Skip to main content

Table 2 Patient level factors associated with pretreatment loss to follow up in a multilevel logistic regression model

From: Patient and health system factors associated with pretreatment loss to follow up among patients diagnosed with tuberculosis using Xpert® MTB/RIF testing in Uganda

Characteristic

Initiated on Rx

N = 410

Not Initiated on Rx

N = 100

Crude Odds Ratio

(95% CI)

Adjusted Odds Ratio

(95% CI)

Sex

 Male

256 (82.1)

57 (17.9)

reference

 Female

154 (77.8)

43 (22.2)

1.30 (0.84–2.04)

Age

 15–24

92 (80.0)

23 (20.0)

reference

 25–34

130 (80.3)

32 (19.7)

0.98 (0.54–1.79)

 35–44

91 (81.9)

20 (18.1)

0.88 (0.45–1.70)

 45–54

51 (77.2)

15 (22.8)

1.17 (0.56–2.46)

  > 55

46 (82.1)

10 (17.9)

0.88 (0.39–2.00)

Phone No.

 Yes

298 (92.3)

25 (8.7)

reference

reference

 No

112 (59.9)

75 (40.1)

8.99 (5.17–15.64)

7.93 (3.93–13.05)

aDistance from health facility (n = 469)

  > 35 km

64 (87.7)

9 (12.3)

reference

 21-35 km

124 (87.3)

18 (12.7)

1.04 (0.44–2.49)

 6-20 km

106 (86.9)

16 (13.1)

1.12 (0.47–2.69)

  < =5 km

103 (81.9)

29 (18.1)

2.08 (0.93–4.67)

bHIV status (n = 479)

 HIV negative

281 (88.4)

37 (11.6)

reference

reference

 HIV positive

129 (80.1)

32 (19.9)

1.86 (1.10–3.12)

1.88 (1.09–3.26)

cART Status (n = 138)

 On ART before TB diagnosis

91 (90.1)

10 (9.9)

reference

 Not on ART before TB diagnosis

34 (91.9)

3 (8.1)

0.61 (0.14–2.61)

  1. a 41 patients (13 who were and 28 who were not initiated on treatment) did not have data on distance from health facility
  2. b31 patients who were not initiated on TB treatment did not have HIV status recorded
  3. c23 patients (4 who were and 19 who were not initiated on TB treatment) did not have data on ART status