Effect of zidovudine on survival of patients with AIDS in Australia

Abstract

Since the first cases of AIDS in Australia was diagnosed in December 1982, there have been substantial improvements in the treatment of AIDS-related conditions. In particular, zidovudine was widely introduced into clinical practice in Australia in June 1987. In order to evaluate its effect, we compared the survival of patients diagnosed before and after July 31, 1987 using data available in early 1989. Survival distributions were compared by means of Kaplan-Meier curves and by fitting exponential survival models incorporating a specieal feature of the data. Before August 1, 1987 the overall distribution of survival times for patients with AIDS in Australia is well described by an exponential distribution with a mean of 1.04 years. The corresponding median survival time for this period was 8.8 months. For patients diagnosed with AIDS after July 31, 1987 the median survival time had not been attained by December 31, 1988. However, the estimated means survival time increased to 2.7 years. Survival times were found to be remarkably stable over the different regions of Australia. We have shown that substantial improvements in survival of patients diagnosed with AIDS in Australia are associated with the widespread availability of zidovudine from mid 1987. To the best of our knowledge this study is the first of its kind to show a major shift in the distribution of survival associated with the introduction of antiviral therapy.