Running head: THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
The Prevalence of Diabetes in the Hispanic Community Mariah Delaire National University
October 4, 2015
1
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
2
The Prevalence of Diabetes in the Hispanic Community Introduction Diabetes in the United States is an increasing public health concern due to the large number of individuals being diagnosed with it annually. It has been reported by the CDC that 10.2% of non-Hispanic whites and 18.7% of non-Hispanic blacks 20 years and older in the United States have diabetes (Schneiderman, 2014). Among this population are about 30 percent of undiagnosed individuals with the rest being racial/ethnic minority populations. It is even more of a public health concern to have insufficient data on different populations, such as the Hispanic population. The Hispanic population has a higher prevalence of diabetes along with higher rates of mobility and mortality in relation to diabetes, making it even more important to understand population’s susceptibility to certain diseases (Hatcher, 2007). The need for data on the prevalence of certain diseases, such as diabetes, helps to plan for healthcare needs. The future of disease control is prevention, which can be made possible by understanding the population along with associated risk factors that contribute to the emergence of a disease.
Research question/objective The objective of this proposal is to understand the prevalence of diabetes in the Hispanic population by identifying common risk factors that contribute to the susceptibility of this disease. The key to this study is to discover associated risk factors that can be well understood to allow for proper prevention, lowering the overall prevalence in the population. The variables that will be analyzed are body mass index, diet, exercise, stress, insurance coverage, and how proactive an individual is about their health/wellness.
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
3
Secondary data analysis will be conducted using the Hispanic Community Health Study/ Study of Latinos (HCHS/SOL) and the National Health and Nutrition Examination Survey (NHANES). These resources will supplement the study to help reach the objective of understanding the prevalence of diabetes in the Hispanic population.
Null Hypothesis: There is no relationship between the independent variables (body mass index, exercise, stress, insurance coverage, proactivity in health, and education) and the prevalence of diabetes.
Alternative hypothesis: There is a relationship between the independent variables and the prevalence of diabetes.
Literature review Diabetes is a major public health concern, where 9.3% of the overall population in the United States have diabetes along with it being the 7th leading cause of death (National Center for Chronic Disease Control and Prevention , 2014). However, there are gaps in information in regards to minority populations. Due to the lack of data, the CDC is unable to estimate the prevalence of diabetes among different minority populations (Schneiderman, 2014). Some major barriers to gathering data include poor patient-physician communication, language barriers, lack of education, diabetes awareness and lack of health insurance (Aranda, 2004). Currently, there are various methods for assessing the risk of having undiagnosed diabetes, but there is a shortage in methods for assessing the risk of developing diabetes. This could be a major indicator as to why there is a gap in knowledge, especially concerning minority populations. If risk factors can
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
4
be determined for individuals who are susceptible to the disease, then prevention can take place to stop the emergence of diabetes. With the Hispanic population in the United States growing rapidly, it is vital to focus on their healthcare needs. In the United States, Mexican-Americans have a two to three fold higher chance of developing diabetes than do non-Hispanic whites, and it is suggested that the prevalence is rapidly increasing (Hunt, 2011). A study that tested how lethal diabetes is in the Mexican American population showed that factors associated with diabetes mortality included biological differences in the severity of diabetes, differences in health care access, treatment practices and major differences in on-going prevention efforts (Hunt, 2011). Although Hispanics have a higher prevalence of diabetes with a greater risk of complications, there was also a lower chance that they have a regular place of care or health insurance. It is extremely important for health care providers to understand diabetes, it symptoms, and treatments along with conveying this information to the patient. Lack of comprehension in regards to health is more common among individuals with low education which has been shown to be associated with worse glycemic control (Hunt, 2011). It was found that individuals in a lower socioeconomic class along with less education had a higher chance of getting diabetes (Schneiderman, 2014). Many individuals who were unware that they had diabetes had poor glycemic control and lack of health insurance, creating even more negative effects on their health (Schneiderman, 2014). The difference in glycemic control between Hispanics and non-Hispanic whites could be greatly influenced by lack of health insurance between the two groups, creating the major differences in diabetes cases (Schneiderman, 2014). It is important to understand the diversity between different populations to allow for proper prevention and treatment of certain diseases.
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
5
Sample The research will be conducted using the Hispanic Community Health Study/ Study of Latinos (HCHS/SOL) and the National Health and Nutrition Examination Survey (NHANES). The survey population for HCHS/SOL consists of 16,000 people of Hispanic origin that were recruited in four field centers located at San Diego State University, Northwestern University in Chicago, Albert Einstein College of Medicine in the Bronx and University of Miami. The participants ranged in age from 18-74 years old and underwent an extensive clinical exam and assessments to determine baseline and then were followed over time to assess changes (National Heart, 2011). The NHANES survey consisted of about 12,000 people every two year cycle where an average of 10,500 people agreed to complete an interview (U.S. Department of Health and Human Services, 2013). The individuals of this study were located in different counties all over the United States who ranged in ages 12 years old to over 70 years old. To produce statistics that are reliable, NHANES over samples individuals 60 years and older, African-Americans, and Hispanics (U.S. Department of Health and Human Services, 2013). The Hispanic Community Health Study is planned to last six and a half years which will collect baseline comprehensive health data from 16,000 Hispanics and have annual follow ups to allow information on changes in health (National Heart, 2011). This is the largest longitudinal epidemiological study in the Latino population in the U.S. for health and disease. This study will provide much insight for prevalence of diabetes in the Hispanic community along with associated risk factors. NHANES is a program designed to gain information about nutritional status and health in adults and children. It is extremely informational in that it combines both interviews and physical examinations to produce vital health statistics for the U.S.
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
6
The HCHS/SOL participants were selected randomly from households recruited during 2008-2011 in New York, Chicago, Miami, and San Diego. The cohort was selected through a stratified multistage area probability sample from these four different communities (Schneiderman, 2014). These four areas have the largest number of Hispanics/Latinos which consists of high concentrations of specific Latino background. Population based sampling allowed for recruitment of participants and rosters of eligible participants were compiled followed by screening through telephone calls or in person interviews (Schneiderman, 2014). Participants were eligible if they were community dwelling and identified as Hispanic/Latino, were able to travel to sites for examinations, not on active military or pregnant, and did not plan on moving for three years (Schneiderman, 2014). The eligible individuals were again sampled by age which resulted in 9,714 participants between the ages of 45-74 years old and 6,701 participants between the ages of 18-44 years old. After screening and eligibility was determined, 42% of the eligible participants enrolled in the study (Schneiderman, 2014). NHANES sampling was done by a multistage probability sample design which selected a sample representative of a civilian population in the United States (U.S. Department of Health and Human Services, 2013). The sample selection process consisted of four stages. The first stage focused on selection of primary sampling units that consisted of counties or small groups. The second selection created segments within the primary sampling units that resulted in a block or group containing a sum of households. The third stage involved selection of specific households within each segment. Lastly, the fourth stage selected specific individuals (U.S. Department of Health and Human Services, 2013). 12,000 people every 2 years were asked to participate. The number of participants who responded averaged around 10,500 people. These
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
7
individuals agreed to take part in a household interview who are located in counties all over the United States (U.S. Department of Health and Human Services, 2013).
Data Collection Data from the HCHS/SOL study was collected through clinical examinations, interviews, and questionnaires which included measurements of height and weight. Standardized instruments were used to collect information about health behaviors, medical history, and demographics (Schneiderman, 2014). A two hour glucose tolerance test was performed for participants who did not have fasting plasma glucose >150 mg/dL and those who already reported having diabetes. After participants fasted for eight hours before the physical examination, blood was collected, processed and frozen on site. Information from the self-report interview was used to determine personal and family medical history, awareness of diabetes, age, sex, Hispanic background, income, education, and length of Residence in the United States. The NHANES data was collected in a variety of different ways. The household interview consisted of four parts which included screener questionnaire, relationship questionnaire, sample participant questionnaire and a family questionnaire (U.S. Department of Health and Human Services, 2013). The questionnaires were administered in the participant’s homes with rare cases being done in a field office or public location. A computer-assisted personal interview system and Blaise software was utilized to help conduct the interviews (U.S. Department of Health and Human Services, 2013). Another component of the data collection was a physical examination that took place in a mobile examination center (MEC). The eligibility for certain components of the exam were determined from age and gender along with interviews and laboratory tests. The physical exam measured a variety of things from hearing, body measurements, balance, to blood
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
8
pressure, muscle strength, respiratory health, heart health and vision. Blood specimens were also taken to measure a variety of different components. One last component was the post MEC data collection questionnaire which consisted of phone interviews regarding current health status and examination results. NHANES used automated systems that integrated biomedical equipment and questionnaire items to provide efficient data capture.
Data Analysis The data for this study will be analyzed used SAS studio. Variables taken from the HCHS/SOL and NHANES study will be analyzed and organized to properly fit the objective of this study. The table below shows what types of variables are going to be measured along with their names for the purpose of SAS. Since this project focuses on disease occurrence in the Hispanic community, only individuals who answer Hispanic/Latino will be eligible to continue forward with the study. Questions
Please Describe your race/ethnicity.
Have you ever been diagnosed with type I or type II diabetes? In a typical week, how many days do you exercise?
Possible Answers
White African American Hispanic/Latino Asian Pacific Islander Native American I don’t know Refused Yes No I don’t know Refused 5 to 7 days a week 2 to 4 days a week Once a week I do not regularly
SAS Variable Name
RACE2
AB51
EXERCISE2
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
9
exercise
How many meals a week consist of fruits and vegetables?
In the last 12 months, did you make any appointments for a check-up or routine care with your health care provider? How many times a week do you feel stressed on a scale of 0-7? (7 being stressed every day)
Do you currently have health insurance coverage? What is your height and weight?
15 to 20+ meals 9 to 14 meals 5 to 8 meals 1 to 4 meals 0 meals I don’t know Yes No I don’t know Refused 7 6 5 4 3 2 1 0 Yes No I don’t know Refused Refused
DIET2
PROACTIVE2
STRESS2
HLTHCOV2
BMI_P
Descriptive Statistics To analyze each variable, descriptive statistics will be used which will help describe, show and summarize data in a way to visually examine it. PROC UNIVARIATE procedures will be used to measure mean, median, mode, and standard deviation for BMI_P and DIET2. To understand distributions, PROC FREQ will be used for the categorical/ordinal variables which
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
10
include HLTHCOV2, STRESS2, PROACTIVE2, EXERCISE2, AB51, and HISPANIC2. Gender and age will also be determined to allow for proper stratifying of data.
Inferential Statistics In order to analyze relationships between the independent and dependent variables, inferential statistics needs to take place. Bivariate analysis will be used to identify relationships between the dependent and all the independent variables with age, gender, and ethnicity used as a control. ANOVA testing (PROC GLM) will be used to analyze differences between BMI (independent variable) and type I or type II diabetes (dependent variable) as well as diet and diabetes. One way ANOVA works is by analyzing differences among group means and their variation. Generally ANOVA works by testing the differences between multiple groups, in this case it was used to test the difference between two variables. The Chi-squared procedure (PROC FREQ) will also be used to understand the relationships between diabetes and health coverage, stress, how proactive individuals are, and exercise, with the test only being ran with the dependent and one independent variable at a time. To fully understand if the relationship between extraneous variables influences the dependent variable, PROC CORR will be used to identify the correlations between BMI and diet. It is critical to understand the relationships between all the variables in order to fully discern if they impact one another. Multivariate analysis is used to study three or more variables at once. Multi-regression using PROC REG will be used to explore the independent effects on one dependent variable (diabetes). As previously discussed, the control variables will remain the same, and multivariate regression will be used to see how body mass index, diet, exercise, stress,
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
11
insurance coverage, and how proactive an individual is about their health/wellness affects diabetes status. Univariate, bivariate and multivariate analysis used together will help understand the relationship between diabetes prevalence and the extraneous variables.
Ethical considerations To help protect the confidentiality of individuals, the NHANES protocol was developed to be in compliance with the HHS Policy for Protection of Human Research Subjects (U.S. Department of Health and Human Services, 2013). The Privacy Act of 1974, Section 308(d) of the Public Health Service Act and Confidential Information Protection and Statistical Efficiency Act were three federal laws that protected all the data (U.S. Department of Health and Human Services, 2013). To ensure protection of information and prevent disclosing information to unauthorized persons, all staff affiliated with NHANES read and signed a nondisclosure affidavit (U.S. Department of Health and Human Services, 2013). Ethics Review Board also protects the rights and welfare of individuals enrolled in the study including vulnerable populations such as children, pregnant women, and the elderly. These same actions will be taken to ensure protection of any additional information gained outside of the NHANES survey and the HCHS/SOL study.
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY
12
References Aranda, J. V. (2004). Awareness of Hypertenstion and Diabetes in the Hispanic Community. Clinical Cornerstone, 7-15. Hatcher, E. W. (2007). Hispanic adults' beliefs about type 2 diabetes: clinical implications. American Academy of Nurse Practioners, 536-545. Hunt, K. G. (2011). Diabetes is more lethal in Mexicans and Mexican Americans compared to non-Hispanic Whites. Ann Epidemiol, 899-906. Mainous, A. B. (2007). Impact of the population at risk of diabetes on projections. Diabetologia, 934-940. Mainous, A. M. (2006). Acculturation and Diabetes Among Hispanics: Evidence from the 1999– 2002 National Health and Nutrition Examination Survey. Public Health Reports, 60-66. National Center for Chronic Disease Control and Prevention . (2014). National Diabetes Statistic Report. Atlanta: Center for Disease Control . National Heart, L. a. (2011). The Hispanic Community Health Study . Department of Health and Human Services. Schneiderman, N. L.-C.-S. (2014). Prevalence of Diabetes Among Hispanics/Latinos From Diverse Backgrounds: The Hispanic Community Health Study/Study of Latinos . Diabetes Care, 2233-2239.
THE PREVALENCE OF DIABETES IN THE HISPANIC COMMUNITY U.S. Department of Health and Human Services. (2013). National Health and Nutrition Examination Survey. Vital and Health Statistics , 1-37.
13