Statement of the Problem: Diabetes mellitus is a common chronic disease that is characterized by insulin resistance.1-3 This insulin resistance is potentially disabling, since individuals affected are at increased risk for a number of serious complications including retinopathy, renal disease and cardiovascular disease.4-7 Minority populations, such as Hispanics and African Americans, are disproportionately affected by diabetes.8,9 Some of the disparity in the prevalence of diabetes may be based on genetically mediated metabolic differences by race/ethnicity, as several studies show that minorities are more prone to insulin resistance even after controlling for other risk factors such as obesity and body fat distribution. 10, 11 Cultural differences in dietary intake may also be relevant, since diet is an important factor in the prevention and treatment of diabetes, but in general minorities have less healthy diets. 12-21. The disparity in the prevalence of diabetes probably has a multi-factorial etiology, based on differences in cultural factors, health beliefs and access to care as well as genetics. Metabolic changes that occur with aging also may affect the prevalence and impact of insulin resistance. For instance, the prevalence of obesity, which is associated with insulin resistance, increases with age for both men and women, and is highest in the age range of 60-74 years.22 In menopausal women, a constellation of symptoms characterized by midlife weight gain, increased waist circumference, hypertension and dyslipidemia has been association with insulin resistance.23 Furthermore, aging is also associated with a loss of lean body mass and gain in adipose tissue, which can increases the risk of insulin resistance even in individuals who do not gain weight.24-26 It is possible that these aging changes have a larger impact on minority populations, since these groups are initially more prone to insulin resistance. However, this has not been established. Therefore in this study we propose to describe the prevalence of insulin resistance by age group and race/ethnicity, which will help define the groups at highest risk for insulin resistance and its associated complications. We will also evaluate differences in the association between dietary factors and insulin resistance by race/ethnicity and age group, in order to identify the dietary factors most relevant to insulin resistance. This will help in the development of more appropriate dietary recommendations for these groups.

Methods: Study Design: Secondary Data analysis of the NHANES 1999-2000 Survey Description Data from the 1999-2000 National Health and Nutrition Examination Survey (NHANES) will be analyzed. The NHANES is a product of the National Center for Health Statistics. It is a continuous, annual survey involving participants from a nationally representative sample of non-institutionalized residents of the United States. Minorities were over-sampled to ensure adequate numbers for analysis. The NHANES includes a detailed household interview and physical examination, plus laboratory information obtained through mobile examination centers. Samples are weighted so they are representative of the US population. Sampling weights are calculated taking into account unequal probabilities of selection due to sample design, non-response and planned over-sampling, then matched to known population control totals to be representative of the US population. Sample Adults (>18 years old) will be included in this sample. Respondents with a self-reported history of stroke, myocardial infarction, hypertension, diabetes, high cholesterol, congestive heart failure and/or coronary heart disease will be excluded. Demographic Data We will categorize individuals by race/ethnicity based on self-report as non-Hispanic whites, AA or Hispanics. Personal history of disease is also based on self-report. Measurement of Insulin Resistance Insulin resistance can be measured using a variety of surrogate markers, since although several measurement methods are available, there is no universally accepted and clinically useful definition, and no specific guidelines exist on the measurement of insulin resistance.27,28 However, the HOMA method has been widely employed in clinical research, and thus will be used in this analysis.29 Dietary Variables The NHANES quantifies dietary intake in the 24 hours prior to the interview via dietary recall interviews that were conducted in person by trained dietary interviewers fluent in Spanish and English. If necessary, translators were available for respondents who speak other languages. A “multiple pass” method was used to obtain dietary information. This entailed obtaining an initial list of foods consumed, after which respondents were asked about the time and place of consumption. A list of frequently forgotten foods was then displayed, and a complete description of the foods eaten obtained. Finally, the foods were reviewed in chronological order with amendments made as appropriate. A standard set of measuring guides, tools used to help the respondent report the volume and dimensions of the food items consumed, were available during interviewing to simplify portion size estimation. The dietary recalls were further characterized as reliable and meeting the minimum criteria if <25% of foods were missing descriptive information, <15% were missing amounts and the respondent remembered at least one food item per meal. Data considered unreliable was not included in the analysis. The daily total caloric intake for each respondent was quantified and used in the models as a continuous variable. The percent of daily calories obtained from saturated fat, carbohydrates and protein will be calculated. Daily intake of dietary fiber in grams will also be identified. All these dietary factors were used simultaneously as separate control variables in the logistic regression models. Other Control Variables Other variables were included as control variables. These included current smoking status, age, gender, physical activity over the last 30 days and body mass index (BMI). BMI is calculated from measured weight and height.

Results: Based on the previous findings that aging, dietary factors and race/ethnicity affect insulin resistance, this study seeks to establish whether the impact of aging on insulin resistance is different based on an individual’s race/ethnicity. This might help identify high risk populations that need to be targeted for health promotion interventions. Furthermore, we seek to describe differences with aging in the association between dietary factors and insulin resistance by race/ethnicity. Establishing the presence of a differential response to dietary factors by age and race/ethnicity would aid the development of more appropriate and effective dietary recommendations for minorities, and the implementation of these dietary recommendations could potentially be a factor in decreasing health disparities.