Volume 18, Number 9—September 2012
Dispatch
Infectious Diseases in Children and Body Mass Index in Young Adults
, Catherine Ley, and Julie ParsonnetTable 2
Independent predictors of BMI in a cohort of persons during childhood and during adulthood, the Philippines, 1983–2005 *†
| Predictor | BMI, age ≈24 mo |
BMI, age ≈22 y |
|||
|---|---|---|---|---|---|
| Parameter estimate | p value | Parameter estimate | p value | ||
| Visits with reported URI‡ | –0.25 | 0.14 | 1.13§ | 0.01 | |
| Visits with current fever | –1.41 | <0.001 | –1.61 | 0.11 | |
| Visits with reported diarrhea | 0.23 | 0.27 | –1.05 | 0.06 | |
| Visits during which child was observed crawling in unsanitary conditions | –0.05 | 0.65 | –0.72 | 0.02 | |
*BMI, body mass index; URI, upper respiratory infection.
†Models were adjusted for maternal schooling/training (<7 or >7 y), sex and birth weight. R2: BMI age 24 mo, 0.08; BMI age 22 y, 0.03.
‡A model that used number of visits instead of proportion of visits identified very similar findings, as did a model that limited analysis to those with >10 visits only.
§The parameter estimate corresponds to the change in BMI when participants with no visits (0%) with infection are compared to those with all visits (100%) with infection. Alternatively, a 10% increase in reported visits with infection corresponds to an increase of 10% of the parameter in BMI in adulthood. For example, a 10% increase in visits with reported URI corresponds to a 0.113 kg/m2 increase in adult BMI.
