Drs. Katharine Bradley and Gwen Lapham studied drinking habits and hospital readmission. They found that underlying social determinants might be the connection.

by Group Health Research Institute (GHRI) Senior Investigator and Group Health Physician Katharine Bradley, MD, MPH, and GHRI Research Associate Gwen Lapham, PhD, MPH, MSW

Primary care is all about keeping people healthy and avoiding medical emergencies if possible. Of course, sometimes people need to go to the hospital for surgery or a serious injury or illness. But an important health care goal is ensuring patients don’t have to go back to the hospital after discharge. We recently found a link between a few simple questions that primary care physicians can ask their patients about alcohol use and patients’ risk for hospital readmissions. Our full analysis suggested that social support may also be an influence.

Our results are part of a long-standing collaboration with clinician leaders and researchers at the U.S. Department of Veterans Affairs (VA) Veterans Health Administration. We’ve helped them add an alcohol-use screen to primary care visits and have studied its effects. The screen is the Alcohol Use Disorder Identification Test-Consumption (AUDIT-C), which consists of three short questions people can complete when they see their primary care providers. The answers help providers care for patients who might have risky drinking behaviors.

What can AUDIT-C tell us about hospital readmission?

Our colleague, Laura Chavez, PhD, designed this study when she was a graduate student in the University of Washington Department of Health Services. She wanted to know if AUDIT-C scores that suggest high-risk drinking might indicate risk of hospital readmission or emergency room visits within 30 days of discharge from a medical or surgical hospitalization. We focused on older adults, since no studies have looked at whether unhealthy alcohol use and readmission are connected in this population. We used a VA database to find nearly 580,000 patients over age 65 with AUDIT-C scores and a medical or surgical hospitalization in their electronic VA medical records. We also obtained data from Medicare for hospitalizations outside VA. We then looked for associations between AUDIT-C scores and readmission within 30 days of discharge.

We found that people with AUDIT-C scores suggesting high-risk drinking had a slightly increased risk (1 percent) of hospital readmission for a medical condition compared to people with low-risk AUDIT-C scores. Our other results, intriguingly, suggest that the link between drinking and readmission is complex. Rather than simply reflecting the negative effects of high-risk alcohol use, our analyses show that risky drinking might reflect poor social and economic conditions and social support, including not being married. Nondrinking was also linked to increased hospital readmission in our study. Other studies show that people who don’t use alcohol at all can have lower overall health. For example, they might be avoiding alcohol because of a health condition, including prior alcohol use disorder. Primary care physicians might consider this possibility when asking patients about their drinking behavior.

Can comprehensive care help?

Our study, which was funded by a dissertation grant Dr. Chavez received from the Agency for Healthcare Research and Quality, suggests the importance of comprehensive patient care. Primary care physicians might consider their patients’ alcohol use in the context of their family support, general health, and socioeconomic conditions. Initiatives at the VA, Group Health, and GHRI all support primary care providers in having screens for unhealthy alcohol use be part of routine care.

We’re glad to be collaborating with the VA in research to identify opportunities to help patients with high-risk drinking behavior. We also support Group Health as a leader in comprehensive care. Finally, we hope you’ll join us in June, when GHRI will encourage conversations about social determinants of health by hosting Baltimore Health Commissioner Dr. Leana Wen for the 2016 Birnbaum lecture.

You can read more about our work in papers by Chavez et al. and Rubinsky et al. You might also be interested in a post by Ed Wagner, MD, MPH, director emeritus of GHRI’s MacColl Center for Health Care Innovation at GHRI, about integrating mental and behavioral health into primary care.