Elsevier

Social Science & Medicine

Volume 55, Issue 6, September 2002, Pages 1055-1068
Social Science & Medicine

Gradients in risk for youth injury associated with multiple-risk behaviours: a study of 11,329 Canadian adolescents

https://doi.org/10.1016/S0277-9536(01)00224-6Get rights and content

Abstract

This study used the Canadian version of the World Health Organization-Health Behaviour in School-Aged Children (WHO-HBSC) Survey to examine the role of multiple risk behaviours and other social factors in the etiology of medically attended youth injury. 11,329 Canadians aged 11–15 years completed the 1997–1998 WHO-HBSC, of which 4152 (36.7%) reported at least one medically attended injury. Multiple logistic regression analyses failed to identify an expected association between lower socio-economic status and risk for injury. Strong gradients in risk for injury were observed according to the numbers of multiple risk behaviours reported. Youth reporting the largest number (7) of risk behaviours experienced injury rates that were 4.11 times (95% CI: 3.04–5.55) higher than those reporting no high risk behaviours (adjusted odds ratios for 0–7 reported behaviours: 1.00, 1.13, 1.49, 1.79, 2.28, 2.54, 2.62, 4.11; ptrend<0.001). Similar gradients in risk were observed within subgroups of young people defined by grade, sex, and socio-economic level, and within restricted analyses of various injury types (recreational, sports, home, school injuries). The gradients were especially pronounced for severe injury types and among those reporting multiple injuries. The analyses suggest that multiple risk behaviours may play an important role in the social etiology of youth injury, but these same analyses provide little evidence for a socio-economic risk gradient. The findings in turn have implications for preventive interventions.

Introduction

While there is considerable evidence that childhood injury is a leading public health problem internationally, the social etiology surrounding these injuries is not well understood. Gradients in risk for childhood injury have been demonstrated for a variety of social indicators, including those describing levels of household income (Baker, O’Neil, Ginsburg, & Li, 1992; Health Canada, 1997; Nolan & Penny, 1992; Dougherty, Pless, & Wilkins, 1990; Durkin, Davidson, Kuhn, O’Connor, & Barlow, 1994), education (Health Canada, 1997; Nolan & Penny, 1992; Jolly, Moller, & Volkmer, 1993), unemployment (Nersesian, Petit, Shaper, Lemieux, & Naor, 1985; Scholer, Mitchel, & Ray, 1997) and social status (Roberts, Marshall, Norton, & Borman, 1992; Harris & Kotch, 1994). Explanations for these differences have generally focused on material deprivation and differential exposures to environmental hazards in lower socio-economic groups (Black, Morris, Smith, & Townsend, 1982; Mare, 1982; Roberts, 1997; Wise, Kotelchuck, Wilson, & Mills, 1985). Some investigators have challenged the existence of injury gradients among populations of adolescents (youth; young people) (West, 1997; Anderson et al., 1994; Williams, Currie, Wright, Elton, & Beattie, 1997) suggesting that explanations for the observed differences in adolescents are speculative since data are often derived from observations of younger children, or are not present at all for specific forms of injury (e.g. Lyons et al., 2000)

The emphasis in the literature on socio-economic explanations for injury gradients ignores the potential for behavioural explanations. Adolescence is a period of change and transition (Feldman & Elliott, 1990), and developmental models may be useful in understanding injury risk during adolescence. Several such models have been identified in the past: cumulative events or simultaneous change, accentuation, and trajectory models.

Cumulative events or simultaneous change models of development are based upon the idea that psychological, social and biological changes in adolescence are often happening at the same time. In combination, these factors affect the ability of adolescents to adjust and cope during adolescence (Simmons & Blyth, 1987). Due to changes in the school and peer environments, renegotiations in family roles and physical changes associated with puberty, the ability of young people to cope may be overtaxed. This in turn has been linked to depressed affect, lowered self-esteem, and a drop-off in school performance (Petersen, Sarigiani, & Kennedy, 1991; Simmons, Carlton-Ford, & Blyth, 1987). It is not known whether risk-taking that results in injury is influenced by these simultaneous and rapid changes in adolescents’ lives.

According to the accentuation model of development, puberty and other times of transition act to accentuate existing difficulties or reinforce pre-existing behavioural tendencies (Block, 1982; Caspi & Moffitt, 1991). This model differs from personality theory which suggests that youth adapt to transitional events by re-organizing their behaviours and even their personalities. In contrast, the accentuation model suggests that existing individual behaviours are reinforced or magnified by challenging situations. Thus, young people may cope with stresses by assimilating new risk-taking situations into existing patterns and cognitive structures, mainly as a way of minimizing change. For example, early patterns of rough play among boys may lead to increased risk for injury in adolescent males.

Both Rutter (1989) and Elder (1985) have also discussed the use of trajectories, or long term influences, to explain patterns of development during adolescence. The social environment in which transition takes place is thought to be an important influence on how a young person develops. For example, the trajectory model suggests that the amount of time that a youth associates with older peers at an early age will influence later risk-taking. This long-term influence affects the age and extent to which youth experiment with drinking and sexual behaviours (Stattin & Magnusson, 1990), and perhaps also affects the degree to which youth put themselves at risk for injury.

The identification of behaviours that place an adolescent or groups of adolescents at risk for poor health has potential significance for the prevention of injuries. Hazardous adolescent lifestyles may include engagement in multiple risk behaviours (Irwin & Millstein, 1986) including unsafe sex, substance use and abuse, risk-taking leading to accidents and injury, and disengagement from school (Graber & Brooks-Gunn, 1995). When these behaviours occur simultaneously they place the youth at a very high risk for illness and injury. If these multiple risk behaviours occur early in adolescence (trajectory model) or cluster together temporally (simultaneous change model), the likelihood of morbidity is even greater (Jessor, 1992; Millstein, Petersen, & Nightingale, 1993.)

The ability to predict the occurrence of adverse health events associated with multiple risk behaviours may be tempered somewhat by the existence of protective factors in the life of an adolescent. For example, a young person may have high rates of risk behaviours but also have high rates of protective factors such as family, school or material supports. Health outcomes experienced by that young person may be better than those experienced by another adolescent with fewer risk behaviours, but also fewer protective factors. In addition, some risk behaviours pose both risks and benefits. Excessive drinking is clearly a risk behaviour but it may also be a potential source of social support, since experimentation represents normal activity during teenage years. This experimentation reflects a willingness to engage in behaviours that may be normative among friends and acquaintances, as the youth moves towards a peer orientation and independence as a replacement for the orientation provided by the family.

Recent analyses of the 1996 National Population Health Survey (Canada) showed that the majority of young Canadians engage in at least one high-risk behaviour such as smoking, binge drinking, and/or sex with multiple partners (Galambos & Tilton-Weaver, 1998). Large percentages (33% of females, 43% of males) engage in two or more of these same behaviours, and youth from lower socio-economic backgrounds tend to report more health risk behaviours. This pattern is analogous to the observations of Connop, King, & Boyce (1999) whose work with young Canadians identified several distinct peer groups of youth. The prevalence of high-risk behaviours including smoking tended to be clustered among groups of youth who were socially alienated from mainstream society.

Immediate and long-term health consequences of multiple risk behaviours are still undefined, although it is reasonable to assume that risk behaviours translate into elevated risks for morbidity. Whether or not injuries occur may also depend on the co-occurrence of protective factors, the intensity of engagement in risk behaviours, and the age of initiation of these behaviours (Irwin & Millstein, 1986; Jessor, 1992).

We had the opportunity to explore relationships between multiple risk behaviours and the occurrence of injury using Canadian records from the 1997–1998 World Health Organization Health Behaviour in School-Aged Children Survey (WHO-HBSC). The WHO-HBSC is a collaborative research project involving countries in Central and Eastern Europe, the Middle East, and North America (Currie, 1998). Large and representative samples of youth are drawn from school-based settings in each of the countries. Canada has participated in the WHO-HBSC on three occasions. During the latest iteration (1997–1998), a series of questions about medically attended injuries was asked, in addition to core questions about health behaviours, risk-taking, demographics, and socio-economic status (King, Boyce, & King, 1999).

The objective of the present study was to analyze medically attended injuries and their association with multiple risk behaviours among young Canadians. The factors examined included a new index that describes the occurrence of multiple-risk behaviours indicators, socio-economic status, and demographic factors. In performing this analysis, we hoped to determine: (1) whether a socio-behavioural gradient in risk for adolescent injury could be demonstrated in Canada; (2) if such a gradient between multiple risk behaviours and injury was observed, whether it could be explained by the reported socio-economic status or other related factors.

Section snippets

The WHO-HBSC

The 1997–1998 WHO-HBSC involved study of a sample of Canadian students from elementary and high schools. A cluster sample design was used with the school class being the basic cluster. The sample was designed according to the international WHO-HBSC protocol (Currie, 1998), in that the distribution of the students reflected the distribution of Canadians in the selected grades (6–10, representing youth with average ages of 11–15 years), and the sample was considered self-weighting. Within each

Results

11,416 Canadian young people participated in the 1997/1998 WHO-HBSC and of these, 11,329 answered the initial question about medically treated injuries. These constituted the sample available for study, with 36.7% reporting at least one injury event.

Table 1 provides a description of the study population according to the confounders selected a priori, as well as the three socio-economic indicators. (Note: not all respondents answered every question; hence the apparent discrepancies when summing

Discussion

There were two major findings from this national, population-based analysis. First, strong gradients in the occurrence of injury were found among Canadian youth in association with the reporting of multiple-risk behaviours. These gradients were observed consistently within subgroups defined by grade, gender, socio-economic status, injury type, severity and number of medically treated injuries. They also remained after adjustment for all available factors that were considered as potential

Acknowledgements

Health Canada funds the Canadian version of the World Health Organization-Health Behaviour in School-Aged Children Survey, under the co-ordination of Mary Johnston. The WHO-HBSC is a WHO/EURO collaborative study. International Coordinator of the 1997/98 study: Candace Curry, University of Edinburgh, Scotland; Data Bank Manager: Bente Wold, University of Bergen. This publication reports on data solely from Canada. This analysis was funded, in part, by the Queen's University Department of

References (88)

  • S Baker et al.

    The injury fact book

    (1992)
  • P Bijur et al.

    Behavioural predictors of injury in school-age children

    American Journal of Diseases of Children

    (1988)
  • P Bijur et al.

    Persistence of occurrence of injuryCan injuries of preschool children predict injuries of school-aged children?

    Pediatrics

    (1988)
  • P.E Bijur et al.

    Parent-adolescent conflict and adolescent injuries

    Journal of Developmental Behavioral Paediatrics

    (1991)
  • P.E Bijur et al.

    Child behavior and accidental injury in 11,966 pre-school children

    American Journal of Diseases of Children

    (1986)
  • D Black et al.

    Inequalities in healththe black report

    (1982)
  • J Block

    Assimilation, accommodation, and the dynamics of personality development

    Child Development

    (1982)
  • A Caspi et al.

    Individual differences are accentuated during periods of social changeThe sample case of girls at puberty

    Journal of Personality and Social Psychology

    (1991)
  • M.F Cataldo et al.

    Behavior of injured and uninjured children and their parents in a simulated hazardous setting

    Journal of Pediatric Psychology

    (1992)
  • K.K Christoffel et al.

    Psychosocial factors in childhood pedestrian injuryA matched case-control study. Kid's’n’ cars team

    Pediatrics

    (1996)
  • Connop, H.L., King, A.J.C., & Boyce, W. (1999). Youth smoking and the role of the peer group. Working Paper Series No....
  • Currie, C.E. (1998) Health behaviour in school-aged children: research protocol for the 1997-98 survey. Edinburgh,...
  • R Cushman et al.

    Helmet promotion in the emergency room following a bicycle injuryA randomized trial

    Pediatrics

    (1991)
  • L.L Davidson

    Hyperactivity, antisocial behavior, and childhood injuryA critical analysis of the literature

    Journal of Developmental and Behavioral Pediatrics

    (1987)
  • L.L Davidson et al.

    Pre-school behaviour problems and subsequent risk of injury

    Pediatrics

    (1988)
  • L.L Davidson et al.

    Hyperactivity in school-age boys and subsequent risk of injury

    Pediatrics

    (1992)
  • G Dougherty et al.

    Social class and the occurrence of traffic injuries and deaths in urban children

    Canadian Journal of Public Health

    (1990)
  • M.S Durkin et al.

    Low-income neighborhoods and the risk of severe pediatric injuryA small-area analysis in Northern Manhattan

    American Journal of Public Health

    (1994)
  • G.H Elder

    Perspectives on the life course

  • T Faelker et al.

    Socio-economic differences in childhood injuryA population-based epidemiologic study in Ontario, Canada

    Injury Prevention

    (2000)
  • Feldman, S., & Elliott, G. (Eds.) (1990). At the threshold: The developing adolescent. Cambridge, MA: Harvard...
  • A.J Flisher et al.

    Risk-taking behaviour of Cape Peninsula high-school students. Part III. Cigarette smoking

    South African Medical Journal

    (1993)
  • A.J Flisher et al.

    Risk-taking behaviour of Cape Peninsula high-school students. Part IV. Alcohol use

    South African Medical Journal

    (1993)
  • A.J Flisher et al.

    Risk-taking behaviour of Cape Peninsula high-school students. Part V. Drug use

    South African Medical Journal

    (1993)
  • A.J Flisher et al.

    Risk-taking behaviour of Cape Peninsula high-school students Part VI. Road-related behaviour

    South African Medical Journal

    (1993)
  • N.L Galambos et al.

    Multiple-risk behaviour in adolescents and young adults

    Health Reports

    (1998)
  • N Giesbrecht

    Reducing risks associated with drinking among young adultsPromoting knowledge-based perspectives and harm reduction strategies

    Addiction

    (1999)
  • J.A Graber et al.

    Models of developmentUnderstanding risk in adolescence

    Suicide and Life-Threatening Behavior

    (1995)
  • M.J Harris et al.

    Unintentional infant injuriesSociodemographic and psychosocial factors

    Public Health Nursing

    (1994)
  • Health Canada. (1997). For the safety of Canadian children and Youth: From injury data to preventive measures. Minister...
  • R.J Hempel

    Intervention to increase seat belt use at a primary care center

    Journal of the American Board of Family Practice

    (1992)
  • S.M Horwitz et al.

    Determinants of pediatric injury

    American Journal of Diseases of Children

    (1988)
  • C.E Irwin et al.

    Biopsychosocial correlates of risk-taking behaviours during adolescence

    Journal of Adolescent Health Care

    (1986)
  • D.L Jaquess et al.

    Previous injuries and behaviour problems predict childrens injuries

    Journal of Pediatric Psychology

    (1994)
  • Cited by (89)

    • Clusters of Contemporary Risk and Their Relationship to Mental Well-Being Among 15-Year-Old Adolescents Across 37 Countries

      2020, Journal of Adolescent Health
      Citation Excerpt :

      However, given the significant changes in adolescents' lives and the lack of an up-to-date theoretically driven and empirically tested model of how traditional and newer risk behaviors and factors cluster together, the study takes a data-driven approach for identifying how risks cluster. Moreover, considering gender differences across these behaviors (e.g., girls report lower mental well-being and tend to engage in different risk behaviors than boys) [25], the present study also examined whether clusters of risk and their relationship with mental well-being are consistent across genders. We addressed the following research questions:

    • A systematic review of tobacco use among adolescents with physical disabilities

      2016, Public Health
      Citation Excerpt :

      Selection criteria for articles included those that employed a quantitative research methodology with clearly specified definitions of disability and tobacco use. Studies were excluded if they assessed only cognitive, learning or behavioral disabilities, or examined short-term disabilities,19–21 or did not examine tobacco use as the primary outcome,22,23 or did not have a non-disability comparison group.24,25 In addition, only studies that included children and adolescents below the age of 18 years were considered eligible.

    View all citing articles on Scopus
    View full text