Patients with alcohol use disorders who develop health care–associated infections have a 71% greater mortality risk, as well as longer hospital stays and higher hospital costs, than do those without alcohol use disorders who develop such infections, according to an analysis of data from a large inpatient database.

An alcohol use disorder diagnosis was found to be an independent predictor of a 2-day increase in hospital length of stay.

The researchers studied the more than 32.7 million hospital discharges in the Nationwide Inpatient Sample for 2007, of which 149,982 developed a health care–associated infection and met inclusion criteria for the study. From the overall group, 4.7% were patients with an alcohol use disorder (AUD), while 5.9% of the cohort with a health care–associated infection also had an AUD codiagnosis, the researchers wrote. Having an AUD was found to be an independent predictor of increased mortality in the latter group (odds ratio 1.71), Dr. Marjolein de Wit of Virginia Commonwealth University, Richmond, and her colleagues reported online April 15 in Alcoholism: Clinical and Experimental Research.

An AUD diagnosis was also found to be an independent predictor of a 2-day increase in hospital length of stay (13 days in those with an AUD vs.11 days in those without). Hospital costs were about $7,500 higher ($34,826 vs. $27,167) for those AUD patients, the investigators said (Alcohol. Clin. Exp. Res. 2011 April 15 [doi: 10.1111/j.1530-0277.2011.01475.x]).

Prior studies have shown that the increased risk of health care infections in this population might be explained by cytokine abnormalities, impaired cell-mediated immune function, and aberrant innate immunity, they noted.

"Studies evaluating therapies aimed at decreasing the risk of developing health care–associated infections in patients with AUD are warranted," they concluded.

In the meantime, steps can be taken to reduce risk of developing health care–associated infections, according to Dr. Claudia Spies, of University Hospital Charite Universitaetsmedizin Berlin, whose earlier research was cited in the current study.

In a statement on the findings, Dr. Spies stressed the importance of external measures such as hand washing and head-of-bed elevation, and of the need for intrinsic measures such as therapeutic interventions specifically targeting hospitalized patients with AUDs. AUDs are factors intrinsic to patients, and thus preventing health care–associated infections in these patients requires such interventions, she explained.

Dr. Spies, along with Dr. de Wit, also emphasized the need for candid discussion about alcohol use with patients – despite the stigma they may feel.

"This is important both when a hospital admission is scheduled as well as at the time of an emergency hospital admission. In the case with scheduled surgeries, such as an elective surgery, one month preoperative abstinence may decrease the risk of health care–associated infections," Dr. de Wit said.

Follow-up care is also a necessity, she added, noting that patients will need anonymous data handling and help finding adequate intervention and treatment for their AUD.

The Nationwide Inpatient Sample for 2007 included hospital discharge data on adults aged 18 years and older. The investigators focused on those who developed pneumonia and/or sepsis, and excluded those who transferred from another health care facility, as well as those with immune suppression, cancer, or a community-acquired infection. Pneumonia was the most common health care–associated infection, occurring in 135,267 AUD patients, while sepsis occurred in 12,046, and both pneumonia and sepsis occurred in 2,579.

Alcohol use disorder designation was "based on an ICD diagnosis and encompassed alcohol dependence syndrome, alcohol abuse, and alcohol-induced mental disorders," the researchers explained.

The greater proportion of AUD patients in the subgroup of individuals with health care–associated infections, compared with the entire cohort, was not surprising; prior studies have demonstrated that AUD is common in hospitalized patients, and that it is associated with an increased risk of health care–associated infections. In fact, some studies have demonstrated rates of AUD up to 42% in hospitalized patients. Thus, given that the rates in this study were only 4.7% in all discharged patients and 5.9% in those with health care–associated infections (likely due to underdiagnosis, according to the investigators), it is likely that the effects of AUD in these patients are actually underestimated by this study, they said.

Although the study is limited by its retrospective design and the fact that the sample used has not been validated, the sample is large and likely representative, and the results indicate that patients with an AUD are disproportionately affected when it comes to mortality, hospital stay, and hospital costs, the investigators said.

The authors reported that they had no disclosures.

Alcohol Use Disorders Up Death Risk for Inpatients With Infections