Worldwide, alcohol use disorder (AUD) is the most prevalent substance use disorder outside of tobacco use (Collaborators, 2021), and 4.2% of the total global burden of disease has been attributed to alcohol use (Alcohol and Drug Use, 2018). While 1.3% of people globally have AUD (Alcohol and Drug Use, 2018), far more drink more than the recommended amounts (e.g., per day: one drink or less for women and two drinks or less for men) (Hallgren et al., 2022a; Grant et al., 2017; O'Connor et al., 2018; The basics, 2022; Drinking levels defined, 2022). Unhealthy alcohol use—reflecting a spectrum of drinking above the recommended amounts to AUD (Saitz, 2005)—can lead to numerous negative health outcomes (Briasoulis et al., 2012, Conigrave et al., 1995, Klatsky et al., 1989, McHugh and Weiss, 2019, Rubinsky et al., 2013a, Williams et al., 2018). Multiple studies have shown that people with unhealthy alcohol use are more likely to be hospitalized than those who drink less (Chew et al., 2011, Klatsky et al., 1989, Williams et al., 2012). These hospitalizations can indicate acute, severe illness for patients and lead to higher costs for patients, healthcare systems, and payers (Briasoulis et al., 2012, Clark et al., 2016, Harris et al., 2010).

The increased rate of hospitalizations and medical comorbidities among patients with unhealthy alcohol use may be preventable if these patients are appropriately identified and treated (Oslin et al., 2014, Watkins et al., 2017, Watkins et al., 2018). Among patients with unhealthy alcohol use detected on alcohol screens, medications and brief counseling offered within primary care can improve alcohol use outcomes (Jonas et al., 2014, Kranzler and Soyka, 2018, Oslin et al., 2014). However, medical providers may miss patients with unhealthy alcohol use unless they have effective tools to screen for these conditions (Venner et al., 2018, Williams et al., 2011, Williams et al., 2014).

The AUDIT-C is a freely available three-question, well-validated (Bradley et al., 2007, Bradley et al., 2016, Bush et al., 1998) screening tool that is used to identify unhealthy alcohol use (Bradley et al., 2007, Bush et al., 1998). The AUDIT-C asks about alcohol consumption, with higher scores indicating more risk of unhealthy alcohol use. Several studies have examined the association between the AUDIT-C completed in routine care (i.e., outside of a research context) and subsequent hospitalizations (Bradley et al., 2016, Chavez et al., 2016, Chavez et al., 2017, Chew et al., 2011, Clark et al., 2016, Williams et al., 2012). These found that increasing AUDIT-C scores, except very low scores, are associated with increased incidence of GI hospitalizations (Bradley et al., 2016), trauma hospitalizations (Williams et al., 2012), 30-day readmission (Chavez et al., 2016, Chavez et al., 2017), and ICU admission and readmission (Clark et al., 2016). However, none of these studies have taken place in routine care outside of the United States Veterans Affairs (VA) healthcare system, which sees an older (majority age 65 or older in some samples) and more male population (94% male) than other healthcare systems (Bradley et al., 2007, Bradley et al., 2016), restricting the generalizability of these studies. Additionally, although unhealthy alcohol use is associated with a wide range of medical conditions (Sterling et al., 2020), none of these prior studies evaluated all-cause hospitalizations in a general primary care sample, but rather focused on specific types of hospitalizations or specific populations (Chavez et al., 2017). Several quality concerns have also been raised about initial AUDIT-C data gathered within the VA (Bradley et al., 2011, Hawkins et al., 2007, Williams et al., 2015), which have informed subsequent AUDIT-C implementation in other systems (Glass et al., 2018, McNeely et al., 2021). The Alcohol Symptom Checklist is a newer tool that has been validated for assessing AUD symptoms in primary care patients who report high-risk drinking on alcohol screens (e.g., AUDIT-C scores ≥ 7) (Hallgren et al., 2022a, Hallgren et al., 2021). The association of the Alcohol Symptom Checklist with hospitalizations has not been examined.

We performed a retrospective cohort study within a single healthcare system in the United States to examine the association between the AUDIT-C and subsequent one-year risk of hospitalization. In a subsample with high-risk drinking (which we define as AUDIT-C scores ≥ 7), we also evaluated the association of scores on the Alcohol Symptom Checklist and one-year hospitalizations. We hypothesize that 1) higher AUDIT-C scores will be associated with greater risk of all-cause hospitalization (with the exception of scores of 0) (Chavez et al., 2016, Chew et al., 2011) and with greater risk of alcohol-attributable hospitalizations in the subsequent year; and 2) higher Alcohol Symptom Checklist scores will be associated with greater risk of all-cause hospitalization. In addition to these primary outcomes, we also examined the association between AUDIT-C and Alcohol Symptom Checklist scores and number of days hospitalized and hospitalizations with alcohol-attributable diagnoses. Our findings may help patients and clinicians understand the health risks at different levels of drinking and adds to a larger body of literature (Bradley et al., 2016, Clark et al., 2016, Rubinsky et al., 2013b, Williams et al., 2018) on how the AUDIT-C relates to health outcomes, further illustrating the clinical utility of this measure.