       Document 1806
 DOCN  M94A1806
 TI    Predictors of hospitalization in persons with HIV-disease.
 DT    9412
 AU    Woodfall B; Hogg RS; Strathdee SA; Le R; Schechter MT; Montaner JS;
       O'Shaughnessy MV; British Columbia Centre for Excellence in HIV/AIDS,
       Vancouver,; Canada.
 SO    Int Conf AIDS. 1994 Aug 7-12;10(1):447 (abstract no. PD0396). Unique
       Identifier : AIDSLINE ICA10/94370769
 AB    OBJECTIVE: To determine demographic and clinical predictors of
       in-patient hospitalization, overall and in high (HEH) experienced
       hospitals, among men and women in British Columbia (B.C.) with
       HIV-disease. METHODS: A population-based province-wide cohort study was
       based on the Centre's HIV/AIDS drug program which distributes
       HIV-related drugs free of charge to all eligible individuals in the
       province with HIV-specific indications. Record linkage was conducted
       with Hospital Programs (B.C. Ministry of Health) to obtain all records
       of in-patient hospitalizations in 1990-93 for participants who
       consented. HEH were defined a priori as the 5 tertiary hospitals with
       the highest number of HIV/AIDS discharges. Logistic regression
       determined which of gender, age, CD4 count, KS, PCP, and geographic
       locality were predictive of the most recent hospitalization, overall and
       in HEH. RESULTS: Analysis was restricted to 773 persons (735 men; 38
       women); the majority (65.4%) were aged 30-44. A total of 335 persons
       (310 men; 25 women) were hospitalized between 1990-93. 3-year rates of
       hospitalization for men and women were 422 and 658 per 1,000 (see
       figure). Of those hospitalized, the mean number of visits was 2.2 (range
       1-13) with 58.5% of these stays being 3 days or less. Most visits (77%)
       occurred in HEH. Logistic regression revealed that hospitalization was
       independently associated with gender (p = 0.0034), low CD4 counts (p =
       0.008), and PCP (p = 0.0001) while hospitalization in a HEH hospital was
       associated with low CD4 counts (p = 0.0047) and PCP (p = 0.0001).
       Univariate analysis of most recent encounters indicated that
       hospitalized men were more likely to have lower CD4 counts (p = 0.015),
       to be older (p < 0.001), and AIDS-defining illnesses (p = 0.019)
       compared to hospitalized women. CONCLUSIONS: This analysis demonstrates
       higher hospitalization rates in HIV+ women than men independent of CD4
       count. Hospitalized women were less likely to have an AIDS-defining
       illness than hospitalized men. Gender differences likely reflect overall
       sex-specific variations in morbidity. TABULAR DATA, SEE ABSTRACT VOLUME.
 DE    Adult  British Columbia  Cohort Studies  Female
       Hospitalization/*STATISTICS & NUMER DATA  Human  HIV
       Infections/IMMUNOLOGY/*THERAPY  Leukocyte Count  Male  T4 Lymphocytes
       MEETING ABSTRACT

       SOURCE: National Library of Medicine.  NOTICE: This material may be
       protected by Copyright Law (Title 17, U.S.Code).

