Skip to main content

Table 3 Prevalence of child growth indicators in the initial dataset and after removing population and longitudinal outliers according to different cutoffs - boys. Brazilian Food and Nutrition Surveillance System (SISVAN), 2008–2017

From: Identifying biologically implausible values in big longitudinal data: an example applied to child growth data from the Brazilian food and nutrition surveillance system

 

Initial

dataset

Dataset after

removing POs

Dataset after removing

children with only

one measurement

Dataset after removing LOs according to different cutoffs

 

< -3/ > +3

< -4/ > +4

< -5/ > +5

< -6/ > +6

Severe stunting

4.75

4.36

4.32

3.87

4.19

4.29

4.32

Moderate stunting

7.70

7.93

7.92

7.98

7.95

7.93

7.93

Adequate

84.60

87.71

87.75

88.15

87.85

87.78

87.76

BIV

2.89

-

-

-

-

-

-

Missing

0.06

-

-

-

-

-

-

Severe underweight

1.17

0.99

0.98

0.82

0.91

0.95

0.97

Moderate underweight

3.19

3.19

3.19

3.11

3.17

3.18

3.19

Adequate

86.91

89.66

84.79

90.73

90.16

89.90

89.90

Overweight

6.94

6.15

6.09

5.34

5.76

5.96

6.04

BIV

1.63

-

-

-

-

-

-

Missing

0.16

-

-

-

-

-

-

  1. Notes: POs: population outliers, LOs: longitudinal outliers, L/HAZ: length/height-for-age z-score, WAZ: weight-for-age z-score