Analyzing Disparities in Childhood Obesity

The descriptive analysis was used to summarize the sample's characteristics. Graphical methods were used to visualize how overweight/obesity in children differed by ethnicity over time. The probability values were determined using column proportion tests. We used Equiplot charts to illustrate the prevalence of overweight/obesity in each wealth quintile by sex and ethnicity for each survey year. The line between the prevalence of the first and last wealth quintiles shows the degree of dispersion; the longer the line, the greater the socioeconomic dispersion. The slope inequality index (SII) and concentration index (CIX) were then calculated for each survey year and group data to estimate wealth inequality in overweight/obesity by ethnicity (25 (link)). To analyze health inequalities across wealth quintiles, it is recommended that both absolute and relative measures of inequality be used simultaneously (26 ). When the results of both inequality measures are significant, inequality between quintiles is asserted. The SII is a weighted, absolute measure of inequality that uses a logistic regression to represent the absolute difference in estimated values of a health indicator between the poorest and richest quintiles, while controlling for all other wealth quintiles (25 (link), 27 (link)). The CIX is a weighted, relative measure of inequality that is related to the Gini coefficient. It calculates the magnitude and direction of health inequality. It is defined as twice the area between the concentration curve and the diagonal, ranging from 1 to 1. The concentration curve represents the cumulative percentage of the health variable relative to the cumulative percentage of the sample, ranked by socioeconomic status from the most disadvantaged group to the most advantaged group (9 (link), 13 (link)). In the CIX analysis, we used the Erreygers correction, as suggested by other researchers who have studied health inequalities (28 –30 (link)). Understanding and interpreting the SII and CIX is easy. With values ranging from −1 to 1, the SII or CIX is a composite description of inequality across the population. The index is zero when there is perfect equality. When the values are negative, children from the poorest quintile are most affected by overweight/obesity; When the values are positive, children from the richest quintile are most affected by overweight/obesity. The magnitude of the index reveals the level of inequality (25 (link)). The CIX allowed comparisons of wealth inequality between surveys for annualized change, plots of the CIX indices and 95% confidence intervals were made. When the CIX value of the annualized change is positive or negative, wealth inequality is said to have decreased or increased, respectively. The analyses were weighted and found to be appropriate for the complex NHANES survey design. Probability values for statistical tests, where 2-sided p-values <0.05 were considered significant. Analyses were performed using Stata (STATA Corp., LP, College Station, Texas), and graphical representations were created using GraphPad Prism 9.