Table 4 Poisson regression with multilevel analysis, with health facility as cluster variable in Northern Tanzania. Interaction effects for mortality among patients with HIV adjusted for the significant effects included in Table 3 (N = 82,925)
From: Effect of tuberculosis infection on mortality of HIV-infected patients in Northern Tanzania
Characteristics | ARR (95%CI) PLHIV without TB | ARR (95%CI) PLHIV/TB co-infection * | Effect of TB infection in baseline ** |
|---|---|---|---|
Age | |||
15–24 | 1 | 2.06 (1.42–2.99) | |
25–34 | 1.27 (1.07–1.52) | 0.97 (0.66–1.43) | |
35–44 | 1.36 (1.14–1.61) | 0.91 (0.63–1.33) | |
45–55 | 1.34 (1.12–1.60) | 0.93 (0.63–1.37) | |
Above 55 | 1.98 (1.65–2.37) | 1.26 (0.84–1.90) | |
Sex | |||
Male | 1 | 1.41 (1.22–1.62) | |
Female | 0.57 (0.53–0.61) | 0.63 (0.53–0.75) | |
Body weight | |||
Below 40 kg | 1 | 0.89 (0.70–1.13) | |
40–60 kg | 0.46 (0.42–0.52) | 0.80 (0.62–1.02) | |
Above 60 kg | 0.22 (0.20–0.26) | 0.58 (0.42–0.80) | |
HIV WHO stage | |||
Stage 1 | 1 | 1.49 (1.01–2.19) | |
Stage 2 | 1.35 (1.19–1.53) | 1.53 (0.99–2.36) | |
Stage 3 | 1.50 (1.33–1.68) | 1.42 (0.96–2.09) | |
Stage 4 | 2.22 (1.95–2.54) | 1.74 (1.16–2.62) | |
Region | |||
Arusha | 1 | 2.02 (1.58–2.59) | |
Kilimanjaro | 0.89 (0.77–1.03) | 0.63 (0.46–0.86) | |
Tanga | 0.86 (0.76–0.97) | 0.61 (0.48–0.78) | |
Functional status | |||
Bedridden | 1 | 1.15 (0.90–1.48) | |
Ambulatory | 2.57 (1.88–3.52) | 1.11 (0.54–2.31) | |
Working | 0.59 (0.52–0.67) | 0.81 (0.64–1.03) | |
Nutritional status | |||
Ok | 1 | 1.61 (1.45–1.79) | |
Moderate | 1.67 (1.50–1.86) | 1.28 (1.03–1.61) | |
Severe | 2.40 (1.71–3.35) | 0.63 (0.35–1.12) | |
Year of follow-up | |||
2012 | 1 | 1.36 (1.05–1.74) | |
2013 | 0.83 (0.74–0.93) | 0.87 (0.63–1.20) | |
2014 | 0.73 (0.65–0.82) | 0.91 (0.68–1.24) | |
2015 | 0.71 (0.64–0.80) | 0.71 (0.52–0.97) | |
2016 | 0.56 (0.50–0.63) | 0.46 (0.33–0.64) | |
2017 | 0.44 (0.38–0.51) | 0.72 (0.51–1.01) | |