Transactions of the Royal Society of Tropical Medicine and Hygiene (2008) 102, 645—652

avai lab le at www.sc iencedi rec t .com

journa l homepage: www.e lsev ierhea l th .com/ journa ls / t rs t

Anaemia and malaria in Yanomami communities withdiffering access to healthcare

P. Grenfell a, C.I. Fanelloa,∗, M. Magrisb, J. Goncalvesb, W.G. Metzgerb,c,S. Vivas-Martınezb,d, C. Curtis a, L. Vivasa

a London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UKb CAICET (Centro Amazonico para la Investigacion y Control de Enfermedades Tropicales ‘‘Simon Bolıvar’’), Puerto Ayacucho,Amazonas, Venezuelac Eberhard Karls Universitat Tubingen, Institut fur Tropenmedizin, Humanparasitologie, Wilhelmstrasse 27, 72074 Tubingen,Germanyd Universidad Central de Venezuela, Escuela de Medicina Luis Razetti, Departamento de Medicina Preventiva y Social,Caracas, Venezuela

Received 5 September 2007; received in revised form 26 February 2008; accepted 26 February 2008Available online 10 April 2008

KEYWORDSAnaemia;Malaria;Access to healthcare;Yanomami;Indigenouspopulations;Venezuela

Summary Inequitable access to healthcare has a profound impact on the health ofmarginalised groups that typically suffer an excess burden of infectious disease morbidity andmortality. The Yanomami are traditionally semi-nomadic people living in widely dispersed com-munities in Amazonian Venezuela and Brazil. Only communities living in the vicinity of a healthpost have relatively constant access to healthcare. To monitor the improvement in the devel-opment of Yanomami healthcare a cross-sectional survey of 183 individuals was conducted toinvestigate malaria and anaemia prevalence in communities with constant and intermittentaccess to healthcare. Demographic and clinical data were collected. Malaria was diagnosedby microscopy and haemoglobin concentration by HemoCue. Prevalence of malaria, anaemia,splenomegaly, fever and diarrhoea were all significantly higher in communities with inter-mittent access to healthcare (anaemia 80.8% vs. 53.6%, P < 0.001; malaria 18.2% vs. 6.0%,P = 0.013; splenomegaly 85.4% vs.12.5%, P < 0.001; fever 50.5% vs. 28.6%, P = 0.003; diarrhoea30.3% vs.10.7% P = 0.001). Haemoglobin level (10.0 g/dl vs. 11.5 g/dl) was significantly asso-

ciated with access to healthcare when controlling for age, sex, malaria and splenomegaly(P = 0.01). These findings indicate a heavy burden of anaemia in both areas and the need forinterventions against anaemia and malaria, along with more frequent medical visits to remoteareas.© 2008 Royal Society of Tropicareserved.

∗ Corresponding author. Tel.: +44 (0)207 927 2339; fax: +44 (0)207 580E-mail address: caterina.fanello@lshtm.ac.uk (C.I. Fanello)

0035-9203/$ — see front matter © 2008 Royal Society of Tropical Medicindoi:10.1016/j.trstmh.2008.02.021

l Medicine and Hygiene. Published by Elsevier Ltd. All rights

9075.

e and Hygiene. Published by Elsevier Ltd. All rights reserved.

6

1

IowtbaAate2haaibiasa1wtmshwtrhadad

polo2(a2fcoaSras2ftip1m12

odrdlhoioeu

2

2

Toeocisanh(atwcjmf

2

IsOucvsyuolaip2003; Magris et al., 2007) to allow detection of a 1 g/dldifference in haemoglobin concentration between the two

46

. Introduction

nequitable access to healthcare has a profound impactn the health of marginalised groups worldwide. Coupledith low income and poor sanitation, these populations

ypically suffer an excess burden of infectious disease mor-idity and mortality. Undernutrition is particularly commonmong indigenous groups; recent research in four Latinmerican countries has linked malnutrition and stuntingmong indigenous children to a lack of maternal educa-ion, road accessibility and access to healthcare (Buitront al., 2004; Larrea and Freire, 2002; Larrea and Kawachi,005). However, general information on the current state ofealth of indigenous populations in Latin America is sparsend based on disaggregated national data (Montenegrond Stephens, 2006). In Venezuela, although healthcares free and access has recently improved (PAHO, 2006),arriers remain, particularly among indigenous populationsn remote areas. One of the largest and least culturallyssimilated ethnic groups is the Yanomami, a traditionallyemi-nomadic people inhabiting an extensive geographicrea spanning the Venezuelan—Brazilian border (Lizot,988). In Venezuela the majority of Yanomami live inidely dispersed communities where healthcare is delivered

hrough basic primary health centres, each attended by aedical doctor, a community health worker and a micro-

copist. As such, communities living in the vicinity of aealth post have relatively constant access to healthcarehilst more remote populations depend upon the intermit-

ent visits of a mobile medical team. Communities out ofeach of these medical teams have virtually no access toealthcare. Although Yanomami beliefs relating to healthnd well-being differ greatly from western views and tra-itional medicine remains strongly relied upon, medicalttention may be sought for the treatment of infectiousiseases.

Prior research has documented an extremely highrevalence of anaemia among Yanomami communitiesf 91% (Perez Mato, 1998). Anaemia can have seriousong-term health effects including reduced cognitive devel-pment (Grantham-McGregor and Ani, 2001; Lozoff et al.,000), growth (Lawless et al., 1994), immune functionOppenheimer, 2001) and physical endurance, whilst severenaemia is associated with increased mortality (WHO,001). The aetiology of anaemia is multifactorial and riskactors often occur concurrently, such as nutritional defi-iencies (particularly iron), malaria and infection withther pathogens, e.g. hookworm, bacteraemia and HIV,s well as predisposing genetic disorders (Crawley, 2004;chellenberg et al., 2003; Shamah-Levy et al., 2003). Recur-ent episodes of malaria contribute to the development ofnaemia and severe anaemia through haemolysis of para-itized red blood cells, dyserythropoeisis (Giribaldi et al.,004; Layez et al., 2005) and accelerated clearance of unin-ected erythrocytes (Evans et al., 2006). Malaria is endemico the Amazon basin and a considerable cause of morbid-ty in Yanomami communities, recent reports indicating a

revalence of 14—24% (Marcano et al., 2004; Perez Mato,998). Whilst Plasmodium vivax is the predominant cause ofalaria in this region, P. falciparum is responsible for about

6% of cases (Ministerio del Poder Popular para la Salud,007).

anpid

P. Grenfell et al.

The aim of this study was to determine the prevalencef anaemia and malaria among Yanomami communities withiffering access to healthcare. As part of national healtheforms, the Venezuelan Ministry of Health is currentlyeveloping a health plan specific to the Yanomami popu-ation. By addressing the relationship between access toealthcare and community anaemia prevalence, the resultsf the study may provide a suitable indicator for monitoringmprovement during the development and implementationf the Yanomami health plan, particularly in relation to thequitable provision of health services to the Yanomami pop-lation.

. Methods

.1. Study area and population

his study was conducted in the Alto Orinoco regionf south Venezuela, an area of tropical perennial for-st intersected by a dense river system. In the absencef roads all travel is by light aircraft, motor boat oranoe. The predominantly Yanomami population inhab-ts geographically dispersed communities (INE, 2001). Thetudy population was drawn from 11 communities in tworeas with differing access to healthcare. Five commu-ities were categorised as having constant access toealthcare (Ocamo area), being located within one km20 min by boat) of a primary health post attended by

resident physician, a community health worker andwo trained microscopists. The remaining six communitiesere assessed as having intermittent access to health-are, being located within a one-to-four hour motorboatourney of the health post (Alto Ocamo area). These com-unities are visited by a medical team approximately

ortnightly.

.2. Study design and methodology

n July 2005, 183 Yanomami individuals were randomlyelected to participate in a cross-sectional survey. In thecamo area, a recent census was provided, which waspdated by the community health worker. In Alto Ocamoommunities, a census was taken before conducting the sur-ey due to a lack of available demographic information. Thetudy population was divided into three age groups (0—4ears, 5—14 years and ≥15 years), a stratification commonlysed in malaria and anaemia studies since these groupsften differ in terms of malaria immunity and haemoglobinevel. Age is typically not documented among Yanomamind was therefore estimated by locally experienced med-cal staff. Sample size calculations were based upon datareviously collected in the region (Kirkwood and Sterne,

reas, within each age group (90% power at 0.05 level of sig-ificance). Although demographic and logistical constraintsrevented strict adherence to these estimates, sample sizesn each age group were considered sufficient to detect thisifference.

mwp

mhuwcsarahma

3

3

Trvstipw

hiw(vO

Anaemia and malaria in Yanomami communities

2.3. Data collection

After receiving informed community and individual consent(written or oral), participants were interviewed regard-ing recent history of fever, chills, diarrhoea and vomiting.Spleen palpability was assessed with the physician seated orcrouched in front of the standing participant (Wernsdorferand McGregor, 1988). The presence of splenomegaly was notassessed in children under two years of age (spleen pal-pability does not necessarily indicate splenomegaly in thisage group) or pregnant women (since the fetus obstructsexamination). Malaria was diagnosed by thin and thick filmmicroscopy using Giemsa-stained finger-prick blood samplestaken by the physician (this also applied to non-participantspresenting with fever). Haemoglobin concentration wasdetermined by the HemoCue method (HemoCue AB, Angel-holm, Sweden), in which a drop of capillary blood is drawninto a glass microcuvette containing dried reagent andthe haemoglobin concentration of the blood is read by aportable photometer. The photometer was calibrated atthe beginning and at the end of each day. Anaemia wasdefined according to the WHO age-specific cut-off pointsfor haemoglobin concentration (WHO, 2001) (infants <6months: 9.5 g/dl; 6 months—4 years: 11 g/dl; 5—11 years:11.5 g/dl; 12—14 years: 12 g/dl; non-pregnant women ≥15years: 12 g/dl; men ≥15 years: 13 g/dl; pregnant women:11 g/dl). Malaria infections were treated according to cur-rent regional guidelines (primaquine and chloroquine forP. vivax; mefloquine and artesunate for P. falciparum) andanaemic individuals were prescribed iron supplements orfolic acid.

This study was conducted under the auspices of the Cen-tro Amazonico de Investigacion y Control de EnfermedadesTropicales ‘‘Simon Bolivar’’ (CAICET), Puerto Ayacucho,Venezuela.

2.4. Statistical analysis

Data were double-entered into EpiData 3.0 (Odense, Den-mark) and analysed using Stata 8.2 (Stata Corp., CollegeStation, TX, USA). Comparison of anaemia and malariaprevalence according to gender or access to healthcare was

ecnyn

Table 1 Demographic and clinical characteristics of the Yanomam

Ocamoa

Demographic characteristicsTotal study population 84Females:males 43:41Mean (SD) age (years) 15.7 (15.6)Median age 10

Clinical characteristics (prevalence (sample size))History of fever 28.6% (84)History of fever and chills 21.4% (84)Diarrhoea 10.7% (84)Vomiting 4.8% (84)Splenomegaly 12.5% (72)Malaria infection (all species) 6.0% (84)

a Access to healthcare is constant in Ocamo and intermittent in Alto O

647

ade using �2 and Fisher’s exact tests. �2 tests for trendere applied to detect trends of increasing or decreasingrevalence across the three age groups.

As haemoglobin concentration was normally distributed,eans and SDs are reported. The difference in mean

aemoglobin level between two groups was investigatedsing unpaired t-tests. Univariate simple linear regressionas performed to assess associations between haemoglobinoncentration and access to healthcare (classified as con-tant or intermittent) and potential confounders. Variablesssociated with haemoglobin level (at P < 0.1) were incorpo-ated into a stepwise multiple linear regression model, withccess to healthcare as the main explanatory variable andaemoglobin as the response variable. A generalised esti-ating equation was used to control for clustering within

nd between communities.

. Results

.1. Clinical characteristics

wenty-nine percent of the study population (53/183)eported fever and shivering in the week prior to inter-iew, with no differences according to age or gender. Theseymptoms were more commonly reported in Alto Ocamohan Ocamo (35.4% vs. 21.4%; P = 0.038). Fever and shiver-ng during the past week were positively associated with theresence of P. falciparum (Fisher’s exact P = 0.003), but notith that of P. vivax (Fisher’s exact P = 0.405).

Thirty-nine of 183 individuals (21.3%) reported havingad diarrhoea in the previous week. Prevalence was highern children aged under five years (37.8%; 17/45) comparedith those aged 5—14 years (15.7%; 11/70) and with adults

≥15 years, 16.2%; 11/68) (P = 0.008). The proportion of indi-iduals reporting diarrhoea was significantly higher in Altocamo compared to Ocamo (Table 1).

The prevalence of splenomegaly was 51.3% (79/154),

xcluding under two-year-olds and pregnant women. Thisondition appeared to be more common in individuals overine years (56.3%; 54/96) than those between two and nineears old (43.1%; 25/58), though the difference was not sig-ificant (P = 0.114). Splenic enlargement was seven times

i study population

Alto Ocamoa P-value

9940:59 0.14417.5 (16.7)

0.68412

50.5% (99) 0.00335.4% (99) 0.03830.3% (99) 0.00112.2% (98) 0.07685.4% (82) <0.00118.2% (99) 0.013

camo.

648 P. Grenfell et al.

Table 2 Prevalence of anaemia by age group and access to healthcare in the Yanomami study population

Age group (years) Prevalence (sample size) P-value

Ocamoa Alto Ocamoa

<5Male 33.3% (9) 88.9% (18) 0.006Female 11.1% (9) 88.9% (9) 0.003Total 22.2% (18) 88.9% (27) <0.001

5—14Male 55.0% (20) 88.2% (17) 0.036Female 77.3% (22) 63.6% (11) 0.438Total 66.7% (42) 78.6% (28) 0.280

≥15Male 8.3% (12) 58.3% (24) 0.005Female 100% (12) 100% (20) —Total 54.2% (24) 77.3% (44) 0.049

lto O

mtptd

3

Ttccwmwwmse

3

Asvt

wpcfiyiibntPOmP

3c

Oh(

Total 53.6% (84)a Access to healthcare is constant in Ocamo and intermittent in A

ore common in Alto Ocamo than in Ocamo (Table 1). Six-een out of 182 patients reported a history of vomiting in therevious week (8.8%). Vomiting was more common amonghe Alto Ocamo population than that of Ocamo (Table 1). Noifferences were observed in relation to age or gender.

.2. Anaemia and haemoglobin

he proportion of anaemic individuals in the study popula-ion was 70.5% (129/183). Anaemia was significantly moreommon in communities with intermittent access to health-are than those with constant access, a difference whichas greatest among under five-year-olds (Table 2). Theean haemoglobin concentration in the study populationas 10.73 g/dl (SD = 1.89 g/dl). Haemoglobin level increasedith age (� = 0.75, P < 0.001), but did not vary by gender. Theean concentration was greater in communities with con-

tant access to healthcare and this effect persisted withinach age group (Table 3).

.3. Malaria

total of 23 malaria infections were detected during thistudy (prevalence 12.6%), 14 (60.9%) of which were due to P.ivax, seven (30.4%) to P. falciparum and two mixed infec-ions (P. vivax and P. falciparum). All individuals infected

i(Imt

Table 3 Mean haemoglobin (g/dl) by age group and access to he

Age group (years) Ocamoa

Mean haemoglobin (SD) n

<5 11.0 (0.7) 185—14 11.3 (1.0) 42≥15 12.3 (1.9) 24

Total 11.5 (1.4) 84a Access to healthcare is constant in Ocamo and intermittent in Alto O

80.8% (99) <0.001

camo.

ith P. falciparum reported having been febrile in theast two days, whereas only eight of fourteen P. vivaxases did. The prevalence of malaria was greatest in underve-year-olds and declined with age: 20.0% in under five-ear-olds (9/45); 14.3% in 5—14 year olds (10/70) and 5.9%n adults (4/68) (test for trend odds ratio = 0.52 per unitncrease in age group; P = 0.027). Prevalence did not differetween males (14/100) and females (9/83) but was sig-ificantly higher in communities with intermittent accesso healthcare than those with constant access (Table 1).lasmodium falciparum cases were detected only in Altocamo (together with seven cases of P. vivax and the twoixed infections) whereas all five infections in Ocamo were

. vivax.

.4. Factors associated with haemoglobinoncentration

n univariate analysis, malaria infection and lowaemoglobin concentration were strongly associatedTable 4); mean haemoglobin level was 2 g/dl lower in

ndividuals infected with malaria than those not infectedmean haemoglobin 8.97 g/dl and 10.98 g/dl, respectively).ndividuals with splenomegaly also showed a reducedean haemoglobin concentration (10.35 g/dl) compared to

hose without (11.51 g/dl) (� = −1.42, P < 0.001; Table 4).

althcare in the Yanomami study population

Alto Ocamoa P-value

Mean haemoglobin (SD) n

8.9 (1.4) 27 <0.00110.1 (2.1) 28 0.00310.7 (2.0) 44 0.002

10.0 (2.0) 99 <0.001

camo.

Anaemia and malaria in Yanomami communities 649

Table 4 Results of simple and multiple linear regression analyses on factors associated with haemoglobin concentration inYanomami study communities

Variable Univariate analysesa Multivariate analysisa

Regression coefficient (�) P-value Regression coefficient (�) P-value

Age 0.824 <0.001 0.618 <0.001Gender −0.596 0.069 −0.843 0.012Fever −0.431 0.302 — —Diarrhoea 0.065 0.900 — —Splenomegaly −1.417 <0.001 −0.690 0.050

r(

tciaMtclhhePfCmrnhRdct

cttrbdawavtsmyo

Malaria infection −1.872Access to healthcare −1.444a All models were adjusted for clustering at the community level.

Fever, shivering and diarrhoea were not associated withhaemoglobin concentration.

Access to healthcare remained strongly associated withhaemoglobin concentration when controlling for age, gen-der, malaria, splenomegaly and clustering at the communitylevel (� = −0.857, P = 0.010) (Table 4).

4. Discussion

The high prevalence of anaemia among the communitiessurveyed meets the WHO definition of a severe publichealth problem (>40% community prevalence) (WHO, 2001).Haemoglobin concentration was considerably lower in com-munities dependent on a mobile medical team than in areaswith constant access to healthcare, a relationship which per-sisted after accounting for differences in demographics andmalaria prevalence. This haemoglobin deficit was most pro-nounced in under five-year-olds, 90% of whom were anaemicin Alto Ocamo (intermittent access).

The finding that all women above the age of 14 years,the majority of the child-bearing population, were anaemicis also of particular concern. Young children and pregnantwomen are typically at greater risk of developing anaemia,the long-term effects of which can lead to stunting andreduced cognitive and motor development in children, aswell as an increased risk of low body iron and anaemia ininfants born to anaemic mothers (Crawley, 2004).

The prevalence of malaria in this study population is com-parable to recently documented levels among Yanomamicommunities (Marcano et al., 2004; Metzger et al., 2007;Perez Mato, 1998). Children under five years of age wereat increased risk of infection, characteristic of malaria-endemic areas where immunity is acquired with age. Asmight be expected, malaria infection was more commonin individuals with intermittent access to healthcare thanthose with constant access. In this region malaria con-trol depends almost entirely upon early case detectionand treatment, rendering communities without good accessto healthcare more vulnerable to malaria mortality andmorbidity. Although hammock nets are used in most com-

munities, to a lesser extent in more remote areas, these aretypically torn and not treated with insecticide, offering littleprotection against infective mosquito bites. Similarly, indoorresidual insecticide spraying is largely ineffective due to thewood and palm leaf structure of the traditional Yanomami

poHl1

<0.001 −1.445 <0.001<0.001 −0.857 0.010

esidence and the exophilic nature of local malaria vectorsMagris et al., 2007).

In the absence of effective preventive tools in this region,imely, accurate diagnosis and effective treatment are criti-al to malaria control. Although P. vivax is typically dominantn the Amazon basin, this study and others have documented

relatively high proportion of P. falciparum cases (Perezato, 1998). In an area where chloroquine remains first-line

reatment against P. vivax but is ineffective against P. fal-iparum (Magris, 1996), accurate diagnosis to the speciesevel is essential. Although rapid diagnostic tests for malariaave been extensively field tested in Africa, little researchas focused on their effectiveness in Latin America. How-ver, recent trials of a rapid diagnostic test in Mexico anderu have demonstrated sufficient sensitivity and specificityor diagnosis of P. vivax in rural communities (Gonzalez-eron et al., 2005; Soto Tarazona et al., 2004). Whilsticroscopy remains the gold standard for malaria diagnosis,

apid tests can provide a quick, cheap and simple alter-ative in communities without direct access to a primaryealthcare facility and their use requires minimal training.esearch into the effectiveness of rapid tests for malariaiagnosis in Yanomami communities may enable their appli-ation in areas currently dependent on mobile medicaleams (Metzger et al., 2007).

Diarrhoea was also more commonly reported amongommunities with intermittent access to healthcare thanhose within reach of the primary healthcare facility. Inhe absence of sanitation and potable water supplies, diar-hoeal disease is common in this region. Reducing theurden of such infections therefore requires effective caseetection and management, which is dependent upon goodccess to medical care. The prevalence of splenomegalyas considerably higher in communities with intermittentccess to healthcare, as would be expected given the ele-ated prevalence of malaria and anaemia in this area. Inhe absence of other infectious diseases that can causeplenomegaly, such as visceral leishmaniasis and schistoso-iasis, the occurrence of splenic enlargement among 9—14

ear-olds, the age-group exhibiting the lowest prevalencef malaria infection, supports previous studies reporting a

revalence of hyperreactive malarial splenomegaly (HMS)f 23% in Yanomami communities (Torres et al., 1988).MS is an aberrant immune response resulting from pro-

onged exposure to malaria infection (Hoffman et al.,984).

6

slicfeaociowlsbmg(icutrAagitha

etta(mcsgbitTchta

iwr2ihopirab

stPiysh

5

Trsmcrrnoa

iooItH(

otbitmsr

AJifsttcfi

Apt

FHof

50

In this study population, malaria infection andplenomegaly remained independent risk factors forow haemoglobin concentration, as would be expected inndividuals experiencing repeated or intense episodes oflinical malaria (Crawley, 2004). Additional risk factorsor anaemia, such as intestinal helminth infection (Evet al., 1998; Ferrari et al., 1992; Miranda et al., 1998)nd nutritional deficiencies (Hidalgo, 2002; Holmes, 1985)ccur in the Amazon region and are likely to be importantontributory factors in this area. The absence of sanitationn Yanomami communities is likely to favour transmissionf intestinal helminths such as hookworm, infections ofhich contribute to anaemia through gastrointestinal blood

oss and malabsorption (Gomez et al., 2000). Nutritionaltatus among formerly nomadic populations is influencedy increased competition for food as communities becomeore sedentary, as well as a changing diet resulting from

reater accessibility to carbohydrate-rich western foodsDufour, 1994; Hidalgo, 1997; Holmes, 1984, 1985). Thisssue is currently being researched among Piaroa indigenousommunities in Venezuela and the results may improve ournderstanding of the relationship between social assimila-ion and nutritional status (G. Hidalgo et al., unpublishedeport). Diet is likely to differ somewhat between remotelto Ocamo communities and the more populated Ocamorea, as are related factors such as income. Although mostoods are exchanged in this area, income can be generatedn Ocamo by driving motorboats etc. Further research intohe epidemiology of nutritional deficiencies and intestinalelminth infections in these communities will be required tossess their relative contribution to the burden of anaemia.

There is a clear need to develop targeted, culturally rel-vant interventions to reduce the prevalence of anaemia inhis region, particularly in communities without good accesso medical care. The WHO emphasizes the importance ofdopting a multifactorial approach to controlling anaemiaWHO/UNICEF, 2004), which in this region requires effectiveanagement of malaria cases, treatment of other infectious

auses of anaemia (e.g. helminth infections) and nutritionalupplementation where necessary. In order for these strate-ies to be delivered effectively, access to healthcare muste improved. The high level of anaemia and splenomegalyn Yanomami communities has been described as an indica-or of ineffectively controlled malaria (Perez Mato, 1998).he difference in anaemia prevalence between areas withonstant and intermittent access to healthcare suggests thataemoglobin level would be a useful indicator to evaluatehe implementation of a health plan aimed at improvingccess to healthcare.

The Yanomami have a complex set of beliefs regard-ng the aetiology, nosology and curing of illness. However,hether they have an ethnobotanical pharmacopoeia is cur-

ently a controversial issue (Kelly and Carrera, 2006; Lizot,006) and the authors’ experience in the area is that theres no specific traditional treatment for malaria, anaemia orelminthiasis. In the current Venezuelan legal frameworkf health, healthcare and indigenous rights, the training

rocess for community health workers and indigenous aux-liaries relies on an intercultural approach that includes theecognition, respect and promotion of traditional medicinend shamanism. The two medical systems do not compete,ut rather they complement each other. Ethnographic work

C

Ee

P. Grenfell et al.

upports this observation, which seems to be the samehroughout the Amazon region (Buchillet, 1991; Kelly, 2003).ersonnel such as indigenous auxiliaries, in similar scenariosn Colombia, have provided examples of being able to anal-se the differences between the two medical systems and toearch for paths to optimise the care services of the officialealth systems (Herrera, 1991).

. Conclusions

he prevalence of anaemia in this study population rep-esents a severe public health problem (WHO, 2001). Theituation is particularly serious in communities with inter-ittent access to healthcare services. Although young

hildren and pregnant women can be identified as high-isk groups, further information on geographically specificisk factors for anaemia, such as helminth infections andutritional deficiencies is needed. This will assist the devel-pment of targeted and effective strategies to controlnaemia in this area.

There is an urgent need to improve access to healthcaren Alto Ocamo and other Yanomami communities dependentn mobile medical teams in order to reduce the burdenf anaemia and other causes of morbidity and mortality.mproving access to healthcare and understanding barrierso access are fundamental priorities of the Pan Americanealth Organisation’s Indigenous Peoples’ Health Initiative

PAHO, 2007).In Venezuela, the health rights of indigenous groups have

nly recently been recognised under the national constitu-ion and significant changes in service provision have yet toe implemented. If access to healthcare is to be improvedn this region, the development, planning and implemen-ation of culturally appropriate public health interventionsust involve all stakeholders, including Ministry of Health

taff, healthcare professionals and indigenous communityepresentatives.

uthors’ contributions: MM conceived the study; MM, PG,G, WGM, SVM and LV contributed to the study design,mplementation and/or coordination; PG was responsibleor data acquisition; PG, CIF and MM were responsible fortatistical analysis and data interpretation; CIF preparedhe manuscript; CC made an important intellectual con-ribution and critically revised the manuscript. All authorsontributed to the manuscript and read and approved thenal version. MM is guarantor of the paper.

cknowledgements: We are grateful to the Yanomamiatients and the collaborating staff for their contributiono this study.

unding: The study was funded by WHO (Grant No.Q/05/131401) and the Ministry of Sciences and Technol-gy of Venezuela (Grant No. PC125-018). C.I. Fanello wasunded by the Sir Halley Stewart Trust.

onflicts of interest: None declared.

thical approval: Ethical approval was obtained from thethical committees of the Centro Amazonico de Inves-

I

K

K

K

L

L

L

L

L

L

L

M

M

M

M

M

Anaemia and malaria in Yanomami communities

tigacion y Control de Enfermedades Tropicales ‘SimonBolivar’, Puerto Ayacucho, Venezuela (17/02/05) and theLondon School of Hygiene and Tropical Medicine, London,UK (Reference No. 3011).

References

Buchillet, D., 1991. Medicinas Tradicionais e Medicina Occidental naAmazonia. Edicoes CEJUP, Belem.

Buitron, D., Hurtig, A.K., San Sebastian, M., 2004. Nutritional statusof Naporuna children under five in the Amazon region of Ecuador.Rev. Panam. Salud Publica 15, 151—159.

Crawley, J., 2004. Reducing the burden of anemia in infants andyoung children in malaria-endemic countries of Africa: from evi-dence to action. Am. J. Trop. Med. Hyg. 71 (2 Suppl.), 25—34.

Dufour, D.L., 1994. Diet and Nutritional Status of Amazonian Peo-ples, in: Roosevelt, A. (Ed), Amazonian Indians from Prehistoryto the Present: Anthropological perspectives. The University ofArizona Press, Tucson.

Evans, K.J., Hansen, D.S., van Rooijen, N., Buckingham, L.A.,Schofield, L., 2006. Severe malarial anemia of low parasite bur-den in rodent models results from accelerated clearance ofuninfected erythrocytes. Blood 107, 1192—1199.

Eve, E., Ferraz, E., Thatcher, V.E., 1998. Parasitic infections in vil-lagers from three districts of the Brazilian Amazon. Ann. Trop.Med. Parasitol. 92, 79—87.

Ferrari, J.O., Ferreira, M.U., Camargo, L.M., Ferreira, C.S., 1992.Intestinal parasites among Karitiana Indians from RondoniaState, Brazil. Rev. Inst. Med. Trop. Sao Paulo 34, 223—225.

Giribaldi, G., Ulliers, D., Schwarzer, E., Roberts, I., Piacibello,W., Arese, P., 2004. Hemozoin- and 4-hydroxynonenal-mediatedinhibition of erythropoiesis. Possible role in malarial dyserythro-poiesis and anemia. Haematologica 89, 492—493.

Gomez, J., Magris, M., Marın, A., Frontado, H., Rangel, T., Botto,C., 2000. Estudio del efecto de Ivermectına en helmintosintestinales en comunidades yanomami del Alto Orinoco, EstadoAmazonas. Bol. Soc. Venez. Microbiol. 20, 131—134.

Gonzalez-Ceron, L., Rodriguez, M.H., Betanzos, A.F., Abadia, A.,2005. Efficacy of a rapid test to diagnose Plasmodium vivax insymptomatic patients of Chiapas, Mexico. Salud Publica Mex. 47,282—287.

Grantham-McGregor, S., Ani, C., 2001. A review of studies on theeffect of iron deficiency on cognitive development in children.J. Nutr. 131, 649S—666S, Discussion 666S—668S.

Herrera, X., 1991. Medicina Tradicional y Medicina Institucional: ElPromotor de Salud Investiga los Puntos de Conflicto, in: Bouch-illet, D. (Ed.), Medicinas Tradicionais e Medicina Occidental naAmazonia, Vol. 1. Edicoes CEJUP, Belem, pp. 247—266.

Hidalgo, G., 1997. Habitos nutricionales de las comunidadesindıgenas asentadas en el eje carretero de Puerto Ayacu-cho, in: Salud y Ambiente: Contribuciones al Conocimientode la Antropologıa Medica y Ecologıa Cultural en Venezuela.FACES/Vicerrectorado Academico, Universidad Central deVenezuela, Caracas, pp. 105—128.

Hidalgo, G., 2002. Vitamina A, Anemia y Antropometrıa Nutricionalen Preescolares y Escolares Piaroa, Estado Amazonas. MSc The-sis, Universidad Simon Bolıvar, Caracas.

Hoffman, S.L., Piessens, W.F., Ratiwayanto, S., Hussein, P.R., Kur-niawan, L., Piessens, P.W., Campbell, J.R., Marwoto, H.A.,1984. Reduction of suppressor T lymphocytes in the tropicalsplenomegaly syndrome. New Engl. J. Med. 310, 337—341.

Holmes, R., 1984. Modificadores no dieteticos del estado nutricionalen poblaciones de la selva tropical de Venezuela. Interciencia 9,389—391.

Holmes, R., 1985. Nutritional status and cultural change inVenezuela’s Amazon territory, in: Hemming, J. (Ed), Change in

M

M

651

the Amazon Basin. Manchester University Press, Manchester, pp.237—255.

NE, 2001. Proyecciones de la poblacion indıgena 2001—2005. Insti-tuto Nacional de Estadıstica, Venezuela.

elly, J.A., 2003. Relations within the health system among theYanomami in the Upper Orinoco, Venezuela. PhD Thesis, Uni-versity of Cambridge.

elly, J., Carrera, J., 2006. Relaciones con la Biomedicina, in:Freire, G., Tillet, A. (Eds.), Salud Indıgena en Venezuela, Vol.1. Ediciones de la Direccion de Salud Indıgena del Ministerio delPoder Popular para la Salud, Caracas, pp. 325—381.

irkwood, B., Sterne, J., 2003. Essential Medical Statistics, seconded. Blackwell Science, London.

arrea, C., Freire, W., 2002. Social inequality and child malnutri-tion in four Andean countries. Rev. Panam. Salud Publica 11,356—364.

arrea, C., Kawachi, I., 2005. Does economic inequality affect childmalnutrition? The case of Ecuador. Soc. Sci. Med. 60, 165—178.

awless, J.W., Latham, M.C., Stephenson, L.S., Kinoti, S.N.,Pertet, A.M., 1994. Iron supplementation improves appetite andgrowth in anemic Kenyan primary school children. J. Nutr. 124,645—654.

ayez, C., Nogueira, P., Combes, V., Costa, F.T., Juhan-Vague, I.,da Silva, L.H., Gysin, J., 2005. Plasmodium falciparum rhoptryprotein RSP2 triggers destruction of the erythroid lineage. Blood106, 3632—3638.

izot, J., 1988. Los Yanomami, in: Coppens, W., Escalante, B. (Eds),Los Aborigenes de Venezuela Vol. III: Etnologia Contemporanea.Fundacion La Salle/Monte Avila, Caracas.

izot, J., 2006. Los Yanomami: El Mundo Intelectual de losYanomami, in: Freire, G., Tillet, A. (Eds.), Salud Indıgena enVenezuela, Vol. 1. Ediciones de la Direccion de Salud Indıgenadel Ministerio del Poder Popular para la Salud, Caracas, pp.269—323.

ozoff, B., Jimenez, E., Hagen, J., Mollen, E., Wolf, A.W., 2000.Poorer behavioral and developmental outcome more than 10years after treatment for iron deficiency in infancy. Pediatrics105, E51.

agris, M., 1996. Susceptibilidad in vitro del Plasmodium fal-ciparum a la cloroquina, amodiaquina, quinina, mefloquina.Ocamo, Alto Orinoco, Edo. Amazonas. MSc Thesis, Universidadde Carabobo, Maracay.

agris, M., Rubio-Palis, Y., Alexander, N., Ruiz, B., Galvan, N., Frias,D., Blanco, M., Lines, J., 2007. Community-randomized trial oflambdacyhalothrin-treated hammock nets for malaria controlin Yanomami communities in the Amazon region of Venezuela.Trop. Med. Int. Health 12, 392—403.

arcano, T.J., Morgado, A., Tosta, C.E., Coura, J.R., 2004.Cross-sectional study defines difference in malaria morbid-ity in two Yanomami communities on Amazonian boundarybetween Brazil and Venezuela. Mem. Inst. Oswaldo Cruz 99,369—376.

etzger, W.G., Vivas-Martinez, S., Rodriguez, I., Goncalves, J.,Bongard, E., Fanello, C.I., Vivas, L., Magris, M., 2007. Malariadiagnosis under field conditions in the Venezuelan Amazon.Trans. R. Soc. Trop. Med. Hyg. 102, 20—24.

inisterio del Poder Popular para la Salud, 2007. Reporte decasos de malaria de la Direccion de Vigilancia EpidemiologicoAmbiental del Estado Amazonas 2004-2006. Boletın Epi-demiologico Semanal, Ano 56, Semana Epidemiologica N( 09.http://www.mpps.gob.ve/ms/modules.php?name=Downloads&op=getit&lid=179 [accessed 23 January 2008].

iranda, R.A., Xavier, F.B., Menezes, R.C., 1998. Intestinal para-sitism in a Parakana indigenous community in southwestern ParaState, Brazil. Cad. Saude Publica 14, 507—511.

ontenegro, R.A., Stephens, C., 2006. Indigenous health in LatinAmerica and the Caribbean. Lancet 367, 1859—1869.

6

O

P

P

P

S

S

S

T

W

W

52

ppenheimer, S.J., 2001. Iron and its relation to immunity andinfectious disease. J. Nutr. 131 (2S-2), 616S—633S, discussion633S—635S.

AHO, 2006. Barrio Adentro: Derecho a la Salud e Inclusion Socialen Venezuela. Organication Panamericana de la Salud, Caracas.

AHO, 2007. Health of Indigenous Peoples.http://www.paho.org/English/AD/THS/OS/Indig-home.htm[accessed 16 January 2008].

erez Mato, S., 1998. Anemia and malaria in a Yanomami Amerindianpopulation from the southern Venezuelan Amazon. Am. J. Trop.Med. Hyg. 59, 998—1001.

chellenberg, D., Schellenberg, J.R., Mushi, A., Savigny, D.,

Mgalula, L., Mbuya, C., Victora, C.G., 2003. The silent burden ofanaemia in Tanzanian children: a community-based study. Bull.World Health Organ. 81, 581—590.

hamah-Levy, T., Villalpando, S., Rivera, J.A., Mejia-Rodriguez, F.,Camacho-Cisneros, M., Monterrubio, E.A., 2003. Anemia in Mex-

W

P. Grenfell et al.

ican women: a public health problem. Salud Publica Mex. 45(Suppl. 4), S499—S507.

oto Tarazona, A., Solari Zerpa, L., Mendoza Requena, D., Llanos-Cuentas, A., Magill, A., 2004. Evaluation of the rapid diagnostictest OptiMAL for diagnosis of malaria due to Plasmodium vivax.Braz. J. Infect. Dis. 8, 151—155.

orres, J., Noya, O., Mondolfi, A., Peceno, C., Botto, C., 1988.Hyperreactive malarial splenomegaly in Venezuela. Am. J. Trop.Med. Hyg. 39, 11—14.

ernsdorfer, W.H., McGregor, I., 1988. Malaria: Principles and Prac-tice of Malariology. Churchill Livingstone, Edinburgh.

HO, 2001. Iron Deficiency Anaemia: Assessment, Prevention, and

Control. A Guide for Programme Managers. World Health Orga-nization, Geneva.

HO/UNICEF, 2004. Focusing on Anaemia: Towards an IntegratedApproach for Effective Anaemia Control. World Health Organi-zation, Geneva.