Trends in Hyperinsulinemia and Insulin Resistance Among Nondiabetic US Adults, NHANES, 1999-2018

Abstract

Introduction: Hyperinsulinemia and insulin resistance are strong predictors of cardiometabolic diseases, which disproportionately affect individuals across gender, racial/ethnic, and socioeconomic groups. We aim to estimate and test the temporal trends in the prevalence of hyperinsulinemia and insulin resistance (IR) by sociodemographic groups among nondiabetic adults in the United States from 1999 to 2018. Methods: We used data from the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018. We fitted linear and joinpoint logistic regression models to test the sample weighted and age-standardized time trends for linear and nonlinear trends in the prevalence of hyperinsulinemia and IR, respectively. Results: We included 17,310 nondiabetic men and nonpregnant women aged 20 years or older. The age-standardized prevalence of hyperinsulinemia increased from 28.2% in 1999-2000 to 41.4% in 2017-2018, with IR prevalence similarly rising from 24.8% in 1999-2000 to 38.4% in 2017-2018. Across the entire period examined, individuals who were male; non-Hispanic Black; Hispanic; or had a lower educational level or lower family income consistently had a higher prevalence of hyperinsulinemia and IR than other groups. We found increasing temporal trends in the prevalence of hyperinsulinemia and IR for all the sociodemographic subgroups, at least in some periods from 1999 to 2018. Conclusions: There was an increased age-standardized prevalence of hyperinsulinemia and IR among nondiabetic adults in the US across each defined sociodemographic group from 1999 to 2018. The difference in prevalence across subgroups underscores the need for designing personalized and targeted interventions to address disparities.

Keywords: NHANES; diabetes; hyperinsulinemia; insulin resistance; metabolic disorders; sociodemographic disparities; temporal trend.

Figure 1

Study population flow diagram. Figure legend: flowchart illustrating the inclusion and exclusion criteria applied to the NHANES 1999–2018 dataset, resulting in a final sample of 17,310 adult participants as the study population of interest. Abbreviations: NHANES, National Health and Nutrition Examination Survey; MEC, Mobile Examination Center.

Figure 2

Age-standardized prevalence and 95% confidence interval of hyperinsulinemia stratified by sex, race/ethnicity, and educational and income level, National Health and Nutrition Examination Survey (NHANES), 1999 to 2018. Figure legend: showing the trends in the age-standardized prevalence of hyperinsulinemia and the disparities across sociodemographic groups. Hyperinsulinemia was defined as fasting insulin levels greater than 10 µU/mL. The prevalence of hyperinsulinemia and the corresponding 95% confidence interval were survey-sample-weighted and age-standardized to 2010 U.S. Census adult population. The sample size for the total, sex-stratified, and race/ethnicity (without Asian)-stratified trends was 17,310. The sample size for the stratified non-Hispanic Asian subgroup was 7066. A total of 19 participants were excluded from the education-stratified trend analyses. A total of 1538 participants were excluded from the income-stratified trend analyses.

Figure 3

Age-standardized prevalence and 95% confidence interval of insulin resistance stratified by sex, race/ethnicity, and educational and income level, National Health and Nutrition Examination Survey (NHANES), 1999 to 2018. Figure legend: showing the trends in the age-standardized prevalence of insulin resistance and the disparities across sociodemographic groups. Insulin resistance was defined as a Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) index greater than 2.6 (the 66.7th percentile). The prevalence of insulin resistance and the corresponding 95% confidence interval were survey-sample-weighted and age-standardized to 2010 U.S. Census adult population. The sample size for the total, sex-stratified, and race/ethnicity (without Asian)-stratified trends was 17,310. The sample size for the stratified non-Hispanic Asian subgroup was 7066. A total of 19 participants were excluded from the education-stratified trend analyses. A total of 1538 participants were excluded from the income-stratified trend analyses.