Predictors of HIV/AIDS among individuals with tuberculosis: health and policy implications

The purpose of this report was retrospectively to assess the application of clinical predictors (i.e. signs and symptoms) for the diagnosis of the human immunodeficiency virus (HIV) infections, including the acquired immunodeficiency syndrome (AIDS), among tuberculosis (TB)-positive subjects enrolled within an urban heterosexual sero-discordant couple (one partner HIV-positive and the other HIV-negative) cohort study at an HIV prevention and research centre in Lusaka, Zambia. Available medical, clinical and laboratory records from 1994 to mid-1998 were reviewed, with follow-up visits to the local chest clinics and sputum analyses laboratories instituted as additional quality assurance measures. A nested case-control study was employed. The World Health Organization clinical case definition for HIV/AIDS (WHOCCDA) was observed to be significantly more likely to predict HIV/AIDS among TB-positive HIV-positive subjects than TB-positive HIV-negative subjects. Therefore, this report documents the relative effectiveness of the WHOCCDA in predicting HIV/AIDS cases and discusses the health and policy implications, especially for the developing world which accounts for a significant proportion and impact of HIV/AIDS. Further research studies based on prospective randomized controlled trails or other rigorous evaluation strategies are warranted to substantiate these findings.