Predicting hospitalization among HIV-infected antiretroviral naïve patients starting HAART: determining clinical markers and exploring social pathways

AIDS Care. 2008 Mar;20(3):297-303. doi: 10.1080/09540120701561296.

Abstract

In the era of highly active antiretroviral therapy (HAART), hospitalization as a measure of morbidity has become of increasing interest. The objectives of this study were to determine clinical predictors of hospitalization among HIV-infected persons initiating HAART and to explore the impact of gender and drug use on hospitalization. The analysis was based on a cohort of HIV-positive individuals initiating HAART between 1996 and 2001. Information on hospitalizations was obtained through data linkage with the BC Ministry of Health. Cox-proportional hazard models were used to assess variables associated with time to hospitalization. A total of 1,605 people were eligible and 672 (42%) were hospitalized for one or more days. The final multivariate model indicated that there was an increased risk of hospitalization among those with high baseline HIV RNA (HR for > 100,000 copies/mL: 1.26; 95%CI: 1.16-1.59) or low CD4 cell counts (HR [95% CI] compared to > or = 200 cells/mm3: 1.62 [1.28-2.06] and 1.29 [1.07-1.56] for < 50 and 50-199 cells/mm(3), respectively). Other factors, including adherence, previous hospitalization, gender and injection drug use remained predictive of hospitalization. These findings highlight the importance of closely monitoring patients starting therapy with low CD4 cell counts in order to mediate or prevent outcomes requiring hospitalization.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Antiretroviral Therapy, Highly Active / economics*
  • Antiretroviral Therapy, Highly Active / trends
  • Biomarkers
  • CD4 Lymphocyte Count / statistics & numerical data
  • Cost-Benefit Analysis
  • Female
  • HIV Infections / drug therapy
  • HIV Infections / economics*
  • HIV Infections / immunology
  • Hospitalization / economics*
  • Hospitalization / trends
  • Humans
  • Male
  • Socioeconomic Factors
  • Substance-Related Disorders / complications
  • Substance-Related Disorders / economics
  • Viral Load / economics*
  • Viral Load / statistics & numerical data

Substances

  • Biomarkers