Table 3. Associations between principal component (PC) scores and outcome.
PC1 | PC2 | ||||||
---|---|---|---|---|---|---|---|
Mean (SE) | Delta † | p-value | Mean (SE) | Delta † | p-value | ||
Mortality (2 weeks) † | Alive | 0.09 (0.12) | -0.95 | 0.01 | 0.01 (0.09) | -0.05 | 0.8 |
Dead | -0.49 (0.44) | -0.08 (0.13) | |||||
IRIS ¶ | No IRIS | 0.14 (0.12) | -0.08 | 0.3 | 0.12 (0.09) | -0.63 | 0.004 |
IRIS | -0.25 (0.48) | -0.75 (0.18) |
†The adjusted p values shown were derived from a linear regression model adjusting for treatment group, CD4+-cell count, and the previously described risk factors for mortality, baseline fungal burden and altered mental status. The adjusted delta is the β-coefficient derived from the adjusted linear regression model. Unadjusted p-values were 0.09 (PC1) and 0.7 (PC2).
¶The association between PC2 and IRIS remained significant in both the linear regression model adjusting for treatment group, CD4+-cell count, and the previously described risk factors for mortality, baseline fungal burden and altered mental status, as described above (p = 0.004), and a linear regression model (results shown in table) adjusting for treatment group and the known risk factors for IRIS, baseline fungal burden and CSF white cell count (p = 0.004). Unadjusted p-value = 0.001. The adjusted delta is the β-coefficient derived from the adjusted linear regression model.