Table 3 Cox proportional-hazards analysis of the association between anthropometric deficits at discharge and risk of relapse
Exposure variables | Proportion of relapse (%) | Univariate models | Multivariate models | |||
|---|---|---|---|---|---|---|
Hazard Ratio | p | Hazard Ratio | p | |||
Models with binary exposure variables | ||||||
WHZ at discharge | WHZ<-2 WHZ≥-2 | 45.3 (24/53) 20.0 (11/55) | 3.2 Ref. | 0.003 | 2.8 Ref. | 0.01 |
MUAC at discharge | MUAC < 125 mm MUAC ≥ 125 mm | 41.7 (15/36) 27.8 (20/72) | 1.5 Ref. | 0.26 | 1.4 Ref. | 0.32 |
HAZ at discharge | HAZ<-2 HAZ≥-2 | 32.8 (20/61) 31.9 (15/47) | 1.1 Ref. | 0.75 | NA | NA |
Duration of treatment | < 10 weeks ≥ 10 weeks | 22.8 (13/57) 43.1 (22/51) | 0.5 Ref | 0.051 | 0.7 Ref. | 0.27 |
Age at discharge | < 24 months ≥ 24 months | 32.8 (21/64) 31.8 (14/44) | 0.9 Ref. | 0.74 | NA | NA |
Sex | Male Female | 40.0 (20/50) 25.9 (15/58) | 1.7 Ref. | 0.14 | 1.6 Ref. | 0.23 |
Models with continuous exposure variables | ||||||
WHZ at discharge | NA | 0.19 | < 0.001 | 0.2 | 0.001 | |
MUAC at discharge in mm | NA | 0.94 | 0.046 | 1.0 | 0.10 | |
HAZ at discharge | NA | 0.81 | 0.24 | NA | NA | |
Duration of treatment in days | NA | 1.0 | 0.10 | 1.0 | 0.86 | |
Age at discharge in months | NA | 1.0 | 0.94 | NA | NA | |
Sex | Male Female | 40.0 (20/50) 25.9 (15/58) | 1.7 Ref. | 0.14 | 1.7 | 0.19 |