Table 2.
Univariate and multivariate Cox proportional hazards regression (CPHR) of possible risk factors for below-target viral suppression
| Characteristic | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| Hazard ratio | P value | 95% confidence interval | Hazard ratio | P value | 95% confidence interval | |
| Pre-UTT period (March to August 2016): risk factors for below-target viral suppression | ||||||
| Optimal adherence to ART | 1.0 | 1.0 | ||||
| Suboptimal adherence to ART | 5.32 | 0.0006 | 2.75–10.29 | 2.29 | 0.007 | 1.21–3.39 |
| UTT period (September 2016 to February 2017): risk factors for below-target viral suppression | ||||||
| Optimal adherence to ART | 1.0 | 1.0 | ||||
| Suboptimal adherence to ART | 14.63 | 0.003 | 3.18–104.38 | 6.24 | 0.01 | 2.63–17.82 |
| Pre-UTT period (March to August 2016): risk factors for suboptimal adherence | ||||||
| Two parent homes | 1.0 | 1.0 | ||||
| One parent homes | 1.13 | 0.42 | 0.84–1.53 | 1.07 | 0.67 | 0.79–1.45 |
| Guardian homes | 1.81 | 0.08 | 0.97–3.77 | 1.09 | 0.60 | 2.80–1.48 |
| 2–5 years old | 1.0 | 1.0 | ||||
| 6–10 years old | 0.50 | 0.03 | 0.34–0.73 | 0.88 | 0.19 | 0.72–1.07 |
| 11–14 years old | 1.45 | 0.01 | 0.81–2.69 | 0.84 | 0.042 | 0.71–0.99 |
| 15–18 years old | 0.40 | 0.05 | 0.27–0.60 | 0.73 | 0.02 | 0.56–0.89 |
| Female gender | 1.0 | 1.0 | ||||
| Male gender | 0.66 | 0.02 | 0.53–0.82 | 0.81 | 0.03 | 0.73–0.91 |
| UTT period risk factors for suboptimal adherence | ||||||
| Two parent homes | 1.0 | 1.0 | ||||
| One parent homes | 1.08 | 0.03 | 1.44–9.90 | 1.11 | 0.90 | 0.22–5.64 |
| Guardian homes | 2.27 | 0.041 | 1.12–6.72 | 1.13 | 0.80 | 2.19–6.62 |
| 1–5 years old | 1.0 | 1.0 | ||||
| 6–10 years old | 1.17 | 0.73 | 0.48–2.85 | 1.11 | 0.90 | 0.22–5.64 |
| 11–14 years old | 2.16 | 0.018 | 1.03–5.73 | 1.02 | 0.20 | 0.07–14.47 |
| 15–18 years old | 1.84 | 0.95 | 0.33–3.28 | 1.20 | 0.27 | 0.01–3.59 |
| Female gender | 1.0 | 1.0 | ||||
| Male gender | 0.73 | 0.56 | 0.26–2.07 | 1.02 | 0.20 | 0.07–14.47 |
The variables considered were adherence to ART, caregiver status, and gender. Another CPHR analysis was conducted to determine risk factors for suboptimal adherence to ART. The variables considered were caregiver status, age group, and gender. The CPHR analyses were done for both the Pre-UTT period (September 2016 to February 2017) and the UTT period (March to August 2016)
Italic values indicate significance of P value (P < 0.05)