TABLE 5.
Estimated effect of each additional month until switching therapy, given current CD4 T-cell count (CD4(j)) and elapsed months since failure occurred (j), among subjects with virologic failure, sampled from a cohort of HIV*-infected persons in San Francisco, California, 2000–2004
| Term† | Estimate (cells/μl) per month of waiting to switch therapy‡ | 95% confidence interval§ |
|---|---|---|
| β1 | −9.2 | −18.1, −0.3 |
| β4 | 0.05 | 0.02, 0.08 |
| β5 | 1.5 | −0.4, 3.3 |
| β7 | −0.01 | −0.02, −0.002 |
HIV, human immunodeficiency virus.
Based on the model E(YĀ(j−1),c(j)(j + 8)|A(j − 1) = 1, Supp(j) = 0, CD4(j)) = β0 + β1c(j)+ β2CD4(j) + β3j + β4c(j) × CD4(j)+ β5c(j) × j + β6j × CD4(j) + β7c(j) × CD4(j) × j, where c(j)denotes the counterfactual switching time in number of months after time j.
Coefficient from a history-adjusted marginal structural model for the first nine time points (j = 0, …, 8).
Based on robust standard error estimates using generalized estimating equations.