Menu ENN Search
Language: English Français

Effect of higher prevalence on statistical associations

This question was posted the Assessment and Surveillance forum area and has 1 replies.

» Post a reply


This question relates to COVID-19.

Lenka Blanarova

Senior Nutrition Assessment Coordinator

Normal user

2 Sep 2020, 16:10

Dear colleagues,

we are currently conducting a study on causes of malnutrition and due to Covid-19 we are unable to collect primary data (cross-sectional quantitative survey coupled with anthropometric survey) in the study area. We have an access to comparable datasets from a previous year but the key stakeholder opposes their use as the measured GAM prevalence was very high and they fear this might influence our analyses (bivariate and multivariate analyses). Do you think this argument is justified if the debated survey was previously validated by NIWG in country? I understand that high GAM prevalence would be an issue if children were incorrectly measured and/or classified but if all quality assurance procedures were followed, could high prevalence still be an issue in relation to our analyses?

Many thanks in advance for your inputs,


Dr. Trenton Dailey-Chwalibóg

Universiteit Gent

Normal user

10 Sep 2020, 09:18

Hi Lenka, 

If there is any reason to believe that the GAM prevalence from last year dramatically decreased (which it probably didn't), then you could reasonably argue to use these data. However, it would only be a simulation, if you are not able to collect primary data.

If anything, I think that GAM prevalence during COVID would be higher than usual. Which would make your estimates drawn from last year’s dataset either close to spot on or conservative.

Hope this helps.


If you have any problem posting a response, please contact the moderator at

Back to top

» Post a reply