At-a-glance – Sex-based differences in heavy episodic drinking among adults in Canada, 2017–2023: a repeated cross-sectional analysis

Introduction

Regularly consuming a high volume of alcohol in a single occasion is associated with an increased risk of dependence, violence, and premature mortality and disability from both acute and chronic conditions.^{Footnote 1Footnote 2} In Canada, heavy episodic drinking (HED) is prevalent, with 20% of individuals 12 years and older reporting this behaviour in 2022.^{Footnote 3} Historically, males have reported higher prevalence of HED across all age groups in Canada; however, emerging evidence suggests a narrowing sex gap.^{Footnote 3Footnote 4} The COVID-19 pandemic, commencing in March 2020, and ensuing lockdown restrictions introduced significant changes in lifestyle, stress, and alcohol consumption patterns in some populations.^{Footnote 5Footnote 6} Examining pre-pandemic and early post-pandemic periods can provide valuable insights into shifts in alcohol consumption behaviours. A recent United States study found that from 2017 to 2019 and 2021 to 2023, sex-based differences in HED narrowed in all age groups, with HED prevalence among young adult females surpassing males.^{Footnote 7} The current study examined recent trends in sex-based differences in HED by age group in Canada and hypothesized that the male-female gap has narrowed among young and middle-aged adults.

Methods

We performed a repeated cross-sectional analysis among individuals 18 years and older in the annual Canadian Community Health Surveys (CCHS) from 2017 to 2023.

The CCHS is a population-representative survey conducted by Statistics Canada that targets individuals 12 years and older, or 18 years and older as of 2023, living in private dwellings, representing more than 96% of the population in Canada.^{Footnote 8} The CCHS does not survey Indigenous Peoples living on reserves, members of the Canadian Armed Forces, people living in institutions, children in foster care, and people living in northern Quebec and the Canadian territories. We excluded individuals younger than 18 years, respondents who were pregnant or breastfeeding, and those who did not provide responses to questions on alcohol consumption.

Consistent with Canadian public health surveillance, including the CCHS, HED was defined as binge drinking at least once per month in the past 12 months.^{Footnote 3} Binge drinking was defined as five or more standard drinks for males and four or more for females on the same occasion.^{Footnote 9} HED was derived from a question on the frequency of binge drinking in the past 12 months and was categorized as HED (once a month, 2 to 3 times a month, once a week, more than once a week) or no HED (less than once a month or never). Before 2019, sex was measured based on respondents' self-report as being "male or female," representing sex-at-birth. Starting in 2019, CCHS introduced questions allowing for the separation of sex (assigned at birth) and gender (man, woman, or non-binary). Consistent with CCHS guidance for analyses spanning before and after 2019, we used the sex variable. Age group was defined as ages 18 to 25, 26 to 49, 50 to 64, and 65 years and older.

We used logistic regression models to examine associations between HED and sex in pooled 2017 to 2019 (pre-COVID-19) and 2022 to 2023 (post-COVID-19) CCHS data across age groups. Three-way interactions across sex, age group, and time period (pre- or post-COVID-19) were modelled. We estimated predicted prevalence and prevalence differences using marginal effects. Data from 2020 and 2021 were included in the results describing the prevalence of HED over time yet excluded from model-based analyses due to lockdowns and disruptions to CCHS data collection during the COVID-19 pandemic. CCHS survey weights (divided by the corresponding number of years in the time period) were used for all models, and 95% confidence intervals (CI) were estimated using bootstrap survey weights with 1000 replicate weights. All analyses were conducted using R Version 4.4.2 (R Foundation for Statistical Computing, Vienna, AT).

In sensitivity analyses, we additionally adjusted for education (less than secondary school graduation, secondary school graduation, postsecondary certificate/diploma/university degree; at the household level for individuals younger than 30 years, otherwise individual-level), rurality (population centre vs. rural area), and immigration status (citizen by birth vs. landed/non-permanent resident).

This study did not require a research ethics board review as it relied exclusively on secondary use of anonymous information as per Article 2.4 of the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans.

Results

The unweighted sample included 150 630 respondents in 2017 to 2019 and 87 227 in 2022 to 2023. The overall prevalence of HED in Canada was 20.5% (95% CI: 20.1, 20.8) in 2017 to 2019 and 20.3% (95% CI: 19.9, 20.8) in 2022 to 2023. As shown in Figure 1, HED prevalence remained relatively stable across age groups for both sexes in 2017 to 2019. However, declines were observed in 2020 to 2021 among males and females aged 18 to 25 and 26 to 49 years. Rates rebounded in 2022 to levels comparable to 2017 to 2019, before softening again among males. Among older age groups, HED prevalence was stable, with a slight increase noted in 2022. Across the study period, the most pronounced decline occurred among males aged 18 to 25 years, with prevalence dropping from 33.7% (95% CI: 30.6, 36.7) in 2017 to 22.0% (95% CI: 18.5, 25.4) in 2023. In contrast, females in the same age group experienced a slight decrease in HED prevalence from 28.7% (95% CI: 25.9, 31.6) in 2017 to 27.8% (95% CI: 23.2, 32.4) in 2023.

Figure 1 - Text description

Group	2017 Prevalence (95% CI)	2018 Prevalence (95% CI)	2019 Prevalence (95% CI)	2020 Prevalence (95% CI)	2021 Prevalence (95% CI)	2022 Prevalence (95% CI)	2023 Prevalence (95% CI)
18–25
Male	33.7 (30.6, 36.7)	31.3 (28.0, 34.7)	30.4 (27.2, 33.5)	23.6 (19.8, 27.5)	20.5 (17.0, 24.0)	26.5 (23.1, 29.9)	22.0 (18.5, 25.4)
Female	28.7 (25.7, 31.7)	25.3 (22.2, 28.5)	25.4 (22.3, 28.5)	21.8 (17.6, 25.9)	19.2 (15.8, 22.6)	28.0 (24.2, 31.8)	27.8 (22.9, 32.7)
26–49
Male	30.8 (29.3, 32.3)	30.9 (29.1, 32.7)	30.0 (28.4, 31.7)	25.4 (23.4, 27.4)	24.1 (22.4, 25.9)	30.5 (28.8, 32.2)	26.5 (24.8, 28.2)
Female	21.6 (20.2, 22.9)	20.7 (19.3, 22.1)	18.2 (16.9, 19.4)	17.8 (16.1, 19.4)	16.7 (15.3, 18.0)	22.3 (20.9, 23.7)	20.4 (18.9, 21.9)
50–64
Male	23.5 (21.8, 25.1)	23.8 (22.1, 25.5)	23.9 (22.2, 25.7)	22.5 (20.2, 24.7)	21.2 (19.2, 23.1)	24.6 (22.9, 26.2)	22.5 (20.7, 24.2)
Female	14.6 (13.2, 16.0)	14.4 (13.0, 15.8)	14.8 (13.4, 16.2)	14.3 (12.7, 16.0)	14.3 (12.5, 16.0)	17.2 (15.7, 18.8)	16.7 (15.2, 18.3)
65+
Male	10.9 (9.9, 12.0)	10.7 (9.6, 11.7)	10.6 (9.7, 11.5)	10.3 (9.2, 11.4)	10.7 (9.7, 11.8)	12.9 (11.9, 13.8)	12.5 (11.4, 13.5)
Female	5.0 (4.3, 5.7)	4.7 (4.0, 5.5)	5.0 (4.4, 5.6)	4.8 (4.2, 5.5)	5.7 (4.9, 6.5)	7.5 (6.8, 8.3)	7.6 (6.9, 8.3)

Abbreviation: CCHS, Canadian Community Health Survey.

Table 1 presents model-based prevalence estimates in HED. In 2017 to 2019, males had a higher HED prevalence than females overall (24.2% vs. 16.5%) with a sex-based prevalence difference of 7.7% (95% CI: 5.6, 9.7) (Table 1). Males had higher HED prevalence across all age groups including among adults aged 18 to 25 years (prevalence: 31.8% vs. 26.5%; prevalence difference: 5.3%, 95% CI:1.6, 8.9).

In 2022 to 2023, males continued to have a higher HED prevalence than females overall (22.2% vs. 18.4%; prevalence difference: 3.8%, 95% CI: 0.9, 6.6); however, sex differences in HED appeared smaller across all age groups compared to 2017 to 2019. Notably, among those aged 18 to 25 years, males had a lower HED prevalence than females (prevalence: 24.2% vs. 27.9%) with a prevalence difference of −3.7% (95% CI: −9.4, 1.9; Table 1), although the confidence interval includes zero, indicating uncertainty about whether a meaningful difference exists.

Similar results were observed in sensitivity analyses in models adjusting for education, rurality, and immigration status (Appendix 1—available upon request from corresponding author).

Discussion

HED is a key indicator of harmful alcohol consumption and an important marker for alcohol-related health risks. This analysis identified a narrowing of sex-based differences in the prevalence of HED in Canada across all age groups, including a convergence and possible reversal among young adults. In 2022 to 2023, females aged 18 to 25 years exhibited a higher observed prevalence of HED than males; however, the confidence interval for this difference included zero, leaving the true magnitude and direction of the sex-based difference uncertain. Further research is needed to clarify whether HED among young adult females is indeed surpassing that of males in Canada.

In young adults, the observed shifts from 2017 to 2023 were driven by declines in HED among both sexes from 2017 through 2021. This was followed by a rebound in 2022 to levels comparable to those observed in 2017 to 2019, with a subsequent softening in 2023 among males relative to females. These findings align with recent data from the United States and Canada which also found a convergence of sex-based differences in HED among young adults driven by both declining rates in HED post-pandemic among young adult males and increasing rates among females.^{Footnote 4Footnote 7Footnote 10} Similar sex-based trends in HED have been observed internationally, including recent data from Europe showing that adolescent girls are now equalling or surpassing boys in alcohol use, and OECD reports indicating rising rates of high-volume use among young, highly educated women in high-income countries.^{Footnote 11Footnote 12} These findings point to a global shift in alcohol consumption patterns, characterized by diminishing sex-based differences and emerging evidence of a potential sex-based reversal in HED among young people.

The high sustained prevalence of HED in young adult females in this study reflects longer‑term secular shifts in alcohol use and shorter‑term pandemic‑related disruptions. Over the past 15 years, increases in alcohol consumption and HED in Canada and internationally have been largely attributable to rising rates among women. ^{Footnote 4Footnote 13Footnote 14Footnote 15} Several factors likely contribute to this pattern, including intensified alcohol industry marketing directed toward women—particularly for ready‑to‑drink beverages, flavoured coolers, and certain wines—which frequently emphasize social identity, lifestyle appeal, and the normalization of heavier drinking.^{Footnote 16Footnote 17} In parallel, traditional social sanctions discouraging female HED have weakened.^{Footnote 18} Emerging evidence also suggests that, relative to men, women may be more likely to report alcohol use during and after the pandemic as a strategy to cope with stress, anxiety, depression, or experiences of trauma.^{Footnote 17Footnote 19Footnote 20} These social and contextual factors may help explain the observed shifts in HED among young adult females in Canada.

In contrast, HED remains prevalent among young adult males in Canada, yet overall patterns during the study period indicate a decline. The factors contributing to this decline are not fully understood, but several plausible mechanisms warrant consideration. One possibility is substitution toward cannabis, as population‑level cannabis use in Canada has increased since legalization and remains more prevalent and frequent among young adult males than females.^{Footnote 21Footnote 22} In addition, alcohol consumption is highly sensitive to affordability and disposable income. Broader affordability pressures may therefore be constraining discretionary spending among both sexes, particularly among young adult males, who are more likely to drink in social settings (e.g. bars, restaurants).^{Footnote 23} Taken together, these findings suggest that the observed convergence in HED reflects both modest increases in young adult females alongside declining trends among young adult males, shaped by evolving gender norms, market forces, substitution effects, and economic conditions. Qualitative research could be valuable in understanding sex- and gender-related differences and the intersections of these identities in HED patterns, views and attitudes toward alcohol.

Although the Guidance on Alcohol and Health adopts a universal risk-based approach, the 2023 report highlights the importance of sex-based differences in alcohol consumption as alcohol affects female and male bodies differently.^{Footnote 24} Females are biologically more susceptible to alcohol-related health harms—both acute and chronic—at lower levels of consumption and more rapidly than males.^{Footnote 25Footnote 26} In Canada, rates of alcohol-associated liver disease and alcohol-attributable hospitalizations have increased more sharply among young adult females than males in recent years.^{Footnote 27Footnote 28Footnote 29} HED is also a well-established risk factor for female breast cancer, even after accounting for cumulative alcohol intake, with exposures between menarche and first full-term pregnancy more strongly linked to premenopausal breast cancer.^{Footnote 30} These trends in HED underscore the urgent need for comprehensive sex-specific and gender-responsive strategies to prevent and manage alcohol use. Population-level strategies that address alcohol price, availability, and marketing are critical for reducing per capita alcohol consumption and HED and minimizing alcohol-related harms.^{Footnote 31} Integrating routine alcohol screening and brief interventions—including personalized feedback and information on alcohol and health—into family, community, and postsecondary campus health settings offers a feasible and impactful approach for identifying and reducing consumption, particularly for young adult females who may present for other health concerns.^{Footnote 32Footnote 33} Mandatory health warning labels on alcohol containers have also demonstrated effectiveness in increasing consumer knowledge of alcohol-caused health risks, reducing alcohol consumption, and countering industry marketing efforts.^{Footnote 34Footnote 35Footnote 36} Finally, strengthening surveillance of alcohol consumption across demographic groups will further enable the development of targeted, equitable public health policies and practices.

Strengths and limitations

Our study has limitations. First, the reliance on self-reported HED introduces the potential for recall and social desirability bias.^{Footnote 37Footnote 38} However, this bias is expected to be relatively consistent across survey cycles and unlikely to affect comparisons across time. Second, nonresponse bias (which may be larger among individuals with HED) and selection bias (due to the exclusion of high-risk populations such as those institutionalized or homeless) are possible in the CCHS.^{Footnote 39} While we used survey weights to address these biases, our results may underestimate HED and are only generalizable to the CCHS target population. Third, prior to 2019, the CCHS did not distinguish between sex and gender, as respondents were only asked to self-identify as male or female. This single item likely captures a mix of biological and sociocultural factors.^{Footnote 40}

Conclusion

National data indicate that the prevalence of HED among adults in Canada remained stable at approximately 20% from 2017 to 2023. Adult males consistently reported higher HED prevalence than females. However, sex-based differences narrowed in 2022 to 2023 compared with 2017 to 2019 across age groups, including evidence of convergence—and possible reversal—in young adults. The narrowing sex-based differences are driven both by increasing rates in HED among adult females, alongside stable or declining rates in males overall and in most age groups. Future research should further examine HED using both sex and gender variables to better capture heterogeneity in alcohol consumption patterns, experiences, and associated needs and to inform more targeted and equitable public health policies and practices.

Conflicts of interest

None.

Funding

The study is funded by the Canadian Cancer Society (grant #707616) and the Canadian Institutes of Health Research.

Authors' contributions and statement

EH conceptualized the study, organized resources, supervised trainees and staff, and wrote the original draft. LB led the formal analysis, contributed to writing the original draft, and reviewed and edited the manuscript. NS contributed to the analysis, contributed to writing the original draft, and reviewed and edited the manuscript. SHF, SF, and BS contributed to the conceptualization and analysis, and reviewed and edited the manuscript.

The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.

Previous | Table of Contents