Table 2. Univariate analysis of potential risk factors for pneumonia in HIV-positive and HIV-negative sub-groups.
| Exposures | HIV-positive subgroup | HIV-negative subgroup | ||||||
|---|---|---|---|---|---|---|---|---|
| Cases
(n= 117) |
Controls
(n= 169) |
Unadjusted OR
(95% CI) |
p-value | Cases
(n= 28) |
Controls
(n= 84) |
Unadjusted OR
(95% CI) |
p-value | |
| Participant characteristics | ||||||||
| Age (years) †, median (IQR) | 36 (31-43) | 36 (32-44) | -- | -- | 39 (30-64) | 35 (26-42) | -- | -- |
| Gender
†, n (%)
Male (reference) Female |
68 (58.1) 49 (41.9) |
93 (55.0) 76 (45.0) |
-- -- |
-- -- |
23 (82.1) 5 (17.9) |
54 (64.3) 30 (35.7) |
-- -- |
-- -- |
| Alcohol intake
††, n (%)
Never (reference) Previous drinker Current drinker |
63 (53.9) 46 (39.3) 8 (6.8) |
94 (55.6) 40 (23.7) 35 (20.7) |
1 1.72 (1.01-2.92) 0.34 (0.15-0.78) |
-- 0.046 0.011 |
11 (39.3) 13 (46.4) 4 (14.3) |
52 (61.9) 15 (17.9) 17 (20.2) |
1 4.10 (1.53-11.00) 1.11 (0.31-3.96) |
-- 0.005 0.869 |
| Smoking status (all forms)
††, n (%)
Never smoked (reference) Ex-smoker Current smoker |
85 (72.7) 28 (23.9) 4 (3.4) |
123 (72.8) 27 (16.0) 19 (11.2) |
1 1.50 (0.83-2.73) 0.30 (0.10-0.93) |
-- 0.182 0.036 |
13 (26.4) 10 (35.7) 5 (17.9) |
68 (81.0) 7 (8.3) 9 (10.7) |
1 7.47 (2.41-12.2) 2.9 1 (0.84-10.08) |
-- 0.001 0.093 |
| Socioeconomic status quintile
†††
Highest (reference) High Middle Low Lowest |
19 (16.2) 25 (21.4) 26 (22.2) 22 (18.8) 25 (21.4) |
26 (15.4) 40 (23.7) 33 (19.5) 37 (21.9) 33 (19.5) |
1 0.86 (0.39-1.86) 1.08 (0.49-2.36) 0.81 (0.37-1.80) 1.04 (0.48-2.28) |
-- 0.692 0.851 0.610 0.929 |
6 (21.4) 2 (7.1) 7 (25.0) 3 (10.7) 10 (35.7) |
28 (33.3) 14 (16.7) 15 (17.9) 16 (19.1) 11 (13.1) |
1 0.67 (0.12-3.74) 2.18 (0.62-7.66) 0.86 (0.19-3.98) 4.24 (1.24-14.50) |
-- 0.645 0.225 0.863 0.021 |
| Participant health characteristics | ||||||||
| Body mass index (kg/m 2) ††, mean (STD) | 19.9 (2.5) | 21.6 (3.9) | 0.85 (0.78-0.92) ‡ | < 0.001 | 20.9 (3.8) | 23.2 (4.9) | 0.84 (0.72-0.98) ‡ | 0.023 |
| CD4 count (cells/µl) ††, median (IQR) | 129 (49-209) | 355 (236-492) | 0.99 (0.99-0.99) ‡ | < 0.001 | -- | -- | ||
| Antiretroviral therapy
††, n (%)
No (reference) Yes |
49 (41.9) 68 (58.1) |
39 (23.1) 130 (76.9) |
1 0.42 (0.25-0.70) |
-- 0.001 |
-- -- |
-- -- |
-- -- |
-- -- |
| Cotrimoxazole prophylaxis
††, n (%)
No (reference) Yes |
50 (42.7) 67 (57.3) |
42 (24.9)
127 (75.2) |
1 0.44 (0.27-0.73) |
-- 0.002 |
-- -- |
-- -- |
-- -- |
-- -- |
| Chronic respiratory disease
††, n (%)
No (reference) Yes |
6 (5.13) 111 (94.9) |
89 (52.7) 80 (47.3) |
1 20.58 (8.58-49.38) |
-- < 0.001 |
1 (3.6) 27 (96.4) |
67 (79.8) 17 (20.2) |
1 106.41 (13.49-839.66) |
-- < 0.001 |
| Previous respiratory diagnosis, n (%)
No (reference) Yes |
14 (12.0) 103 (88.0) |
103 (61.0) 66 (39.1) |
1 11.48 (6.07-21.73) |
-- < 0.001 |
2 (7.1) 26 (92.9) |
31 (86.9) 11 (13.1) |
1 86.27 (17.92-415.40) |
-- < 0.001 |
| Previous chronic respiratory
symptoms, n (%) No (reference) Yes |
8 (6.8) 109 (93.2) |
92 (54.4) 77 (45.6) |
1 16.28 (7487-35.01) |
-- < 0.001 |
1 (3.6) 27 (96.4) |
69 (82.1) 15 (17.9) |
1 124.20 (15.63-986.79) |
-- < 0.001 |
| FEV
1 % of predicted
*, median (IQR)
(n=349) |
60.2 (54.2-73.3) | 70.3 (62.0-81.0) | 0.98 (0.96-0.99) ‡ | 0.006 | 57.0 (41.7-66.0) | 71.4 (61.6-79.7) | 0.93 (0.90-0.97) ‡ | < 0.001 |
| FVC % of predicted
*, median (IQR)
(n=349) |
74.7 (65.4-82.6) | 81.5 (71.7-88.6) | 0.97 (0.95-0.99) ‡ | 0.001 | 73.0 (64.6-80.4) | 82.2 (73.1-89.6) | 0.93 (0.89-0.97) ‡ | 0.002 |
| Spirometric classification
*, n (%)
(n=349) Normal (reference) Obstructive Restrictive |
28 (27.5) 17 (16.7) 57 (55.8) |
76 (51.7) 17 (11.6) 54 (36.7) |
1 2.71 (1.22-6.04) 2.87 (1.61 – 5.07) |
-- 0.014 <0.001 |
6 (26.1) 8 (34.8) 9 (39.1) |
37 (48.1) 10 (13.0) 30 (39.0) |
1 4.93 (1.39-17.54) 1.85 (0.59-5.78) |
-- 0.014 0.290 |
| Pollution exposures | ||||||||
| Mean ambulatory PM
2.5 exposure
(µg/m 3) ††, median (IQR) |
60.4 (41.0-103.0) | 55.2 (34.6-89.1) | 1.00 (1.00-1.00) ‡ | 0.145 | 70.7 (50.3-109.4) | 56.7 (41.3-92.3) | 1.00 (1.00-1.01) ‡ | 0.410 |
| Mean Ambulatory CO exposure
(ppm) ††, median (IQR) |
6.0 (2.7-11.3) | 4.5 (2.5-9.2) | 1.03 (1.00-1.07) ‡ | 0.047 | 3.1 (1.2-7.4) | 4.7 (1.0-11.5) | 0.93 (0.85-1.01) ‡ | 0.079 |
| Mean household PM
2.5 exposure
* (µg/m 3) †††, median (IQR) (n=258) |
125.2 (77.4-254.9) | 167.1 (90.6-311.9) | 1.00 (1.00-1.00) ‡ | 0.559 | 189.5 (132.4-344.3) | 132.4 (69.5-292.1) | 1.00 (1.00-1.00) ‡ | 0.074 |
| Mean household CO exposure
(ppm) ††, median (IQR) |
6.9 (2.8-13.6) | 5.4 (2.9-11.8) | 1.02 (1.00-1.04) ‡ | 0.109 | 4.5 (2.4-8.7) | 7.5 (3.6-16.1) | 0.96 (0.90-1.01) ‡ | 0.121 |
| Cooking with solid fuel frequency
††, n (%)
Cooks rarely (reference) Cooks occasionally Cooks sometimes Cooks often Cooks frequently |
29 (24.8) 36 (30.8) 40 (34.2) 12 (10.3) 0 (0) |
22 (13.0) 53 (31.4) 73 (43.2) 20 (11.8) 1 (0.6) |
1 0.52 (0.26-1.03) 0.42 (0.21-0.82) 0.46 (0.18-1.13) 1 |
-- 0.062 0.011 0.088 -- |
4 (14.3) 16 (57.1) 8 (28.6) 0 (0) 0 (0) |
16 (19.1) 23 (27.4) 34 (40.5) 10 (11.9) 1 (1.2) |
1 2.78 (0.78-9.89) 0.94 (0.25-3.59) 1 1 |
-- 0.114 0.929 -- -- |
| Primary cooking fuel
†††, n (%)
Electricity (reference Wood Charcoal Plastic Bottles |
9 (7.7) 17 (14.5) 91 (77.8) 0 (0) |
13 (7.7) 26 (15.4) 130 (76.9) 0 (0) |
1 0.94 (0.33-2.69) 1.01 (0.41-2.46) -- |
-- 0.915 0.981 -- |
1 (3.6) 8 (28.6) 18 (64.3) 1 (3.6) |
10 (11.9) 6 (7.1) 68 (81.0) 0 (0) |
1 13.33 (1.32-134.61) 2.65 (0.32-22.06) 1 |
-- 0.028 0.368 -- |
| Ventilation whilst cooking
†††, n (%)
Mainly cooks outside or only uses electricity (reference) Mainly cooks inside with ventilation Mainly cooks inside without ventilation |
22 (18.8) 73 (62.4) 22 (18.8) |
28 (16.5) 108 (63.5) 34 (20.0) |
1 0.86 (0.46-1.62) 0.82 (0.38-1.79) |
-- 0.641 0.623 |
2 (7.14) 19 (67.9) 7 (25.0) |
14 (16.7) 66 (78.6) 4 (4.8) |
1 2.02 (0.42-9.66) 12.25 (1.79-83.95) |
1 0.381 0.011 |
| Pollution from heating/lighting
†††, n (%)
No Yes |
105 (89.7) 12 (10.3) |
155 (91.2) 15 (8.8) |
1 1.18 (0.53-2.62) |
-- 0.683 |
23 (82.1) 5 (17.9) |
78 (92.9) 6 (7.1) |
1 2.83 (0.79-10.11) |
-- 0.110 |
*Missing data was not imputed. ‡ Per unit change. † A priori forced variable included in the logistic regression model. †† A priori potential confounder with Likelihood Test Ratio p-value <0.2 therefore entered into the logistic regression model (note PM 2.5 and CO exposures included as potential confounders for CRD analysis only). ††† A priori potential confounder with Likelihood Test Ratio p-value >0.2 therefore not entered into the logistic regression model.
OR: odds ratio; CI: confidence interval; IQR: interquartile range; STD: standard deviation; FEV 1: forced expiratory volume in 1 second; FVC: forced vital capacity; PM 2.5: particulate matter <2.5µm; CO: carbon monoxide; ppm: parts per million.