Lifecourse socioeconomic predictors of midlife drinking patterns, problems and abstention: Findings from the 1958 British Birth Cohort Study

https://doi.org/10.1016/j.drugalcdep.2008.01.014Get rights and content

Abstract

Background

Research suggests that outcomes associated with drinking may differ depending upon patterns of consumption, drinking related symptoms and social problems. This paper investigated socioeconomic predictors (measuring multiple indices, period and consistency of disadvantage) of midlife drinking patterns.

Methods

Socioeconomic information from the 1958 British Birth Cohort Study (n = 9146) included: manual socioeconomic position and owner/buyer residential tenure (7, 11, 16, 33 and 42 y), and educational attainment (33 y). At 45 y, the overlap between drinking patterns was explored using the Alcohol Use Disorders Identification Test. Patterns included: ‘Moderate-binge’ (binge drinkers with low-problem scores, consuming within UK sensible drinking weekly guidelines); Low-Problem Heavy (LPH) drinkers (regardless of binge); ‘Problem’ (and heavy or binge) and ‘Non-/occasional’ (≤monthly) drinkers. These categories were compared to ‘Low-risk’ drinkers.

Results

Socioeconomic disadvantage was consistently associated with moderate-binge, non-/occasional and problem but not LPH drinking. The highest risk was associated with multiple and persistent disadvantage across childhood and adulthood; this risk was partially accounted for education. Non-/occasional and moderate-binge drinking was predicted by disadvantage during childhood alone. The socioeconomic disadvantage of non-/occasional drinkers was not explained by past problem or heavy drinking.

Conclusions

Socioeconomic disadvantage across the lifecourse was consistently linked to specific drinking patterns. Furthermore, associations linking socioeconomic disadvantage with drinking patterns will typically be underestimated if multiple and persistent disadvantage is not investigated. The role of persistent socioeconomic disadvantage in the poor health of non-drinkers and moderate-binge drinkers needs investigation. The findings support current initiatives targeting the reduction of social and individual costs associated with specific drinking patterns.

Introduction

Poor health behaviours, such as diet and smoking are frequently associated with disadvantaged socioeconomic circumstances (Graham and Power, 2004). However, findings on the socioeconomic predictors of drinking are more mixed. For instance, heavy drinking (Karlamangla et al., 2006), alcohol use disorders and binge drinking (Jefferis et al., 2007, Laatikainen et al., 2003, Naimi et al., 2003) have been associated with lower qualifications, but high average alcohol consumption has also been linked to higher education and income (Moore et al., 2005), and heavy drinking to professional compared to non-professional employment (Jonas et al., 2000). Other studies report no differences in socioeconomic circumstances relating to drinking problems (Marmot, 1997). Recently, Casswell et al. (2003) found that socioeconomic circumstances differentially predicted quantity and frequency measures of drinking. Higher income was associated with more frequent consumption, but lower qualifications were associated with greater quantity. While studies demonstrate inconsistency across alcohol measures, researchers, other than Casswell et al., have not tested whether socioeconomic predictors vary across patterns of alcohol use.

While alcohol research frequently investigates predictors or outcomes of alcohol disorders, heavy drinking, and binge drinking, the relationships linking these drinking patterns and problems remain relatively undocumented in the general population. For instance, Roche (1999, p. 121) noted that ‘most survey instruments are constructed in such a way that infrequent heavy drinking (binge drinking) disappears and becomes indistinguishable from frequent light or moderate drinking, thus masking an important pattern of consumption’. Indeed, Bennett et al. (1991) found that 32% of men and 16% of women who drank within the safe weekly limits reported binge drinking at least once or twice a month. These two studies are rare in that they acknowledge the importance of investigating overlap and distinction between alcohol patterns and problems. To date there has been no research contrasting predictors of different patterns and problems, including binge drinking amongst otherwise moderate drinkers.

Inequalities in health behaviours such as drinking may have origins in early childhood. Indeed, parental occupation in early childhood has been linked with increased risk of substance use disorders (Timms, 1998), alcohol use disorders (Poulton et al., 2002) and higher levels of adolescent alcohol consumption (Droomers et al., 2003). Hemmingsson et al. (1999) investigated alcohol diagnoses and intergenerational social mobility, in terms of social class of origin (father) and destination (own) amongst young male conscripts. Stable manual social class and entry into manual employment predicted alcohol diagnoses. In Scotland, less advantaged childhood social class, own current social class, and residence in a disadvantaged area, predicted heavier drinking when aged 35–64 y (Davey Smith and Hart, 2002). However, in a recent review of longitudinal studies, Wiles et al. (2007) concluded that there was a surprising lack of research, and weak and inconsistent evidence to support an association between childhood socioeconomic status and later alcohol use. Furthermore, most studies only include a single measure of socioeconomic circumstances or use information collected at a single age. Such limited scope and duration of exposure provides a weak proxy for disadvantage throughout childhood (Poulton et al., 2002).

Some longitudinal studies report that education may be an important mediator in the link between childhood socioeconomic status and adult alcohol and other drug problems (Fothergill and Ensminger, 2006) and binge drinking (Lawlor et al., 2005). For instance, Lawlor et al. (2005) noted that factors related to educational achievement most likely explain the association between childhood socioeconomic status and behavioural risk factors in adulthood. However, few longitudinal studies have investigated the role of education in mediating the association linking childhood socioeconomic status with later alcohol use, and researchers have not contrasted findings across different patterns of drinking (Wiles et al., 2007).

Lastly, non-drinkers need to be considered alongside other levels of consumption because they report higher levels of socioeconomic disadvantage relative to moderate drinkers (Jonas et al., 2000, Marmot, 1997). An American study undertaking five sweeps over 20 years, found that people who consistently reported consuming <12 drinks in the previous year had lower levels of education, employment and income than consistent drinkers (Moore et al., 2005). It is possible that the socioeconomic disadvantage of non-drinkers is due to past heavy or problem drinking; but it is also feasible that socioeconomic disadvantage polarises drinking behaviour, causing some people to abstain and others to drink heavily. Studies have not tested these hypotheses using data collected across the lifecourse.

Overall, there is a lack of research quantifying the strength of associations linking multiple indices and periods of disadvantage with different patterns of midlife drinking and abstaining. The main aim of this study was therefore to investigate whether socioeconomic circumstances across childhood and adulthood predicted ‘binge’, ‘heavy’ and ‘problem’ consumption patterns in midlife. We also investigated: (1) the overlap between binge, heavy and problem drinking; (2) whether socioeconomic circumstances relate to specific drinking patterns; (3) the effect of multiple indices and periods of socioeconomic disadvantage on midlife drinking patterns; (4) whether education is a potential mediator of the associations between socioeconomic circumstances and drinking patterns; and (5) whether past drinking is a feasible explanation for non-/occasional drinkers’ socioeconomic disadvantage.

Section snippets

The 1958 British Birth Cohort Study

The 1958 cohort study commenced as a perinatal study and included 98% of all births in England, Scotland and Wales during a week in March, 1958. Individuals have been interviewed when aged 7, 11, 16, 23, 33, 42 and 45 y. Ethical approval for the 45 y survey was given by the South East Multi-Centre Research Ethics Committee; for the 42 y survey, approval was obtained from the North Thames Multi-Centre Research Ethics Committee. For the 33 y survey, ethical committee approval was not sought, and for

Results

Table 1 describes socioeconomic status and alcohol consumption of the sample (n = 9146). In early childhood most participants had a manual SEP and lived in non-owner/buyer accommodation, but by 33 y most were owner/buyers. Nearly 30% of the sample had manual SEP or non-owner/buyer tenure during childhood only and 12% had consistently reported (9–10 times) manual SEP and non-owner/buyer residential tenure across both adulthood and childhood.

Table 1 also shows the proportion of the sample who

Discussion

In this study, lower levels of education, manual SEP and non-owner/buyer residential tenure were consistently associated with increased risk of midlife binge and problem drinking, but not heavy consumption per se. More specific drinking patterns were explored and non-/occasional, moderate-binge and problem drinkers were consistently found to have lower levels of education, and occupational and residential disadvantage across the lifecourse relative to low-risk drinkers. Risk ratios for

Conflicts of interest

All the authors declare that they have no conflicts of interest.

Acknowledgements

We thank the data providers: Centre for Longitudinal Studies, Institute of Education and National Birthday Trust Fund, National Children's Bureau, City University Social Statistics Research Unit (original data producers).

Role of funding source: Research at the Institute of Child Health and Great Ormond Street Hospital for Data collection at age 45 y and statistical analysis were funded by the United Kingdom Medical Research Council, grant G0000934, awarded under the Health of the Public

References (42)

  • G. Corrao et al.

    Alcohol and coronary heart disease: a meta analysis

    Addiction

    (2000)
  • G.D. Davey Smith et al.

    Life-course socioeconomic and behavioral influences on cardiovascular disease mortality: the collaborative study

    Am. J. Public Health

    (2002)
  • Department of Health, 1995. Sensible drinking: the report of an interdepartmental Working Group. Department of...
  • R. Doll et al.

    Mortality in relation to alcohol consumption: a prospective study among male British doctors

    Int. J. Epidemiol.

    (2005)
  • M. Droomers et al.

    Occupational level of the father and alcohol consumption during adolescence; patterns and predictors

    J. Epidemiol. Community Health

    (2003)
  • K.M. Fillmore et al.

    Moderate alcohol use and reduced mortality risk: systemic error in prospective studies

    Addict. Res. Theory

    (2006)
  • B. Galobardes et al.

    Indicators of socioeconomic position (part 1)

    J. Epidemiol. Community Health

    (2006)
  • B. Galobardes et al.

    Indicators of socioeconomic position (part 2)

    J. Epidemiol. Community Health

    (2006)
  • J. Gill et al.

    Practical demonstration of personal daily consumption limits: a useful intervention tool to promote responsible drinking among UK adults?

    Alcohol Alcohol.

    (2007)
  • H. Graham et al.

    Childhood disadvantage and health inequalities: a framework for policy based on lifecourse research

    Child Care Health Dev.

    (2004)
  • B.J.M.H. Jefferis et al.

    Social gradients in binge drinking and abstaining: trends in a cohort of British adults

    J. Epidemiol. Community Health

    (2007)
  • Cited by (43)

    • Alcohol drinking in one's thirties and forties is associated with body mass index in men, but not in women: A longitudinal analysis of the 1970 British Cohort Study

      2021, Preventive Medicine
      Citation Excerpt :

      Alcohol drinking (Britton et al., 2015) and BMI (Johnson et al., 2015) may vary with time. Potential confounders may also vary with time, including smoking habit (Daly et al., 2016), socioeconomic status (Caldwell et al., 2008), longstanding illness (Ng Fat et al., 2015), and leisure time physical activity (Dregan and Gulliford, 2013). Therefore, a longitudinal analysis with repeated measurements would provide a better understanding of the independent associations of alcohol drinking with obesity in the UK (Britton et al., 2015).

    • Disparities in drinking patterns and risks among ethnic majority and minority groups in China: The roles of acculturation, religion, family and friends

      2016, Drug and Alcohol Dependence
      Citation Excerpt :

      People who have more drinking friends tend to drink alcohol and have risky drinking patterns (Donovan, 2004; McKay, 2015; Newman et al., 2004; Ryan et al., 2010; Simons-Morton, 2004). Except for the above mentioned factors, female gender (Holmila and Raitasalo, 2005; Ma et al., 2006; Wilsnack et al., 2009), age (Kim et al., 2013; Wilsnack et al., 2006; Wolff et al., 2014), marital status (Bogart et al., 2005; Kearns-Bodkin and Leonard, 2005; O’Malley, 2004), education (Caldwell et al., 2008; Casswell et al., 2003; Cutler and Lleras-Muney, 2006; Huerta and Borgonovi, 2010) and people’s perceptions of the risk of alcohol consumption on the body (Bühler et al., 2015; Grevenstein et al., 2015; Henry et al., 2005) are related to the level of risky drinking. Our study hypothesis is that aboriginal culture enculturation is associated with a reduced risk of drinking and mainstream culture acculturation is associated with an increased risk of drinking.

    View all citing articles on Scopus
    View full text