C. Scheidegger et W. Zimmerli, INCIDENCE AND SPECTRUM OF SEVERE MEDICAL COMPLICATIONS AMONG HOSPITALIZED HIV-SERONEGATIVE AND HIV-SEROPOSITIVE NARCOTIC DRUG-USERS, AIDS, 10(12), 1996, pp. 1407-1414
Objectives: To examine differences in the incidence and spectrum of di
seases, as well as duration of inpatient stay, between HlV-seronegativ
e and HIV-seropositive narcotic drug users (NDU). Design: Retrospectiv
e analysis of 9 years of experience. Data collection by chart review u
sing pre-set criteria for diagnoses. Estimation of hospital admission
densities by assuming a dynamic but stable population of 2000 NDU (wit
h a mean HIV-seroprevalence of 25%) throughout the study period. Patie
nts: Comprising 314 HIV-seronegative NDU, 217 HIV-seropositive NDU, an
d 10 NDU with admissions registered in either group (from a total of 1
011 admissions). Results: The overall admission incidence density was
35 and 120 per 1000 person-years among HIV-seronegative NDU and HIV-se
ropositive NDU, respectively [risk ratio (RR) 3.5, 95% confidence inte
rval (CI) 3.2-3.7]. Compared with seronegative NDU, HIV-seropositve ND
U were more frequently admitted for various nonopportunistic infection
s (RR 7.2, 95% CI 6.1-8.4), including pneumonia (RR 10.9, 95% CI 7.6-1
6.6), tuberculosis (RR 30.0, 95% CI 3.6-233.8), soft-tissue infections
(RR 3.5, 95% CI 1.7-7.2), osteoarticular infections (RR 6.0, 95% CI 1
.5-23.9), endocarditis (RR 5.3, 95% CI 1.5-17.9), and various other in
fections (RR 5.8, 95% CI 3.2-10.5). HIV-seropositive NDU were also mor
e frequently admitted for non-infectious medical complications (RR 2.3
, 95% CI 1.8-3.0). Seronegative NDU had a shorter median inpatient sta
y (2 versus 9 days, P < 0.00001). HIV infection accounted for an estim
ated excess burden of at least 2700 inpatient care days in 9 years amo
ng the 500 local HIV-seropositive NDU. Conclusions: Among NDU, HIV inf
ection adds considerable excess burden in terms of severe complication
s needing inpatient care.