In 1979, Michael P.
Stern, professor of medicine at the University of Texas Health Science Center in San Antonio, spearheaded two epidemiological studies of cardiovascular risk factors among Mexican Americans living in Texas.
The first, located in Laredo, a town on the border with Mexico, began in February.
Six months later, he started a second study in San Antonio, which was considered at the time to have the highest concentration of persons of Mexican ancestry of any city in the country.
Stern expressed concern that this rapidly growing population remained woefully understudied, despite accumulating evidence of serious health problems among them, which he deemed of considerable “public health relevance.” He also argued for the scientific value of studying Mexican Americans, whom he described as being of “mixed European and native American ancestry.” This “unique bicultural population,” he argued, offers “an ideal ‘laboratory’ ” for exploring the relative contributions of genes and sociocultural factors to disease patterns in different ethnic populations.
His assumption was that since Native Americans appeared to have “a genetic predisposition to diabetes,” and since “the Mexican American population was estimated to have a 30 to 40 percent native American contribution to its gene pool,” a similarly high rate of diabetes among Mexican Americans would likely be attributable to their genes.
Armed with these assumptions, Stern and his team traveled to Laredo.
Focusing on two neighborhoods where low-income Mexican Americans resided, they used Spanish surnames and Spanish-language ability to identify the 389 individuals who ended up participating in the study.
Participants filled out a questionnaire and agreed to have measurements taken of their weight, height, blood pressure, fasting blood glucose, serum cholesterol, and serum triglycerides.
Stern then compared the results to data he had for both the Akimel O’odham and the nation at large.
Although he exercised caution because the studies of these various populations had employed different sampling strategies and experimental designs, he still felt confident enough in the strength of the research to use it “to help place the Laredo Project results in perspective.”
Stern concluded that poor Mexican Americans in Laredo had levels of “overweight” and hyperglycemia that fell between those of the Akimel O’odham and the nation as a whole. What he could not determine was whether this pattern was “primarily due to sociocultural or to genetic factors.”
The problem, as he saw it, was that diabetes and obesity were known to be linked, and sociocultural factors were known to affect levels of adiposity. But the Laredo study, he pointed out, had not been designed to distinguish between genetics and sociocultural factors. After all, the study participants were all poor Mexican Americans. Fortunately, that was not the case in San Antonio.
The San Antonio Heart Study, introduced earlier, was a much larger enterprise.
The first phase alone involved interviews with roughly four thousand people and lasted from 1979 to 1982.
Designed to tease out the relative contributions of “socioeconomic, cultural, life-style, and genetic factors” in explaining disease patterns among Mexican Americans, the study collected data from three different neighborhoods in the city: an affluent “suburb,” whose population was roughly 90 percent Anglo American; a middle-class “transitional” neighborhood that was about 60 percent Mexican American and 40 percent Anglo American; and a low-income “barrio,” which was primarily Mexican American.
The interviewers went door to door gathering information about smoking, diet, exercise, and other personal habits, as well as family background, attitudes and beliefs about obesity and dieting, access to health care, and general knowledge of health and disease.
Participants also received a free medical examination in a nearby mobile unit where the same measurements were taken as in Laredo.
The results once again indicated that Mexican Americans had a higher prevalence of both obesity and diabetes than Anglo Americans, but not as high as the Akimel O’odham.
Stern and his team believed they could do more with the data from San Antonio, since they now had information about Mexican Americans and Anglo Americans living in the same neighborhood, and information about each group from three different socioeconomic neighborhoods.
In a series of papers they began publishing in 1982, the researchers showed that regardless of neighborhood, Mexican Americans had higher rates of obesity than Anglo Americans, although the gap decreased with increasing wealth.
They also found that within each weight category (they had divided the subjects into “lean,” “average,” and “obese”), Mexican Americans had higher rates of diabetes.
In other words, lean Mexican Americans had a higher prevalence of diabetes than lean Anglo Americans.
They therefore concluded that something beyond socioeconomic status and obesity must be contributing to high rates among Mexican Americans.
Their best guess was that the “residual” factor was the concentration of Native American genes.
A paper published in 1984 engaged this question directly.
The lead investigator, Lytt I.
Gardner Jr., was an epidemiologist from the San Antonio research team.
Lacking a genetic marker to determine Native American ancestry, which he would have preferred, Gardner followed standard practice among researchers at the time and turned to traits, like skin color, that were known to be “predominantly under genetic control.” Based on the tests he ran, he concluded that the Mexican Americans he randomly selected from the barrio had a genetic admixture that was roughly 46 percent Native American; those from the transitional neighborhood had about 27 percent, and those from the suburbs had around 18 percent.
Given that the diabetes rates declined from 14.5 percent to 10 percent to 5 percent in the same neighborhoods, respectively, and given that prior research had demonstrated that obesity alone could not explain this pattern, Gardner “speculated” that genetic factors played an important role. “The association of genetic admixture with NIDDM rates,” he wrote, “suggests that much of the epidemic of NIDDM in Mexican Americans is confined to that part of the population with a substantial native American heritage.”
Note that Gardner no longer referred to “diabetes” in general, but rather to NIDDM.
His use of this term reflected major changes, initiated by the National Diabetes Mellitus Research and Education Act, aimed at improving diabetes treatment and research.
Along with the creation of the National Diabetes Advisory Board and diabetes research and training centers, stricter guidelines had been put in place to establish uniform diagnostic criteria.
As mentioned earlier, one of the major goals was to eliminate the single umbrella term, diabetes mellitus, for the various forms of the disease, along with such vague descriptors as mild diabetes, severe or acute diabetes, or juvenile-onset and adult-onset diabetes, and to replace them with more clearly defined categories.
Thus after 1979, clinicians and researchers referred either to NIDDM (type 2), which was the most common form, making up 90 to 95 percent of all cases; IDDM (type 1), which accounted for 5 to 10 percent of all cases; GDM (gestational diabetes mellitus); and several other minor types.
This also meant that after 1979, studies of diabetes among “nonwhites” focused almost exclusively on NIDDM, since IDDM appeared to be far less of a problem in their communities. As a result, negative judgments about diabetes, which quickly became associated almost exclusively with NIDDM, also became associated more readily with minority populations.
Such judgments were evident in the papers produced by the scientists working in San Antonio. This happened despite the sophistication of their research, their acknowledgment that skin color was not a totally reliable indicator of genetic admixture, and their admission that absence of a genetic marker for NIDDM rendered their conclusions tentative. At times their judgments and condescension were obvious, as when Stern referred to Mexican Americans struggling with diabetes as an “ideal ‘laboratory.’ ”
At other times, researchers’ prejudices were less transparent, as in their obsession with finding out the percent Native American ancestry of the different socioeconomic groups they studied.
Stern and his team may have presented Mexican Americans as a “bicultural population,” but at no point in their many publications did they explore, or even mention, the impact of the population’s European ancestry on disease patterns.
They did not even bring this up for the suburban residents who, according to their own data, had an 82 percent European admixture.
Why not write that European ancestry “protected” Mexican Americans from diabetes?
Far from a trivial point, it reflects an overall tethering of Mexican Americans to Native Americans that made it difficult for Stern and his peers to imagine Mexican Americans as white.
In fact, Mexican Americans’ claim to “whiteness,” like that of Jews in the early twentieth century, was far from settled.
Although officially classified as white in 1848 at the end of the Mexican-American War, Mexican immigrants encountered considerable discrimination when they attempted to purchase land, move into “white” neighborhoods, attend good schools, or seek remunerative employment, in short, when they tried to assimilate.
Much of the prejudice they encountered derived from the color of their skin, which suggested ancestral ties to Native Americans.
During a congressional hearing in 1926 on whether to extend the 1924 Restrictive Immigration Act to immigrants from countries in the Western Hemisphere, Congressman John Box of Texas, who favored stricter controls, argued that Mexicans were basically Indians “and very seldom become naturalized.” To him, this meant that “they know little of sanitation, are very low mentally, and are generally unhealthy.” Box’s description, which both revived older images of Mexicans as disease vectors and expressed unabashedly the eugenic sentiments of his day, took on particular meaning during the Depression, when the economic downturn intensified animosity toward the influx of people from south of the border.
Such racial animus resulted in the reclassification of Mexicans as a distinct race in the 1930 U.S.
Census, although that was dropped by the next census.
Still, in the very decades when national differences between Europeans gradually lost their association with race and all Americans of European ancestry simply became “white,” Mexican Americans and others of Hispanic ancestry continued to be singled out.
In the 1950 and 1960 U.S.
Census, they were grouped under the heading “Persons of Spanish Mother Tongue”; in 1970 this was changed to “Persons of Both Spanish Surname and Spanish Mother Tongue”; and in 1980 to “Hispanic.”
Publications from the San Antonio study had none of the animus that radiated from Box’s comment.
What remained, however, was a sense of distance and confusion about how to think of the racial makeup of the study’s subjects.
This is evident in the way Stern and his team used quotation marks when comparing Mexican Americans to what they called “ ‘other whites.’ ” The punctuation suggests that while they accepted Mexican Americans’ whiteness, they still considered it somewhat different.
They were not unusual in this regard.
Some Mexican Americans even embraced an image of themselves as “another white race,” finding the suggested hybridity useful for building group identity, fighting discrimination, and making sure that their health needs were met.