Adolescent smoking and tertiary education: opposing pathways linking socio-economic background to alcohol consumption.

Michael J Green,Alastair H Leyland,Michaela Benzeval,Helen Sweeting

doi:10.1111/add.13365

Abstract

Background and AimsIf socio‐economic disadvantage is associated with more adolescent smoking, but less participation in tertiary education, and smoking and tertiary education are both associated with heavier drinking, these may represent opposing pathways to heavy drinking. This paper examines contextual variation in the magnitude and direction of these associations.DesignComparing cohort studies.SettingUnited Kingdom.ParticipantsParticipants were from the 1958 National Child Development Study (NCDS58; n = 15 672), the British birth cohort study (BCS70; n = 12 735) and the West of Scotland Twenty‐07 1970s cohort (T07; n = 1515).MeasurementsParticipants self‐reported daily smoking and weekly drinking in adolescence (age 16 years) and heavy drinking (> 14/21 units in past week) in early adulthood (ages 22–26 years). Parental occupational class (manual versus non‐manual) indicated socio‐economic background. Education beyond age 18 was coded as tertiary. Models were adjusted for parental smoking and drinking, family structure and adolescent psychiatric distress.FindingsRespondents from a manual class were more likely to smoke and less likely to enter tertiary education (e.g. in NCDS58, probit coefficients were 0.201 and –0.765, respectively; P < 0.001 for both) than respondents from a non‐manual class. Adolescent smokers were more likely to drink weekly in adolescence (0.346; P < 0.001) and more likely to drink heavily in early adulthood (0.178; P < 0.001) than adolescent non‐smokers. Respondents who participated in tertiary education were more likely to drink heavily in early adulthood (0.110 for males, 0.182 for females; P < 0.001 for both) than respondents with no tertiary education. With some variation in magnitude, these associations were consistent across all three cohorts.ConclusionsIn Britain, young adults are more likely to drink heavily both if they smoke and participate in tertiary education (college and university) despite socio‐economic background being associated in opposite directions with these risk factors.

Highlights

Socio-economic inequalities in excessive alcohol consumption are inconsistent in both adolescence [1,2,3] and early adulthood [3,4,5,6], which are key developmental periods for drinking [7,8,9]
It has been suggested that these inconsistent findings result from pathways associated with socio-economic position (SEP) working in opposing directions [4]; while some pathways leading to increased drinking are more common among more disadvantaged adolescents, others may be more common among more advantaged adolescents
For parental drinking: in NCDS58 interviewers at age 7 indicated whether the family suffered from problems with alcoholism; in BSC70 parent or child reports of a parent drinking ‘three or four times a week’ or more or on ‘most days’ were coded as heavy parental drinking; and for Twenty-07 1970s cohort (T07) either parent consuming more than 14 units and 21 units was coded as heavy parental drinking

Summary

Introduction

Socio-economic inequalities in excessive alcohol consumption are inconsistent in both adolescence [1,2,3] and early adulthood [3,4,5,6], which are key developmental periods for drinking [7,8,9]. It has been suggested that these inconsistent findings result from pathways associated with socio-economic position (SEP) working in opposing directions [4]; while some pathways leading to increased drinking are more common among more disadvantaged adolescents, others may be more common among more advantaged adolescents. This paper explores two probable opposing pathways between parental socio-economic position and drinking in adolescence and early adulthood—smoking and tertiary education. If socio-economic disadvantage is associated with more adolescent smoking, but less participation in tertiary education, and smoking and tertiary education are both associated with heavier drinking, these may represent opposing pathways to heavy drinking. Measurements Participants self-reported daily smoking and weekly drinking in adolescence (age 16 years) and heavy drinking (> 14/21 units in past week) in early adulthood (ages 22–26 years). For NCDS58, psychiatric distress was indicated by scores of 2 or more on the neuroticism component of the Rutter behavioural scale [31]

Objectives

Methods

Results

Discussion

Conclusion