Skip to main content

Advertisement

Table 2 Age-adjusted logistic regression models predicting respiratory distress (N = 130)

From: Plasmodium falciparumantigenic variation: relationships between widespread endothelial activation, parasite PfEMP1 expression and severe malaria

   Respiratory distress  
Models Explanatory variables OR (95% CI) P
1 $ Rosette frequency 5.7(1.6, 20.6) 0.008
2 $ Group A-like 1.76(0.49, 6.34) 0.39
3 ang-2 4.1(2.0, 8.55) 0.0001
4 $ parasite density (peripheral) 2.59(1.56, 4.3) 0.0002
5 PfHRP2 1.58(1.18, 2.11) 0.002
6 $ IE surface antibodies 0.98(0.96, 1.0) 0.1
7 Rosette frequency 3.0(0.7, 12.9) 0.14
ang-2 3.5(1.67, 7.3) 0.001‡
8 $ Rosette frequency 2.0(0.51, 8.3) 0.3
parasite density (peripheral) 2.4(1.4, 4.0) 0.001‡
9 Rosette frequency 3.54(0.93- 13.42) 0.06
  PfHRP2 1.5(1.11, 2.03) 0.008
10 ang-2 3.2(1.5, 6.99) 0.003‡
parasite density (peripheral) 2.17(1.30, 3.62) 0.003‡
11 Ang-2 3.35(1.56, 7.20) 0.002‡
  PfHRP2 1.37(1.0, 1.87) 0.044
12 Rosette frequency 1.25(0.26, 6.0) 0.78
Parasite density (peripheral) 2.1(1.24, 3.6) 0.006‡
ang-2 3.1(1.4, 6.9) 0.005‡
13 rosette 2.18(0.49, 9.70) 0.31
  PfHRP2 1.34(0.98, 1.83) 0.07
  Ang-2 3.01(1.38, 6.56) 0.005‡
  1. The analysis in this table is similar to that in Table 1 except the outcome of interest is respiratory distress. Ang-2 seems to be in the causal pathway linking rosetting frequency to respiratory distress (models 7, 12 and 13). ‡ = variables that improved the fit of the model using LR χ2 improvement test. The analysis in this table is based on 130 samples with rosette frequency data. PfHRP2 data was obtained for 126 out of the 130 samples. $ indicate models published in [10] and repeated here for comparison.