Alcohol consumption is a modifiable risk factor for several cancers. Previous studies have also shown a positive association between alcohol and cancer progression, recurrence and cancer-related death. Few studies have been conducted to characterize alcohol consumption patterns among cancer survivors compared to cancer-free individuals.

This cross-sectional study combined data from NHANES between 1999-2016 to compare alcohol consumption behaviors among adults with a history of cancer to cancer-free population controls. Multinomial logistic regression was used to calculate the adjusted odds ratios (aOR) and 95% confidence intervals (CI) of drinking status among survivors (former drinkers; current drinkers; never drinkers) compared to controls. Multivariable logistic regression models were also used within current drinkers to calculate the adjusted odds of binge drinking and exceeding moderate drinking. Models were adjusted for demographic and socioeconomic factors.

A total of 3,113 survivors and 39,527 controls were included. Compared with controls, survivors were less likely to be current drinkers (63.4% vs. 72.6% in controls) and were more likely to be former drinkers (24.4% vs. 15.5% in controls). The odds of being current drinkers vs. never drinkers was significantly lower among survivors compared with controls (aOR: 0.84, 95% CI: 0.71-0.99, p-value: 0.04), while no significant odds of being former drinkers vs. never drinkers was observed. When comparing survivors to controls by cancer type, breast cancer survivors had lower odds of being current drinkers vs. never drinkers (aOR: 0.67, 95% CI: 0.51-0.88, p-value<0.01), and they also had significant lower odds of binge drinking when compared with controls (aOR: 0.43, 95% CI: 0.20-0.95, p-value: 0.04). Prostate cancer survivors had higher odds of being current drinker vs. never drinker compared with controls; however, the result was not significant (aOR: 1.33, 95% CI: 0.92-2.11, p-value: 0.18) and the multivariable logistic regression models showed that current drinkers had lower odds of being exceeding moderate drinkers (aOR:0.71, 95% CI: 0.50-1.00, p-value: 0.05) and being binge drinkers (aOR: 0.35, 95% CI: 0.16-0.74, p-value<0.01) compared with controls. We also observed significant higher odds of being binge drinkers among cervical cancer survivors (aOR:2.51, 95% CI: 1.27-4.92, p-value<0.01). No significant association between cancer status and drinking behaviors were observed by the number of cancers diagnosed and the years since diagnosis.

Our findings suggest that cancer survivors are at lower odds of being current drinkers when compared with cancer-free population controls. Given the high odds of binge drinking among cervical cancer survivors, public health strategies are needed to reduce alcohol consumption in this group.

Citation Format: Junrui Lyu, Maneet Kaur, Avonne E. Connor. A national study of alcohol consumption patterns among population-based U.S. cancer survivors compared with cancer-free controls [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr LB089.