Factors Associated with Mortality among Patients on TB Treatment in the Southern Region of Zimbabwe, 2013

Table 3

Factors associated with mortality among TB patients in southern Zimbabwe, 2013.

Characteristics		Died, (%)	Univariate RR (95% CI)	Multivariate-adjusted RR
				Model 1	Model 2	Model 3	Model 4
Total	1,847	428 (23.2)	—	—	—	—	—
Age (in years)
<5	63	11 (17.5)	Reference	Reference	Reference	Reference	Reference
5–14	69	11 (15.9)	0.91 (0.43–1.96)	0.91 (0.42–1.99)	0.89 (0.43–2.00)	1.76 (0.66–4.66)	1.07 (0.39–2.94)
15–44	1,257	290 (23.1)	1.32 (0.77–2.28)	1.17 (0.65–2.10)	1.16 (0.64–2.09)	2.17 (0.97–4.84)	1.12 (0.47–2.65)
45–64	347	74 (21.3)	1.22 (0.69–2.17)	1.10 (0.60–2.01)	1.12 (0.61–2.4)	2.02 (0.91–4.51)	0.95 (0.39–2.32)
65+	107	41 (38.3)	2.19 (1.22–3.95)	2.48 (1.35–4.55)	2.47 (1.35–4.57)	3.13 (1.4–6.98)	1.37 (0.5–3.8)
Missing data	4	1 (25)	—	—	—	—	—
Type of health facility	1847
Rural health clinic	870	214 (24.6)	Reference	Reference	Reference	Reference	Reference
Mission/rural hospital	319	79 (24.8)	1.01 (0.80–1.26)	0.96 (0.77–1.20)	1.04 (0.62–1.76)	0.94 (0.94-0.94)	0.91 (0.68–1.21)
Polyclinic	254	35 (13.8)	0.56 (0.40–0.78)	0.56 (0.40–0.78)	0.22 (0.07–0.7)	0.55 (0.39–0.77)	0.57 (0.39–0.85)
District/provincial hospital	398	98 (24.6)	1.00 (0.81–1.23)	0.99 (0.81–1.21)	0.33 (0.12–0.9)	0.9 (0.9-0.9)	0.78 (0.59–1.02)
Missing data	6	2 (33.3)	—	—	—	—	—
TB category
New TB	1557	337 (21.6)	Reference	Reference	Reference	Reference	Reference
Retreatment TB	288	90 (31.3)	1.44 (1.19–1.76)	1.34 (1.10–1.63)	1.2 (0.65–2.22)	1.53 (1.39–1.68)	1.59 (1.24–2.04)
Missing data	2	1 (50)	—	—	—	—	—
HIV status
HIV-negative	396	59 (14.9)	Reference	Reference	Reference	Reference	Reference
HIV-positive	1439	364 (25.3)	1.70 (1.32–2.18)	1.87 (1.44–2.42)	—	—
HIV unknown	1	1 (100)	—	—	—	—	—
Missing data	11	4 (36.4)	—	—	—	—	—
WHO clinical staging at ART
I	27	3 (11.1)	Reference	Reference	Reference	Reference	Reference
II	70	9 (12.9)	1.16 (0.34–3.95)	—	0.65 (0.17–2.43)	—
III	470	69 (14.7)	1.32 (0.44–3.93)	—	1.22 (0.45–3.31)	—
IV	39	5 (12.8)	1.15 (0.30–4.43)	—	1.29 (0.36–4.58)	—
Missing data	831	278 (33.4)	—	—	—	—	—
CD4 count at ART initiation(cells/mL)
≤50	94	28 (29.8)	Reference	Reference	Reference	Reference	Reference
51–200	132	15 (11.4)	0.38 (0.22–0.67)	—	0.40 (0.23–0.7)	—
201–350	68	9 (13.2)	0.44 (0.22–0.88)	—	0.47 (0.24–0.91)	—
351–500	27	1 (3.7)	0.12 (0.02–0.87)	—	0.13 (0.02–0.88)	—
>500	29	5 (17.2)	0.58 (0.25–1.36)	—	0.61 (0.26–1.46)	—
Missing data	1,089	306 (28.1)	—	—	—	—	—
ART use
No	117	90 (76.9)	Reference	Reference	Reference	Reference	Reference
Yes	1,322	274 (20.7)	0.27 (0.23–0.31)	—	—	0.25 (0.22–0.29)
ART initiation in relation to start of TB
>3 months before TB treatment	392	83 (21.2)	Reference	Reference	Reference	Reference	Reference
0–3 months before TB treatment	165	56 (33.9)	0.84 (0.63–1.12)	—	—	—	1.8 (1.34–2.43)
<2 weeks after TB treatment	515	73 (14.2)	1.34 (0.99–1.82)	—	—	—	0.73 (0.54–0.98)
2–8 weeks after TB treatment	13	2 (15.4)	0.56 (0.41–0.76)	—	—	—	0.88 (0.24–3.19)
Not recorded	237	60 (25.3)	—	—	—	—	—

RR = relative risk; CI = confidence interval; TB = tuberculosis; PTB = pulmonary TB; EPTB = extrapulmonary PTB; HIV = human immunodeficiency virus; WHO = World Health Organisation; ART = antiretroviral therapy.
variables refer only to those who were diagnosed as HIV-positive.
generalized linear regression models 2, 3, and 4 exclude HIV-negative and HIV unknown patients.
Model 1 assesses HIV status whilst adjusting for the potential confounding effects of age, type of health facility, and TB category.
Model 2 assesses WHO clinical staging and CD4 count at ART initiation whilst adjusting for the potential confounding effect of age, type of health facility, and TB category.
Model 3 assesses ART use (yes/no) whilst adjusting for the potential confounding effect of age, type of health facility, and TB category.
Model 4 assesses “ART timing in relation to start of TB treatment” whilst adjusting for potential confounding effect of age, type of health facility, and TB category.