students in Year 7 (ages 11 to 12) registered in the 36 schools (n = 5120). Students complete an extensive questionnaire annually, as well as being weighed and measured and providing a saliva sample for cotinine assay.

Description of sample

Students (5120) were registered in Year 7 (U.S. Grade 6) at the 36 participating schools. The achieved sample comprised 4320 adolescents (1742 girls, 2578 boys), representing an 84% coverage rate; 515 (10%) were absent from the class on the day of the visit and 285 (5.5%) opted out of the study, either by parental or self-exclusion. Numbers in each analysis vary slightly due to missing data on that item.

Students (1617) (1040 boys, 577 girls) came from mixed-sex schools, and 2702 (1536 boys,

Discussion

The link between socioeconomic deprivation and cancer is now well established, as are the inequalities in cancer-risk factors in adults [3]. The stage in the life course at which these inequalities emerge is less clear. The present analyses examine evidence for a deprivation gradient in cancer risk factors in early adolescence (11–12 years old); a stage when young people take increasing control over their lifestyle. The data reported here are from the first wave of a large longitudinal study

Acknowledgements

This study was funded by Cancer Research UK and the Department of Health, and was initiated by the lead author. The participation of the 36 schools and 4320 students is gratefully acknowledged. We also thank Dr. David Boniface, who advised on the analysis of the data. Postcode-to-enumeration district matching was carried out via UK area look-up tables, by Ludi Simpson, CCSR, University of Manchester. Enumeration district deprivation scores were derived by James Harris from the 1991 Census,

References (53)

  • D. Freedman et al.

    Black/White differences if relative weight and obesity among girlsThe Bogalusa Heart Study

    Prev Med

    (2000)
  • C.L. Hupkens et al.

    Class differences in the food rules mothers impose on their childrena cross-national study

    Soc Sci Med

    (1998)
  • F. Faggiano et al.

    Socioeconomic differences in cancer incidence and mortality

    IARC Sci Publ

    (1997)
  • J. Peto

    Cancer epidemiology in the last century and the next decade

    Nature

    (2001)
  • Wardle J, Jarvis MJ. Health related behaviours. Independent inquiry into inequalities in health report. In: Acheson D,...
  • Bridgwood A, Lilly R, Thomas M, Bacon J, Sykes W, Morris S. Living in Britain: results from the 1998 General Household...
  • MMWR

    (2000)
  • A.E. Cavelaars et al.

    Educational differences in smokinginternational comparison

    BMJ

    (2000)
  • W.P. James et al.

    Socioeconomic determinants of health. The contribution of nutrition to inequalities in health

    BMJ

    (1997)
  • M.K. Serdula et al.

    Fruit and vegetables intake among adults in 16 statesresults of a brief telephone survey

    Am J Public Health

    (1995)
  • Boreham R, Erns B, Falaschetti E, Hirani V, Primatesta P. Risk factors for cardiovascular disease. Health Survey for...
  • M. Fogelholm et al.

    Determinants of energy balance and overweight in Finland 1982 and 1992

    Int J Obes Relat Metab Disord

    (1996)
  • L. Hill et al.

    Fruit and vegetables as adolescent food choices in New Zealand

    Health Prom Int

    (1998)
  • K.R. Allison et al.

    Age and sex differences in physical inactivity among Ontario teenagers

    Can J Public Health

    (1997)
  • P. Gordon-Larsen et al.

    Determinants of adolescent physical activity and inactivity patterns

    Pediatrics

    (2000)
  • B.P. Zhu et al.

    Cigarette smoking and its risk factors among elementary school students in Beijing

    Am J Public Health

    (1996)
  • Cited by (130)

    • Disparities in mortality among 25–44-year-olds in England: a longitudinal, population-based study

      2018, The Lancet Public Health
      Citation Excerpt :

      For almost all types of cancer, risk of death is higher and death is likely to occur more quickly in more deprived areas.33,34 Particularly striking deprivation gradients are observed for cancers of the oropharynx and lung, which are primarily attributed to smoking.35 Similar disparities have been reported for cancer of the cervix, which is caused by human papillomavirus, and is largely determined by sexual behaviour risk factors, including age at first intercourse, number of sexual partners, failure to use a condom during intercourse with multiple partners, and screening programme non-engagement.36

    View all citing articles on Scopus
    View full text