Skip to main content

Table 3 Multivariate analysis of potential risk factors for the suspected toxic syndrome

From: Suspected paracetamol overdose in Monrovia, Liberia: a matched case–control study

 

Cases

(N = 30)

Hospital Controls (N = 53)

p-value

Community Controls (N = 48)

p-value

Pooled Controls (N = 101)

n(%)

n(%)

OR

(95% CI)

aORa

(95%CI)

n(%)

OR(95% CI)

aOR*

(95%CI)

n(%)

OR

(95% CI)

aORa

(95%CI)

p-value

Paracetamol (total)

26(87)

28(53)

   

46(96)

   

74(73)

   

Paracetamol supratherapeutic dose

9(30)

5(9)

3.8 (1.1–12.4)

3.7 (1.0–13.1)

0.046

2(4)

13.6 (1.7–110.1)

13.6 (1.7–110.1)

0.014

7(7)

5.6 (1.9–16.9)

6.6 (2.1–21.3)

0.001

Antibiotics

16(53)

16(30)

   

42(88)

   

58(57)

   

Co-trimoxazole

7(23)

3(6)

4.4 (1.0–17.1)

4.3 (1.0–18.6)

0.048

17(35)

0.6 (0.2–2.0)

  

20(20)

1.3 (0.5–3.3)

  

Erythromycin

2(7)

0(0)

   

4(8)

0.9 (0.2–4.9)

  

4(4)

1.9 (0.3–10.4)

  

Anti-malaria Medication

11(37)

12(23)

   

36(75)

  

48(48)

   

Antimalarial drugs supratherapeutic doses

1(3)

0(0)

   

6(13)

0.3 (0.4–2.8)

  

6(6)

0.6 (0.1–5.3)

  

Traditional Medicine

20(67)

18(34)

   

22(46)

   

40(40)

   

“Tonton” Leaf

4(13)

4(8)

1.7 (0.4–6.8)

  

5(10)

1.4 (0.3–5.9)

  

9(9)

1.4 (0.4–5.3)

  

“Monkey Apple” Leaf

3(10)

0(0)

   

4(8)

1.4 (0.3–6.1)

  

4(4)

2.4 (0.5–10.8)

3.4 (0.7–17.3)

0.148

“Wild Wild” Leaf

2(7)

3(6)

1.2 (0.2–7.0)

  

2(4)

2.0 (0.3–14.1)

  

5(5)

1.4(0.3–7.5)

  

Pre-prepared traditional treatments

3(10)

1(2)

5.2 (0.5–50.4)

10.9 (0.7–160.9)

0.081

0(0)

   

1(1)

11.2 (1.2–108.1)

  

“Lynch Street” Medicine

1(3)

2(4)

1 (0.0–11.0)

  

2(4)

0.6 (0.1–7.0)

  

4(4)

0.9 (0.1–8.0)

  

Almond Bark

0(0)

0(0)

   

4(8)

  

4(4)

   

Other traditional treatments

7(23)

10(19)

1.4 (0.5–4.6)

  

10(21)

0.9 (0.3–3.0)

  

20(20)

1.3 (0.5–3.5)

  
  1. aAdjusted odds ratios (aORs) are derived from stepwise conditional logistic regression models. The initial multivariate models included all the variables which were included in the univariate analysis presentedby the odds ratio (OR). Non-significant variables were eliminated to reach the final model as presented