\doc\web\99\06\alz.txt Gregory M. Cochran wrote: Two points: First, At least in the U.S. blacks have a higher risk of Alzheimer's than whites and by quite a bit. A quote from a recent abstract "In the absence of the APOE-epsilon4 allele, the cumulative risks of AD to age 90 years, adjusted for education and sex, were 4 times higher for African Americans (RR, 4.4; 95% CI, 2.3-8.6) and 2 times higher for Hispanics (RR, 2.3; 95% CI, 1.2-4.3) than for whites. In the presence of an APOE-epsilon4 allele, the cumulative risk of AD to age 90 years was similar for individuals in all 3 ethnic groups." ________ REPLY Here are some interesting responses to the Tang et al study from the Journal of the American Medical Association: To the Editor.-While Dr Tang and colleagues[1] report an increased incidence of AD in African Americans and Hispanics compared with whites, there is "no accepted definition of race,"[2] since the "classification into races has proved to be a futile exercise."[3] Ethnicity has been suggested as a substitute for race or even as an alternative to it, but the federal designation of "race/ethnicity" has unhelpfully combined what can be neither lumped nor split. Consider the ethnic heterogeneity of whites with varying genetic combinations and environmental influences on cultural practices. Add to that consideration the West African-restricted heterogeneity of African Americans, not to mention Hispanics, and a dizzying morass of culture, history, and DNA emerges. To those who respond that perhaps there is indeed a "sickling gene for AD" and that the contribution by Tang et al[1] allows us to consider that possibility, it must be pointed out that it is precisely the intraethnic and interethnic heterogeneity among those studied that so confounds that premise as to preclude it. In addition, one must also properly wonder about the fact that these 2 "high-risk" groups also have been historically powerless and systematically disadvantaged in our society compared with the power status of the "low-risk" whites. The authors do allow for the potential importance of environmental factors in the pathogenesis of AD, but in the process, they have offered weak methods for elucidating it. The challenge in research will be to control for societal status and cultural practice. Only then will the dangers of bias and confounding be appropriately addressed. Barry L. Farkas, MD McKeesport, Pa References 1. Tang M-X, Stern Y, Marder K, et al. The APOE-4 allele and the risk of Alzheimer disease among African Americans, whites, and Hispanics. JAMA. 1998;279:751-755. 2. Osborne NG, Feit MD. The use of race in medical research. JAMA. 1992;267:275-279. 3. Cavalli-Sforza LL, Menozzi P, Piazza A. The History and Geography of Human Genes. Princeton, NJ: Princeton University Press; 1994. (JAMA. 1998;280:1662) --------------------------------------------------------------------------- ----- To the Editor.-Dr Tang and colleagues[1] find a discrepancy between the presence of the APOE-4 allele and the risk of AD among African Americans, whites, and Hispanics in the United States and suggest that other genetic or environmental factors were involved. In fact, dietary risks for AD have been discussed in the medical literature during the past year. Based on the findings that African Americans have 4 times the AD prevalence for those aged 65 years or older as native Nigerians (6.2% vs 1.4%)[2] and that Japanese Americans have 2.2 times the AD prevalence as native Japanese (4.2% vs 1.9%),[2] I conducted a multicountry analysis of AD prevalence vs national dietary supply of macronutrients.[2] Data from 11 countries, including 7 European and North American countries, were analyzed. The study found that dietary energy intake and fat supply 4 years prior to the AD prevalence study were the highest dietary risk factors, while dietary fish reduced the risk of AD.[2] The dietary analysis also found that the prevalence of AD for those aged 65 years or older in the United States is 5.2%, not 10% as is often quoted. (The 10% figure is based on a study finding that 95% of those with senile dementia had AD.[2]) These findings have now been supported by 2 case-control studies. In one,[3] patients who developed AD had an average intake of 1674 J/d more after the age of 60 years than controls, while before the age of 60 years, the diets were more nearly comparable. Those with AD also consumed fewer antioxidants than controls throughout life. The other study, from Zutphen in the Netherlands, found a positive correlation for both dietary total fat and saturated fat with AD for those aged 55 years or older and an inverse correlation between fish consumption or linoleic acid intake and AD.[4] In addition to oxidative stress, these findings suggest that cerebral inflammation caused by prostaglandins derived from omega-6 fatty acids is involved. This is consistent with the roles of fish oils (omega-3) and nonsteroidal anti-inflammatory drugs in reducing the risk of AD.[2,4] The APOE-4 allele also has been identified as a risk factor for heart disease,[5] yet dietary risks for heart disease are well known. Thus, diet is likely the primary environmental influence on the development of AD. Many articles have been published regarding excess obesity and heart and other disease rates among African Americans compared with other Americans. To advance the understanding of the etiology of AD, more attention should be directed to dietary links, perhaps starting with differences among the various ethnic groups in the United States and abroad. William B. Grant, PhD Yorktown, Va References 1. Tang M-X, Stern Y, Marder K, et al. The APOE-4 allele and the risk of Alzheimer disease among African Americans, whites, and Hispanics. JAMA. 1998;279:751-755. 2. Grant WB. Dietary links to Alzheimer's disease. Alzheimer Dis Rev. 1997;2:42-55. Available at: http://www.coa.uky.edu/ADReview/vol2-97.htm). Accessed April 2, 1998. 3. Smith MA, Petot GJ, Perry G. Diet and oxidative stress: a novel synthesis of epidemiological data on Alzheimer's disease. Alzheimer Dis Rev. 1997;2:58-59. 4. Kalmijn S, Launer U, Ott A, Witteman JCM, Hofman A, Breteler MMB. Dietary fat intake and the risk of incident dementia in the Rotterdam study. Ann Neurol. 1997;42:776-782. 5. Pasagian-Macaulay A, Aston CE, Ferrell RE, McAllister AE, Wing RR, Kuller LH. A dietary and behavioral intervention designed to lower coronary heart disease: risk factors are unaffected by variation at the APOE gene locus. Atherosclerosis. 1997;132:221-227. (JAMA. 1998;280:1662-1663) --------------------------------------------------------------------------- ----- To the Editor.-Dr Tang and colleagues[1] may be forgiven for their careless use of the appellative Hispanic, since the US Census Bureau itself has muddled that term to the point of uselessness. However, as defined by the 1990 US Census, the only common characteristic of people of Spanish/Hispanic origin is their language. As recognized in the Editorial[2] accompanying the article, the issue of ethnicity is a complicated one but crucial to the interpretation of the results. The ethnic makeup of so-called Hispanics runs such a diverse gamut as to make the term useless for any scientific purpose other than linguistic. On the one hand, European Spaniards are themselves of diverse ethnicity such as Goths, Celts, and Semites. On the other, Latin Americans cover an incredibly wide spectrum of ethnic origins, from the mostly unmixed European populations of Argentina and Uruguay to the 80% to 90% Indian countries such as Bolivia and Paraguay. Central America and the Caribbean offer interesting contrasts: the former countries' populations are about two thirds of Indian descent, while the Spanish-speaking peoples of Cuba and Puerto Rico are approximately 50% black, with essentially no Indian ancestry. In the "Comment" section, the authors stated that the majority of Hispanics in the study were of Dominican origin. The Dominican Republic, which shares the island of Hispaniola with Haiti, is perhaps the most extreme example of the dangers of ethnic labeling. Mulattoes make up the majority of the Dominican population, while the poorest classes are most often black. The article's generalization of the study's results to Hispanics is misleading and unwarranted. Researchers seeking information regarding the ethnic makeup of the various peoples of the world are referred to 2 good resources: the Encyclopedia of World Cultures [3] and the Encyclopedia of the Peoples of the World.[4] Horacio A. Méndez, PhD Marquette University Milwaukee, Wis References 1. Tang M-X, Stern Y, Marder K, et al. The APOE-4 allele and the risk of Alzheimer disease among African Americans, whites, and Hispanics. JAMA. 1998;279:751-755. 2. Kukull WA, Martin GM. APOE polymorphisms and late-onset Alzheimer disease: the importance of ethnicity. JAMA. 1998;279:788-789. 3. Levinson D. Encyclopedia of World Cultures: Sponsored by the Human Relations Area Files at Yale University. Boston, Mass: GK Hall Co; 1995. 4. Gonen A, ed. Encyclopedia of the Peoples of the World. New York, NY: Henry Holt; 1993. (JAMA. 1998;280:1663) --------------------------------------------------------------------------- ----- In Reply.-Dr Barker and colleagues suggest that recruitment methods might explain differences in the association between AD and the APOE-4 polymorphism reported by us compared with other studies. However, studying clinic patients can result in a form of bias, referred to as Neyman's fallacy,[1] such that the relationship between a risk factor and disease is different for those who participate (self-selected cases, specialty clinic or hospital cases, or prevalent cases) than for individuals who would have been eligible to participate but were otherwise not in the study.[1] Because prevalence is the product of incidence multiplied by duration, risk factors identified in prevalent cases may reflect factors that promote survival with the disease. Indeed, the mortality rate among patients with AD may be lower among patients with an APOE-4 allele.[2,3] Barker et al also infer that they may have greater accuracy in diagnosis of AD. Using autopsy confirmation as the criterion standard, the sensitivity of the clinical diagnosis of AD in our community series was 91% and the specificity was 68%, similar to those found at 26 Alzheimer's Disease Research Centers.[4] Dr Farkas challenges our study by stating that there is "no accepted definition of race," but he offers no alternative. Allowing individuals to determine their own ethnic group affiliation is superior to assigning a race or an ethnic group by some other arbitrary method. That the frequency of disease varies significantly among individuals who belong to different ethnic or cultural groups becomes important in determining the causes of disease. Farkas also suggests that we believe there is a specific gene for AD in blacks and Hispanics. However, any gene or genes associated with AD in these 2 ethnic groups will likely also be associated with the disease in whites. Dr Grant raises the interesting possibility that dietary differences among the ethnic groups might contribute to the higher frequency of AD among blacks and Hispanics compared with whites in our study. We have not conducted a formal study of diet and AD, but we previously found that the most robust predictor of plasma lipid levels was the APOE genotype, not ethnic group membership.[5] Dr Méndez suggests that we used the term Hispanic inappropriately and should have been more restrictive. Using the 1990 census questionnaire as a guide, we asked individuals if they consider themselves white, black, or other, and we then asked if they are Hispanic. If they answered Hispanic, the country in which they were born was queried. We indicated that 84% of those identified as Hispanic were of Caribbean origin, predominantly from the Dominican Republic. It seems that Méndez wants to further distinguish groups by race or to use a different method to define ethnic group. We do not support that view and choose to remain consistent with the US Census definitions for the reasons stated above. Ming-Xin Tang, PhD Richard Mayeux, MD, MSc Columbia University New York, NY References 1. Lasky T, Stolley PD. Selection of cases and controls. Epidemiol Rev. 1994;16:6-17. 2. Stern Y, Brandt J, Albert M, et al. The absence of the apolipoprotein-E 4 allele is associated with a more aggressive form of Alzheimer's disease. Ann Neurol. 1997;41:615-620. 3. Van Duijn CM, de Knijff P, Wehnert A, et al. The apolipoprotein E allele is associated with an increased risk of early-onset Alzheimer' s disease and reduced survival. Ann Neurol. 1995;37:605-610. 4. Mayeux R, Saunders AM, Shea S, et al. Utility of the APOE genotype in the diagnosis of Alzheimer's disease. N Engl J Med. 1998;338:506-511. 5. Pablos-Méndez A, Mayeux R, Ngai C, Shea S, Bergland L. Association of apo-E polymorphism with plasma lipid levels in a multi-ethnic elderly population. Arterioscler Thomb Vasc Biol. 1997;17:3534-3541. (JAMA. 1998; 280:1663) ------------------------------------------------------------------------ eGroups eLerts! Exclusive discounts on personal finance products and services Join Now! http://clickhere.egroups.com/click/27 eGroup home: http://www.eGroups.com/list/h-bd Free Web-based e-mail groups by eGroups.com .