Abstract
BackgroundEnvironmental Enteropathy (EE) is a subclinical condition caused by constant fecal-oral contamination and resulting in blunting of intestinal villi and intestinal inflammation. Of primary interest in the clinical research is to evaluate the association between non-invasive EE biomarkers and malnutrition in a cohort of Bangladeshi children. The challenges are that the number of biomarkers/covariates is relatively large, and some of them are highly correlated.MethodsMany variable selection methods are available in the literature, but which are most appropriate for EE biomarker selection remains unclear. In this study, different variable selection approaches were applied and the performance of these methods was assessed numerically through simulation studies, assuming the correlations among covariates were similar to those in the Bangladesh cohort. The suggested methods from simulations were applied to the Bangladesh cohort to select the most relevant biomarkers for the growth response, and bootstrapping methods were used to evaluate the consistency of selection results.ResultsThrough simulation studies, SCAD (Smoothly Clipped Absolute Deviation), Adaptive LASSO (Least Absolute Shrinkage and Selection Operator) and MCP (Minimax Concave Penalty) are the suggested variable selection methods, compared to traditional stepwise regression method. In the Bangladesh data, predictors such as mother weight, height-for-age z-score (HAZ) at week 18, and inflammation markers (Myeloperoxidase (MPO) at week 12 and soluable CD14 at week 18) are informative biomarkers associated with children’s growth.ConclusionsPenalized linear regression methods are plausible alternatives to traditional variable selection methods, and the suggested methods are applicable to other biomedical studies. The selected early-stage biomarkers offer a potential explanation for the burden of malnutrition problems in low-income countries, allow early identification of infants at risk, and suggest pathways for intervention.Trial registrationThis study was retrospectively registered with ClinicalTrials.gov, number NCT01375647, on June 3, 2011.
Highlights
Environmental Enteropathy (EE) is a subclinical condition caused by constant fecal-oral contamination and resulting in blunting of intestinal villi and intestinal inflammation
The median relative model error (MRME) of Smoothly clipped absolute deviation (SCAD) were closest to the Oracle estimator compared to other methods
When the correlation was moderate (ρ = 0.2, 0.5), Stepwise did slightly better than Adaptive Least absolute shrinkage and selection operator (LASSO) and Minimax concave penalty (MCP), but when correlation was large (ρ = 0.8), Adaptive LASSO performed better than Stepwise and MCP
Summary
Environmental Enteropathy (EE) is a subclinical condition caused by constant fecal-oral contamination and resulting in blunting of intestinal villi and intestinal inflammation. To deal with large number of covariates or predictors, one common approach is testing the association between each covariate and the outcome of interest through univariate regression model; a subset of those covariates are selected based on their significance for subsequent multivariable analysis. This framework is a common method in biomedicine for variable selection, but it can be a great challenge when the number of covariates is large in massive datasets. The essential problems with such method remain, that is, the parameter estimates tend to be highly biased in absolute values, their standard errors tend to be incorrect, and pvalues tend to be too low due to multiple comparisons and are difficult to correct [3]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
