A METAANALYSIS OF ESTIMATES OF THE AIDS INCUBATION DISTRIBUTION

Information from 12 studies is combined to estimate the AIDS incubatio n distribution with greater precision than is possible from a single s tudy. The analysis uses a hierarchy of parametric models based on a fo ur-parameter generalized F distribution. This general model contains f our standard two-parameter distributions as special cases. The cases a re the Weibull, gamma, log-logistic, lognormal distributions. These fo ur special cases subsume three distinct asymptotic hazard behaviors. A s time increases beyond the median of approximately 10 years, the haza rd can increase to infinity (Weibull), can plateau at some constant le vel (gamma), or can decrease to zero (log-logistic and lognormal). The Weibull, gamma and 'log-logistic distributions' which represent the t hree distinct asymptotic hazard behaviors, all fit the data as well as the generalized F distribution at the 25 percent significance level. Hence, we conclude that incubation data is still too limited to ascert ain the specific hazard assumption that should be utilized in studies of the AIDS epidemic. Accordingly, efforts to model the AIDS epidemic (e.g., back-calculation approaches) should allow the incubation distri bution to take several forms to adequately represent HIV estimation un certainty. It is recommended that, at a minimum, the specific Weibull, gamma and log-logistic distributions estimated in this meta-analysis should all be used in modeling the AIDS epidemic, to reflect this unce rtainty.