LAUSR

The authors thank the statistics consultants of Research Development Center of Sina Hospital, Tehran, Iran, for technical assistance. This work was not supWe were interested to read the paper by Murimi et al that was published in the Journal. The authors aimed to evaluate the difference in the prevalence of food insecurity between rural and urban households in West Texas. Fifty-eight participants and 133 participants from rural and urban households, respectively, were recruited into the study. Results demonstrated that the prevalence of household food insecurity in rural and urban households was 43% and 71%, respectively (P 1⁄4 .001). These figures for childhood food insecurity were 26% and 52%, respectively (P 1⁄4 .002). Although the statistical methodwas correct and the data were interesting, some methodological and statistical issues should be considered. We wonder why the authors performed a chi-square test to assess the association between variables when powerful and interesting statistical tests were available. The chi-square test is used to determinewhether there is a significant difference between variables; this test did not provide information about the strength of association. We reanalyzed the data presented in the study ofMurimi et al.We conducted binary logistic regressionmodels to assess the effect of rural and urbanhouseholds as an independent variable on household food insecurity and childhood food