Addressing Disparities in Minority Populations Starts with Better Data

Two studies presented at the American Diabetes Association's 74th Scientific Sessions show that evaluating diabetes risk and patient health by nation of origin and ethnic background yields richer insights into how the disease affects populations.​

Evaluating the health of minority populations with broad categories like “Hispanics,” or “Asians” fails to capture the diversity of what is happening within these demographic groups, according to presenters who spoke Friday at the 74th Scientific Sessions of the American Diabetes Association, which is taking place at the Moscone Center in San Francisco, California.

The symposium, “Racial/Ethnic Disparities in Diabetes and its Complications,” featured 2 sets of data that tracked the prevalence of diabetes and other diseases, in part by looking at a patient’s nation of origin or ethnic background. Both presenters said more precise data can provide richer clues on how diet, cultural traditions, and genetics affect health.

Neil Schneiderman, PhD, principal investigator at the Miami Field Center of the Hispanic Community Health Study/Study of Latinos (HCHS/SOL), reviewed findings first released in February 2014 by the National Heart, Lung, and Blood Institute.1 The study of more than 16,400 persons of Hispanic origin, the largest of its kind ever in the United States, followed patients ages 18-74 from October 2006 to May 2013. A follow-up study now under way will conclude in 2019.2

Notably, the study breaks down the Hispanic groups, who already make up 15% of the US population, into subgroups based on place of origin: Puerto Rico, Mexico, Cuba, Dominican Republic, Central America, and South America. Besides Miami, the study gathered data from patients in Chicago, the Bronx in New York City, and San Diego.2 Patients received free health exams in exchange for taking medical tests, answering questions, and providing blood samples. The study goes beyond just recording health statistics. Patients were asked what they ate, Dr Schneiderman said, and researchers bought groceries from local stores and suppliers based on these reports, which were sent to the US Department of Agriculture for evaluation for nutritional value.

Already, the study is providing important insights into ways that diet and adaptation to the American culture affect health. Although diabetes was more common among persons at least 70 years old, Dr Schneiderman said, “In each age group, the longer you live here, the higher the prevalence of diabetes.”

In fact, prevalence of diabetes — almost all of it type 2 diabetes mellitus (T2DM) – showed up at the same rate among persons of Hispanic ethnicity who were born in the United States, 14.5%, as it did among those born abroad who had been in this country 6-10 years.

Not surprisingly, persons with better education and higher incomes had lower rates of diabetes. More troubling, only 59% of those with diabetes were aware of their condition, Dr Schneiderman said. “This becomes important, because we know we have people walking around with diabetes who don’t know it,” he said.

Many patients in the study do not have health insurance, and Dr Schneiderman said this is seen in the outcomes: 52% of the patients had coverage, while 48% had adequate glycemic control. In Miami, he said, among those ages 18-65, 70% had no health insurance, and glycemic control was higher among older patients, who have access to Medicare. “There are good indications that having health insurance helps patients achieve glycemic control, because control increases significantly with age,” he said. “Those who had no health insurance were significantly more likely to show up at baseline without the knowledge that they had diabetes.”

In coming years, the study will be able to evaluate how the Affordable Care Act affects health measures, because researchers will have access to data both before and after its implementation. Of note, 3 of the sites are in states that have expanded Medicaid, while the Miami site is in Florida, which thus far has not expanded Medicaid. “We will have data from how many were insured before, and how many after, and we will be able to ask about the consequences of having become insured,” Dr Schneiderman said.

In response to a question, Schneiderman said future papers on the results will examine how being undocumented affects diabetes prevalence.

Gertraud Maskarinec, MD, PhD, an epidemiologist at the University of Hawaii, shared results from studies among Hawaiians, who frequently are of mixed racial and ethnic origin. The original data was gathered for a study examining the relationship between diet and cancer; the data on the Hawaiian cohort was compared with data on Asians living in Los Angeles County, California.3 Dr Maskarinec shared a pie chart breaking down the population of Hawaii along traditional demographic lines: whites, African-Americans, Hispanics, and Asians and Pacific Islanders; the last group made up 65% of the state’s population. But when she shared finer population data, grouped according to ethnic backgrounds that include Japanese-Americans, persons of Korean, Chinese, Filipino, or other Asian descent, as well as native Hawaiians, no single group constitutes a majority. And it’s these distinctions that appear important in diabetes care, she said.

Data first gathered for cancer research was quickly applied to studies involving diabetes, she said, “because it’s probably the biggest health we have in the state of Hawaii.” While the first data set was based on a mailed survey, later work has involved gathering blood and urine samples, and linking her results with data provided by Hawaiian health plans. Through this process, more than 11,800 persons with diabetes have been identified. Among the findings:

  • Diabetes incidence cannot be completely explained by obesity rates or body mass index (BMI), because it occurs among certain groups of Asians at much lower BMI than among whites.
  • Japanese-Americans and native Hawaiians have twice as high diabetes incidence as whites.
  • Persons of mixed Asian descent had the highest diabetes incidence, at 24.9%, while whites had the least at 9.9%, prior to controlling for BMI. But even after controlling for BMI, risk among persons of Asian descent remains 2 to 2.5 times higher than it did for whites.
  • Examinations of visceral and subcutaneous fat among white and Asian women showed that certain Asian-Americans, including persons of Japanese ethnicity, were more likely to have visceral fat, which appears to have a stronger link to diabetes.

Dr Maskarinec cited research that suggests higher insulin sensitivity among Japanese-Americans, which may account for higher levels of diabetes.4 She said emerging work in genetics may reveal more about the biological factors behind the disease, because when she controlled for factors like BMI and smoking, “Those factors do not explain the risk — there is something else there that is responsible for the differences.”


  1. Tanner L. Landmark Hispanic study may offer longevity clues. Federal News Radio website. Published February 24, 2014. Accessed June 14, 2014.
  2. Hispanic Community Health Study. National Heart, Lung, and Blood Institute website. Accessed June 14, 2014.
  3. Kolonel LK, Henderson BE, Hankin JH, et al. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol. 2000;151(4):346-357.
  4. Morimoto A, Tatsumi Y, Deura K, et al. Impact of impaired insulin secretion and insulin resistance on the incidence of type 2 diabetes mellitus in a Japanese population: the Saku study. Diabetologia. 2013;56(8):1671-1679.
Related Videos
View All
Related Content
© 2023 MJH Life Sciences
All rights reserved.