Table 3. Univariate and multiple logistic regression analysis of factors associated with cure from TB (n = 519).
Factor | Level | Proportion cured | Univariate Regression | Multiple Regression | ||
Freq/n (%) | OR | 95% CI | OR | 95% CI | ||
Gender | Male | 277/299 (92.6) | 1 | 1 | ||
Female | 204/220 (92.7) | 1.01 | 0.52–1.98 | 1.32 | 0.56–3.11 | |
Age Group (in years) | <30 | 169/180 (93.9) | 1 | 1 | ||
30–39 | 146/164 (89.0) | 0.53 | 0.24–1.15 | 0.72 | 0.25–2.06 | |
40–49 | 72/79 (91.1) | 0.67 | 0.25–1.80 | 0.59 | 0.18–1.92 | |
≥50 | 78/80 (97.5) | 2.54 | 0.55–11.73 | 1.09 | 0.21–5.66 | |
Missing | 16/16 (100) | |||||
Actual days treatment missed | None | 319/336 (94.9) | 1 | |||
< 15 | 149/159 (93.7) | 0.79 | 0.36–1.78 | 0.84 | 0.34–2.08 | |
15–29 | 8/19 (42.1) | 0.04 | 0.01–0.11 | 0.04 | 0.01–0.14 | |
≥ 30 | 5/5 (100.0) | - | - | - | - | |
+Missed Treatment in Intensive Phase | No | 390/412 (94.7) | 1 | |||
Yes | 91/107 (85.1) | 0.32 | 0.16–0.64 | - | - | |
Missed Treatment in Continuation Phase | No | 368/400 (92.0) | 1 | |||
Yes | 113/119 (95.0) | 1.64 | 0.67–4.02 | - | - | |
DOTS Option | Guardian | 362/393 (92.1) | 1 | |||
Hospital | 85/91 (93.4) | 1.21 | 0.49–3.00 | - | - | |
Health Centre | 29/30 (96.7) | 2.48 | 0.33–18.85 | - | - | |
Missing | 5/5 (100) | |||||
Patient Category | New Case | 425/455 (93.4) | 1 | 1 | ||
Relapse | 54/60 (90.0) | 0.64 | 0.25–1.60 | 0.60 | 0.20–1.84 | |
Failure | 2/4 (50.0) | 0.07 | 0.01–0.52 | 0.09 | 0.01–1.18 | |
Whether the patient was on ART | Yes | 70/78 (89.7) | 1 | |||
No | 173/190 (91.1) | 1.16 | 0.48–2.82 | - | - | |
Missing | 238/251 (94.8) | |||||
HIV status | Negative | 140/144 (97.2) | 1 | 1 | ||
Positive | 247/273 (90.5) | 0.27 | 0.09–0.79 | 0.35 | 0.11–1.10 | |
Unknown | 94/102 (92.2) | |||||
Initial smear positivity | 1+ | 127/139 (91.4) | 1 | |||
2+ | 113/124 (91.1) | 0.97 | 0.41–2.29 | - | - | |
3+ | 193/205 (94.2) | 1.52 | 0.66–3.49 | - | - | |
Scanty | 44/46 (95.7) | 2.08 | 0.45–9.65 | - | - | |
Missing | 4/5 (80.0) |
Note: + - was not included in the adjusted model as it is highly correlated with actual number of days missed treatment.