Table 5. The effect of changing from a normal to a skew-normal distribution on the random-effects on the regression coefficients, with 95% credible intervals, in the asymptote model.
| Parameter | cVL2 adjusted model 6 | cVL2 adjusted model 7 | ||
|---|---|---|---|---|
| Estimate | 95% CI | Estimate | 95% CI | |
| Female-sex | 6.35 | (3.019, 14.382) | 6.52 | (3.004, 14.397) |
| Baseline age | 0.54 | (0.389, 0.740) | 0.54 | (0.386, 0.748) |
| Baseline CD4 count | 3.00 | (2.175, 4.145) | 3.04 | (2.235, 4.242) |
| Baseline log viral load | 1.26 | (0.905, 1.799) | 1.27 | (0.904, 1.785) |
| Time on treatment | 4.08 | (3.019, 5.579) | 4.04 | (3.034, 5.425) |
| Time on treatment—squared | 0.65 | (0.550 0.745) | 0.65 | (0.564, 0.746) |
| cumulative log viral load | 0.42 | (0.236, 0.730) | 0.41 | (0.236, 0.718) |
cVL2 –cumulative HIV log10 viral load. All estimates reported as posterior odds ratios with 95% credible intervals (CI).