A recent study published in Addiction investigates the long-term trends in alcohol abstention and consumption in England from 2001 to 2019. Despite a decrease in alcohol consumption over the decades, alcohol-specific death rates have remained relatively stable. This study utilises Age–Period–Cohort (APC) models to understand these paradoxical trends.

Study Design and Participants

The research is based on repeat cross-sectional survey data from the Health Survey for England, conducted between 2001 and 2019. The study includes residents of England aged 13 years and older.

Measurements and Variables

Outcome Variables

• Alcohol Abstention: Whether participants abstained from alcohol.
• Alcohol Consumption: The quantity of alcohol consumed, measured in units.

Age Groups

Participants were divided into nine age groups:

• 13–15
• 16–17
• 18–24
• 25–34
• 35–44
• 45–54 (reference group)
• 55–64
• 65–74
• 75+

Time Periods

The study covered four time periods:

• 2001–04
• 2005–09 (reference period)
• 2010–14
• 2015–19

Birth Cohorts

Participants were also categorised into 18 five-year birth cohorts ranging from 1915–19 to 2000–04, with the reference cohort being 1960–64.

Proxy Variables

To better understand the APC effects, the study used several proxy variables:

• Age Effects: Systolic and diastolic blood pressure.
• Period Effects: Alcohol affordability, internet usage, and household alcohol expenditure.
• Birth Cohort Effects: Prevalence of smoking and overweight.

Key Findings

Alcohol Abstention

• Young Ages: The odds of abstaining were significantly higher at younger ages, particularly 13–15 years (OR = 5.38; 95% CI = 4.50–6.43).
• Time Periods: The lowest odds were observed during the first period (2001–04) with an OR of 0.83 (95% CI = 0.79–0.86).
• Birth Cohorts: A U-shaped pattern was observed across different birth cohorts.

Alcohol Consumption

• Age Groups: The consumption rate increased until ages 18–24 (IRR = 1.41, 95% CI = 1.34–1.48) and decreased thereafter.
• Time Periods: The highest incidence rate ratio was during 2001–04 (IRR = 1.07; 95% CI = 1.05–1.08).
• Birth Cohorts: An inverted J-shape pattern was observed by birth cohort.

Proxy Variable Analysis

• Blood Pressure Measures: Using systolic and diastolic blood pressure as proxies resulted in APC effects that differed from the base-case model.
• Similar to Base-Case Model: Internet usage, household expenditure on alcohol, and smoking prevalence produced APC effects similar to the base-case model.
• Differing from Base-Case Model: Alcohol affordability and the prevalence of overweight showed different APC effects compared to the base-case model.

The study concludes that the discrepancy between decreasing alcohol consumption and stable alcohol-related deaths in England from 2001 to 2019 may be partly explained by the halt in abstention trends since 2010 and a slight decline in consumption since 2001.

Source: Wiley Online Library