Skip to main content
. Author manuscript; available in PMC: 2011 Jul 27.
Published in final edited form as: JAMA. 2009 Jan 21;301(3):277–285. doi: 10.1001/jama.2008.1018

Table 3.

Effect of RUTF supplementation on mortality and morbidity

Non-intervention Intervention
Mortality
 N1 1,862 1,671
 No events / child-year 18 / 1,099 7 / 986
 Incidence rate / child year2 0.016 (0.011, 0.026) 0.007 (0.003, 0.015)
 Incidence rate ratio (95% CI) 1.00 0.43 (0.18, 1.04)
 Adjusted Hazard Ratio3 (95% CI) 1.00 0.51 (0.25, 1.05)
Malaria
 N1 1,862 1,671
 No. visits with diagnosis / Total no. of visits 721 / 12,789 330 / 11,542
 Prevalence4 (%) 5.64 (1.53, 9.74) 2.86 (0.78, 4.94)
 Prevalence ratio (95% CI) 1.00 0.51 (0.45, 0.58)
 Adjusted OR (95% CI)5 1.00 0.76 (0.51, 1.13)
Diarrhea
 N1 1,862 1,671
 No. visits with diagnosis / Total no. of visits 170 / 12,789 156 / 11,542
 Prevalence4 (%) 1.33 (1.03, 1.63) 1.35 (0.74, 1.96)
 Prevalence ratio (95% CI) 1.00 1.02 (0.82, 1.26)
 Adjusted OR (95% CI) 5 1.00 1.07 (0.88, 1.28)
Respiratory Infection
 N1 1,862 1,671
 No. visits with diagnosis / Total no. of visits 114 / 12,789 117 / 11,542
 Prevalence4 (%) 0.89 (0.37, 1.41) 1.01 (0.44, 1.59)
 Prevalence ratio (95% CI) 1.00 1.14 (0.88, 1.47)
 Adjusted OR (95% CI)5 1.00 1.21 (0.89, 1.63)
1

Number of children contributing to crude analysis.

2

Incidence rates by intervention group were estimated by taking the mean of the corresponding village incidence rates, weighted by the person-months of observation from each village that contributed to the mean.

3

Adjusted hazard ratios estimated from a marginal Cox proportional hazards model with time from recruitment to the event as the outcome and predictors that included intervention group, child age at recruitment, sex, baseline HAZ, and district.

4

Prevalence was calculated by summing the number of visits the child had the morbidity diagnosis divided by the number of visits. Mean prevalence is calculated by taking the mean of the village prevalence weighted by the person-months of observation from each village.

5

Adjusted odds ratios estimated from generalized linear mixed effect models with presence of the morbidity as the outcome and predictors included intervention group, child age at recruitment, sex, baseline HAZ score, district, and calendar month.