Skip to main content
. 2021 Dec 10;25(10):2855–2863. doi: 10.1017/S1368980021004808

Table 2.

The generalised linear model of log-binomial regression of the probability of diarrhoea in children aged 6–59 months

Model Diarrhoea in children PR SE P > |Z| 95 % CI
1 Drinking water
  No 1·00
  Yes 0·91 0·03 0·011 0·852 0·979
Fe supplementation (12 months)
  No 1·00
  Yes 1·19 0·03 <0·0001 1·125 1·267
Anaemia
  No 1·00
  Yes 1·17 0·03 <0·0001 1·141 1·288
2 Drinking water
  No 1·00
  Yes 0·93 0·02 0·006 0·887 0·980
Fe supplementation (7 d)
  No 1·00
  Yes 1·09 0·01 <0·0001 1·067 1·116
Anaemia
  No 1·00
  Yes 1·20 0·02 <0·0001 1·149 1·253

The General Linear Model is controlled by drinking water, sanitation service, sex, place of residence (rural or urban) and region (coastal, highlands and jungle), rotavirus vaccine, grade of education (no education, elementary, high school and college) and age (6–11, 12–23, 24–35 and 36–59 months).

Drinking water, sanitation service, Fe supplementation, rotavirus vaccination, where 0: No and 1: Yes.

The variable sex corresponds to 0: female and 1: male. Place of residence 0: urban, 1: rural grade of education, 0: no education, 1: primary, 2: secondary and 3: university. Geographic region, 1: Coast, 2: Highlands and 3: Jungle.

Ages in months are categorical variables.

Model 1 = Fe supplementation during 12 months before the survey (sample = 98 321).

Model 2 = Fe supplementation during 7 d prior to the survey (sample = 114 121).

Survey: DHS 2009–2019 (Peru).