Trends & Associations of HIV-related admission and mortality proportions in Princess Marina Hospital, Gaborone from 2000-2006

Abstract

Background. Human Immunodeficiency Virus / Acquired Immunodeficiency Syndrome (HIV/AIDS) has been recognised in Botswana for the last two decades, however, facility-based trends and associations of Hospital admissions and mortality proportions due to HIV/AIDS and determinant factors, have not been studied in settings such as Princess Marina Hospital. . Objective. The aim of this study was to analyse the HIV-related admission and mortality proportions at Princess Marina Hospital in the years 2000, 2003 and 2006, compare and establish the trends Methods. Patient records for the years 2000, 2003and 2006 were reviewed. Cases were identified by documented HIV status or evidence of clinical immunosuppression by having any of the conditions listed in section B20-B24 of the International Classification of Diseases (ICD 10 B20-B24).HIV-related admissions, HIV-related mortality proportions and HIV- Case fatality Rates were calculated for each of the specified periods. Chi-square test for linear using ptrend in STATA was used. A Log binomial regression method was used to identify important risk factors for HIV-associated mortality and admission. Results. A total of (N= 24 541) records were reviewed for the years 2000-2006. The HIV-admission proportions (HIV related admission/Total admissions) were 0.101(988/9748), 0.112(868/7745) and 0.123(754/6148) admissions for the years 2000, 2003 and 2006, respectively. The HIV mortality proportions (HIV-related mortality/Total mortality) were 0.381(289/759), 0.474(309/652) and 0.371(252/680) deaths for the years 2000, 2003 and 2006, respectively. The HIV-Case Fatality Rates (HIV-admission/HIV-related mortality) were 29.2% (289/988), 35.6% (309/868) and 33.4% (252/754) for the years 2000,2003 and 2006, respectively. Chi-square test for linear trend in STATA was only statistically significant for HIV-related admissions (p<0.05).The log binomial regression model indicated that the relative risks of both HIV-related admission and death were lower in females (aRR 0.54 CI 0.32,0.93) and (aRR 0.53 CI 0.34,0.86), but increased in the age-group 14-49 years (aRR 2.08 CI 1.07,4.04) and (aRR 1.21 CI 1.12,3.94), respectively. Conclusion. The proportion of HIV-related admissions has increased, HIV related deaths and the HIV-Case Fatality Rate were notable since 2000-2006. Moreover, there was a linear trend in the proportion of HIV-related admission but none evident in HIV-related deaths and HIV-Case Fatality Rate. Female gender and age group 14-49 years were the two most important determinants for both HIV-related admissions and deaths. Future research focusing on recent trends and determinants of both HIV-related admissions and deaths are required.