Secondary Logo

Share this article on:

Hypertension, the Kuna, and the Epidemiology of Flavanols

McCullough, Marjorie L. ScD, RD*; Chevaux, Kati BS; Jackson, Lilian BS, RD; Preston, Mack RN; Martinez, Gregorio MD; Schmitz, Harold H. PhD; Coletti, Caroline BA, MS§; Campos, Hannia PhD; Hollenberg, Norman K. MD§

Journal of Cardiovascular Pharmacology: June 2006 - Volume 47 - Issue - p S103-S109
Original Articles

A low sodium diet has often been implicated in the protection of low blood pressure populations from hypertension, but several other dietary factors, including those as yet unidentified, may also be involved. The Kuna Indians of Panama are free of hypertension and cardiovascular disease, but this is changing with migration to urban areas. We compared the indigenous diet of Kuna Indians living on remote islands in Panama (Ailigandi), whose lifestyle is largely hunter-gatherer, with those who have moved to a suburb of Panama City (Vera Cruz). Between April and October 1999, members of a Kuna research team administered a 118-item food frequency questionnaire to133 adult Kuna from Ailigandi and 183 from Vera Cruz. Single 24-hour urine collections and nonfasting blood samples were obtained. The Kuna in Ailigandi reported consuming a 10-fold higher amount of cocoa-containing beverages, 4 times the amount of fish, and twice the amount of fruit as urban Kuna (P<0.05 by t test). Salt added was ample among those living in Ailigandi and Vera Cruz according to both self-report (7.1±1.1 and 4.6±0.3 tsp weekly) and urinary sodium levels (177±9 and 160±7 mEq Na/g creatinine), respectively. The low blood pressure of island-dwelling Kuna does not seem to be related to a low salt diet. Among dietary factors that varied among migrating Kuna, the notably higher intake of flavanol-rich cocoa is a potential candidate for further study.

*Epidemiology and Surveillance Research, American Cancer Society, Atlanta, GA

Mars, Incorporated, Hackettstown, NJ

Panamanian Ministry of Health, Panama City, Panama

§Department of Medicine and Radiology at Brigham and Women's Hospital and Harvard Medical School

Department of Nutrition, Harvard School of Public Health, Boston, MA

Supported by National Institutes of Health Grants T32 HL-07609, NCRR GCRC M01RR026376, 1 P50ML53000-01, 1 R01 DK54668-01, the Baxter Foundation and Mars, Incorporated.

Reprints: Marjorie McCullough, ScD, RD, Epidemiology and Surveillance Research, American Cancer Society, 1599 Clifton Road, NE, Atlanta, GA 30309-4251 (e-mail:

Approximately 25% of US adults have hypertension, a major risk factor for coronary heart disease, stroke and premature death.1,2 Even high normal blood pressure is associated with graded, increased risk of cardiovascular disease.2,3 The “normal” increase in blood pressure seen in the United states is not observed in all populations, and blood pressure has been shown to increase with migration within populations, implicating environmental etiology.

Based largely on observational data in vegetarian and migrating populations, several dietary factors have been hypothesized to influence blood pressure. Fairly established dietary risk factors for hypertension include excessive body weight, salt, and alcohol. In contrast, higher intakes of potassium, magnesium, calcium, and perhaps fiber and protein, as found in fruit, vegetables, low fat dairy products, are associated with lower risks.4 However, even these factors are inconsistently associated or minimally associated with changes in blood pressure when examined in isolation. In the mid 1990s, the Dietary Approaches to Stop Hypertension (DASH) trial hypothesized that combinations of dietary factors would have additive or interactive effects on blood pressure lowering.5 The DASH feeding study, which was conducted in US men and women with borderline hypertension, found that a diet high in fruits and vegetables, whole grains, low fat dairy products, and a higher ratio of unsaturated to saturated fats lowered blood pressure as much as a single antihypertensive agent.6 A subsequent study, the DASH Sodium trial,7 observed further lowering of blood pressure with a low sodium diet (∼65 mEq/d) in combination with the DASH diet. The current JNC VII lifestyle guidelines for hypertension prevention recommend weight management, physical activity, limiting alcohol consumption and salt intake, and consuming a DASH diet pattern high in potassium and calcium.8

Nevertheless, many dietary hypotheses in relation to hypertension and other chronic diseases remain inadequately explored. Factors in foods with putative cardiovascular benefits such as the vasodilator action of certain plant-based flavonoids, are rapidly gaining attention.9,10 However, many populations do not consume diets varied enough in certain factors of interest to enable meaningful epidemiologic analyses.

At approximately the same time that the DASH trial was evaluating diet patterns to lower blood pressure in borderline hypertensives, we had the opportunity to evaluate diet in a migrating, low blood pressure population. The Kuna Indians of Panama living in their indigenous island setting were documented to be free of hypertension and cardiovascular disease in the 1940s11 and again in the 1990s12, even though salt intake is comparable to that in the US.12 Their freedom from cardiovascular disease seems not to be genetic, but rather influenced by environmental exposures, because higher blood pressure has been documented among migrant Kuna.12 Migration can influence lifestyle behaviors dramatically, including dietary choices, owing to changes in food accessibility. Most “unacculturated” populations free of chronic diseases,13 including the Kuna, have traditionally resided in remote areas, and logistic and cultural practises have made quantitative estimates of dietary intake challenging. For the Kuna, these challenges include an unwritten native language, culturally specific cooking and eating practises, and incomplete food composition tables. With the participation of a Kuna registered dietitian (L.J.), who has intimate knowledge of cooking and eating practises of the Kuna, and other local health care workers and community members, we developed a food frequency questionnaire (FFQ) to assess usual dietary intake among the Kuna residing in different areas. Herein we describe and compare dietary patterns in the indigenous, island-dwelling Kuna community and in Kuna that had migrated to the suburbs of Panama City as part of an ongoing study on cardiovascular risk among the Kuna.

Back to Top | Article Outline


Kuna, Nutrition, Health and Hypertension is an ongoing, cross-sectional study of correlates of blood pressure in the Kuna who are undergoing cultural transition. This is a collaborative effort of investigators from the Ministry of Health in Panama, Mars, Incorporated, the Brigham and Women's Hospital and Harvard Medical School, and the Harvard School of Public Health, Boston, MA. The study protocol was approved by the investigational review boards of the Brigham and Women's Hospital, Boston, MA and the Panamanian Ministry of Health.

Back to Top | Article Outline

Data Collection

All data were collected in Ailigandi (an island on the Caribbean coast of Panama) and Vera Cruz (a suburb of Panama City) between April and October 1999. Residents of Vera Cruz had migrated on average approximately 20 years earlier. Team leaders from Ailigandi participated in protocol development and data collection training for all aspects of the study. The leaders consisted of a Kuna physician (G.M.), registered nurse (M.P.), registered dietitian (L.J.), and 6 trained hospital workers. These individuals subsequently trained local health workers, bilingual in Spanish and Kuna, who recorded all information in Spanish. The family was the sampling unit so that prepared dishes and commonly consumed foods could be sampled for analysis with the assistance of the Jefe (cook, usually mother). Kuna families were identified from census information, and then a subset was randomized to participation order. Local community members described the study, obtained consent, and acted as liaisons during data collection. Modest monetary remuneration was supplemented with sewing equipment for female participants, and fishing supplies for men.

After informed consent was obtained, the FFQ and medical questionnaires were administered. Within 2 weeks of this date, urine collection jugs and detailed directions for 24-hour collections were provided, with the Jefe acting as overseer. On the following day, completed 24-hour urine samples were collected, and remaining questionnaire data, and height (without shoes, in cm), weight (in light clothing, in kg), and blood pressure (mm Hg, sphygmomanometry) were measured. Urine collections were measured, fractionated into aliquots, and stored in a dedicated, diesel-powered freezer. On the same day, a blood sample was drawn to measure serum electrolytes and cholesterol. Blood and urine specimens were transported on dry ice by chartered flight to Panama City and then to Boston for assay. Thermometers designed to record the highest temperature experienced during storage were employed to ensure sample stability. Measurements of plasma and urine electrolytes, urea, and creatinine were carried out by autoanalyzer.

Back to Top | Article Outline


Development of the Kuna FFQ was a multidisciplinary effort that spanned several years.14 Food and beverage items for inclusion were initially identified by discussion among US and Kuna investigators. Additional items were identified from foods commonly reported on 24-hour recalls,12 by interview with a Kuna dietitian and other team members, and through pilot testing of the FFQ in which participants listed commonly consumed items not provided on the FFQ. The involvement of the Kuna in FFQ development and administration was instrumental.15 Typical portion sizes were estimated using food models, standard portions, and several measured sizes of local “noidars” (half-gourds) used for bowls. Kuna food preparation methods and culturally sensitive questions were considered during the development of forms and training materials. The final FFQ contained 118 items, asked about usual food intake patterns over the previous year (ranging from never to number of times per week), and included a separate section for seasonal items. For this, the questionnaire included the section, “when in season, how often do you consume this item?” When it seemed that the respondent did not understand the seasonality component, or other part of the questionnaire, a registered dietitian clarified the response. Throughout pilot testing of the FFQ, several changes were made to include dietary habits in various regions where Kuna reside. This paper compares FFQ responses between 2 regions (Ailigandi and Vera Cruz) in which the FFQ versions were most similar. For purposes of analysis, a season was assumed to last 3 months. All reported portion sizes were converted to similar units for comparison. Computerization of the FFQ has not yet been completed; hence, nutrient estimates from the FFQ are not available. A full-scale FFQ validation has not yet been feasible. However, the Pearson correlation between (natural) log-transformed urinary potassium (mg potassium/g creatinine) and log-transformed number of servings of fruits and vegetables per week reported on the FFQ among participants was high (r=0.66).

Food intakes and population characteristics are provided as means with standard error of the mean as the index of dispersion. Unpaired t tests were used to compare food intake and other characteristics in the 2 groups (SAS version 8.01, SAS Institute, Cary, NC). A P value of ≤0.05 was considered statistically significant.

Back to Top | Article Outline


Demographic and health characteristics of participants from Ailigandi and Vera Cruz are shown in Table 1. Individuals studied in Ailigandi were slightly older than those interviewed in Vera Cruz. Both study samples were comprised of approximately two-thirds women (more women were available to study because they were at home during the day). Blood pressure levels were lower on Ailigandi, although hypertension was uncommon in both groups. The Kuna residing in Vera Cruz weighed more than those in Ailgandi, but the difference in body mass index did not achieve statistical significance. Serum cholesterol was similar in both groups. Urinary biomarkers were normalized to gram of urinary creatinine to adjust for potentially incomplete 24-hour urine samples. Urinary sodium values from both locations were comparable, and indicate intakes similar to that in the United states, corroborating our earlier report.12 Urinary potassium was slightly, and statistically significantly, lower in Vera Cruz than in Ailigandi; other electrolytes and urea levels were similar.



Figures 1A–D provide a comparison of intakes of fruit, protein sources, salt/sweets/fat, and beverages in the 2 regions. Our findings indicate roughly twice the intake of fruit, especially plantains, bananas, and mangos on the islands compared with Vera Cruz (Fig. 1A). A higher consumption of both fresh and preserved (smoked, canned) fish was reported on the islands (43 versus 10 oz/wk, or 6 versus 1.5 oz/d) (Fig. 1B). The overall difference in intake of protein sources is in contrast with the estimates of urinary urea (Table 1), which suggests similar intakes of protein in both locations, but an FFQ and a single urine sample cover different time periods. Kuna residing in Ailigandi reported adding approximately 7 tsp of salt to food weekly, although those in Vera Cruz reported 4.6 tsp weekly (Fig. 1C). Consumption of home-prepared corn beverages and cocoa beverages in Ailigandi exceeded that in Vera Cruz by 10-fold (Fig. 1D). On Ailigandi, cocoa (either processed brand obtained from Colombian trade boats or home-grown and ground) is a staple: most Kuna reported drinking an average of at least three 10-oz servings of cocoa beverage (plain, or with corn) per day. Other traditional beverages (eg, inna, a traditional corn beverage) were consumed more regularly on the island, and tea and soda were consumed more regularly in Vera Cruz.



In addition, margarine, mayonnaise, and cookies are consumed more regularly in Vera Cruz, although added sugar (obtained from Columbian trade boats) was higher on the islands. The main vegetables consumed by the Kuna include pimento, yucca, corn, kidney beans, and some leafy greens. Patterns of intake varied slightly for certain vegetables (average number of pimento per week in Ailigandi vs. Vera Cruz, 4.3±0.7 vs. 1.0±0.1/wk, P<0.001; number of avocado per week, 3.2±0.5 vs. 0.01±0.01, P<0.001) (not shown). Lettuce was consumed more frequently in Vera Cruz: 2.3±0.2 1/2 cup servings/wk versus 0.8±0.1 servings in Ailigandi, P<0.001. Milk and ice cream are consumed more regularly in Vera Cruz than in Ailigandi (1.2±0.1 cups/wk vs. 0.3±0.1 cups/week, P<0.001 and 1.5±0.4 cups/wk vs. 0.2±0.06 cups/wk, P<0.05, respectively), but intakes of dairy products are low in both locations. Alcoholic beverage consumption was twice as high in Vera Cruz compared with Ailigandi, although intakes are still low (86±21 vs 42±11 mL/wk, P=0.07), as is typical of Kuna tradition (not shown).

Back to Top | Article Outline


Kuna Indians living a traditional lifestyle in the San Blas Island chain consume a diet rich in fish, cocoa, and fruit. Those living in the suburban village of Vera Cruz and work in Panama City consume primarily store-bought foods and their diet is lower in fruit and much lower in fish and cocoa. Salt is available on the islands, and the FFQ and urinary sodium levels indicate ample intake in both locations. Kuna residing in the Panama City suburb also had a slightly greater weight than those residing on Ailigandi, but the difference in body mass index was not statistically significant. An FFQ specifically developed for this population was well accepted and was able to distinguish large differences in several key food groups thought to influence cardiovascular health. Although this was a cross-sectional analysis, these findings indicate that departures from the traditional Kuna diet occur with migration, and provide ideas for additional hypotheses on how diet may influence hypertension and heart disease.

The low blood pressure of isolated societies is often attributed primarily to low sodium intake,16–18 although our studies12 and those of others19 suggest that low blood pressure can be maintained even with a sodium intake similar to that in the United states. A higher potassium intake, greater physical activity, and low alcohol consumption has also been documented in these populations in some,13,18,20 but not all21 studies. We observed a slight difference in urinary potassium and alcohol between Kuna residing in the 2 locations.

The diets of indigenous, traditional societies and migrating, “low blood pressure populations” have been studied by many investigators.13,22 Previous dietary assessment methodologies employed in similar remote populations included observation,17,23 assessment of per capita intake,24 one to three 24-hour recalls,12,20,21,25 urine collections,12,16,18,20,26 or the diet history method.19 However, assessment of long-term diet (eg, 1 y) is of particular interest in the study of chronic disease etiology because important variations in seasonal intakes may be missed with only a few days’ worth of intake data. Further, multiple recall methods are often cost-prohibitive in large, ongoing studies, making the FFQ a reasonable alternative. The FFQ has not been employed previously to study diet in low blood pressure populations, but it has been used successfully to document changing diet quality in populations undergoing health transition, including indigenous populations in Canada27 and rural studies in Mexico and Central America.28 The Kuna FFQ development spanned years of testing and feedback; the method was well accepted by the Kuna, and seems to have captured important differences in food intake.

The diet patterns we observed in this low blood pressure population share some aspects in common with the DASH diet, which reduced blood pressure as much as drug monotherapy in US hypertensives.6 The traditional diet in Ailigandi is high in fruit (especially bananas, plantains, and mangos when in season), and fish, and low in animal products and animal fat. However, unlike the DASH diet, the Kuna consume few dairy products, or calcium, and their intake of nuts is infrequent. They also consume fewer and different vegetables (beans, pimento, green lettuce, and tomato) than DASH (green leafy, leguminous, etc).29 Regardless, blood pressure on the “ideal” DASH diet did not reach levels enjoyed by island-dwelling Kuna, indicating the potential for discovery of further protective factors.

An unexpected finding early in our research was the unusually high amount of cocoa consumed by the Kuna in their traditional setting. Although the majority of cocoa consumed is purchased from Columbian traders, the Kuna also grow their own cocoa. In a previous paper, Kuna cocoa sources (home-grown and Columbian cocoa powder) were shown to be high in certain flavonoids, especially the flavanols and procyanidins.30 At 10% weight per volume, we estimate that the Kuna may consume approximately 1880 mg procyanidins per day from 4, 8 ounce cups of cocoa beverage (Columbian brand).30 Thus, the several glasses of cocoa consumed daily by island-dwelling Kuna may contribute importantly to intake of flavanols and procyanidins. Accumulating evidence suggests that flavanol-rich foods, present in high quantities in some plant-based foods, have favorable cardiovascular effects, including vasodilation which could contribute to blood pressure lowering.9,10,31–35 Indeed, flavanol-rich cocoa has been shown to activate vascular nitric oxide synthase in human beings, providing a plausible mechanism for a blood pressure effect.34

Several reviews suggest a potential reduction in CVD mortality with higher intakes of a variety of flavonoid compounds.10,36–39 However, few epidemiologic studies specifically examined flavanol compounds (catechins, epicatechins, and their oligomers known as procyanidins) or cocoa in relation to cardiovascular disease risk. A meta-analysis of high flavanol tea consumption concluded that consumption of 2 to 3 cups of tea daily reduced coronary heart disease (CHD) mortality by 11%.40 However, tea is not an exact proxy for cocoa owing to the presence of different flavanol compounds, different polyphenol compounds, and different concentrations of other nutrients: cocoa is higher in magnesium, calcium, and potassium than tea.

Two prospective studies have specifically evaluated cocoa or certain flavanols in relation to CVD risk.41,42 In a study of 805 elderly Dutch men followed for 10 years, Arts et al41 observed a significantly lower risk of fatal ischemic heart disease, but not of incident MI, stroke or stroke, mortality, among those with the highest compared with lowest total catechin intake. During a 12-year follow-up of over 34,000 women from the Iowa Women's Health study, women with a high intake of certain catechin-based compounds were at a nonstatistically significant lower risk of CVD risk than those with lowest intakes.42 When specific compounds were examined, epicatechin plus catechin was associated with a borderline 24% lower risk, compared with total gallate, which was null.42 When foods were examined, pears and wine were associated with significantly lower risk, but tea was not.42 Chocolate was nonsignificantly inversely associated with risk, but “chocolate” included a variety of dessert sources that may not be concentrated sources of flavanols such as catechin or epicatechin. One case-control study conducted in Greece43 used a recent USDA flavonoid database44 to examine the relationship between all 6 flavonoid classes and CHD risk. The only statistically significant association observed was a 24% decrease in risk with each 21 mg increase in flavanol intake (as monomers). Finally, in a cross-sectional analysis of the SU VI MAX prospective trial, a higher intake of flavanol-containing foods, including cocoa, was inversely related to estimated CVD risk in women, but not in men.45 Thus, the few epidemiologic studies of cocoa or flavanols conducted thus far suggest a possible lowering of CVD risk, but more research is needed.

To move the field forward, more research on the absorption and metabolism of flavanols is needed. In addition, researchers need to clearly identify which compounds are being quantified and presented (eg, small molecular weight compounds such as epicatechin vs. larger molecular weight epicatechin-based procyanidins), and employ similar terminology to enhance interstudy comparisons. Additional studies in populations with wide ranges of intakes of a variety of flavanol consuming foods are also needed. The recent publication of the USDA flavonoid database44 and proanthocyanidin database,46 and inclusion of relevant food sources on epidemiologic study questionnaires, will facilitate more thorough investigations of flavonoid compounds in the future. In addition, careful consideration must be given to the fact that common agricultural, postharvest handling, and manufacturing practises associated with plant-based food products can lead to significant reduction or total elimination of flavanols that may have been present in freshly harvested food crops. The fate of flavanols in cocoa and cocoa-based products provides a useful case study to learn from, as does the data sets obtained from different apple cultivars.9,47 Investigators need to be aware that erroneous conclusions can be reached based on faulty assumptions made about flavanol content of food categories.

In our study of the Kuna, there were some limitations that deserve mention. We did not have resources to computerize and fully validate the FFQ. However, reported total fruit and vegetable consumption was highly correlated with urinary potassium (r=0.66), suggesting reasonable validity for these exposures. A high priority in the future will be to undertake a comprehensive validation of the FFQ using several recalls throughout the year, and to evaluate reproducibility of the FFQ during different seasons.

Major strengths of this study include the close involvement of Kuna Indians in the study design, development of tools, and administration of FFQ. For example, the Kuna registered dietitian who reviewed all FFQs has worked intimately with Kuna families in both locations. She flagged potentially inaccurate reports and followed up by clarifying reports with data collectors or participants. Assessment of the diet and lifestyle of remote populations must involve individuals from the population under study to enhance accurate reporting.

Traditional, indigenous, migrating populations offer unique opportunities to study relationship between diet, other lifestyle factors, and chronic disease development, for 2 primary reasons: first, they are more likely to be genetically homogeneous, allowing the examination of environmental separate from genetic factors, and secondly, because large variations in exposure accompany migration. Unfortunately, with urbanization and modernization encroaching on most of these groups, the opportunity to study these important factors is quickly disappearing.

Our study was cross-sectional and thus can only be considered hypothesis-generating, although cross-sectional assessments of the diet of migrating populations are considered appropriate during rapid transition.48 These data would suggest that the low blood pressure of traditional-dwelling Kuna and the potential changes with acculturation would not be explained primarily by differences in sodium. The most interesting and potentially protective feature of the Kuna diet may be its entirety: a plant-based and fish-based diet pattern supplemented with foods rich in certain flavonoids. Other lifestyle factors, including changes in physical activity, smoking, weight gain, and stress, should also be studied in relation to acculturation among the Kuna. Future prospective studies should examine whether maintenance of this diet pattern in migrating Kuna will help to maintain cardiovascular health.

Back to Top | Article Outline


1. Burt VL, Whelton P, Roccella EJ, et al. Prevalence of hypertension in the U.S. adult population. Results from the Third National Health and Nutrition Examination Survey, 1988–1991. Hypertension. 1995;25:305–313.
2. Stamler J. Blood pressure and high blood pressure. Aspects of risk. Hypertension. 1991;18:I95–I107.
3. Stamler J, Neaton JD, Wentworth DN. Blood pressure (systolic and diastolic) and risk of fatal coronary heart disease. Hypertension. 1989;13:I2–I12.
4. McCullough M, Lin PH. Nutrition, diet, and hypertension. In: Coulston AM, Rock CL, Monsen ER, eds. Nutrition in the Prevention and Treatment of Disease. San Diego: Academic Press; 2001:303–324.
5. Sacks F, Obarzanek E, Windhauser M, et al. Rationale and design of the dietary approaches to stop hypertension trial. Ann Epidemiol. 1995;5:108–118.
6. Appel L, Moore T, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med. 1997;336:1117–1124.
7. Sacks FM, Svetkey LP, Vollmer WM, et al. Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. DASH-Sodium Collaborative Research Group. N Engl J Med. 2001;344:3–10.
8. NIH. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Bethesda: National Heart Lung and Blood Institute; 2004.
9. Hammerstone JF, Lazarus SA, Schmitz HH. Procyanidin content and variation in some commonly consumed foods. J Nutr. 2000;130(Suppl):2086S–2092S.
10. Kris-Etherton PM, Keen CL. Evidence that antioxidant flavonoids in tea and cocoa are beneficial for cardiovascular health. Curr Opin Lipid. 2002;13:14–49.
11. Kean BH. The blood pressure of the Cuna Indians. A J Trop Med. 1944;24:341–343.
12. Hollenberg NK, Martinez G, McCullough M, et al. Aging, acculturation, salt intake, and hypertension in the Kuna of Panama. Hypertension. 1997;29:171–176.
13. James GD, Baker PT. Human population biology and hypertension. Evolutionary and ecological aspects of blood pressure. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis and Management. New York: Raven Press, Ltd; 1990:137–145.
14. Teufel NI. Development of culturally competent food-frequency questionnaires. Am J Clin Nutr. 1997;65(Suppl):1173S–1178S.
15. Jerome NW. Culture-specific strategies for capturing local dietary intake patterns. Am J Clin Nutr. 1997;65(Suppl):1166S–1167S.
16. Oliver WJ, Cohen EL, Neel JV. Blood pressure, sodium intake, and sodium related hormones in the Yanomamo Indians, a “no-salt” culture. Circulation. 1975;52:146–151.
17. Page LB, Damon A, Moellering RC. Antecedents of cardiovascular disease in six Solomon Islands societies. Circulation. 1974;XLIX:1132–1146.
18. Carvalho JJM, Baruzzi RG, Howard PF, et al. Blood pressure in four remote populations in the INTERSALT study. Hypertension. 1989;14:238–246.
19. Connor WE, Cerqueira MT, Connor RW, et al. The plasma lipids, lipoproteins, and diet of the Tarahumara Indians of Mexico. Am J Clin Nutr. 1978;31:1131–1142.
20. Poulter NR, Khaw KT, Hopwood BEC, et al. The Kenyan Luo migration study: observations on the initiation of a rise in blood pressure. BMJ. 1990;300:967–972.
21. He J, Tell GS, Tang Y-C, et al. Effect of migration on blood pressure: the Yi People study. Epidemiology. 1991;2:88–97.
22. Lowenstein FW. Blood-pressure in relation to age and sex in the tropics and subtropics. Lancet. 1961;1:389–392.
23. Ree GH. Arterial pressures in a West African (Gambian) rural population. J Trop Med Hyg. 1973;76:65–70.
24. Prior IAM. Cardiovascular epidemiology in New Zealand and the Pacific. NZ Med J. 1974;80:245–252.
25. Eason RJ, Pada J, Wallace R, et al. Changing patterns of hypertension, diabetes, obesity and diet among Melanesians and Micronesians in the Solomon Islands. Med J Australia. 1987;146:465–473.
26. Kaufman JS, Owoaje EE, James SA, et al. Determinants of hypertension in West Africa: contribution of anthropometric and dietary factors to urban-rural and socioeconomic gradients. Am J Epidemiol. 1996;143:1203–1218.
27. Receveur O, Boulay M, Kuhnlein HV. Decreasing traditional food use affects diet qualtiy for adult Dene/Metis in 16 communities of the Canadian Northwest Territories. J Nutr. 1997;127:2179–2186.
28. Romieu I, Hernandez-Avila M, Rivera JA, et al. Dietary studies in countries experiencing a health transition: Mexico and Central America. Am J Clin Nutr. 1997;65(Suppl):1159S–1165S.
29. Karanja N, Obarzanek E, Lin P-H, et al. Descriptive characteristics of the dietary patterns used in the Dietary Approaches to Stop Hypertension trial. J Am Dietetic Assoc. 1999; 99(Suppl):S19–S27.
30. Chevaux KA, Jackson L, Villar ME, et al. Proximate, mineral and procyanidin content of certain foods and beverages consumed by the Kuna Amerinds of Panama. J Food Comp Anal. 2001;14:553–563.
31. Holt RR, Schramm DD, Keen CL, et al. Chocolate consumption and platelet function. JAMA. 2002;287:2212–2213.
32. Taubert D, Berkels R, Roesen R, et al. Chocolate and blood pressure in elderly individuals with isolated systolic hypertension. JAMA. 2003;290:1029–1030.
33. Heiss C, Dejam A, Kleinbongard P, et al. Vascular effects of cocoa rich in Flavan-3-ols. JAMA. 2003;290:1030–1031.
34. Fisher NDL, Hughes M, Gerhard-Herman M, et al. Flavonol-rich cocoa induces nitric-oxide-dependent vasodilation in healthy humans. J Hypertension. 2003;21:1–6.
35. Keen CL, Holt RR, Oteiza PI, et al. Cocoa antioxidants and cardiovascular health. Am J Clin Nutr. 2005;81:298S–303S.
36. Hollman PCH, Katan MB. Health effects and bioavailabaility of dietary flavonols. Free Rad Res. 1999;31:S75–S80.
37. Maron DJ. Flavonoids for reduction of atherosclerotic risk. Curr Atheroscl Rep. 2004;6:73–78.
38. Ross JA, Kasum CM. Dietary flavonoids: bioavailability, metabolic effects, and safety. Annu Rev Nutr. 2002;22:19–34.
39. Arts ICW, Hollman PCH. Polyphenols and disease risk in epidemiologic studies. Am J Clin Nutr. 2005;81:317S–325S.
40. Peters U, Poole C, Arab L. Does tea affect cardiovascular disease? A meta-analysis. Am J Epidemiol. 2001;154:495–503.
41. Arts ICW, Hollman PCH, Feskens EJM, et al. Catechin intake might explain the inverse relation between tea consumption and ischemic heart disease: the Zutphen Elderly Study. Am J Clin Nutr. 2001;74:227–232.
42. Arts ICW, Jacobs DRJ, Harnack LJ, et al. Dietary catechins in relation to coronary heart disease death among postmenopausal women. Epidemiology. 2001;12:668–675.
43. Lagiou P, Samoli E, Lagiou A, et al. Intake of specific flavonoid classes and coronary heart disease—a case-control study in Greece. Eur J Clin Nutr. 2004;58:1643–1648.
44. Nutrient Data Laboratory. USDA Database for the Flavonoid Content of Selected Foods. Beltsville, MD: US Department of Agriculture; 2003.
45. Mennen LI, Sapinho D, de Bree A, et al. Consumption of foods rich in flavonoids is related to a decreased cardiovascular risk in apparently healthy French women. J Nutr. 2004;134:923–926.
46. Nutrient Data Laboratory. USDA Database for the Proanthocyanidin Content of Selected Foods. Beltsville, MD: US Department of Agriculture; 2004.
47. Fisher ND, Hollenberg NK. Flavanols for cardiovascular health: the science behind the sweetness. J Hypertens. 2005;23:1453–1459.
48. Whiting SJ, Mackenzie ML. Assessing the changing diet of indigenous peoples. Nutr Rev. 1998;56:248–250.

diet; migration; blood pressure; cocoa; flavonoids; salt; flavanols

© 2006 Lippincott Williams & Wilkins, Inc.