Hospital utilization and costs in a cohort of injection drug users

Abstract

Background: Many injection drug users (IDUs) seek care at emergency departments and some require hospital admission because of late presentation in the course of their illness. We determined the predictors of frequent emergency department visits and hospital admissions among community-based IDUs and estimated the incremental hospital utilization costs incurred by IDUs with early HIV infection relative to costs incurred by HIV-negative IDUs.

Methods: The Vancouver Injection Drug User Study (VIDUS) is a prospective cohort study involving IDUs that began in 1996. Our analyses were restricted to the 598 participants who gave informed consent for our study. We used the participants' responses to the baseline VIDUS questionnaire and, from medical records at St. Paul's Hospital, Vancouver, we collected detailed information about the frequency of emergency department visits, hospital admissions and the primary diagnosis for all visits or hospital stays between May 1, 1996, and Aug. 31, 1999. The incremental difference in hospital utilization costs by HIV status was estimated, based on 105 admissions in a subgroup of 64 participants.

Results: A total of 440 (73.6%) of the 598 IDUs made 2763 visits to the emergency department at St. Paul's Hospital during the study period. Of these 440, 265 (160.2%) made frequent visits (3 or more). The following factors were associated with frequent use: HIV-positive status (seroprevalent: adjusted odds ratio [OR] 1.7, 95% confidence interval [CI] 1.2-2.6; seroconverted during study period: adjusted OR 3.0, 95% CI 1.6-5.7); more than 4 injections daily (adjusted OR 1.5, 95% CI 1.1-2.1); cocaine use more frequent than use of other drugs (adjusted OR 2.0, 95% CI 1.2-3.6); and unstable housing (adjusted OR 1.5, 95% CI 1.1-2.2). During the study period 210 of the participants were admitted to hospital 495 times; 118 (56.2%) of them were admitted frequently (2 or more admissions). The 2 most common reasons for admission were pneumonia (132 admissions among 79 patients) and soft-tissue infections (cellulitis and skin abscess) (90 admissions among 59 patients). The following factors were independently associated with frequent hospital admissions: HIV-positive status (seroprevalent: adjusted OR 5.4, 95% CI 3.4-8.6; seroconverted during study period: adjusted OR 2.9, 95% CI 1.4-6.0); and female sex (adjusted OR 1.8, 95% CI 1.1-3.1). The incremental hospital utilization costs incurred by HIV-positive IDUs relative to the costs incurred by HIV-negative IDUs were $1752 per year.

Interpretation: Hospital utilization was significantly higher among community-based IDUs with early HIV disease than among those who were HIV negative. Much of the hospital use was related to complications of injection drug use and may be reduced with the establishment of programs that integrate harm reduction strategies with primary care and addiction treatment.