[Skip to Content]
[Skip to Content Landing]
July 14, 1997

The Impact of Alcohol-Related Diagnoses on Pneumonia Outcomes

Author Affiliations

From the Research Unit and Clinical, Addiction, Research, and Education Unit (Dr Saitz) and the Health Care Research Unit (Drs Ghali and Moskowitz), Section of General Internal Medicine, Department of Medicine, Boston Medical Center, Boston University School of Medicine, Boston, Mass. Dr Ghali is now with the University of Calgary Faculty of Medicine, Calgary, Alberta.

Arch Intern Med. 1997;157(13):1446-1452. doi:10.1001/archinte.1997.00440340078008

Background:  There is controversy regarding the role of alcoholism as a prognostic factor in hospitalized patients with pneumonia.

Objective:  To assess the impact of alcohol abuse on hospitalization charges, length of hospital stay, intensive care unit use, and in-hospital mortality.

Methods:  We studied a cohort of all adults hospitalized in 1992 in Massachusetts with a principal diagnosis of pneumonia, and all Massachusetts residents hospitalized for pneumonia in 6 bordering states.

Results:  For the 23 198 pneumonia cases the mean total hospitalization charges were $9925, mean length of hospital stay was 9.6 days, 12% of the cases had intensive care unit stays, and 10% of the cases died during the hospitalization. In bivariate analyses, pneumonia cases with alcoholrelated diagnoses had higher charges (mean, $11 232 vs $9877, P=.07), had shorter length of hospital stay (9.2 vs 9.6 days, P=.02), were more likely to experience an intensive care unit stay (19% vs 12%, P<.001), and had lower in-hospital mortality (6.0% vs 10.2%, P<.001). Multivariable analyses adjusting for comorbidity, pneumonia etiology, and demographics revealed that for pneumonia cases with alcohol-related diagnoses, risk-adjusted hospital charges were $1293 higher (adjusted mean, $11 179 vs $9888, P<.001), length of hospital stay was 0.6 days longer (10.1 vs 9.5 days, P=.001), intensive care unit use was higher (18% vs 12%; adjusted odds ratio, 1.63; 95% confidence interval, 1.33-1.98), and mortality was no different (10% with or without an alcohol-related diagnosis).

Conclusions:  Having an alcohol-related diagnosis is associated with more use of intensive care, longer inpatient stays, and higher hospital charges. To understand resource utilization in cases of pneumonia, alcohol abuse is a comorbid factor that must be considered.Arch Intern Med. 1997;157:1446-1452