Table 2 Associations between prior antiretroviral drug use, pre-treatment drug resistance and virological failure.
N | Events | Unadjusted OR (95% CI) | P-value | Adjusted OR (95% CI) | P-value | |
|---|---|---|---|---|---|---|
Effect of prior ARV use on PDR a | ||||||
Any prior ARV use | ||||||
No | 2534 | 288 | 1.0 | 1.0 | ||
Yes | 119 | 29 | 6.8 (4.3–10.8) | <0.001 | 7.2 (4.4–11.7) | <0.001 |
Type of prior ARV use | ||||||
None | 2534 | 288 | 1.0 | 1.0 | ||
ART | 58 | 16 | 8.3 (4.5–15.3) | <0.001 | 9.1 (4.8–17.2) | <0.001 |
sdNVP | 39 | 7 | 4.4 (1.9–10.2) | 0.001 | 3.3 (1.4–8.1) | 0.008 |
Other | 22 | 6 | 8.1 (3.1–21.2) | <0.001 | 15.1 (5.3–42.5) | <0.001 |
Effect of prior ARV use on VF b | ||||||
Any prior ARV use | ||||||
No | 1953 | 190 | 1.0 | 1.0 | ||
Yes | 86 | 20 | 2.7 (1.6–4.6) | <0.001 | 3.1 (1.6–6.1 | 0.001 |
Type of prior ARV use | ||||||
None | 1953 | 190 | 1.0 | 1.0 | ||
ART | 4 4 | 13 | 3.6 (1.7–7.7) | 0.001 | 3.9 (1.6–9.1) | 0.002 |
sdNVP | 29 | 5 | 1.9 (0.6–6.0) | 0.256 | 3.1 (0.9–11.0) | 0.088 |
Other | 13 | 2 | 1.7 (0.6–4.7) | 0.317 | 1.4 (0.5–4.0) | 0.562 |
Logistic regression model including interaction term (prior ARV use* PDR)c | 0.485 | |||||
PDR as intermediate on causal pathway d Causal mediation analysis | ||||||
Natural direct effects (NDE) | 2.1 (1.2–3.8.) | 0.013 | 2.7 (1.5–5.0) | 0.002 | ||
Natural indirect effects (NIE) | 1.4 (0.9–2.3) | 0.1340 | 1.8 (1.0–3.1) | 0.059 | ||
Controlled direct effect (CDE) | 2.1 (1.1–4.1) | 0.031 | 2.5 (1.2–5.1) | 0.010 | ||
Total effects (TE)e | 3.0(1.6–5.7) | 0.001 | 4.8 (2.3–10.0) | <0.001 | ||
Proportion mediated PM = NIE/TE = 38% | ||||||
Proportion eliminated PE = (TE − CDE)/TE = 48% | ||||||
