Insulin resistance and cognitive test performance in elderly adults: National health and nutrition examination survey (NHANES)

Objectives: To examine the relationship between homeostatic model of insulin resistance (HOMA-IR) and cognitive test performance among population 60 years in a national database. Hypothesis: Higher insulin resistance is associated with lower cognitive test performance score in the population >= 60 years. Participants: We analyzed data from the National Health and Nutrition Examination Survey (NHANES) 1999-2000 and 2001-2002. Measurements: Cognitive test performance was measured by the Digit Symbol Substitution (DSS) exercise score. The main independent variable was the homeostasis model assessment of insulin resistance (HOMA-IR). We used bivariate analysis and generalized linear model adjusting for age, gender, race, education, body mass index, and systolic and diastolic blood pressures; total cholesterol, low density lipoprotein (LDL), high density lipoprotein (HDL) and triglyceride levels; and physical activity, diabetes mellitus, stroke, and congestive heart failure. STATA 14 was used to analyze the data taking into consideration the design, strata and weight. Results: Of the 1028 participants, 44% were male and 85% were white. The mean age was 70.0 +/- 0.28 (SE) years. Their average HOMA-IR was 3.6 +/- 0.14 and they had a mean of 49.2 +/- 0.8 correct DSS score in the cognitive test. Adjusting for the confounding variables, HOMA-IR was associated with decline in DSS score (B = -0.30, 95% confidence interval = -0.54 and -0.05, p = 0.01). The model explained 44% of the variability of the DSS score (R-2 = 0.44). Significant predictors of decline in DSS score were age, gender, race, and education (p = 0.01). Conclusion: Insulin resistance as measured by HOMA-IR was independently associated with lower cognitive test performance score among elderly participants aged >= 60 years. Longitudinal studies are needed to test the mechanism and the causal relationship.