Subsistence patterns and blood pressure variation in two rural Caboclo communities of Marajó Island, Pará, Brazil

zy zyxwvuts zy zyx zyxwvuts zyxwv zyxwv AMERICAN JOURNAL OF HUMAN BIOLOGY 7:535-542 (1995) Subsistence Patterns and Blood Pressure Variation in Two Rural Caboclo Communities of Marajo Island, Para, Brazil HILTON P SILVA,’ DOUGLAS E CREWS,’ ’AND WALTER A NEWS3 ‘Department of Anthropology and Department of Preventive Medicine, The Ohio State Uniuersitv. Columbus. Ohm 43210. 3Departmento de Biologtn, Unwer.sidade de Sao Paulo, Sao Paulo, SP 05422-970, Brazil ABSTRACT Blood pressure (BP) increases with age in westernized societies, is higher in men, and is correlated with the body mass index (BMI). Traditional societies present more variable patterns of BP. In 1991, BP and anthropometric data from two “Caboclo”(rural populations of mixed ancestry) groups from Marajo Island, Brazil, were collected: The Paricatuba group, (N = 20: 12 women), with a subsistence base of fishing, collection of palm fruits, and traditional gardening; and the Praia Grande group (N = 26; 14 women), where subsistence is based on mechanized agriculture. In Paricatuba, mean BP is 109/74 mmHg in men and 101/70 mmHg in women. There are no significant differences between BP of men and women, and systolic blood pressure (SBP) increases with age. Both SBP and diastolic blood pressure (DBP) are associated with weight, but only DBP is associated with the BMI, while SBP is associated with stature. In Praia Grande, mean BP is 120/76 mmHg in men and 118/70 mmHg in women, with no significant differences between the sexes. In Praia Grande, SBP is higher than in Paricatuba, and both SBP and DBP are associated with age. Compared with urban groups, both Caboclo samples have low BP. Still, differences in BP and body habitus between the two groups support a hypothesis that degree of westernization influences mean levels of BP in rural Amazonian populations. Further, the results also may be interpreted as suggesting that associations of sex, age, and BMI with BP, commonly reported in urban samples, are a byproduct of westernization rather than a result of genetic factors. Q 1995 Wiley-Liss. Inc zyxwv Cardiovascular diseases are presently the leading cause of adult death in cosmopolitan societies such as the United States and Great Britain (Lopes, 1993). Large differences between cardiovascular disease prevalence and mortality rates occur even in these developed countries (Higgins, 1991). In recent decades (1968-1981), a marked decline in coronary heart disease mortality has occurred in the United States and several other developed countries (Stamler, 1985) Causes for the observed differences and the declines in mortality rates from cardiovascular diseases continue to elude and in. triweresearchers rCalifano, lg7’; 1985). Differences in the distribution Of average blood pressure among national populations, ethnic groups, and medwomen and youngold members Of the Same population or ethnic group suggest that many factors, in addition to genes, such as diet, lifestyle, body habitus, and culture, influence blood pressure (Crews and Mancilha-Carvalho, 1993; James, 1991; Lewis, 1990; Silva and Eckhardt, 1994; Ward, 1983). Studies among traditional-living native populations in South America and elsewhere consistently report low average blood pressures compared with urban populations (Crews and Mancilha-Carvalho, 1993; De Lima, 1950; Friedlaendcr et al., 1987; Glanville and Geerdink, 1972; Harper et al., &ceivcd July 16,1994; accepted February 8,1995, Address reprint requests t o Hilton P. Silva, Department of Anthropology, The Ohio State University, 244 Lord Hall, 124 W. 17thAve., Colurnbus, OH 43210. This paper received the Edward E. Hunt, Jr. Award for the best student poster at the 1994 meetings of the Human Biology Council. zyxwvutsrqp 0 1995 Wiley-Liss, Inc. 536 zyxwvuts zyxwvuts H.P. SILVA ET AL. zyxwvutsrqpon zyxwvu Fig. 1. Detail of Marajo Island showing geographic location of Praia Grande and Paricatuba. LAdapted and redrawn from Murrieta RSS, Brondizio E, Siqueira A, (19891, Bol. Mus. Par. E. Goeldi, Serie Antropol., 5:147, with permission.] 1994; Nee1 et al., 1964; Oliver et al., 1975; Oliveira, 1952; Salzano and CallegariJacques, 1988). Conversely, native populations the world over who are experiencing in situ alterations in their immediate environment appear to show higher blood pressures with increased exposure to cosmopolitan lifestyles (He et al., 1991; Lewis, 1990; Lowenstein, 1961). Unfortunately, longitudinal data are not available to document a rise in blood pressure among the majority of native populations. In addition, little data are available on blood pressure of non-Indian populations residing outside Latin American urban centers. Many admixed populations, with combined Amerindian, African, and European ancestry, reside in rural and isolated areas of South America. In the Brazilian Amazon, the common name for such groups is Caboclo. The purpose of this paper is to report blood pressures and anthropometric dimensions from two Caboclo samples who reside in the same general environment but who practice somewhat different subsistence techniques. The specific aims are to determine whether more westernized subsistence patterns are associated with higher blood pressures or poorer body habitus, as assessed by anthropometry, in two samples of Caboclos. The second aim is to determine the degree to which blood pressures are age related in the two samples. MATERIALS AND METHODS Sample zy zyxwvutsr As part of a larger research project on Caboclo populations residing in the Amazon estuary (Neves, 1992), blood pressures and anthropometric dimensions were taken on 46 individuals from two Caboclo communities, Paricatuba and Praia Grande, of the Marajo Island, Para, Brazil (1Oo22'54S, 48°50'10W) (Murrieta, 1994; Neves, 1992) (Fig. 1). zyxw zy zyxwvutsr SUBSISTENCE AND BLOOD PRESSURE IN CABOCLO Caboclo is the term commonly used to describe populations of African, European (in this case mostly Portuguese), and Native American ancestry in Brazil. Other Caboclo populations of the Amazon estuary have been reported to show about 53% European, 22% Native American, and 25%)African ancestry in their present gene pool (Salzano, 1986). Caboclo populations of Marajo Island have inhabited the area for more than a century (Moran, 1974; Murrieta et a]., 1991). Since 1974, some Caboclo communities of Marajo Island have experienced in situ changes in lifestyles, from traditional subsistence agriculture to cash-based mechanized agriculture. The data for this study come from two conimunities representing these extremes of lifestyle. Paricatuba has a more traditional subsistence lifestyle, while Praia Grande has a lifestyle more dependent on cash cropping and the regional market economy. Since the culture, environment, and economy of this area are well described elsewhere (Murrieta et al., 1991; Murrieta, 1994; Neves, 1992; Siqueira et al., 1993), only a brief description is provided. In Paricatuba, subsistence is based on aqai fruit (Euterpe oleracea), slash and burn agriculture, fishing, and occasional small game hunting. Families in Paricatuba plant mostly manioc, sugarcane, and bananas. Choice of crops is often dictated by the nature of the land on which the gardens are situated; either terra firme, that is, areas not flooded by the river's daily tides, or ucirzea, that is, areas that are flooded daily. In Paricatuba the majority of daily activities are family based and subsistence directed. When extra fish are caught or an abundance of aqai is collected, these are often sold in Ponta de Pedras and Belem (Neves, 1992; Siqueira et al., 1993). In 1991, Paricatuba had a population of 144 individuals (Neves, 1992). As in most Caboclo populations, the majority of the population is children and adolescents (Siqueira et al., 1993). Of 144 people residing in the area, only about 30 were over the age of 17 years; of these, 12 women and 8 men present in the area at the time of the study participated in the measurement protocols. In Praia Grande, subsistence is based on mechanized agriculture of coconuts, beans, rice, and corn for both consumption and for sale in Ponta de Pedras and Belem (Murrieta et al., 1991). Various individuals may also do some fishing and occasional collect- 537 ing of aqai; however, these activities and some animal husbandry (chickens, turkeys, pigs) play only a minor role in the local economy. For all intents and purposes, inhabitants of Praia Grande are completely integrated into the market economy and media culture of the area, including access to electricity, television, and radio, and reliance on processed foods and other commercially produced consumer products (Murrieta et al., 1991; Murrieta, 1994; Neves, 1992). In 1991, the Ponta de Pedras community was composed of 22 families totaling 117 individuals, of whom about 35 were 17 years of age and older (Neves, unpublished data); of these, 14 women and 12 men present at the village during the study participated in the research. In all cases, age was determined from birth certificates or government issued identification cards. zyx Methods Blood pressures and anthropometry. Due to the constraints of the research protocol, blood pressures were measured only once for each individual. This was done on the right arm with a conventional TycosB aneroid sphygmomanometer, according to procedures described by Glanville and Geerdink (1972). In general, each participant rested in a seated position for 5 minutes before measurements were taken. Anthropometric dimensions were taken following the techniques described by Weiner and Lourie (1969). Weight was measured with a portable scale and recorded to the nearest 500 g. Stature was measured with a Swiss GPM@ anthropometer and recorded to the nearest millimeter. All measurements of blood pressure were made by the same physician (HPS), and all anthropometric dimensions were also taken by a single individual (WAN).Thus, interobserver error is not a factor. Statistical analysis. In the first phase of the analysis, average blood pressure, age, weight, stature, and the body mass index LBMI = wt(kg)/ht(m)'J were compared within each sample by sex and then between samples with the use of a linear approximation to the t-test. Reported are the means, their standard deviations, 95% confidence intervals for the differences between means, and associated P values. Next, univariate and bivariate associations, adjusted for sex (0 = women, 1 = men), of all available measures (sex, age, weight, stature, BMI) with zyxwvuts zyxw 538 zyxwvutsrq zyxwvuts zyxwvutsrq zyxwvu zyxwv H.P. SlLVA ET AL. zyxwv zyxw zyxwvutsr TABLE 1 . Means, standard deviations, P values, and 95% confidence intervals for the differences between sexes within each moup for all studv variables Age (years) SBP (mmHg) DBP (rnmHgJ Weight (kg) Stature (cm) BMI (wtihtz) Men (N = 8) Mean (SD) Paricatuba Women (N = 12) Mean (SD) P value 40.1 (9.2) 109.4 115.2) 73.9 (9.91 54.8 (9.4) 161.4 (8.0) 21.0 (2.7) 34.8 (10.8) 101.5 113.5) 70.2 (8.9) 44.5 (6.1) 150.3 (5.0) 19.6 (2.5) 0.273 0.241 0.408 0.008 0.002 0.300 Men (N = 12) - 4 5, -58, -54, 3 0, 4 8, -12, 15 1 215 126 17 7 17 4 38 36 9 (15 9) 1197(162) 755(82) 60 0 (1351 161 0 (7 0) 230(48) Praia Grande Women (N = 14) Mean (SD) P value 39.1 (16.2) 118.2 (16.8) 69.8(9.1) 48.4 (6.1) 150.0 (4.0) 21.4(2.4) 0.729 0.885 0.135 0.008 0.001 0.264 95% CI -15.3, -12.5, -1.8, 3.4, 6.6, -1.4, 10.9 12.4 12.4 20.0 15.7 4.8 SBP, systolic blood pressure; DRP, diastnlir blood pressure; BMI, body mass index. blood pressures were examined within each sample with use of linear regression. Total variance explained (R2) in either systolic or diastolic blood pressure, adjusted for the number of variables in the model, and associated P values are reported. In the final phases of this analysis, both samples were combined and linear regression, with dichotomous variables ( 0 , l )for location or for sex was used to examine associations of either location or sex with blood pressures, and finally, to examine the combined effects of location and sex, and then location, sex, weight, stature, age, and BMI, on blood pressures. The purpose was to determine if village of residence remained independently associated with blood pressures after statistical control for the possible confounding variables. Linear regression was then used to evaluate possible influences of locality and age on blood pressures within each sex. R2, adjusted for the number of variables in the model, and associated P values are reported. Finally, stepwise regression was used to determine which independent variables statistically explained the most variance in blood pressures within the combined sample and then for each sex in the combined sample. 11.0 cm). As in Paricatuba, in Praia Grande neither SBP nor DBP is significantly different between men and women. Among men, the Praia Grande sample shows higher SBP and DBP, weight, and BMI than the Paricatuba sample. Similarly, among women, the Praia Grande sample shows higher SBP, weight, and BMI, but not DBP, than the Paricatuba sample (Table 1). Risk factors and blood pressure There were clear univariate associations between age and stature, and a borderline association between weight and SBP, but no associations with either sex or BMI were observed in Paricaturba (Table 2). With use of a bivariate model controlling for sex, stature remains associated with SBP, while weight no longer shows even a borderline association. Weight and the BMI are the only two measures strongly associated with DBP in univariate analyses of the Paricatuba sample (Table 2). Following adjustment for sex with the bivariate model, the magnitude of the associations of weight and the BMI with DBP decrease, although they remain significant (P < 0.10). I n the Praia Grande sample, univariate models with either sex, weight, stature, or the BMI do not explain large amounts of variation in either SBP or DBP (Table 2). However, age alone explains 34%of the variation in SBP, and 28% of the variation in DBP. Statistical control for sex does not greatly alter the observed univariate associations with SBP. However, a strong secondary association of sex with DBP is observed when the effects of age and sex are estimated simultaneously in a bivariate model. Furthermore, a previously unobserved association of stature with DBP is observed in zyxwvut RESULTS Sex differences in body habitus and blood pressure As shown in Table 1, there are large differences between mean weight (wt) and stature (st)of men and women in Paricatuba (wt: 10.3 kg, st: 11.1 cm). However, neither systolic blood pressure (SBP) nor diastolic blood pressure (DBP) differ significantly between men and women in Paricatuba. I n Praia Grande, weight and stature also differ between men and women (wt: 11.6 kg, st: zyxw zyxwvutsrqp zyxwvu zyxwvu zyxw zyxwvu SUBSISTENCE AND BLOOD PRESSURE IN CABOCLO 539 TABLE 2. Regression analysis reporting variance explained (R2)and P values within each group for all i d p e n d e n t variables and second1.y for all independent variables rontrolling for sex Paricatuba Praia Grande DBP SBP Independent Sex Age Weight Stature BMI Age Sex Weight Sex Stature Sex RMI Sex R2 0.07 0.19 0.14 0.25 0.02 0.13 0.05 0.09 P value R2 0.241 0.054 0.096 0.024 0.540 0.095 0.444 0.764 0.237 0.058 0.732 0.732 0.303 SBP 1P P value 0.03 0.07 0.20 0.07 0.19 0.01 0.408 0.228 0.042 0.227 0.050 0.313 0.587 0.065 0.709 0.392 0.954 0.076 0.682 0.12 -0.02 0.00 0.34 0.00 0.01 0.01 0.29 -0.08 -0.04 P value R2 0.884 0.001 0.742 0.535 0.491 0.002 0.674 0.771 0.979 0.322 0.427 0.510 0.991 0.09 0.28 0.02 0.00 0.04 0.35 0.01 0.15 DBP P value 0.135 0.004 0.459 0.829 0.296 0.002 0.045 0.990 0.204 0.068 0.020 0.458 0.199 zyxwvutsrqponm -0.02 0.11 0.06 0.03 zyxwvutsrq BMI = body mass index; SBP = systolic blood pressure, DBP P v a l u r = 0.000 = P -;O.006. R2 = 0.00 = It2 ~:0.006. = diastolic blood pressure. TABLE 3. Regression ann1,yses of systolic and diastolic blood pressure for both Pnricatuha and Praia Grande samples combined (Ar = 46) Model 1 2 3 4 51 61.2 Independent variables Location Sex Age Location Sex Location Sex Age Weight Stature BMI Total Age Location Stature ~~ Systolic blood pressure P value ~ RZ 0.17 0.02 0.25 0.15 0.40 0.25 0.38 0.42 ~~ 0.004 0.348 0.000 0.005 0.403 0.007 0.489 0.000 0.445 0.266 0.41 3 ,004 .002 .056 Independent variables Location Sex Age Location Sex Location Sex Age Weight Stature Place of residence and blood pressure After combining the two samples, place of residence (location) appears to be significantly associated with SBP, but not with DBP (model 1; Table 3). Conversely, sex shows a borderline significant association (P< 0.10) with DBP, but not with SBP (model 2). Age, however, shows clear significant associations with both blood pressures in the combined sample (model 3). In the R' P value 0.00 0.07 0.19 0.824 0.086 0.002 0.903 0.091 0.440 0.294 0.002 0.041 0.040 0.029 0.02 BMI Total Age Location BM1 - body mass index. 'R2(explained variance) values are total in models 1-4 and cumulative in models 5 and 6. 'Model 6 is stepwise regression with all variables the bivariate model with the inclusion of sex (Table 2). Diastolic blood pressure 0.27 0.19 0.24 ,002 .039 zy bivariate model, including location and sex as independent variables (model 4), the pattern observed in univariate analyses is not greatly altered. In the multivatiate model (model 5, Table J), including all possible independent variables, only location and age are strongly predictive of SBP. Conversely, age, weight, stature, and the BMI are strongly associated with DBP, but location and sex are not (model 5). The final analyses in Table 3 (model 6 ) are based on stepwise regression with all independent variables. These analy- 540 zyxwvutsr zyxwvut zyxwvuts zyxwvutsrqpon zyxwvut zyxwvutsr H.P. SlLVA ET AL. zyxw TABLE 4. Regression analyses of systolic and diastolic blood pressures for both Paricatubn and Praia Grand@sampLes curnbined b y sex Model Men (N : 20) 1 2 3’ Women (N = 26) 1 2 3’ Independent variable SBP R2 DBP Independent P value variable RZ P TdlUt? Location 0.09 0.192 Location 0.00 0.756 Location 0.158 Location 0.680 Age 0.05 0.306 Age -0.06 0.379 No variable entered stepwise model with P s.10 for men for systolic or diastolic hlood pressure. Location Location Age Age Location 0.24 0.62 0.50 0.62 0.011 0.005 0.000 0.000 0.005 Location Location Age Age Weight. 0.00 0.34 0.38 0.913 0.459 0.001 0.001 0.075 zyxwvutsrq 0.42 SBP = systolic blood pressure; DBP - diastolic blood pressure. R2 - Variance accounted for by the variable and all preceding variables in the model P value 0.000 = P .:0 0006. ‘Model 3 is stepwise regression with all variables. ses show that age, location, and stature together account for 42% of the variance in SBP after adjustment for all variables in the model. Only age and location entered the model for DBP, jointly explaining 24% of the variance. As a final step, an analysis by sex was done with the samples from both locations combined to examine possible effects of location on blood pressures (Table 4).Among men, no association of location with either SBP or DBP was observed in univariate analysis. However, among women, location was associated with SBP (model 1). In bivariate models, location and age are both strongly associated with SBP in women, but not in men (model 2). Further, among women, age, but not location, is strongly associated with DBP in the bivariate analysis, but this effect is not observed among men (model 2). With stepwise regression, results for SBP among women are identical to those of model 2, with age and location explaining 62% of the variation. For DBP, age and weight jointly explain 42% of the variation observed among the women. No variables entered the stepwise model for either SBP or DBP among men. DISCUSSION There are numerous reports on blood pressure variation among native South American populations (Crews and Mancilha-Carvalho, 1993; Fleming-Moran et a]., 1991; Lowenstein, 1961; Salzano and Callegari-Jacques, 1988); however, few reports describe blood pressure variation among ad- mixed populations (James et al., 1991; Silva and Eckhardt, 1994). Data from native South American samples show low blood pressures compared with urban, cosmopolitan, and westernized populations (FlemingMoran et al., 1991; Lowenstein, 19611,a pattern consistent with reports from most traditional, nonwesternized, populations worldwide (Chin Hong and McGarvey, 1993; James et al., 1991; Schall, 1993). The two groups reported here are admixed and are considered economically and culturally transitional according to the criteria used by Pollard et al. (1991), since they participate to some extent in the regional market economy. However, they show variable participation in this economy. Such differences between Praia Grande and Paricatuba may still be of sufficient magnitude to initiate differences in average SBP and DBP. Since these two communities generally share the same genetic heritage and inhabit similar geographic environments, one possible explanation for any observed differences in average blood pressures is differential participation in market and subsistence activities and concomitant lifestyle differences. Admittedly, the samples from each group are small, but they represent 66% and 72% of the total resident adult population of Paricatuba and Praia Grande, respectively. Thus, although the small samples may not be representative of all Caboclo populations living in more subsistence-based and more marked-based communities, they are a fairly complete sample of adults from these two communities, and are at least representative of this lifestyle contrast. zyxwvut zy SUBSISTENCE AND BLOOD PRESSURE IN CABOCLO In Brazil, as in many other developed or developing countries, cardiovascular diseases are the main cause of death among non-Indian adults (James et al., 1991).However, in the present sample, no individual showed hypertension (BP > 140/90 mmHg). As for age-, sex-, and anthropometry-related differences in blood pressures, most studies in western or westernized populations report an increase in blood pressures with age in both sexes (James et al., 1991; Pollard et al., 1991; Stamler, 1991), with men having, on average, higher blood pressures than women (Crews, 19931, and high correlations between blood pressures and body measurements (Gerber et al., 1993; Mukherjee et al., 1988; Silva et al., 1994). In these two populations, both SBP and DBP were associated, to some extent, with anthropometric dimensions, particularly in Paricatuba. Blood pressures of men and women did not differ in univariate analyses, but when both groups were combined, both sex and location were significant independent variables. Only in women did blood pressures show a positive association with age, particularly in Praia Grande. These findings may be considered indicative of a moderate degree of “westernization.” In addition, the almost complete lack of association of the BMI, weight, or stature with blood pressures in Praia Grande is an unusual finding that needs further study. Patterns of blood pressures in the two groups show that these non-Indian populations respond in the same fashion as Indian populations t o changes in lifestyle. However, in contrast t o most groups previously studied, there is no significant difference between blood pressures of men and women in these two communities. This finding must be considered tentative, since the sample sizes are small. This problem is difficult to overcome since over 65%of the adults in the two populations were study participants. The precise manner in which body habitus may influence blood pressures in these two communities is not clear. This, in part, reflects the fact that only weight and stature are available to assess this relationship. Additional anthropometry might clarify associations of the BMI with blood pressures in these populations. Still, the high association of the BMI with DBP when both groups were combined suggests that there is little difference between western and transitional societies in this relationship. The absence of an 541 zyxwv association with SBP may, however, indicate some degree of difference from more westernized groups. Following control for place of residence and sex, age shows a strong association with both SBP and DBP in both of these transitional societies. Similarly, when the sample was examined by sex with both groups combined, both location and age were associated with SBP in women, whereas neither location nor age were associated with blood pressures in men. Interestingly, in the final stepwise analyses, age shows a strong independent association with both SBP and DBP in both samples combined. Conversely, when the combined sample is dichotomized by sex, age is strongly associated with both SBP and DBP in women, but not in men. This is not true for native American populations of the Amazon (Crews and MancilhaCarvalho, 1993) or transitional living populations elsewhere, such as the Gainj (Harper et al., 1994), among whom no age-related differences in blood pressure have been observed. Investigations of biomedical and physiological characteristics among non-Indian rural populations of the Amazon remain infrequent. This is surprising since these groups comprise a majority of the population inhabiting the area. Together with study of native South Amerindians, study of Caboclo populations of the Amazon should provide important insights into the mechanisms of human biological adaptation t o westernization. ACKNOWLEDGMENTS We thank the populations of Paricatuba and Praia Grande for their kind participation in this study. This research was partly supported by Museu Paraense Emilio Goeldi, and CNPq, Brazil. Hilton Pereira da Silva received a Ph.D. fellowship from CNPqBrazil during the preparation of this work. Presentation of an earlier version of this paper, as a poster, a t the 1994 Scientific Meeting of the Human Biology Council, in Denver, CO, was partly supported by a travel grant from the Hughes Memorial Fund of the Department of Anthropology, The Ohio State University. Thanks are also due to Silvia Melcher for her kind assistance in the organization of the anthropometric data and to Susan Bean for her secretarial services. zyxwvu zyxwvu zyxwvu H.P. SILVA ET AL 542 zyxwvutsrqp zyxwvu LITERATURE CITED Califano JA (1979)Healthy People: The Surgeon General’s Report on Health Promotion and Disease Prevention. Washington DC: U S . Department of Health, Education, and Welfare. Chin Hong W ,McGarvey ST (1993) Lifestyle incongruity and adult blood pressure in Western Samoa. Am. J. Phys. Anthropol. 16:70 (abstract). Crews DE (1993) What do differential associations of risk factors for blood pressure across gender tell us? Am. J . Phys. Anthropol. 16:75 (abstract). Crews DE, Mancilha-Carvalho JJ 11993) Correlates of blood pressure in Yanomami Indians of northwestern Brazil. Enthnic. Dis. .7:362-371. De Lima PE (1950) Niveis tensionais dos indlos Kalapalo e Kamaiura. Rev. Bras. Med. 7:787-788. Fleming-Moran M, Santos RV, Coimbra CEA (1991) Blood pressure levels of the Surui and Zoro Indians of the Brazilian Amazon: Group- and sex-specific effects resulting from body composition, health status, and age. Hum. Biol. 63:83&861. Friedlaender JS, Howells W, Rhoads JG (eds.1 (1987) The Solomon Islands Project: A Long-Term Study of Health, Human Biology, and Culture Change. Oxford: Claredon Press. Gerber LM, Scwartz JE, Schnall PL, Pickering TG (1993) Body fat and its distribution in relation to sex differences in blood pressure. Am. J. Phys. Anthropol. 1 6 3 3 (abstract). Glanville EV, Geerdink RA (1972) Blood pressure of Amerindians from Surinam. Am. J. Phys. Anthropol. 37: 251-254. Harper GJ, Crews DE, Wood JW (1994) Lack of agerelated blood pressure increase in the Gainj, Papua New Guinea: Another low blood pressure population. Am. J. Hum. Biol. 6:121-122 (abstract). He J, Tell GS, Tang YC, Mo PS, He GQ (1991) Effect of migration on blood pressure: The Yi people study. Epidemiology 2:88-97. Higgins M (1991) International studies of cardiovascular disease. Ann. Epidemiol. lt567-569. James GD (1991) Blood pressure response to the daily stressors of urban environments: Methodology, basic concepts, and significance. Yrbk. Phys. Anthropol. 34: 89-110. James SA, Almeida-Pilho N, Kaufman JS (1991) Hypertension in Brazil: A review of the epidemiological evidence. Ethnic. Dis. lr91-98. Lewis DE J r (1990) Stress, migration, and blood pressure in Kiribaty. Am. J. Hum. Biol. 2:139-151. Lopes AD (1993)Assessing the burden of mortality from cardiovascular diseases. World Health Stat. Qu. 46: 91-96. Lowenstein FW (1961) Blood pressure in relation to age and sex in the tropics and subtropics. Lancet 1:389392. Moran EF (1974)The adaptive system ofthe Amazonian Caboclo. In C Wagley (ed.): Man in the Amazon. Gainesville, FL: University Press of Florida, pp. 137159. Mukherjee BN, Byard PJ, Bhattacharya SK, Rao DC (1988) Blood pressure in a rural West Bengal fishing community: An epidemiologic profile. Hum. Biol. 6‘0: 69-79. Murrieta RSS (1994) Diet and Subsistence: Changes in Three Caboclo Populations on Marajo Island, Amazonia, Brazil. Master’s thesis, Department of Anthropology, University of Colorado, Boulder. Murrieta RSS, Siqueira A, Brondizio E, Moran EF (1991) Estrategias de subsistencia da comunidade de Praia Grande, ilha de Marajo, Para, Brasil. Bol. Mus. Par. E. Goeldi, Serie Antropologia. NeelJV, Salzano FM, Junqueira PC, Keiter F, MayburyLewis U (1964) Studies of the Xavante Indians of the Brazilian Mato Grosso. Am. J. Hum. Gen. 16:52-140. Neves WA (1992) Antropologia Ecologica de Populagdes Ribeirinhas do Estuario do Amazonas: Subsistencia e Adaptacao. Brasilia: Relatorio Cientifco Apresentado ao CNPq. Oliveira HC (1952)0 estado de saude dos indios Karaja em 1950. Rev. Mus. Paulista. 6:489-508. Oliver WJ, Cohen EL, Nee1 JV (1975) Blood pressure, sodium intake, and sodium related hormones in the Yanomamo Indians, a “no-salt” culture. Circulation. 52t146-151. Pollard TM, Brush G, Harrison GA (1991) Geographic distributions of within-population variability in blood pressure. Hum. Biol. 6‘5:643-661. Salzano FM (1986) Em busca das raizes. CiBncia Hoje 5:4%53. Salzano FM, Callegari-Jacques SM (1 988) South American Indians: A Case Study in Evolution. New York: Oxford University Press. Schall J I (1993) Sex differences in blood pressure response to modernization. Am. J. Phys. Anthropol. 16: 174 (abstract). Silva HP, Eckhardt RB (1994)Westernization and blood pressure variation in four Amazonian populations. Col. Anthrop. 18:81-86. Silva HP, Crews DE, Neves WA (1994) Blood pressure, nnthropometry and the impact of Westernization on traditional populations from the Brazilian Amazon. Annals ofthe I11 Congress0 Latino American0 de Anthropologia Biologica & I1 Reuniao da Sociedade Brasileira de Paleopatologia, pp. 47 (abstract). Siqueira AD, Brondizio ES, Murrieta RSS, Silva IIP, Neves WA, Viertler RB (1993) Estrategias de suhsistencia da populacao ribeirinha do igarape do Paricatuba, ilha de Marajo, Brasil. Bol. Mus. Par. E. Goeldi, Serie Antropologia. Stamler J (1985) The marked decline in coronary heart disease mortality rates in the United States, 19681981: Summary of findings and possible explanations. Cardiology 72: 11-22. Stamler J 11991) The Yi migrant study: Population exposures influencing blood pressure. Epidemiology 2r83-87. Ward R (1983) Genetic and sociocultural components of high blood pressure. Am. J. Phys. Anthropol. 62:91105. Weiner JS, Lourie JA (1969) Human Biology,A Guide to Field Methods. Oxford: Blackwell Scientific. zyxwvutsrq zyxwvut zyxwvut