Table 3.
Final model with significant covariates and their impact on adherence.
| S. No | Covariates and impact on transitions | OR | 95% CI | Interpretation |
|---|---|---|---|---|
| 1. | Age >25 years on P10 | 0.61 | 0.49–0.77 | Reduced P10 leading to increased adherence |
| 2. | Female sex on P10 | 0.67 | 0.55–0.80 | Reduced P10 leading to increased adherence |
| 3. | Partner on ART for 6 months on P01 | 0.86 | 0.84–0.88 | Reduced P01 leading to reduced adherence |
| 4. | Partner on ART for 6 months on P10 | 1.34 | 1.30–1.38 | Increased P10 leading to reduced adherence |
| 5. | Wants relationship to succeed on P01 | 1.45 | 1.35–1.56 | Increased P01 leading to increased adherence |
| 6. | Wants relationship to succeed on P10 | 0.79 | 0.73–0.85 | Reduced P10 leading to increased adherence |
| 7. | In follow-up for 6 months on P01 | 0.80 | 0.79–0.82 | Reduces P01 leading to reduced adherence |
| 8. | In follow-up for 6 months on P10 | 1.25 | 1.22–1.28 | Increased P10 leading to reduced adherence |
| 9. | Problem alcohol use on P01 | 0.63 | 0.59–0.66 | Reduced P01 leading to reduced adherence |
| 10. | Problem alcohol use on P10 | 1.16 | 1.10–1.24 | Increased P10 leading to reduced adherence |
| 11. | Group with 100% condom use vs no sex group on P10 | 0.84 | 0.81–0.87 | Reduced P10 leading to increased adherence |
| 12. | Group with less than 100% condom use vs no sex group on P01 | 1.21 | 1.18–1.24 | Increased P01 leading to increased adherence |
| 13. | Group with less than 100% condom use vs no sex group on P10 | 0.69 | 0.67–0.72 | Reduced P10 leading to increased adherence |
The 95% confidence intervals were calculated with NONMEM asymptotic standard errors on covariate parameters using the following formula: CI = e θ +1.96(SE). Any covariate, which contained ‘1’ in its confidence interval was considered as non-significant and dropped from the model.