Temporal Trends and Molecular Epidemiology of HIV-1 Infection in Taiwan From 1988 to 1998

Abstract

Objective: To examine the influence of country/region of birth on spectrum of AIDS-defining illness. Methods: National surveillance data for 4,629 adolescents and adults diagnosed with AIDS from 1992 through 1998 were analyzed. Country of birth was grouped into five broad categories (Australia, other predominantly industrialized country regions, sub-Saharan Africa, Asia-Pacific, and other regions with predominantly developing countries). Proportions of AIDS-defining illnesses were calculated and compared by country/region of birth. Role of country/region of birth in the distribution of AIDS-defining illnesses was further assessed using logistic models. Results: Of the 4,488 (97.0%) AIDS cases with country of birth recorded, 1,120 (25.0%) were born outside Australia. In multivariate analyzes, AIDS cases born in sub-Saharan Africa had an increased risk of tuberculosis (odds ratio [OR], 18.7; confidence interval [CI], 9.2-38.2) and cryptococcosis (OR, 2.4; CI, 1.1-5.4), but a decreased risk of esophageal candidiasis (OR, 0.3; 0.1-0.8) and Pneumocystis carinii pneumonia (OR, 0.5; 0.3-0.9) compared with AIDS cases born in Australia. Tuberculosis risk was also elevated among AIDS cases born in Asia-Pacific (OR, 9.6; 5.3-17.5) and other developing country regions (OR, 3.1; 0.9-10.4). Risk of AIDS-defining illnesses was similar for AIDS patients born in Australia and other industrialized country regions. Country of birth had no influence on risk of cytomegalovirus (CMV)-related disease and Mycobacterium avium complex (MAC) infection. Differential AIDS-defining illness risk was more pronounced among AIDS patients born in developing countries who had resided in Australia for less time. Conclusions: Differential risk by country/region of birth for some AIDS-defining illnesses, especially among more recent arrivals from developing countries, suggests that environmental microbial habitats are important determinants of opportunistic infection risk. Similar risk of CMV disease and MAC infection is consistent with the ubiquitous nature of these microbial agents and suggests that previously reported low prevalence from developing countries may reflect poor diagnostic capacity rather than level of risk.