Editorial Credits/ Crédits éditorials

Subscription Rates /Prix d'abonnements

Advertising Rates / Tarifs publicitaires (PDF)

Guest Editorial
From Counting to Understanding: The Evolving Epidemiologic Approach to Dementia
Ian McDowell, PhD
(PDF)

In Review
More Than the Epidemiology of Alzheimer’s Disease: Contributions of the Canadian Study of Health and Aging
Joan Lindsay, Elizabeth Sykes, Ian McDowell, René Verreault, Danielle Laurin
(PDF)

Alzheimer’s Disease, Genes, and Environment: The Value of International Studies
Hugh C Hendrie, Kathleen S Hall, Adesola Ogunniyi, Sujuan Gao
(PDF)

Original Research
Hidden Cardiac Lesions and Psychotropic Drugs as a Possible Cause of Sudden Death in Psychiatric Patients: A Report of 14 Cases and Review of the Literature
Dominique Frassati, Alain Tabib, Bernard Lachaux, Natalie Giloux, Jean Daléry, François Vittori, Dorothée Charvet, Cécile Barel, Bernard Bui-Xuan, Rachel Mégard, Louis Pierre Jenoudet, Jacques Descotes, Thierry Vial, Quadiri Timour
(PDF)

The 5-Factor Model of Personality and Antidepressant Medication Compliance
Nicole L Cohen, Erin C Ross, R Michael Bagby, Peter Farvolden, Sidney H Kennedy
(PDF)

Social, Demographic, and Clinical Factors Related to Disruptive Behaviour in Hospital
Andrea K Boggild, Marnin J Heisel, Paul S Links
(PDF)

The Course of Depressive Illness in General Practice
Frédéric Limosin, PhD, Jean-Yves Loze, Myriam Zylberman-Bouhassira, Mark E Schmidt, Eléna Perrin, Frédéric Rouillon
(PDF)

Review Paper
Prevalence and Incidence Studies of Mood Disorders: A Systematic Review of the Literature
Paul Waraich, Elliot M Goldner, Julian M Somers, Lorena Hsu
(PDF)

Brief Communication
Bupropion Sustained Release Treatment Reduces Fatigue in Cancer Patients
Jodi L Cullum, Agnieszka E Wojciechowski, Guy Pelletier, J Steven A Simpson
(PDF)

Patient Factors Associated With Missed Appointments in Persons With Schizophrenia
Shalom Coodin, Douglas Staley, Barb Cortens, Rob Desrochers, Sandy McLandress
(PDF)

Book Reviews
(PDF)

The Treatment of Anxiety Disorders: Clinical Guides and Patient Manuals. 2nd ed Reviewed by
Richard Swinson, MD

Cognitive-Behavioral Group Therapy for Social Phobia: Basic Mechanisms and Clinical Strategies
Reviewed by
Michael Van Ameringen, MD, FRCPC

Letters to the Editor
(PDF)

Aripirazole–Olanzapine Combination for Treatment of Schizophrenia

Improvement of Torticollis With Quetiapine in a Schizophrenia Patient

Internalizing Antecedents of Conduct Disorder

Travel Time and the Use of Psychiatric Outpatient Clinic Services in Coastal Northern Norway

Respiratory Panic Disorder Treatment With Clonidine

In Review

Alzheimer’s Disease, Genes, and Environment: The Value of International Studies

Hugh C Hendrie, MB, ChB, DSc¹, Kathleen S Hall, PhD², Adesola Ogunniyi, MD³, Sujuan Gao, PhD⁴

Advances in molecular genetics have revolutionized epidemiologic research. Epidemiologists are now able to combine the techniques of population genetics with more traditional risk factor research to formulate etiologic hypotheses. For example, from their studies in hypertension, Cooper and Kaufman have proposed a disease model involving the contributions of genes and environment and their possible interactions to explain disease rates in populations (1; Figure 1).

Figure 1  Modelling the contribution of genes and the environment. Asterisks indicate interactions between factors 

The resolution of these models, however, is likely to vary among populations. One reason for this variation is that environmental exposures involving putative environmental risk factors for disease are also likely to differ among populations.

Comparisons between developing and developed nations offer a unique opportunity for applying this proposed model. Such studies provide a much wider diversity of environmental exposures than do studies of populations drawn solely from industrialized countries. In the latter, important risk factors may be missed because of their very pervasiveness (2). However, genetic variation also clearly influences disease risk. Research involving the human genome project suggests that, while there are few large differences in chromosome structure, there are many variations in small insertions and (or) deletions, in short tandem repeat sequences, and in single nucleotide differences within and among human populations. Any of these differences may influence disease risk, often acting in conjunction with environmental exposures (3–5).

Single nucleotide polymorphisms (SNPs) have proven to be an important tool in genetic studies of complex disease. SNPs are single base pair changes in the DNA sequence that can be found in either the coding or noncoding region of the DNA. Estimates suggest that an SNP may occur on average in every 1000 base pairs, resulting in as many as 3 million potentially identifiable SNPs (6–8). Variations in allele frequencies are also common among populations.

Genetic studies in African populations have assumed a particular pertinence. Several investigators have now studied a wide variety of polymorphic loci to investigate the extent of human genetic diversity and to delineate the relations among modern human populations. Consistent with the anthropological hypotheses that posit an African origin for modern humans, the greatest genetic diversity occurs in sub-Saharan African populations (9–11). For example, it has been established that genetic variations among Nigerian groups appear to exceed the genetic variation among all European populations (12). Thus, epidemiologic research involving African populations, particularly when combined with parallel studies in developed countries, offers the opportunity to evaluate not only wide environmental but also wide genetic diversity in determining disease phenotypes in populations

This paper describes the steps necessary to construct Cooper and Kaufman’s disease model (1) to study Alzheimer’s disease (AD), with particular reference to the results from our ongoing Indianapolis–Ibadan Dementia Project (I-IDP). These steps include the following:

1. A comparison of the rates of illness (preferably incidence rates) between the 2 populations.

2. A comparison of the frequency of putative risk factors, both genetic and environmental, between the populations.

3. An examination of the association of environmental risk factors with AD within each population.

4. An examination of the association of genetic risk factors with AD within each population, including identification of possible genetic or genetic–environmental interactions that may account for differences in the strength of genetic (or environmental) risk between populations, if it occurs.

5. An estimate of the potential significance of these risk factors in explaining the differences in illness rates observed between the populations.

6. Development of a risk factor model, which should account for observed phenotypic variation in the populations under study.

The Indianapolis–Ibadan Dementia Project (I-IDP)

Since 1992, research teams from Indiana University and the University of Ibadan have been collaborating on studies of the prevalence and incidence of dementia in elderly African Americans and Yoruba. The teams use identical methodology incorporating a 2-stage design. So far, they have conducted a prevalence study followed by 2 incidence studies at 2- and 5-year intervals. The I-IDP is supported by the National Institute on Aging and is ongoing.

Comparison of Incidence Rates Between Populations

The following point should be emphasized: for the purpose of risk factor research, determining that illness rates in the study populations are similar may be just as significant as determining that they are different if the populations are exposed to different levels of putative environmental or genetic risk factors.

Many potential methodological pitfalls arise in comparative studies involving populations in countries at different developmental levels and with different cultures. It is therefore essential that comparative studies be conducted by the same experienced group of investigators familiar with the social and cultural standards of the study populations. Further, these investigators must use test instruments that have been harmonized with adequate normative values. Harmonization is the process of translating and adapting test instruments to make them appropriate for the target population’s language and culture. Pilot studies are necessary to establish normative values for the target population.

We have reported significantly lower prevalence rates of dementia in Yoruba (2.29%), compared with African Americans (8.24%), as well as similarly lower rates for AD in Yoruba (1.41%), compared with African Americans (6.24%) (13). Our prevalence rates for African Americans are approximately the same as those reported in the Canadian Study of Health and Aging (14). Our prevalence rates for Yoruba are at the lower end of previously reported rates.

Prevalence rates, however, depend on factors additional to illness incidence rates. Between-site differences in life expectancy or survival of subjects with and without dementia could affect prevalence. Incidence rates (that is, the number of new cases occurring over a fixed period of time) give a better indication of true rates of illness than do prevalence rates. In our 5-year incidence study, the age-standardized incidence rates for both dementia and AD were significantly lower for Yoruba (1.35% and 1.15%, respectively) than for African Americans (3.24% and 2.52%, respectively) (15). It should be noted that, although prevalence and incidence rates were consistently lower among Yoruba, the association with age was identical in both sites; that is, prevalence rates in both sites roughly doubled with every 5 additional years of age.

Our reported incidence rates for dementia and AD in African Americans are in the higher range of previously published incidence rates but are similar to other published rates from African American populations. The incidence rates for both dementia and AD for Yoruba are among the lowest of previously reported rates (16). In the single other study of incidence rates of AD involving a comparison of populations from a developing and developed countries (the Indo-US Cross National Dementia Epidemiology Study), the incidence rate for AD in an Indian population living in Ballabgarh, a rural district of Northern India, was one-sixth the rate found among elderly subjects living in the Monongahela Valley in Pennsylvania (17).

It is tempting to propose from these studies that dementia rates are lower in developing countries or in more traditional societies than in developed countries. However, there are still far too few comparative studies available to make this generalization at this time.

One rather surprising finding from our study is that the increased risk for mortality from dementia is similar in Yoruba and African Americans, despite very different availability of health care (Ibadan relative risk = 2.83, Indianapolis relative risk = 2.05) (18).

Comparison of Frequency of Genetic and Environmental Risk Factors Between Populations

Genetic Variations
The association between the possession of the apolipoprotein E (apoE) e4 allele and AD is one of the most consistent findings in AD research and is being confirmed in many studies throughout the world (19). Variations in allelic frequency are commonly found when populations are compared, and apoE is no exception. The frequency of the e4 allele of apoE has been reported to vary between 5% in Sardinians to over 40% in Pygmies in Africa (20). Indeed, e4 allele frequency varies among populations in sub-Saharan Africa as widely as it varies in the rest of the world. It is not clear yet what effect these variations have on the rates of AD in different countries. In our study, however, the apoE allele frequencies were almost identical in the Yoruba and the African Americans (Table 1).

Environmental Risk Factors
Great cultural and socioeconomic differences exist between the impoverished, predominantly Muslim residents of the Idikan wards of Ibadan and the elderly African Americans living in Indianapolis. The elderly Yoruba have relatively limited access to health care and live for the most part in large extended families. In contrast, most of the African Americans in our study reported good access to health care and were either living in single-family dwellings with immediate family members or were widowed and living alone. Fifty percent of the elderly African Americans reported living alone, compared with only 7.4% of the Yoruba. One rather intriguing preliminary finding of our study is that there was an overall positive relation between indices of social involvement and cognitive function for both African Americans and Yoruba; that is, greater social involvement was associated with better cognitive scores. Educational levels differed between the 2 populations. In Ibadan, 85% of the subjects had received no formal education. In Indianapolis, the mean duration of education for our cohort was 9.6 years (2).

There are many lifestyle differences between these 2 populations. For example, dietary intake varies widely: the elderly Yoruba in the Idikan wards consume a low-calorie, low-fat diet comprising mainly grains, roots, and tubers supplemented with a small amount of fish. Ascorbic acid levels have been reported to be relatively high among the Yoruba, probably because of their high consumption of peppers. The African-American diet is high in fat and sodium and low in fibre. These lifestyle differences are reflected in significant differences in biological and medical variables (Table 2)—variables often associated with risk of circulatory problems such as heart attack and stroke (2).

Self-reports of diabetes and hypertension may underestimate the extent of these diseases in the elderly. An intensive medical evaluation conducted at the clinical assessment phase of our study in Ibadan reported a higher rate of hypertension (27.8%) than did self-reports; however, the rate was still considerably lower than the rates in African Americans (21). Blood pressure measurements showed only small differences between the 2 groups, but approximately 60% of the African Americans with diagnosed hypertension were taking antihypertensive medications.

Examination of the Association Between Environmental Risk Factors and AD Within Each Population

A risk factor, if valid, should modify disease change rates in both populations in the same manner. If the association between the risk factor and disease differs in the 2 populations, a careful investigation to understand the causes for this discrepancy should be undertaken.

For example, consider education. Many studies report that a high level of education protects against the development of AD. In Ibadan, 85% of the subjects had received no education. In Indianapolis, the mean number of years of education for our cohort was 9.6. This finding is counterintuitive in view of the low rates of AD in the Yoruba. It may suggest that education level is not directly related to AD risk but, instead, serves as a marker for other influences in childhood. In the African Americans, for example, the combination of low education and childhood residence in the rural South increased the risk of AD (22). In the Yoruba, education was not significantly associated with AD rates.

Increasing evidence suggests that vascular risk factors and vascular disease are not only associated with increased risk of stroke-related dementia but may also contribute to the development, progression, and clinical severity of AD (23–25). It is noteworthy that the Yoruba had a lower incidence of both vascular disease and vascular risk factors, including hypertension, than did the African Americans. Many studies, but not all, report that hyper- tension, particularly when occurring in middle age, is associated with an increased risk of late-onset AD (26–30). In our study, self-reports of hypertension were not associated with an increased risk for AD in either the Yoruba or the African Americans. However, both cross-sectional and longitudinal studies found that the use of antihypertensive medication in African Americans was associated with a reduced risk of dementia. Moreover, medication reduced the odds of incident cognitive impairment (that is, poor cognitive performance; cognitive impairment, not dementia; and dementia) by 38% (odds ratio 0.62; 95%CI, 0.45 to 0.84). Antihypertensive medication use was very rare in the Yoruba (31,32).

Cholesterol levels were much lower in Yoruba than in African Americans, probably reflecting dietary differences. It has been suggested that lipid-related mechanisms may have a role in the pathogenesis of AD, and there is intriguing evidence that statins may reduce the risk of incidence of the condition (33). We were unable to confirm a protective effect for statins in our study. However, only a small number of African Americans (n = 30) were using them.

There were great differences in diet between the Yoruba and the African Americans. Many reports have suggested a link between certain constituents of diet and a reduced risk of dementia (for example, high levels of vitamin E and low levels of fat) (34–36). While we anticipate that the Yoruba diet conforms more closely to the hypothetical ideal diet for preserving brain function than does the African–American diet, we do not have the detailed nutritional evaluation necessary to test this hypothesis. Nevertheless, a small pilot study of lymphoblastoid cell lines from 8 Yoruba subjects revealed little evidence of DNA damage, supporting the concept that the low-calorie, low-fat, high-antioxidant Yoruba diet produced less oxidative stress than did the high-calorie diets characteristic of Western countries (37).

Examination of the Association Between Genetic Risk Factors and AD Within Each Population

In contrast to reports from other countries, but consistent with other studies of African Americans, we have found only a relatively weak association between the possession of the apoE e₄ allele and AD in elderly African Americans; the association reached significance only for the e₄ homozygotes (Tables 3 and 4). As in other reports, the apoE e₂ allele shows a protective trend in African Americans (38). In the elderly Yoruba, by contrast, we can determine no significant association between AD and the e4 allele, either for a single or a double copy. In fact, in the Yoruba, only possession of the apoE e₃ allele comes close to significance (39).

Possible genetic mechanisms may explain both the lowered risk in African Americans and the absence of risk relating to the apoE e₄ allele in the Yoruba. First, the apoE e₄ alleles may not be the same from population to population. These alleles may differ by nucleotide changes, which may affect gene function, or by nucleotide changes in the noncoding regions, which may affect gene expression (40,41). Second, nucleotide changes in other genes that interact with apoE e₄ (for example, apolipoprotein receptors) may affect apoE e₄ function and, thus, affect how apoE e₄behaves as a risk factor for AD. We are currently exploring all these possibilities (42).

It is also possible that the lack of apoE e₄ effect is due to gene–environment interaction. The differences in cholesterol levels between sites and the role that apoE plays in cholesterol processing make an apoE–cholesterol interaction, resulting in a differential AD risk, an obvious possibility. Two prior studies have reported a significant interaction between apoE and cholesterol in determining risk for AD (43,44). In fact, Notkola and colleagues suggested that cholesterol mediates some of the effects of apoE e₄ on AD (44). In a preliminary study involving African Americans, we did detect a significant interaction between e₄ and cholesterol. Increasing levels of cholesterol increased the risk of AD, but only in those subjects who did not possess an e₄ allele (45).

Carbo and Scacchi have proposed that apoE e₄ may represent a “thrifty” allele (20). In early human environments, when meeting nutritional needs was uncertain and diets tended to be very low in fat, the e₄ allele, which tends to conserve cholesterol, may have conferred an advantage to its possessor, thus ensuring its survival. Now, in an environment where obesity is a problem and diets are rich in fat, the original advantage associated with possession of the e₄ allele may instead become a liability.

Potential Significance of These Risk Factors in Explaining Differences in Illness Rates Between the Study Populations

It is likely that the lower incidence of vascular risk factors and vascular disease in the Yoruba, compared with the African Americans, accounts for a significant proportion of the differences in incidence rates of AD and dementia. However, our measurements of these factors are currently too imprecise to estimate accurately the effect size owing to them. In our continuing study, we are including biochemical assays of risk factors for vascular disease, such as lipid levels, 8-isoprostanes, homocysteine, insulin, and interleukin-6. Together with more comprehensive brain imaging, these assays will allow us to determine more precisely the extent and the effect of vascular disease on AD risk in the 2 populations.

We did attempt to estimate the effect of the difference in risk associated with the possession of the apoE e₄ allele. However, as Table 5 demonstrates, e₄ accounted for only a small percentage of the variation for developing AD in the African-American population, something akin to the risk associated with years of education. Thus, it is unlikely that the apoE e₄–related difference in risk accounts for much of the difference in AD incidence rates observed between the 2 populations.

It is also possible (or probable) that other as-yet-unidentified genetic or environmental risk factors are associated with the phenotypic variation of AD observed between the populations.

Development of a Risk Factor Model That Should Account for Observed Phenotypic Variation Between Populations

If we return now to Cooper and Kaufman’s disease model (1), our current knowledge could be characterized as follows: With regard to the genetic contribution to disease risk, if the relation to apoE e4 is weak in populations of African origin, it strongly suggests that other as-yet-unidentified genes may be involved in AD in these populations. Several possible candidates exist, but so far, none have been identified conclusively.

Our studies and others strongly suggest that vascular risk factors play an important role in the genesis of AD and the other dementias. It is possible that these factors are influenced primarily by diet and that the constituents of the Yoruba diet (which is low-calorie, low-fat, and possibly, high in anti- oxidants) may lower such potential vascular risk factors as lipid levels and also lower levels of oxidative stress. However, vascular risk factors such as high blood pressure also have a significant genetic component, and genetic differences in association with environmental factors may account for the variations between the populations. Further, the possibility that other social or cultural influences may account for disease variation between the 2 populations should not be ruled out. The protective value of living in large, extended families offering much social interaction and support may be significant.

Our exploration of gene × gene or gene × environment inter- action currently focuses on the attempt to explain the lowered risk associated with apoE e₄ in the 2 populations, either through interaction with lipids or with other contributing genes, as described previously (Table 6).

We hope our continuing work in this project will allow us to propose a more comprehensive explanation in the future.

Conclusion

International comparative studies, particularly those involving populations from developing and developed countries, offer a unique opportunity for applying the new information regarding population genetics to traditional AD risk factor research in a comprehensive model aiming to explain phenotypic variations in populations. By the year 2005, about 70% of all the elderly worldwide will be living in developing countries. The burden of caring for AD patients in these countries is likely to be staggering. We hope that, by including comparisons from the developed and developing worlds, models of AD etiology may be constructed that are more generalizable than those constructed primarily from investigation into Western populations.

Funding and Support

This study is supported by the National Institute of Health Grant AG 09956

References

1. Cooper RS, Kaufman JS. Race and hypertension: science and nescience. Hypertension 1998;32:813–6.

2. Hendrie HC. Exploration of environmental and genetic risk factors for Alzheimer’s’s disease: the value of cross-cultural studies. Current Directions in Psychological Science 2001;10:98–101.

3. Barbujani G, Magagni A, Minch E, Cavalli-Sforza LL. An apportionment of human DNA diversity. Proc Natl Acad Sci USA 1997;94:4516–9.

4. Collins FS, Guyer MS, Charkravarti A. Variations on a theme: cataloging human DNA sequence variation. Science 1997;278:1580–1.

5. Collins FS, Brooks LD, Chakravarti A. A DNA polymorphism discovery resource for research on human genetic variation. Genome Res 1998;8:1229–31.

6. Miller RD, Kwok PY. The birth and death of human single-nucleotide poly- morphisms: new experimental evidence and implications for human history and medicine. Hum Mol Genet 2001;10:2195–8.

7. Gray IC, Campbell DA, Spurr NK. Single nucleotide polymorphisms as tools in human genetics. Hum Mol Genet 2000;9:2403–8.

8. Goldstein DB. Islands of linkage disequilibrium. Nat Genet 2001;29:109–11.

9. Tishkoff SA, Dietzsch E, Speed W, Pakstis AJ, Kidd JR, Cheung K, and others. Global patterns of linkage disequilibrium at the DC4 locus and modern human origins. Science 1996;271:1380–7.

10. Ingman M, Kaessmann H, Paabo S, Gyllensten U. Mitochondrial genome variation and the origin of modern humans. Nature 2000;408:708–13.

11. Armour JA, Anttinen T, May CA, Vega EE, Sajantila A, Kidd JR, and others. Minisatellite diversity supports a recent African origin for modern humans. Nat Genet 1996;13:154–60.

12. Kidd JR, Pakstis AJ, Zhao H, Lu RB, Okonofua FE, Odunsi A, and others. Haplotypes and linkage disequilibrium at the phenylalanine hydroxylase locus, PAH, in a global representation of populations. Am J Hum Genet 2000;66:1882–99.

13. Hendrie HC, Osuntokun BO, Hall KS, Ogunniyi AO, Hui SL, Unverzagt FW, and others. Prevalence of Alzheimer’s’s disease and dementia in two communities: Nigerian Africans and African Americans. Am J Psychiatry 1995;152:1485–92.

14. Canadian Study of Health and Aging Working Group. Canadian Study of Health and Aging: study methods and prevalence of dementia. CMAJ 1994;150:899–913.

15. Hendrie HC, Ogunniyi A, Hall KS, Baiyewu O, Unverzagt FW, Gureje O, and others. Incidence of dementia and Alzheimer’s disease in two communities: Yoruba residing in Ibadan, Nigeria and African Americans residing in Indianapolis, USA. JAMA 2001;285:739–47.

16. Gao S, Hendrie HC, Hall KS, Hui S. The relationship between age, sex and the incidence of dementia and Alzheimer’s disease: a meta-analysis. Arch Gen Psychiatry 1998;55:809–15.

17. Chandra V, Pandav R, Dodge HH, Johnston JM, Belle SH, DeKosky ST, and others. Incidence of Alzheimer’s disease in a rural community in India. The Indo-US Study. Neurology 2001;57:985–9.

18. Perkins AJ, Hui SL, Ogunniyi A, Gureje O, Baiyewu O, Unverzagt FW, and others. Risk of mortality for dementia in a developing country: the Yoruba in Nigeria. Int J Geriatr Psychiatry 2002;17:566–73.

19. Roses AD. Apolipoprotein E alleles as risk factors in Alzheimer’s disease. Annu Rev Med 1996;47:387–400.

20. Carbo RM, Scacchi R. Apolipoprotein E (APOE) allele distribution in the world. Is APOE*4 a ‘thrifty’ allele? Ann Hum Genet 1999;63:301–10.

21. Ogunniyi A, Baiyewu O, Gureje O, Hall KS, Unverzagt F, Oluwole OSA, and others. Morbidity patterns in a sample of elderly Nigerians resident in Idikan Community, Ibadan. West Afr Med J 2001;20:227–31.

22. Hall KS, Gao SJ, Unverzagt FW, Hendrie HC. Low education and childhood rural residence: risk for Alzheimer’s disease in African Americans. Neurology 2000;54:95–9.

23. Ellis RJ, Olichney JM, Thal LJ, Mirra SS, Morris JC, Beekly D, and others. Cerebral amyloid angiopathy in the brains of patients with Alzheimer’s disease: the CERAD experience, Part XV. Neurology 1996;46:1592–6.

24. Snowden DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR. Brain infarction and the clinical expression of Alzheimer’s disease: the Nun Study. JAMA 1997;227:813–7.

25. Skoog I. The interaction between vascular disorders and Alzheimer’s disease. Alzheimer’s Dis Related Disord 1999;58:523–30.

26. Launer LJ, Masaki K, Petrovich H, Foley D, Havlik RJ. The association between midlife blood pressure levels and late-life cognitive function: the Honolulu-Asia Aging Study. JAMA 1995;274:1846–51.

27. Kivipelto M, Helkala EL, Laakso MP, Hanninen T, Hallikainen M, Alhainen K, and others. Midlife vascular risk factors and Alzheimer’s disease in later life: longitudinal, population based study. BMJ 2001;322:1447–51.

28. Desmond DW, Tatemichi TK, Paik M, Stern Y. Risk factors for cerebrovascular diasese as correlates of cognitive function in a stroke-free cohort. Arch Neurol 1993;50:162–6.

29. Glynn RJ, Beckett LA, Hebert LE, Morris MC, Scherr PA, Evans DA. Current and remote blood pressure and cognitive decline. JAMA 1999;281:438–45

30. Guo Z, Viitanen M, Fratiglioni L, Winblad B. Low blood pressure and dementia in elderly people: the Kungsholmen project. BMJ 1996;312:805–8.

31. Murray MD, Lane KA, Gao S, Evans RM, Unverzagt FW, Hall KS, and others. Preservation of cognitive function with antihypertensive medications: a longitudinal analysis of a community-based sample of African Americans. Arch Intern Med 2002;162:2090–6.

32. Richards SS, Emsley CL, Roberts J, Murray MD, Hall K, Gao S, and others. The association between vascular risk factor mediating medications and cognition and dementia diagnosis in a community-based sample of African Americans. J Am Geriatr Soc 2000;48:1–7.

33. Jick H, Zornberg GL, Jick SS, Seshadri S, Drachman DA. Statins and the risk of dementia. Lancet 2000;356:1627–31

34. Morris MC, Evans DA, Bienias JL, Tangney CC, Bennett DA, Aggarwal N, and others. Dietary intake of antioxidant nutrients and the risk of incident Alzheimer’s disease in a biracial community study. JAMA 2002;287:3261–3.

35. Perkins AJ, Hendrie HC, Callahan CM, Gao S, Unverzagt FW, Xu Y, and others. Association of antioxidants with memory in multiethnic elderly sample using the Third National Health and Nutrition Examination Survey. Am J Epidemiol 1999;150:37–44.

Butterfield DA, Koppal T, Subramaniam R, Yatin S. Vitamin E as an antioxidant/ free radical scavenger against amyloid beta-peptide-induced oxidative stress in neocortical synaptosomal membranes and hippocampal neurons in culture: insights into Alzheimer’s disease. Rev Neurosci 1999;10:141–9

37. Lahiri DK, Xu Y, Klaunig J, Baiyewu O, Ogunniyi A, Hall K, and others. Effect of oxidative stress on DNA damage and beta-amyloid precursor proteins in lymphoblastoid cell lines from a Nigerian population. Ann N Y Acad Sci 1999;893:331–6

38. Sahota A, Yang M, Gao S, Hui SL, Baiyewu O, Gureje O, and others. Apolipoprotein E - associated risk for Alzheimer’s disease in the African American population is genotype dependent. Ann Neurol 1997;42:659–61.

39. Osuntokun BO, Sahota A, Ogunniyi AO, Gureje O, Baiyewu O, Adeyinka A, and others. Lack of an association between the apolipoprotein E e₄ allele and Alzheimer’s disease in elderly Nigerians. Ann Neurol 1995;38:463–5.

Fullerton SM, Clark AG, Weiss KM, Nickerson DA, Taylor SL, Stengard JH, and others. Apolipoprotein E variation at the sequence haplotype level: implications for the origin and maintenance of a major human polymorphism. Am J Hum Genet 2000;67:881–900

41. Nickerson DA, Taylor SL, Fullerton SM, Weiss KM, Clark AG, Stengard JH, and others. Sequence diversity and large-scale typing of SNPs in the human apolipoprotein E gene. Genome Res 2000;10:1532–45.

42. Hui DY, Innerarity TL, Mahley RW. Defective hepatic lipoprotein receptor binding of beta-very low density lipoproteins from type III hyperlipoproteinemic patients. Importance of apolipoprotein E. J Biol Chem 1984;259:860–9.

43. Jarvik GP, Wijsman EM, Kukull WA, Schellenberg GD, Yu C, Larson EB. Interactions of apolipoprotein E genotype, total cholesterol level, age, and sex in prediction of Alzheimer’s disease: a case-control study. Neurology 1995;45:1092–6.

44. Notkola IL, Sulkava R, Pekkanen J, Erkinjuntti T, Ehnholm C, Kivinen P, and others. Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer’s disease. Neuroepidemiology 1998;17:14–20

45. Evans RM, Emsley CL, Gao S, Sahota A, Hall KS, Farlow MR, and others. Serum cholesterol, APOE genotype and the risk of Alzheimer’s disease in a population-based study of African Americans. Neurology 2000;54:240–2

Author(s)

Manuscript received and accepted November 2003.

1. Professor of Psychiatry, Department of Psychiatry, Indiana University School of Medicine; Research Scientist, The Center for Aging Research, Regenstrief Institute; Indianapolis, Indiana.

2. Associate Professor, Department of Psychiatry, Indiana University School of Medicine, Indianapolis, Indiana

3. Professor and Chair, Department of Medicine, University College Hospital, Ibadan, Nigeria.

4. Associate Professor, Division of Biostatistics, Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana.

Address for correspondence: Dr HC Hendrie, Regenstrief Institute, Indiana University School of Medicine, Indiana University Center for Aging Research, 1050 Wishard Boulevard, RG6, Indianapolis, IN 46202–2872

e-mail: hhendri@iupui.edu

1 | 2