Hello Arno,

And thank you for your interest in the study. Let me try to answer your questions:

1. Marriage status of caretakers. We investigated the relationship between this variable, a household-level risk factor, and SAM *not infection) as this was the objective of our study. Our assumption was that caretakers who live alone (either never married, divorced or widows) have less time/ resources for child care (this could potentially cover increased risk of infection due to less time for hand washing for example) which in return results in increased risk of SAM. Check the Discussion part of this article, I quote "With regard to the caregiver’s marital status and its association with SAM, the very broad confidence interval reflects the paucity of cases we encountered. This result should therefore be considered with caution. While the link between women’s social position and child nutrition is well recognized [Van den Bold M et al., 2013; Na M, Jennings L et al. 2015; Pratley P. 2016], future studies investigating position of women, autonomy and capabilities in the household and in the community could better shed lights on power relations and the mechanisms linking marital status and child undernutrition in this context."

2. Odds ratio (OR) or Crude Odds ratio (COR) are obtained when you are considering the effect of only one predictor variable ie your equation consists of only one independent variable. However when you include more variables in the analysis (confounder variables for the said relationship) you get what is called and Adjusted Odds Ratio (AOR), which takes into account the effect due to all the additional variables included in the analysis. In our study, we reported OR or COR for bi-variate analysis which looked at the relationship between individual variables and SAM. On contrary, we reported AOR for those variables that were included in the multi-variable models and showed to be significantly associated with SAM. Variables that were significant in bivariate analysis were considered for inclusion in the multi-variable conditional logistic regression models. Check Tables 2 and 3 in the Results section to see which variables I am referring to.

Hope this helps!

Best,

Jovana

Andrea
Thanks for informing us about this study. I can see from the results that the highest AORs*related to the SAM symptoms at the individual level were diarrhea (10.7), fever (8.4) and vomiting (7.6). At the household level the AORs were lower eg caretaker’s hand washing habits (1.9) and absence of a toilet (1.9). But one factor stands out from all others at the household level and that is the marriage status of the caretaker (7.7) with a wide range of 2.2-30. Do you have an explanation for this variable as it relates to risk for infection? Is this more related to the age of the caretaker whereby children taking care of babies present the highest risk? Is marriage related to learning about infection?

A methodology question - you report AORs and not ORs. Aren't AORs adjusted based on a combination of variables? What then were the variables that are being co-factored into the results? Can one combine improved water source, presence of a toilet and hand-washing in order to get an overall impact of WASH interventions?

* for more on OR: www.ncbi.nlm.nih.gov/pmc/articles/PMC2938757/

Regards

Dear Members of the WG12,

I am pleased to share with you an article about a study carried out in Chad on individual and household-level risk factors for SAM among under-five children in Mao health district.

You can find the study at the following link:

archpublichealth.biomedcentral.com/artic...86/s13690-018-0281-5

Individual and household risk factors of severe acute malnutrition among under-five children in Mao, Chad: a matched case-control study


Severe acute malnutrition (SAM) is one of the leading causes of morbidity and mortality in Chad. The reasons behind persistently high prevalence of SAM in the Kanem region are still poorly understood, leaving national and international partners without clearly identified drivers to address. Current knowledge of SAM determinants in this context is largely based on very limited data. The aim of this study was thus to investigate individual and household-level risk factors for SAM among under-five children in Mao health district. A matched case-control study was conducted on 411 (137 cases and 274 controls) children aged 6–59 months with their caretakers from mid-February to August 2017. Data were collected by using a structured interviewer administered questionnaire, anthropometric measurements and through direct observations of household environment. Controls were matched to their cases on place of residence and on age (± 3 months). Data were double-entered, processed and analysed using Epi Info 7.2.0.

Best Regards,
Andrea Angioletti