Estimated effects of the date of confirmed infection on the markers of disease progression.

General linear models of the rate of CD4 decline were fitted to the estimated CD4 slopes per patient; mixed-effect models were fitted to the original CD4 count data. Date refers to the earliest date of confirmed infection in each patient. Further co-factors included gender, riskgroup, baseline CD4 count, age and interaction terms; stepwise elimination of non-significant factors was performed based on likelihood ratio tests. Effects not related to date are not shown.CI, confidence interval; HET, heterosexual; MSM, men having sex with men.