What Influences 11-year-olds to Drink? Findings From the Millennium Cohort Study

Yvonne Kelly; Alice Goisis; Amanda Sacker; Noriko Cable; Richard G. Watt; Annie Britton

Disclosures

BMC Public Health. 2016;16(169) 

In This Article

Results

Who Drinks by Age 11?

Overall 13.6 % of cohort members reported having drunk more than a few sips of an alcoholic drink. Cohort members who reported drinking were more likely to be boys (15.7 % vs 11.3 %, p < 0.001), to have started puberty (14.3 % vs 13.2 %), to be a second or later born child (14.0 % vs 12.9 %), to have socioemotional difficulties (18.7 % vs 12.8 %, p < 0.001), to report antisocial behaviours (none = 10.1 %, 1 = 20.7 %, 2 or more = 42.0 %, p < 0.001), report truancy (24.8 % vs 13.2 %, p < 0.001), smoke cigarettes (50.9 % vs 12.4 %, p < 0.001), to be from poorer families (15.4 % in the poorest quintile vs 11.5 % in richest quintile, p < 0.01) and not have any religious affiliation (15.7 % vs 11.6 %, p < 0.001). Table 1 shows the distribution of covariates by cohort member drinking.

Does Parental or Friends' Drinking Matter?

Cohort members whose mothers drank were more likely to drink and these estimates changed little on adjustment for covariates (fully adjusted OR – light/moderate = 1.6, 1.3 to 2.0, heavy/binge = 1.8, 1.4 to 2.3 compared to those with non-drinking mothers). Cohort members for whom data on mother's drinking was missing were also more likely to drink (fully adjusted OR = 2.0, 1.2 to 3.4). Cohort members whose fathers drank were also more likely to drink but these estimates lost statistical significance when covariates were taken into account (fully adjusted OR – light/moderate = 1.3, 0.9 to 1.9, heavy/binge = 1.3, 0.9 to 1.9). Having friends who drank was associated with more 7 times the odds of cohort member drinking, and twice the odds when cohort members reported not knowing whether their friends drank. These estimates changed on adjustment for covariates but remained highly statistically significant (fully adjusted ORs 4.8, 3.9 to 5.9 and 1.8, 1.4 to 2.2 respectively) (Table 2).

What is the Role of Perceptions of Harm, Expectancies Towards Alcohol, Parental Supervision, and Family Relationships?

Perceptions of harm, expectancies towards alcohol, parental supervision, and family relationships were associated with the likelihood of cohort member drinking in the expected direction (Appendix Table 3). Associated with the reduced likelihood of cohort member drinking were: heightened perception of harm from drinking 1–2 drinks daily (OR - some risk = 0.9, 0.7 to 1.1, great risk = 0.6, 0.5 to 0.7); and negative expectancies towards alcohol (OR = 0.5, 0.4 to 0.7). Associated with an increased risk of cohort member drinking were: positive expectancies towards alcohol (OR = 1.9, 1.4 to 2.5); not being supervised by parents on weekends and weekdays (for often OR = 1.2, 1.0 to 1.4); frequent battles of will (OR = 1.3, 1.1 to 1.5); and not being happy with family (OR = 1.2, 1.0 to 1.5).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....