Natural history parameters for enteric pathogens to inform modeling studies of diarrhea among children in low-resource settings: Results from the MAL-ED longitudinal birth cohort
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Policy decisions may rely on mathematical modelling to predict intervention impacts. Information for key model parameters is limited for most enteric pathogens. To support informed modeling, we characterized incidence, severity, and duration for adenovirus 40/41, astrovirus, Campylobacter jejuni and C. coli, Cryptosporidium , norovirus GII, rotavirus, sapovirus, Shigella , heat-stable enterotoxin-producing enterotoxigenic Escherichia coli (ST-ETEC), and typical enteropathogenic E. coli (tEPEC) in a longitudinal birth cohort.
Methods
We analyzed stool specimens from MAL-ED, a multisite birth cohort with active surveillance of children in South America, southeast Asia, and sub-Saharan Africa. We defined unique infections using longitudinal test results, attributed etiologies to diarrheal episodes, calculated infection rates and disease progression probabilities, and characterized age-based trends.
Results
Most pathogens had peak infection rates around 7 to 9 months of age, with incidence gradually decreasing in the second year of life. In contrast, Cryptosporidium and ST-ETEC incidence plateaued after 9 months of age instead of declining. Shigella incidence continually increased in the first two years of life. The likelihood of developing diarrhea increased with age for Shigella but decreased with age for adenovirus 40/41, Campylobacter jejuni and C. coli , and tEPEC. The likelihood of attributable diarrhea becoming severe decreased with age.
Conclusions
The age at peak infection burden and peak disease burden were not necessarily the same for a given pathogen. Each pathogen evaluated had its own distinct age trends. These results could support informed modelling of impacts of interventions for specific enteric pathogens, particularly in low-resource settings.
