Skip to main content
. 2009 Oct 13;88(1):39–48. doi: 10.2471/BLT.08.057901

Table 3. Infant malnutrition at the first immunization visita as a predictor of death between then and 6 months of age, Ghana, India and Peru, 1995–1997.

Malnutrition indicator at first immunization visitb No. of deaths
Sensitivity Specificity AUC (95% CI)
among malnourished among non-malnourished
Ghana (n = 2637)
WHO child growth standards
Stuntingc 9 of 627 21 of 2007 30.0 76.3 0.53 (0.45–0.62)
Wastingd 0 of 21 29 of 2587 0.0 98.9 0.50 (0.49–0.50)
Underweight 7 of 331 23 of 2306 23.3 87.6 0.55 (0.48–0.63)
NCHS growth reference
Stuntingc 5 of 288 25 of 2346 16.7 89.1 0.53 (0.46–0.60)
Wastingd 0 of 3 27 of 2523 0.0 99.9 0.50 (0.50–0.50)
Underweight 3 of 99 27 of 2538 10.0 96.3 0.53 (0.48–0.58)

India (n = 3718)
WHO child growth standards
Stuntingc 31 of 1 049 16 of 2638 66.0 72.0 0.69 (0.62–0.76)
Wastingd 23 of 486 22 of 3114 51.1 85.0 0.65 (0.56–0.74)
Underweight 47 of 1458 10 of 2260 82.5 61.5 0.72 (0.67–0.77)
Severe underweighte 40 of 561 17 of 3157 70.2 85.8 0.78 (0.72–0.84)
NCHS growth reference
Stuntingc 24 of 599 23 of 3088 51.1 84.2 0.68 (0.60–0.75)
Wastingd 7 of 50 23 of 3426 23.3 98.8 0.61 (0.53–0.69)
Underweight 37 of 446 20 of 3272 64.9 88.8 0.77 (0.7–0.83)

Peru (n = 2251)
WHO child growth standards
Stuntingc 1 of 149 12 of 2098 7.7 93.4 0.51 (0.43–0.58)
Wastingd 0 of 16 13 of 2230 0.0 99.2 0.5 (0.5–0.5)
Underweight 1 of 112 12 of 2139 7.7 95.0 0.51 (0.44–0.59)
NCHS growth reference
Stuntingc 0 of 46 13 of 2201 0.0 97.9 0.49 (0.49–0.49)
Wastingd 0 of 0 13 of 2241 0.0 100.0 0.5 (0.5–0.5)
Underweight 0 of 21 13 of 2230 0.0 99.1 0.5 (0.49–0.5)

AUC, area under the receiver operating characteristic curve; CI, confidence interval; NCHS, National Center for Health Statistics; WHO, World Health Organization.
a The first immunization visit took place at 6 weeks in Ghana and India and at 10 weeks in Peru.
b Malnutrition indicators were determined using either WHO child growth standards or the NCHS growth reference.
c Stunting could not be assessed for 3, 31 and 4 infants in Ghana, India and Peru, respectively.
d Wasting could not be assessed for 29, 118 and 5 infants in Ghana, India and Peru, respectively.
e Figures for the severe malnutrition indicator were calculated only when the malnutrition indicator had a high sensitivity and a low specificity.