Analysis of dynamic cohort data

Left-truncated and interval-censored data, termed dynamic cohort data, aris e in longitudinal studies with rolling admissions and only occasional follo w-up. The authors compared four approaches for analyzing such data: a const ant hazard model; maximum likelihood estimation with flexible parametric mo dels; the midpoint method, in which the midpoint of the last negative and f irst positive test result is used in a Cox proportional hazards model that accounts for left truncation; and a semiparametric method that uses imputed failure times in the Cox model. By using a simulation study, they assessed the performance of these approaches under conditions that can arise in obs ervational studies: changes in disease incidence and changes in the underly ing population. The simulation results indicated that the constant hazard m odel and midpoint method were inadequate and that the flexible parametric m odel was useful when enough parameters were used in modeling the baseline h azard. The semiparametric method ensured correct parameter (odds ratio) est imation when the baseline hazard was misspecified, but the trade-off increa sed computational complexity. In this paper, a study of the incidence of hu man immunodeficiency virus in patients repeatedly tested for the virus at a sexually transmitted disease clinic in New Orleans, Louisiana, illustrates the methods used.