Skip to main content

Table 1 Univariate logistic regression analysis to identify factors associated with molecular clusters, i.e. as a surrogate for recent transmission

From: Transmission patterns of rifampicin resistant Mycobacterium tuberculosis complex strains in Cameroon: a genomic epidemiological study

  Ungrouped (n = 110) % Clustered (n = 85) % Univariate logistic regression
OR 95% lower 95% upper P value
Age         
  > 50 18 16.4 6 7.1 REF    
 30–50 51 46.4 44 51.8 2.6 0.9 7.1 0.06
  < 30 34 30.9 30 35.3 2.6 0.9 7.5 0.07
 Unknown 7 6.4 5 5.9 2.1 0.5 9.4 0.31
Gender         
 Female 41 37.3 37 43.5 REF    
 Male 66 60 46 54.1 0.8 0.4 1.4 0.39
 Unknown 3 2.7 2 2.4 0.7 0.1 4.7 0.75
HIV status         
 Negative 50 45.5 51 60 REF    
 Positive 33 30 20 23.5 0.6 0.3 1.2 0.13
 Unknown 27 24.5 14 16.5 0.5 0.2 1.1 0.08
Lineage         
 Others 7 6.4 10 11.8 REF    
 L4.X 49 44.5 26 30.6 0.4 0.1 1.1 0.07
 L4.6 54 49.1 49 57.6 0.6 0.2 1.8 0.39
Genotypic resistances (n) 1.7 (mean)   2.4 (mean)   1.7 1.3 2.3 6.64E−05
(per unit increase)
Compensatory mutation         
 No 85 77.3 53 62.4 REF    
 Yes 25 22.7 32 37.6 2.1 1.1 3.8 0.02
High dose isoniazid         
 Susceptible 39 35.5 5 5.9 REF    
 Resistant 71 64.5 80 94.1 8.8 3.3 23.5 1.53E−05
Ethambutol         
 Susceptible 62 56.4 22 25.9 REF    
 Resistant 48 43.6 63 74.1 3.7 2 6.8 3.02E−05
Pyrazinamide         
 Susceptible 70 63.6 41 48.2 REF    
 Resistant 40 36.4 44 51.8 1.8 1.1 3.3 0.03
Prothionamide         
 Susceptible 87 79.1 69 81.2 REF    
 Resistant 23 20.9 16 18.8 0.9 0.4 1.8 0.72
  1. P-values < 0.05 indicated in bold text
  2. Factors with P ≤ 0.1 were included in backwards exclusion/forward selection procedures to obtain the best supported multiple logistic regression model, which results are reported in the main text only
  3. OR odds ratio