Regression models for CD4⁺ naive and memory cell subpopulations.

Backward elimination regression was used to develop multivariate models for CD4+ T cell subpopulations. Independent variables were divided by their standard deviations (shown in parentheses). Age, gender, race/ethnicity, seasonality, BMI, IL-6, and CMV and H. pylori titers were included as the candidate starting variables. Only significant variables (p<0.05) were retained in the final models. BMI: Body mass index; CMV: Cytomegalovirus; IL-6: Interleukin-6; ns: non-significant.