Correlates of Delayed Diagnosis among Human Immunodeficiency Virus-Infected Pulmonary Tuberculosis Suspects in a Rural HIV Clinic, South Africa

Table 4

Univariate and multivariate Cox proportional hazards regression analysis.

Characteristic	Univariate HR (95% CI)	𝑃-value	Multivariate HR (95% CI)	𝑃 value
Patient age
18 to 40 years	1		1
>40 years	1.8 (0.53–1.32)	0.05	1.57 (0.25–0.90)	0.02
Sex
Female	1
Male	1.26 (0.79–2.03)	0.33	—	—
Education level
No education	1
Primary	2.62 (1.28–5.34)	0.25
Secondary	7.09 (0.87–7.56)	0.17
Tertiary	1.54 (0.79–2.95)	0.19
Method of TB diagnosis
History and CXR	1
Smear positive	0.83 (0.48–2.14)	0.95	—	—
Sputum culture positive	2.09 (1.19–3.65)	0.09
Viral load (copies/mL)
<400	1		1
>400	2.3 (0.87–1.56)	0.07	1.89 (0.74–1.63)	0.06
CD4 Count (cells/mm^3)
<50	1
50–200	0.56 (0.18–1.72)	0.31	—	—
>200	0.64 (0.20–1.64)	0.42
ART use at PTB diagnosis
No	1		1
Yes	0.82 (1.05–1.3)	0.04	0.49 (0.36–0.97)	0.04
TB treatment outcome
Completed treatment	1
Died	2.42 (1.07–5.47)	0.03	—	—
Lost to follow-up	0.92 (0.49–1.73)	0.81
Still on treatment	0.79 (0.43–1.46)	0.45

HR: hazard ratio; CI: confidence interval.