INTRODUCTION

Alcohol use is associated with an increased risk of several cancers, and for some cancers, risk increases even at one drink per day.1 Nevertheless, only 38% of US adults are aware of this association.2 Patients consider clinicians to be the most trustworthy source of health information3; the importance of incorporating more discussion about alcohol and cancer risk into patient visits has been recently emphasized.4 Yet, we know very little about the frequency of such discussions or how they are related to awareness among cancer survivors or in the general population.

The objectives of this study were to assess how often US adults report discussions with their clinicians about the risks of alcohol and to examine the association between such discussions with awareness of alcohol as a risk factor for cancer. We also considered the moderating role of cancer history.

METHODS

Data originated from the Health Information National Trends Survey (HINTS) 5 Cycle 3, a nationally representative survey administered by the National Cancer Institute in 2019. HINTS 5 Cycle 3 included two modes of survey response (mail and online); both modes of response were included. Of the 5438 respondents, n = 889 were excluded due to not having seen a clinician in the past year (n = 477) or not having responded to the awareness item (n = 412), leaving n = 4549 for the analytic sample.

The exposure of interest was clinician discussions of the harms of alcohol (“any” or “none”). The outcome was endorsement of the association between alcohol and cancer (“yes,” aware of the association, “do not know,” or “no”). Cancer-related information seeking, beliefs about cancer and health, demographic characteristics, history of chronic diseases associated with alcohol use (cancer, diabetes, heart disease, and high blood pressure),2 and current alcohol use were included as covariates.

A weighted multinomial multivariable logistic regression model estimated the association between any clinician discussion and reporting “yes” and “do not know” (both compared to “no”) for awareness, adjusting for covariates. To explore effect modification by history of cancer, a second weighted multinomial multivariable logistic regression model was fit including an interaction term between clinician discussion and cancer history.

RESULTS

Fewer than half (44.1%) of US adults reported a clinician discussing the harms of alcohol in the last year (Table 1). A minority of respondents endorsed an association between drinking too much alcohol and cancer (34.9%), consistent with past research. Over a third (39.0%) reported “do not know” and over one quarter (26.2%) reported “no.” The prevalence of clinician discussion among those who reported “yes,” “do not know,” and “no” was 52.0%, 37.1%, and 44.1%, respectively (p < 0.001). In the multivariable model (Table 2), clinician discussion was positively associated with awareness (OR = 1.66, 95% CI: 1.25, 2.20). Cancer history did not moderate this association (interaction term p-values = 0.46 and 0.80 for do not know and yes outcomes, respectively).

Table. 1 Descriptive frequencies HINTS 5 Cycle 3
Full size table
Table. 2 Weighted Multinomial Multivariable Logistic Regression Model Examining the Association Between Clinician Discussion of the Harms of Alcohol and Awareness
Full size table

DISCUSSION

This is the first known evaluation of how clinician discussion regarding the harms of alcohol is associated with awareness of alcohol as a risk factor for cancer and our study uses data from a nationally representative sample. Less than half of adults seeing a clinician in the last year reported discussing the harms of alcohol. Clinician discussion was associated with higher awareness of the carcinogenic nature of alcohol, independent of other known correlates of awareness and irrespective of cancer status. These results support the need to increase counseling about alcohol as a cancer risk factor in the general population, where awareness is consistently suboptimal.

Brief alcohol interventions in primary care have been shown to reduce unhealthy alcohol use; however, these interventions are only recommended by the US Preventive Services Task Force for heavy or problematic alcohol users.5 Thus, discussions regarding potential harms of alcohol from even low levels of consumption are likely underutilized in clinical care. Given the critical role of primary care in promoting cancer prevention activities, clinical interventions and guidance for promoting these potentially difficult conversations are needed. Clinicians are the most trusted source of health information.3 Therefore, promoting discussions about the potential harms of alcohol from a clinician may be an effective way to increase awareness and reduce unhealthy alcohol use in the population, particularly for adults who may believe that cancer risk is immutable.