Skip to main content
. 2015 Dec 9;93(6):1224–1230. doi: 10.4269/ajtmh.15-0256

Table 3.

Bivariate and multivariate analysis by AFT model for time to interruption (of NFX treatment)

Covariate Bivariate analysis Multivariate analysis
Coefficient (βi) TR exp(βi) 95% CI P value Coefficient (βi) TR exp(βi) 95% CI P value
Severity
 Severe Ref Ref Ref Ref
 Mild/moderate 1.82 6.20 (2.84–13.6) < 0.001 1.88 6.52 (3.24–13.1) < 0.001
Doses mg/kg/day
 > 8 Ref Ref Ref Ref
 ≤ 8 1.01 2.77 (0.89–8.59) 0.077 0.57 1.78 (0.90–3.49) 0.095
Age (years)
 ≥ 48 Ref Ref Ref Ref
 < 48 −0.08 0.92 (0.34–2.49) 0.874 0.45 1.57 (0.90–2.74) 0.109
ADRs
 > 3 Ref Ref Ref Ref
 ≤ 3 0.92 2.51 (1.39–4.51) 0.002 0.58 1.78 (1.04–3.03) 0.033
Sex
 Male Ref Ref
 Female 0.55 1.72 (0.74–4.01) 0.203
History of previous pathologies
 Yes Ref Ref
 No 0.67 1.95 (0.82–4.64) 0.129
Alteration of leukocytes
 Yes Ref Ref
 No 0.01 1.01 (0.23–4.40) 0.921
Alteration of lymphocyte
 Yes Ref Ref
 No 0.29 1.33 (0.70–2.56) 0.378
Alteration of neutrophils
 Yes Ref Ref
 No 0.46 1.59 (0.51–5.01) 0.426
Alteration of eosinophil
 Yes Ref Ref
 No 0.08 1.08 (0.57–2.07) 0.795

ADRs = adverse drug reactions; AFT = accelerated failure time model; CI = confidence interval; NFX = nifurtimox; Ref = reference variable; TR = time ratio.

Bold values are statistically significant results P < 0.005.