245

To determine whether there was any difference inchildhood leukemia incidence rates between popula-tions living in the proximity to oil fields and thoseliving in areas free from oil exploitation in the Amazonbasin of Ecuador, 91 cancer cases among children(0–14 years) from the provinces of Sucumbios, Orel-lana, Napo, and Pastaza during the period 1985–2000were studied. The relative risks for all leukemias indi-cated significantly elevated levels in the youngest agegroup (0–4 years), both genders combined (RR 3.48,95% CI 1.25–9.67), and in all age groups (0–14 years)combined for females (RR 2.60, 95% CI 1.11–6.08) andboth genders combined (RR 2.56, 95% CI 1.35–4.86).There was no significant difference between the twogroups in all other cancer sites combined. Study resultsare compatible with a relationship between childhoodleukemia incidence and living in the proximity of oilfields in the Ecuadorian Amazon. Key words: crude oil;leukemia; Amazon; Ecuador.

INT J OCCUP ENVIRON HEALTH 2004;10:245–250

The Amazon basin of Ecuador, known as the “Ori-ente,” consists of more than 100,000 km2 oftropical rainforest lying at the headwaters of the

Amazon river network. The region contains one of themost diverse collections of plant and animal life in theworld.1

In 1967, a Texaco–Gulf consortium discovered a richfield of oil beneath the rainforest, leading to an oilboom that has permanently reshaped the region. Sincethen, foreign companies together with Ecuador’snational oil company have extracted more than two bil-lion barrels of crude oil from the Ecuadorian Amazon.During this process, millions of gallons of untreated

toxic wastes, gas, and oil have been released into theenvironment.2 Indigenous federations, peasants’ move-ments, and environmental groups in Ecuador haveorganized in opposition to unregulated oil develop-ment, charging that contamination has caused wide-spread damage both to people and to the environ-ment.3–5 Oil development activities include severalcontaminating processes. In the Amazon basin ofEcuador, each exploratory well that is drilled producesan average of 4,000 cubic meters of drilling wastes(drilling muds, petroleum, natural gas, and formationwater) from deep below the earth’s surface. Thesewastes are frequently deposited into open, unlined pitscalled separation ponds, from which they either aredirectly discharged into the environment or leach outas the pits degrade or overflow from rainwater.2,3

If commercial quantities of oil are detected, the pro-duction stage starts. During production, oil is extractedin a mixture with formation water and gas and sepa-rated in a central facility. At each facility, over 4.3 mil-lion gallons of liquid wastes are generated every dayand discharged without treatment into pits. Roughly 53million cubic feet of “waste” gas from the separationprocess are burned daily without temperature or emis-sions controls. Additional potential contamination ofthe air is generated at pits and oil spills by hydrocar-bons coming from standing oil slicks.6

Routine maintenance activities at over 300 produc-ing wells discharge an estimated 5 million gallons ofuntreated toxic wastes into the environment every year.Leaks from wells and spills from tanks have beencommon.7 According to a study conducted by the gov-ernment in 1989, spills from flow lines alone weredumping an estimated 20,000 gallons of oil every twoweeks.8

Spills from the main and secondary pipelines arealso common. In 1992, the Ecuadorian governmentrecorded approximately 30 major spills, with an esti-mated loss of 16.8 million gallons of crude oil.3 Cur-rently, it has been estimated two big spills occur perweek from the main oil fields in the region.9 Forinstance, in 1989 at least 294,000 gallons and in 1992,

Original Articles

Incidence of Childhood Leukemiaand Oil Exploitation in theAmazon Basin of Ecuador

ANNA-KARIN HURTIG, DRPH, MIGUEL SAN SEBASTIÁN, PHD

Received from the Instituto de Epidemiología y Salud Comuni-taria “Manuel Amunárriz,” Apdo. 17-10-7410, Quito, Ecuador. Sup-ported by a grant from Medicus Mundi Gipuzkoa and Capuchinos-Navarra.

Address correspondence and reprint requests to: Dr. Anna-KarinHurtig, Geografigränd 14 C, 907 32 Umeå, Sweden; e-mail: <akhurtig@hotmail.com>.

about 275,000 gallons of crude oil caused the Naporiver (1 km wide) to run black during one week.10

Overall, more than 30 billion gallons of toxic wastesand crude oil had been discharged into the land andwaterways of the “Oriente” by 1993.3 This compares tothe 10.8 million gallons spilled in the Exxon Valdezdisaster in Alaska in 1989, one of the most damagingsea oil spills that has ever occurred.

In 1994, the Ecuadorian environmental and humanrights organisation Centro de Derechos Económicos y Sociales[Center for Economic and Social Rights] released areport documenting dangerous levels of toxic contami-nation.11 Concentrations of polynuclear aromatichydrocarbons found in drinking, bathing, and fishingwaters were 10 to 10,000 times greater than allowableunder the United States Environmental ProtectionAgency guidelines. In 1999, the Instituto de Epidemiologíay Salud Comunitaria “Manuel Amunárriz” (IESCMA), alocal health nongovernmental organization, undertookwater analysis for total petroleum hydrocarbons (TPH)in communities in the proximity of oil fields and com-munities far distant from them. Water analyses showedhigh levels of exposure to oil-derived chemicals amongthe residents of the exposed communities.12 In somestreams hydrocarbon concentrations reached 144 and288 times the limit permitted by the European Com-munity regulation.13 The same year, a report from theMinistry of Environment supported these results, whenconcentrations of TPH over 300 times the limit permit-ted were found in the streams of one of the communi-ties of the previous study.14

Although several studies have focused on residentsexposed to major oil spillages,15–17 epidemiologic stud-ies of communities exposed to oil pollutants near oilfields are few.18,19

In a study in the Amazon basin of Ecuador, an excessof cancers was observed among males in a villagelocated in an oil-producing area.20 A later study fromthe Amazon basin of Ecuador found significantly higherincidences of all cancer sites combined in both menand women in counties where oil exploitation had beenon going for at least 20 years. Significantly elevatedlevels were observed for cancers of stomach, rectum,skin (melanoma), soft tissue, and kidney in men and forcancers of the cervix and lymph nodes in women. Anincrease in hematopoietic cancers was also observed inchildren.21 The object of the present study was to exam-ine in detail of the incidence of leukemia in the agegroup 0–14 years to determine whether there was anydifference in incidence rates between populations livingin proximity to oil fields and those living in areas freefrom oil exploitation in the Amazon basin of Ecuador.

POPULATION AND METHODS

Area of Study

The study was carried out in the provinces ofSucumbios, Orellana, Napo, and Pastaza, situated inthe eastern part of Ecuador (Figure 1). Each provinceis divided into counties (cantones). The study area hasa total population of approximately 356,406 indige-nous people and peasants.22 The indigenous peoplelive in small communities scattered along the rivers,making their living by hunting, fishing, and subsistenceagriculture. The peasants arrived to the area in the1970s, following the paths opened by oil companies.They make their living mainly by agriculture and cattle-raising. In oil-producing areas approximately 2% of theworking population is employed by the oil industry.23

Physical infrastructure in the region is poor. Few vil-lages and small towns (10,000–15,000 citizens) haveelectricity and piped drinking water, the majority of theinhabitants live without these facilities. Many of theroads in oil-producing counties are paved by crude oilto reduce the amount of dust otherwise produced inthis tropical climate. In each province there is a provin-cial hospital and the counties have health centers. Thehospitals have no histopathology services and no accessto radiotherapy or chemotherapy. Two mission hospi-tals with well functioning infrastructures are located inthe no–oil-producing counties of Mera and Archidona.Oil-producing areas have no better medical facilitiesthan those areas where no such industry is present.Qualified personnel in the oil industry are contractedfrom the capital or abroad and flown out in case ofhealth problems. Only recently have some oil compa-nies included health expenditures in their contracts

246 • Hurtig, San Sebastián www.ijoeh.com • INT J OCCUP ENVIRON HEALTH

Kilometers

0 50 100

Figure 1—Map showing counties included in the study;exposed counties in grey.

with residents. Two counties, Sachas and Shushufindi,are producing and processing palm oil. The oil indus-try is the only major industry in the region.

Cancer Data

No cancer registry is available in the Amazon region.Suspected cancer cases are referred from theseprovinces to Quito, the capital. All cases diagnosed inQuito are registered in the National Cancer Registry.24

This register was used for the purpose of our study.During 1985–2000, 1,207 cases of cancer were reportedto the National Cancer Registry from the provinces ofSucumbios, Orellana, Napo, and Pastaza among allages. The National Cancer Registry contains personalidentification, gender, age at diagnosis, cancer site, his-tology (the 10th International Classification of Dis-eases), year of diagnosis, residence at diagnosis(county), and education. In the register, eight caseslacked data on age and three cases among the agegroup 0–14 years lacked data on place of residence;those were excluded from the study.

Population Data

Population data from the counties of the fourprovinces by gender and five-year age strata for the year1993 were used. These were the projections of theNational Institute of Statistics and Census based on the1990 National Census.25

Exposure Status

The study was ecologic, and exposure status wasdefined on a county level. Exposed children weredefined as those living in a county where oil exploita-

tion had been ongoing for at least 20 years at the timeof the study. Non-exposed were identified as thosecounties without oil development activities (excludingseismic studies during the late 1990s with no exploita-tion activities). Four counties (Lago Agrio, Shushu-findi, Orellana, and Sachas; 56,202 children; 51.5%males) were defined as exposed and 11 (Cascales, Pto.El Carmen, La Bonita, Lumbaqui, Aguarico, Tena,Archidona, El Chaco, Baeza, Puyo, Mera; 71,970 chil-dren; 50.7% males) as non-exposed.

Statistical Analysis

Incidence rates for overall and specific cancer siteswere calculated. Relative risks (RRs) along with the95% confidence intervals (CIs) were calculated formales and females.

RESULTS

Between 1985 and 2000 a total of 91 cases of cancers,including 42 leukemia cases, were observed in thestudy area among children 0–14 years old. Twenty-eightcases of leukemia and 27 cases of other cancersoccurred in exposed counties.

Data on all leukemia and all other cancer sites com-bined by gender and age group are presented in Table1. The RR for all leukemia indicated significantly ele-vated levels in the youngest age group (0–4 years), bothgenders combined (RR 3.48, 95% CI 1.25-9.67), and inall age groups (0-14 years) combined for females (RR2.60, 95% CI 1.11–6.08) and both genders combined(RR 2.56, 95% CI 1.35–4.86) in the exposed counties.There was no significant difference in relation to expo-sure status in all other cancer sites combined.

VOL 10/NO 3, JUL/SEP 2004 • www.ijoeh.com Childhood Leukemia • 247

TABLE 1. Risks of Leukemia and All Other Cancers for Exposed versus Non-exposed to Oil Pollution by Age Groupand Sex, Amazon Region, 1985–2000

All Leukemias All Other Cancers_____________________________________ ______________________________________Cases in Cases in

Exposed Group RR (95 % CI) Exposed Group RR (95 % CI)

0–4 years oldFemale 6 7.58 (0.91–62.99) 6 1.52 (0.46– 4.97)Male 8 2.45 (0.74– 8.13) 8 1.96 (0.64– 5.98)Both 14 3.48 (1.25– 9.67) 14 1.74 (0.77– 3.92)

5–9 years oldFemale 4 1.72 (0.38– 7.68) 3 0.77 (0.18– 3.24)Male 3 3.78 (0.39–36.37) 3 3.78 (0.39–36.37)Both 7 2.23 (0.65– 7.62) 6 1.28 (0.41– 3.95)

10–14 years oldFemale 6 2.05 (0.58– 7.26) 4 1.82 (0.41– 6.96)Male 1 1.31 (0.08–20.94) 3 1.31 (0.26– 6.49)Both 7 1.87 (0.59– 5.89) 7 1.56 (0.34– 4.64)

0–14 years oldFemale 16 2.60 (1.11– 6.08) 13 1.30 (0.60– 2.81)Male 12 2.52 (0.95– 6.72) 14 1.96 (0.85– 4.53)Both 28 2.56 (1.35– 4.86) 27 1.57 (0.90– 2.76)

Data on the distribution of leukemia cell types bygroup and gender are presented in Table 2. Acute lym-phoblastic leukemia (ALL) accounted for 20 (71.0%)of the leukemia cases in the exposed group and 10(71.0%) of those in the unexposed group. ALL wasfound to be significantly elevated in the exposed coun-ties for females (RR 2.60, 95% CI 1.05–6.46) and forboth genders combined (RR 2.56, 95% CI 1.35–4.86).

DISCUSSION

This study compared incidences of childhood leukemiain counties with oil-development activities and thosewithout in the Amazon basin of Ecuador (1985–2000).The results showed considerable differences in the inci-dences of childhood leukemia according to our expo-sure definition.

Childhood leukemia is the most common canceramong children. In Ecuador, 60% of deaths due tocancer in children less than 14 years of age are attrib-utable to leukemia. The standardized incidence ofleukemia for the Quito population has not changed inthe 15 years of the NCR (5–6/100.000), being similarfor both sexes.24 While low incidences could beexpected in our rural population compared withQuito, the possibility of underreporting must be con-sidered. The reasons for a higher incidence amonggirls are unclear. A possible explanation might be moreexposure to contaminated water during daily activities.

Crude oil is a complex mixture of many chemicalcompounds, mostly hydrocarbons. The petroleumhydrocarbons of most toxicologic interest are volatileorganic compounds (benzene, xylene, and toluene)and polynuclear aromatic hydrocarbons (PAH).26 Ben-zene is a well-known cause of leukemia,27,28 and perhapsother hematologic neoplasms and disorders.29,30 Noadequate data on the incidences of cancers after humanexposures to the other volatile organic chemicalsexist.15 An ecologic study that examined the relation-ship between the incidence of leukemia and volatile

organic chemical (VOC) contamination of drinkingwater supplies in the United States suggested that drink-ing water contaminated with VOCs might increase theincidence of leukemia among exposed females.31

Corresponding studies of the incidence of leukemiain people residing near oil fields are lacking. Moreproblematic, related existing studies tend to be basedon lower levels of exposure than those in Ecuador. Astudy from Wales did not find an association betweenthe incidence of leukemias and lymphomas in childrenand young people and their residence in the areaaround the BP Chemical site at Baglan Bay, SouthWales.32 A report encompassing all industrial com-plexes that include major oil refineries in Great Britainfound no evidence of an association between residencenear oil refineries and leukemia or non-Hodgkin’s lym-phoma.33 However, the relationships between leukemiaand toxic exposures were examined in a case–controlstudy of a cluster of 14 childhood cases in a restrictedarea in The Netherlands. Results showed excessiveexposures both to insecticides and to petroleum prod-ucts among the cases.34 There are also several studiesshowing that petroleum and fuel exhaust exposures areleukemia hazards in industrial workers, and that not allof the toxicity is explained by benzene. Childhoodleukemia and other childhood cancers have been geo-graphically associated with industrial atmosphericeffluent, for example, with petroleum-derived volatilesin Great Britain.35,36 Few studies have been conductedin petroleum-exploration and petroleum-productionworkers. High incidences of leukemia in oil-fieldsworkers have been found in studies carried out in theUnited States and China.37,38

Because they reflect group rather that individual char-acteristics and exposures, ecologic studies must be inter-preted cautiously. The use of aggregated data instead ofthe joint distributions of exposure, outcome, and covari-ates at the individual level may lead to severe bias in eco-logic analyses.39 Using narrow exposure data and smallunits of analysis (parishes) could have minimized the

248 • Hurtig, San Sebastián www.ijoeh.com • INT J OCCUP ENVIRON HEALTH

TABLE 2. Risks of Acute Leukemia cell type for category of exposed versus non-exposed by sex, Amazon region,1985-2000

Cases inExposed Group Incidence Rate RR (95 % CI)

Lymphoblastic leukemiaFemale 14 3.21 2.60 (1.05- 6.45)Male 6 1.30 2.52 (0.63-10.09)Both 20 2.22 2.56 (1.20- 5.47)

Myeloblastic leukemiaFemale 2 0.46 2.60 (0.24-28.69)Male 6 1.30 2.52 (0.63-10.09)Both 8 0.89 2.56 (0.77 -8.50)

All leukemia cell typesFemale 16 3.67 2.60 (1.11- 6.08)Male 12 2.60 2.52 (0.95- 6.72)Both 28 3.11 2.56 (1.35- 4.86)

effect of this bias, but was not feasible in the present studydue to the lack of data. Overall, it is difficult to measurethe impact of the ecologic bias in the study.

Because pf geographic and socioeconomic impedi-ments to accessing adequate health care, it is likely thatmany cases of cancer never got referred to Quito fromthe study area. Health services are poor in bothexposed and non-exposed counties, but factors such asdiagnostic skills and transport facilities might influencereferral patterns. It is also possible that on a countylevel there are differences in racial composition andlife-style patterns between populations that might con-found risk estimates. However, no information aboutthe distribution of such potentially important con-founders was available.

Several limitations in the data and methods need alsoto be considered. Population data relied on countycensus figures estimated from the 1990 NationalCensus. Errors in population estimates, including dif-ferential migration patterns, might bias estimates ofrisk. It is possible that the exposed counties had hadmore rapidly increasing populations compared with thenon-exposed ones, providing a relatively greater under-estimate of population denominators for these coun-ties. Data from the National Institute of Statistics andCensus give no evidence that this is the case.22,25 Cancerrates were based on county of residence at time of diag-nosis without information as to length of time at cur-rent residence. Because the latency period for cancercan be long, an assessment of migration into and out ofcounties as well as residence time in the county wouldhave been useful, but no data were available.

One possibility to explain any excess risk near anindustrial source is that it reflects parents’ occupationalexposures rather than environmental factors. Parents’occupational data were not available. Two exposedcounties have also oil palm industry, where pesticideuse is common. The impact of this exposure on theresults presented could not be measured.

The results suggest a relationship between leukemiaincidence in children and living in the proximity of oilfields, although this ecologic study cannot lead to acausal inference. However, the possibility of a causalrelationship is supported by several criteria. First, thestrength of the association between the outcome andthe exposure; second, the finding that only leukemiawas at increased risk in the exposed area increases theplausibility of the results. Third, by using surrogatedata that are representative of several decades of oilpollution exposure, a plausible time sequence fromexposure to development of disease can be inferred.

Further research is necessary to determine whetherthe observed associations do reflect an underlyingcausal relationship. A next step could be epidemiologicstudies at the individual level. Meanwhile, an environ-mental monitoring system to assess, control, and assistin the elimination of sources of pollution in the area

and a surveillance system to gain knowledge of the evo-lution of cancer incidence and distribution in the areaare urgently recommended.

The authors thank Dr. Yepez from the National Cancer Registry forproviding the data.

References

1. Ecuador rainforest. <www.rainforestweb.org/Rainforest_Regions/South_America/ Ecuador/?state=more> (accessed 25April 2003).

2. Kimerling J. Amazon Crude. New York: Brickfront Graphics, 1991.3. Jochnick C, Normand R, Zaidi S. Rights violations in the

Ecuadorian Amazon: the human consequences of oil develop-ment. Health Human Rights. 1994;1:82-100.

4. Garzón P. Impacto socioambiental de la actividad petrolera:estudio de caso de las comunidades San Carlos y La Primavera.In: Varea A, Ortiz P (eds). Marea negra en la Amazonía: con-flictos socioambientales vinculados a la actividad petrolera en elEcuador. Quito, Ecuador: Abya-Yala, 1995: 265-94. [In Spanish]

5. Kimerling J. Oil development in Ecuador and Peru: law, politicsand the environment. Draft paper presented at Amazonia 2000:Development, Environment and Geopolitics. London, England:Institute of Latin American Studies, University of London, 1998.

6. Centro de Derechos Económicos y Sociales. El petróleo no eseterno. Quito, Ecuador: CDES, 1999. [In Spanish]

7. Almeida A. Reseña sobre la historia ecológica de la Amazoníaecuatoriana. In: Martínez E (ed). Ecuador post petrolero.Quito, Ecuador: Acción ecológica, 2000. [In Spanish]

8. Dirección General de Medio Ambiente. Estudio de impactoambiental 42. Quito, Ecuador: Dirección General de MedioAmbiente, 1989. [In spanish]

9. Dos derrames de petróleo al mes en el campo Auca (2002) ElComercio, 18 October. [In Spanish]

10. Rainforest Action Network News. <www.ran.org/ran/ran_campaigns/amazonia/ texaco_ecuador.html> (accessed 26April 2003).

11. Centro de Derechos Económicos y Sociales. Violaciones dederechos en la Amazonía Ecuatoriana. Quito, Ecuador: Abya-Yala., 1994.

12. San Sebastián M. Informe Yana Curi: impacto de la actividadpetrolera en la salud de poblaciones rurales de la Amazonia ecua-toriana. Quito, Ecuador: Cicame & Abya-Yala, 2000. [In Spanish]

13. Zehner R, Villacreces LA. Estudio de la calidad de aguas de ríoen la zona de amortiguamiento del Parque Nacional Yasuní.Primera fase: monitoreo de aguas—screening Octubre de 1997.Coca, Ecuador: Laboratorio de Aguas y Suelos P. MiguelGamboa-Fepp, 1998. [In Spanish]

14. Ministerio de Medio Ambiente. Informe de inspección ambien-tal al área de las comunidades Flor de Manduro y Centro Man-duro ubicadas en el bloque siete operado por la compañíaOryx. Quito, Ecuador: Ministerio de Medio Ambiente, 1999. [InSpanish]

15. Campbell D, Cox D, Crum J, Foster K, Christie P, Brewster D.Initial effects of the grounding of the tanker Braer on health inShetland. BMJ. 1993;307:1251-5.

16. Palinkas LA, Petterson JS, Russell J, Downs MA. Community pat-terns of psychiatric disorders after the Exxon Valdez oil spill.Am J Psychiatry. 1993;150:1517-23.

17. Lyons RA, Temple MF, Evans D, Fone DL, Palmer SR. Acutehealth effects of the Sea Empress oil spill. J Epidemiol CommunHealth. 1999;53:306-10.

18. San Sebastián M, Armstrong M, Stephens C. La salud demujeres que viven cerca de pozos y estaciones de petróleo en laAmazonía ecuatoriana. Rev Panam Salud Pública. 2001;9:375-84. [In Spanish]

19. San Sebastián M, Armstrong M, Stephens C. Outcome of preg-nancy among women living in the proximity of oil fields in theAmazon basin of Ecuador. Int J Occup Environ Health. 2002;8:312-9.

20. San Sebastián M, Armstrong M, Cordoba JA, Stephens C. Envi-ronmental exposure and cancer incidence near oil fields in theAmazon basin of Ecuador. Occup Environ Med. 2001;58:517-22.

VOL 10/NO 3, JUL/SEP 2004 • www.ijoeh.com Childhood Leukemia • 249

21. Hurtig AK, San Sebastián M. Geographical differences of cancerincidence in the Amazon basin of Ecuador in relation to resi-dency near oil fields. Int J Epidemiol. 2002;31:1021-7.

22. Instituto Nacional de Estadísticas y Censos del Ecuador. VICenso de Población y V de Vivienda. Quito, Ecuador: INEC,2001. [In Spanish]

23. United Nations Childrens Fund. Situación de las madres y losniños en zonas de grandes proyectos. Quito, Ecuador: UNICEF,1992. [In Spanish]

24. Sociedad de Lucha contra el Cáncer. Cáncer en regiones delEcuador. Quito, Ecuador: SOLCA, 2001. [In Spanish]

25. Instituto Nacional de Estadísticas y Censos del Ecuador. V Censode Población y IV de Vivienda. Quito, Ecuador: INEC, 1990. [InSpanish]

26. IARC. Monographs on the evaluation of the carcinogenic risk ofchemicals to man: occupational exposures to petroleum refin-ing; crude oil and major petroleum fuels. Volume 45. Lyon,France: International Agency for Research on Cancer, 1989.

27. Wong O. An industry wide mortality study of chemical workersoccupationally exposed to benzene. I. General results. Br J IndMed. 1987;44:365-81.

28. Austin H, Delzell E, Cole P. Benzene and leukemia. A review ofthe literature and a risk assessment. Am J Epidemiol. 1988;127:419-39.

29. Hayes RB, Yin SN, Dosemeci M, et al. Benzene and the dose-related incidence of hematologic neoplasms in China. J NatlCancer Inst. 1997;89:1065-71.

30. Savitz DA, Andrews KW. Review of epidemiologic evidence onbenzene and lymphatic and hematopoietic cancers. Am J Ind

Med. 1997;31:287-95.31. Fagliano J, Berry M, Bove F, Burke T. Drinking water contami-

nation and the incidence of leukemia: an ecologic study. Am JPublic Health. 1990;80:1209-12.

32. Lyons RA, Monaghan SP, Heaven M, Littlepage BN, Vincent TJ,Draper GJ. Incidence of leukaemia and lymphoma in youngpeople in the vicinity of the petrochemical plant at Baglan Bay,South Wales, 1974 to 1991. Occup Environ Med. 1995;52:225-8.

33. Wilkinson P, Thakrar B, Walls P, et al. Lymphohaematopoieticmalignancy around all industrial complexes that include majoroil refineries in Great Britain. Occup Environ Med. 1999;56:577-80.

34. Mulder YM, Drijver M, Kreis IA. Case–control study on the asso-ciation between a cluster of childhood haematopoietic malig-nancies and local environmental factors in Aalsmeer, TheNetherlands. J Epidemiol Communi Health. 1994;48:161-5.

35. Knox EG, Gilman EA. Hazard proximities of childhood cancersin Great Britain from 1953–80. J Epidemiol Commun Health.1997;51:151-9.

36. Knox EG, Gilman EA. Migration patterns of children withcancer in Britain. J Epidemiol Commun Health. 1998;52:716-26.

37. Sathiakumar N, Delzell E, Cole P, Brill I, Frisch J, Spivey G. Acase–control study of leukemia among petroleum workers. JOccup Environ Med. 1995;37:1269-77.

38. Yang C, Zhang X. Incidence survey of leukaemia in China. ChinMed Sci J. 1991;6:65-70.

39. Dos Santos Silva I. Cancer Epidemiology: Principles and Meth-ods. Lyon, France: International Agency for Research onCancer, 1999.

250 • Hurtig, San Sebastián www.ijoeh.com • INT J OCCUP ENVIRON HEALTH

456

From the Guest Editors:—In theirletter, “Elevating the Level of Scien-tific Discourse,”1 Rothman and Arel-lano claim that “it would be a misuseof science to use weak or flawedstudies as a pretext to support theclaim that the reason to preserve theAmazon environment is to avoid anadverse effect on cancer rates.”However, as those authors know,there is no such thing as perfect sci-ence. Weak or “flawed” studies canbe useful in determining risk. Onthe other hand, as shown by Gen-naro et al.’s article in this issue, thepetrochemical companies, includ-ing Texaco, manipulated studies toavoid liability and protective publichealth regulations. Obviously,manipulated studies should notserve as a basis for public healthdecisions. Such has been the case inthe many studies produced byTexaco and Chevron that they claimshow that their workers, exposed tomany known carcinogens, arehealthier than unexposed popula-tions.2–6 This may in part be due tothe healthy-worker effect, but thecompanies have also been known touse dubious methods to study theirown workers’ disease, includingcounting exposed subcontractors aspart of “unexposed” controlgroups.7 These companies often usestudies to justify workplace and envi-ronmental exposures that are dan-gerous according to both commonsense and scientific evidence.

Texaco has tried to use Rothmanand Arrellano’s work to discreditstudies showing adverse healtheffects of its practices. Unfortu-nately, a magnifying glass taken tothe Ecuador studies cannot makethe risk disappear. This is some-thing that Rothman knows well; hebegins his chapter in his jointlyedited book with a critical analysis

quate environmental monitoring asthey were of the scientists who inter-preted the only data that were avail-able. “Weak and flawed” or not, as aconsequence of their sponsor’s con-duct these were the only data avail-able to identify the public healthimpact of the exploitation of thesenatural resources.

Rothman and Arellano shouldnot be shocked to discover that thecompany that solicited their reviewand personal appearance at a pressconference (entitled “Experts SayHealth Studies Promoted byLawyers and Activists are Flawed,Biased, and Inconclusive”) wasusing them to influence publicopinion, political leaders, and thecourts in Ecuador. It remains to beseen whether these antics will helpthe oil company escape blame forthe health and environmentaldamage in this part of the Amazon.During the more than two decadesthey operated in Ecuador, Texaconever reinjected wastewater fromthe drilling process, even thoughthe company held patents on themost effective way to do so.16,17

Wastewater re-injection has beenrequired since 1911 in Texas and by1971 was described by industry asbeing “integral” to oil productionin the United States.18 Reinjectingthe wastewater in the Amazonwould have cost Texaco only about$1/barrel, according to conserva-tive estimates.19 During most of thetime Texaco was extracting oil fromEcuador, it was sold for US$12–35per barrel.20 Surely the oil giantcould have afforded to take thesame precautions in Ecuador as itdid in the United States.

Despite Rothman and Arellano’sclaim, we do not wish to “contendthat Chevron–Texaco would not beallowed to request an expert evalu-

of Hill’s seminal work on episte-mology,8 pp. 24-–8 yet he does not crit-icize Hill’s statement that, “All sci-entific work is incomplete—whether it be observational orexperimental. All scientific work isliable to be upset or modified byadvancing knowledge. That doesnot confer upon us a freedom toignore the knowledge we alreadyhave, or to postpone the action thatit appears to demand at a giventime.”9 p 300

Hill also advises us to considerseveral “points of view” in consider-ing causal relationships.9 Labora-tory findings can confirm or sub-stantiate meager or equivocalresults in humans or identify car-cinogenic risks to humans withouthaving any other information. Inthe absence of reliable chemicalcarcinogenesis data for humans,the prudent public health path totake is to rely on controlled andwell-conducted long-term carcino-genesis bioassays using laboratoryanimals. Animal study findingshave led to an international con-sensus that a number of the agentsgenerated during oil productionare toxic and/or carcinogenic, andwith foresight may be deemed tohave the potential for constitutingserious public health hazards.

The evidence in this instance isthat containment has been defi-cient and that widespread environ-mental damage has resulted.10–15

The responsible companies haveeither failed to conduct or havefailed to publish the qualitative andquantitative environmental and bio-logical monitoring that Rothmanand Arellano deem valuable. Roth-man and Arellano would appearmore even-handed had they been ascritical of their sponsors’ failures tocontain hazards and conduct ade-

Letters

Chevron–Texaco’s Science

VOL 11/NO 4, OCT/DEC 2005 • www.ijoeh.com Letters • 457

ation of scientific work.” Instead,we urge that scientists truly inter-ested in the public and environ-mental health be aware of the usesto which their work may be put. ForChevron–Texaco science is not atool for pursuing truth or humanhappiness. Instead, it is a means tothe corporation’s goal of maximiz-ing profit. Profit maximization isnot the aim of public health, andpublic health scientists should notbe surprised if they are called totask for serving the interests of cor-porations with such self-interest.

We wish the oil company hadspent its resources to implement 90-year-old hygiene practices thatcould have protected the environ-ment and people. If they had doneso, perhaps they could have avoidedspending so much on hiring anarmy of lawyers, public relationsspecialists, scientific consultants,and expert witnesses, etc., to defendtheir indefensible conduct.

DAVID S. EGILMAN, MD, MPHSUSANNA RANKIN BOHME, AM8 North Main Street, Suite 404Attleboro, MA 02703

References

1. Rothman KJ, Arellano F. Elevating thelevel of scientific discourse (letter). Int

J Occup Environ Health. 2005;11:327-8.

2. Wong O, Morgan RW, Bailey WJ, Swen-cicki RE, Claxton K, Kheifets L. An epi-demiological study of petroleum refin-ery employees. Br J Ind Med. 1986;43:6-17.

3. Divine BJ, Hartman CM. Update of astudy of crude oil production workers1946–94.Occup Environ Med. 2000;57:411-7.

4. Divine BJ, Hartman CM, Wendt JK.Update of the Texaco mortality study1947–93: Part I. Analysis of overall pat-terns of mortality among refining,research, and petrochemical workers.Occup Environ Med. 1999;56:167-73.

5. Divine BJ, Wendt JK, Hartman CM.Cancer mortality among workers at abutadiene production facility. IARC SciPubl. 1993;(127):345-62.

6. Divine BJ, Barron V, Kaplan SD.Texaco mortality study. I. Mortalityamong refinery, petrochemical, andresearch workers. J Occup Med. 1985;27:445-7.

7. Ludwig ER, Madeksho L, Egilman D.Re: Mesothelioma and lung tumorsattributable to asbestos among petro-leum workers. Am J Ind Med. 2000;37:275-82.

8. Rothman KJ, Greenland S (eds).Modern Epidemiology, 2nd ed. Phila-delphia, PA: Lippincott Raven, 1998.

9. Hill AB.The Environment and disease:association or causation? Proc RoyalSoc Med. 1965;58:295-300.

10. San Sebastian M, Armstrong B, Cor-doba JA, Stephens C. Exposures andcancer incidence near oil fields in theAmazon basin of Ecuador. Occup Envi-ron Med. 2001;58:517-22.

11. San Sebastian M, Armstrong B,Stephens C. Health of women livingnear oil wells and oil production sta-tions in the Amazon region of Ecuador.Rev Panam Salud Publica. 2001;9:375-

84. [In Spanish]12. San Sebastian M, Armstrong B,

Stephens C. Outcomes of pregnancyamong women living in the proximityof oil fields in the Amazon basin ofEcuador. Int J Occup Environ Health.2002;8:312-9.

13. Hurtig AK, San Sebastian M. Gyneco-logic and breast malignancies in theAmazon basin of Ecuador, 1985–1998.Int J Gynecol Obstet. 2002;76:199-201.

14. Hurtig AK, San Sebastian M. Geo-graphical differences in cancer inci-dence in the Amazon basin of Ecuadorin relation to residence near oil fields.Int J Epidemiol. 2002;31:1021-7.

15. Hurtig AK, San Sebastian M. Incidenceof childhood leukemia and oilexploitation in the Amazon basin ofEcuador. Int J Occup Environ Health.2004;10:245-50.

16. United States Patent 3,680,389. Freder-ick H. Binkely, Jr., et al., assignor toTexaco Inc. August 1, 1972.

17. United States Patent 3,817,859. Jack F.Tate, assignor to Texaco Inc. June 18,1975.

18. Water Pollution Control Legislation—1971 (Proposed Amendments to Exist-ing Legislation). Hearings before theCommittee on Public Works, House ofRepresentatives, Ninety-Second Con-gress, First Section. July, September,and November 1971. Washington, DC:Committee on Public Works, p 1713.

19. Lawrence AW, Miller JA. A regionalassessment of produced water treat-ment and disposal practices andresearch needs: topical report. Pre-pared by Remediation Technologiesfor Gas Research Institute, ContractNo. 5091-253-2215. August 1995.

20. World Oil Market and Oil PriceChronologies: 1970–2004. March 2005.Energy Information Association.<http://www.eia.doe.gov/emeu/cabs/chron.html>. Accessed 7/14/05.

“Yana Curi”Report

The impact of oil developmenton the health of the people of the Ecuadorian Amazon

Dr. Miguel San Sebastián Dr. Juan Antonio Córdoba

Translation by Kristen Keating

--------------------------------------------------------------------------------------------------Departamento de Pastoral Social del Vicariato de Aguarico

London School of Hygiene and Tropical MedicineMedicus Mundi

For all those who fight to defend the land and the people of theOriente...

“The earth is not dying,she is being killed.

And the people who are killing herHave names and addresses.”

-U. Utah Philips

Vicariato Apostólico de AguaricoEditions: CICAME

Coca, Orellana

Authors: Miguel San Sebastián Juan Antonio Córdoba

June 1999

CONTENTS

INTRODUCTION 1

I. IS THERE REASON FOR CONCERN? 3

1. The people’s response to the situation 4

2. Evidence of the dangers of contamination 5

II. THE CASE OF SAN CARLOS 6

1. Area of study 6

2. How did we conduct the study? 7

3. What were the results? 9

4. What is the significance of these results? 13

5. Some final recommendations 15

APPENDIX 1 16

APPENDIX 2 23

APPENDIX 3 24

BIBLIOGRAPHY 25

1

--------------------------------------------------------------------------------------------a YANA CURI: Quechua for “oil”. Literally “Black Gold.”

INTRODUCTION

Oil has been and continues to be not only one of the principal sources ofincome for Ecuador, but also a destructive force on its environment. Since1972, international oil companies led by Texaco in collaboration with thenational company, Petroecuador (previously CEPE), have extracted almosttwo billion barrels of oil mainly from the region of the Amazon. In thisprocess, millions of gallons of oil and toxic waste have been dumped directlyinto the environment. Indian communities and farmers, as well as nationalenvironmental groups, have spoken out during the past years in clearopposition to the lack of regulation of oil industry development,complaining that the contamination has caused enormous damage to theenvironment as well as to the health of the people1.

Despite the evident impact of oil exploitation on the Amazon region’secosystem, and the potential health risks for the area’s inhabitants, there hasbeen no integral chemical study of its impact on the environment,2 andinformation about the health effects on residents of these oil-extracting areasis scarce 3. But the residents are extremely worried about the risk that thecontamination poses to their health. The oil companies and the governmentitself have challenged the affected communities and environmental groups toprovide evidence of harmful effects on health in order for them to changetheir oil-extracting strategies and control contamination.

As a result, the Departamento de Pastoral Social del Vicariato Apostólico ofAguarico, sensitive to the problem and to the concerns of the communities,has begun, in collaboration with the Department of Tropical Medicine andHygiene of the University of London, to investigate the possibleconsequences that the oil contamination has on the health of the ruralpopulation of the Ecuadorian Amazon.

The two main objectives of this study are:• increase the limited knowledge that exists about the health effects caused

by the contamination surrounding the oil wells.• answer the questions and concerns of the people of the Oriente about the

effects that this contamination has on their health.

This study, called Yana Curia, begins with an initial report on the risk ofcancer that residents of the oil exploitation zones are exposed to. In themonths to come, a second report will be published with the results of

2

--------------------------------------------------------------------------------------------a YANA CURI: Quechua for “oil”. Literally “Black Gold.”

another, more extensive study on the impact of the oil contamination on thehealth of the people.

The current document is divided into two parts:

ð the first part briefly presents the problem of oil contamination in theEcuadorian Amazon and the response of the people of the region to thesituation.

ð the second part examines the case of San Carlos, a town subjected toelevated and continuous exposure to oil contaminants, where a groupingof cancers was discovered. In this section, the development of theinvestigation of this grouping, the significance of the results, and somefinal recommendations are presented.

We would like to especially thank Rosa Moreno for her collaboration andhospitality each time we showed up in San Carlos; the departments ofstatistics of the Society to Fight Cancer (Sociedad de Lucha Contra el Cancer/ SOLCA), the Eugenio Espejo Hospital and the Baca Ortiz Hospital for thedata they provided, Janet Andrade and Dr. Pepe Yépez, for their interest andhelp with the various problems that arose while conducting our study; ourfriends from the associations for health promotion “Sandi Yura” and theFundación Salud Amazónica for all of their logistic support. Finally, we aregrateful to the Spanish non-governmental organization Medicus MundiGipúzkoa, which collaborated in the financing of our study.

3

I. IS THERE CAUSE FOR CONCERN?

Since the start of oil exploitation in the Ecuadorian Oriente, there has beenno clear policy for reducing the devastating contamination that it causes.Even the few areas created for the conservation of ecosystems, such asnational parks and protected areas, have not been respected by the oilcompanies4.

With the development of the oil industry, roads were constructed to makepossible the colonization and deforestation of a million hectares of forest. Inaddition, every day hundreds of oil wells generate more than 4.3 milliongallons of toxic waste, which are almost totally unloaded or spilled into theuntreated environment. The crude oil, poured regularly on the roads inorder to maintain them and to control dust, is scattered throughout theenvironment by the rain. It is estimated that the cracks in the in the Trans-Ecuadorian Pipeline (Sistema del Oleoducto Trans-Ecuatoriano [SOTE]) haveprovoked the spilling of 400,000 barrels of oil. These practices contaminateinnumerable rivers, streams, and estuaries, often the only sources of water forthe inhabitants of this region. Similarly, every day the continuous burningof oil and millions of cubic feet of gas produce highly toxic compounds thatcontribute to the contamination of the air.

The activities of the oil companies during these years have been characterizedby the logic of rapid accumulation with little investment, without taking intoaccount the country’s need for energy or natural resources. Obviously thisattitude has led to carelessness and disrespect towards environmental,cultural, and socioeconomic issues of the towns and areas where theyoperate5.

The companies have defended themselves against these accusations,emphasizing the essential importance of oil for the development of Ecuadorand claiming that all of their actions have complied with environmentalregulations of the country and that their methods followed “the habitualpractices of international industry.” They have also sustained that the healthproblems attributed to this contamination have never been proven6.

Despite the fact that the Ecuadorian laws and the very Constitution of theRepublic guarantee “the right to live in an environment free ofcontamination,” with the State responsible for such protection, until nowthere has been little or no political action by the different governments to

4

follow the law and establish a system of environmental control over thepowerful oil industry7.

1. The people’s response to the situation

From the very beginning, residents of the oil production zones have voicedconcerns about the contamination. Farmers and Indians have reported thatmany local estuaries and rivers, once filled with fish, now lack aquatic life;reports of cattle dying from drinking such water are also frequent. Theresidents of these zones frequently complain that bathing in these waterscauses itchiness and development of rashes on their skin7,8.

The Indian communities and farmers of the Oriente have repeatedlycomplained to the different governments and companies about the situation,demanding better quality of life, attention to their basic needs, technicalassistance, and above all, cleaning-up of the contamination. But the responseof the governments and the oil companies has consisted of “patches”(covering up some of the pools of waste, constructing a school or latrine hereand there, opening a path) without confronting the real root of theproblem9,10.

However, in 1991 the publication of the book “Amazon Crude”2 by theNorth American Judith Kimerling elevated oil contamination in Ecuador tothe status of an international environmental problem. It was the first timethat clear evidence was presented to the media, the government, and the oilcompanies, that supported the claims of the communities. Kimerling showedhow oil development can cause a negative impact on the land and the peoplein each phase of its cycle, from the first seismic studies and drillings to thephases of production and transport.

In 1993, a health promotion association from Sucumbios conducted a studyon the effects of oil on health. The study included communities exposed tothe contamination and communities where there had been no oildevelopment. The study suggested that the exposed communities had moreillnesses, spontaneous abortions, and a higher mortality rate than thecommunities that were not contaminated3.

In this same year, a group of Amazon Indians and farmers representing30,000 affected individuals took legal action in New York against the oil

5

companies, accusing them of irreparable damage to the environment. Afterseveral attempts, the lawsuit has been accepted and in the next few months aUnited States judge will decide if the trial will take place in the U.S. or inEcuador10.

In 1994, the Center for Economic and Social Rights8 published a reportdocumenting the dangerous levels of oil contamination in the rivers of thenorthern region of the Ecuadorian Oriente. This same report documentsnumerous skin problems among the local people apparently related to the oilcontamination. Consequently, the Ecuadorian government has been accusedof violating human rights.

In 1994 the Amazon Defense Front (Frente de Defensa de la Amazonía[FDA]) was created with the participation of numerous organizations offarmers and Indians and local non-governmental organizations (NGOs) withthe objective of supervising the trial against Texaco. Since its formation, thegroup has organized various workshops about the environment, publishedaccusations and reports about oil spills, led community information sessionsand organized the visits of government representatives to the contaminatedareas, in an intent to open the eyes of the authorities to this disaster.

2. Evidence of the dangers of contamination

According to published information, the risk of adverse effects on health isgreatly increased when one is exposed to the contaminants created by oilextraction.

Studies in laboratory animals as well as wild animals show that exposure tooil can cause lesions in various organs, and can provoke birth defects, cancer,and even death.

Different studies on the health effects of oil on humans have demonstratedthat the exposed populations face an elevated risk of grave and irreversibleillnesses, presenting an important public health problem. The effects canpresent themselves with different intensity in each of the phases of the oildrilling process: exploration, drilling, production, and transport.

Appendix 1 presents an extensive review of published information on theimpact of oil on the health of animals as well as humans.

6

7

II. THE CASE OF SAN CARLOS

In October of 1998, the village of San Carlos was visited as a preliminarycontact for its inclusion in the study of the impact of oil drilling practiceson the health of the people of the Oriente. In the first conversations withthe head of nursing at the health center and some residents, the presence ofseveral cases of cancer was observed; they attributed this to elevated andcontinuous exposure to oil. Through his medical team, the Vicariate ofAguarico decided to carry out an investigation of this potential grouping ofcancers.

1. Area of study

San Carlos is a population located about 12 km. from Joya de los Sachas(heading toward Coca) following a road called vía La Parker. San Carlosbelongs to Sachas canton, in the province of Orellana. The town is made upof an urban center (70% of the population) surrounded by a rural zone(Precooperative Abdón Calderón).

San Carlos has approximately 1,000 inhabitants. The majority of thepopulation arrived in the 70’s following the routes of access opened by theoil companies. Their livelihood is based on agriculture and raising cattle.

The deficient infrastructure of San Carlos does not cover the basic needs ofthe people. There are no facilities for drinking water and no drainagesystems. Only the urban center has electricity, running water, and a healthcenter. The roads are covered with toxic waste.

At the entrance to San Carlos is the pumping station Sacha Sur and morethan 30 working wells have been identified in the area. Both the station andthe wells eliminate their toxic waste in the estuaries and rivers that cross thetown. These same water sources are habitually used by the people of thetown for drinking, cooking, bathing, and washing clothing. In Sacha Surthere are four powerful torches that burn gas constantly throughout the dayand night. Almost all of the oil wells in San Carlos have been functioningfor 20 years11.

8

2. How did we conduct the study?

The investigation of this group of cancers was divided into three parts:

• The discovery and confirmation of cases which occurred during the lastten years, 1989-1998;

• The calculation of the standard cancer rate in the area of study; thiscalculation permits us to detect if there is an excess incidence of cancerbeyond the expected rate.

• The investigation of possible contamination of water in the area.

2.1 Discovery and confirmation of cases

The first step in this study was to determine how many cases of cancer hadoccurred in San Carlos, where they had occurred, and when.

Thanks to the collaboration of the head of the local infirmary, a list wascreated with the names, ages, time of residence and place of diagnosis of allpossible cases of cancer in the past ten years, those who had died as well asthose who had survived. Then the registers of the hospitals where thepatients had been cared for were consulted in order to confirm the diagnosisof cancer.

From an initial list of 18 names, after confirming the definitive diagnosis, wewere left with ten cases of cancer (there is one more probable case, but thediagnosis has not yet been confirmed). The hospitals were: SOLCA (6patients), Eugenio Espejo Hospital(3), and Baca Ortiz hospital (1). For eachpatient, these institutions provided the age, place of residence, date ofdiagnosis, and type of cancer.

2.2 Calculation of the standard cancer rate.

Once the diagnoses were confirmed, the objective was to determine if thenumber of cancers found represented a normal occurrence or a true excess.To find out, the standard cancer rate (SCR) was calculated.

9

In order to calculate the SCR, it is necessary to know not only the number ofcancers in San Carlos, but also the following information:

ð The population of San Carlos; since there had not been an officialcensus of the population of San Carlos for the period 1989-1998, anestimate of the census was used based on the distribution of thepopulation of La Joya de los Sachas canton12 and on data collected inSan Carlos as part of the present study. (See Appendix 3). In order tosimplify the calculation, the annual census of San Carlos during theperiod 1989-1998 was assumed to be constant. As we stated earlier, thetotal population of San Carlos for 1998 was estimated at 1,000inhabitants (Table 1).

ð The rate of cancers in a reference population; in this case we decided thatour reference population would be Quito, a city with adequateinformation on cancer through the National Tumor Registry (RegistroNacional de Tumores RNT). The RNT provides information on theincidence of cancer in the population of Quito, divided by age and sex,for the period 1985-199613.

ð Finally, the SCR is calculated by dividing the number of observedcancers (O) in San Carlos during this 10-year period by the number ofexpected cancers (E) according to the rates of the reference population, inthis case Quito (O/E). The calculation of the SCR is explained inAppendix 2.

Table 1. Estimated age distribution of the population of San Carlos,1998.

Age Men Women Total0-4 94 77 1715-14 165 135 30015-44 233 191 42445-64 46 38 84>65 12 9 21

TOTAL 550 450 1000

2.3 Evaluation of the environment

10

In order to evaluate the extent to which the people of San Carlos wereexposed to contamination, samples of water were collected from the areasused by the community to supply drinking water, and water for cooking,bathing, and washing clothes. The analysis of the samples included totalhydrocarbons (HT) and was conducted in the water and soil laboratory ofthe P. Miguel Gamboa School, in Coca.

The method for measuring the HT required extraction with 1.1.2-trichloro-trifloro-ethane and determination by infra-red spectophotometry. The exactorigin of the water samples was hidden from the laboratory investigators inorder to avoid subjectivity in the interpretation of the results.

Because of the lack of economic and technical resources, an evaluation of thesoil and the air in these zones was not possible.

3. What were the results?

3.1 Discovery and confirmation of cases.

Ten patients with cancer were found in San Carlos during the period from1989-1998. Of these 10 cases, 3 suffered from stomach cancer; this numberrepresents almost 40% of the cancers which we consider those only found inmales. The rest of the cancers were of the larynx (1), liver (1), a melanoma (atype of skin cancer), bile duct (1), cervix (1), a lymphoma (a cancer of thelymphatic ganglions) and leukemia (a cancer of the white blood cells).Almost all of the cases of cancer found where diagnosed in males (80%). Thecharacteristics of the patients and the tumors are outlined in Table 2.

Sixty percent of the cancers were diagnosed in the last 3 years. The age inwhich the cancers were diagnosed ranged from 5 to 86 years. Of the 10patients diagnosed with cancer, 6 have died (another patient died in April ofthis year); the majority of these deaths took place a short time after diagnosiswhich could suggest either the aggressive nature of the cancer or the delay ofthe patients in seeking medical help. The patients’ time of residence in SanCarlos ranged from 7 to 30 years, with an average time of 17 years.

11

Table 2. Cases of cancer discovered in San Carlos, Orellana, 1989-1998.

ICDa SEXTYPE ofCANCER

DATE ofDIAGNOSIS

AGE atDIAGNOSIS

DATE ofDEATH

C24 M Bile ductb March 89 68 July 89C16 M Stomach June 91 64 July 92C16 M Stomach August 92 55 Sept. 92C16 M Stomach June 97 65 Oct. 98C32 M Larynx Sept. 97 46 -C22 M Liver August 98 86 Sept. 98172 M Melanoma Nov. 96 52 Aug. 97C42 M Leukemiac July 93 5 -C77 F Lymphoma

dJuly 96 28 April 99

C53 F Cervix May 98 52 -a ICD: International Classification of Diseases.b Others unspecified of the biliar systemc Acute lymphoblastic leukemiad Non-Hodgkins lymphoma

3.2 Determination of the standard rate of cancers.

The results of the calculations of the SCR, adjusted according to age, arepresented in Table 3. Considering all of the cancer cases, an excess of cancerwas found in the male population (8 observed compared to 3.5 expected).This suggests a risk in this population 2.3 times the expected rate, or anexcess risk of 130%. Depending on the type of cancer, the male populationof San Carlos presents a risk 30 times higher than expected for developingcancer of the larynx; 18 times higher for bile duct cancer; 15 times higher forliver cancer and melanoma, 4.6 times higher for stomach cancer, and 2.6times higher for leukemia. All of these cancers, except leukemia, are alsostatistically significant.c In women, after all cancers were considered, anoverall excess was not found. However, the risk of lymphoma is 6.7 timeshigher than expected and the risk of cervical cancer is 2.3 times higher.

c Statistically significant: indicates that there is higher than 95% probability that the obtained results are notdue to chance.

12

Table 3. Incidence of cancer in San Carlos, 1998 (O= number of observedcancers; E= number of expected cancers; SCR=standard cancer rate; CI95%= confidence interval of 95%).

CANCER MEN WOMENO E SCR CI 95% O E SCR CI 95%

Stomach 3 0.64 4.64 1.5-14.4 - - - -Larynx 1 0.03 29.9 4.2-212.5 - - - -Bile duct 1 0.05 18 2.5-127.9 - - - -Liver 1 0.06 14.5 2-102.7 - - - -Melanoma 1 0.06 15.2 2.1-107.7 - - - -Leukemia 1 0.37 2.65 0.3-18.8 - - - -Lymphoma - - - 1 0.14 6.7 0.9-47.5Cervix - - - 1 0.43 2.3 0.3-16.2TOTAL a 8 3.5 2.3 1.1-4.5 2 4 0.5 0.12-1.9a Calculations done with incidence rates of all cancers except skin cancer (C44)

Due to the fact that the majority of the men with cancer had died, using thesame method, the number of expected deaths in San Carlos due to cancerwas calculated and compared with the number observed deaths. The resultsof these calculations are presented in Table 4. When total cancers wereconsidered, the number of deaths observed in San Carlos was much higherthan the expected rate (6 observed, 1.6 expected). If we divide the results forthe different types of cancer, the risk of mortality from stomach cancer is 8times higher in San Carlos than in Quito, liver cancer mortality is 21 timeshigher, bile duct cancer mortality is 26.5 times higher, and melanomamortality is 70 times higher. As stated earlier, this data either suggests theaggressive nature of these tumors or reflects the difficulty posed to the peopleof San Carlos because of the lack of access to hospitals where they can obtainan early diagnosis.

Table 4. Incidence of cancer mortality in San Carlos, 1998 (O= numberof observed cancers; E= number of expected cancers; SCR= standardcancer rate; CI 95%= confidence interval of 95%).

CANCER MENO E SCR CI 95%

Stomach 3 0.36 8.21 2.65-25.46Liver 1 0.046 21.33 3-151.44Melanoma 1 0.0014 69.74 9.82-495.07Bile duct 6 0.037 26.57 3.74-188.66TOTALa 1.67 3.59 1.61-7.99a Calculations done with incidence rates of all cancers except skin cancer (C44)

13

3.3 Evaluation of the environment

The main river that runs through San Carlos is the Huamayacu river. Thepeople use the water of this river for bathing, washing clothing, drinking,and also fishing. In the areas surrounding San Carlos are the Basura river,the Parker river, and other small rivers which are also used by the people ofthe town. During the month of March 1999, samples were taken from theHuamayacu, Basura, and Parker rivers, and from the Iniap estuary. Thesamples were taken in the wintertime when there was no visible waste in therivers. The results are presented in Table 5.

In the Iniap estuaries, the concentration of hydrocarbons was more than 10times the level permitted by the regulations of the European Community, theconcentration was 53 times higher in the Parker river, 144 times higher inthe Huamayacu, and in the Basura the concentration of hydrocarbons was288 times the acceptable limit for drinking water.

In the 1994 study conducted by the CDES8, high levels of contaminationfrom aromatic polycyclic hydrocarbons were found in the rivers whichoriginate from the Sacha Central pumping station and run through SanCarlos.

These data further demonstrate the extent to which the residents of this townare exposed to oil contaminants greatly exceeding internationally recognizedsafety limits, and how this exposure persists throughout the years.

Table 5. Concentration of total hydrocarbons (TH)a in the rivers of SanCarlos, 1999.

IDENTIFICATION TOTAL HYDROCARBONSParker River 0.53Huamayacu River 1.444Basura River 2.888Iniap estuary 0.097a The limit of hydrocarbons permitted in drinking water according to regulations of theEuropean Community is 0.01 parts per million (ppm).

14

4. What is the significance of these results?

This study has shown that the population of San Carlos is subjected to amuch higher risk of developing cancer than should be expected based on thecharacteristics of the population.

The risk in males is particularly high for cancer of the larynx, bile ducts,liver, stomach, melanoma, and leukemia. It is also important to point outthe high risk of dying from cancer in this population, especially from cancerof the stomach, liver, bile duct, and melanoma.

In women, an excess risk of cervical cancer and lymphoma has beenobserved.

It is suspected that the excess risk of cancer is due to continuous andpersistent contamination of the environment by toxins from oil developmentin the area. This potential association between the occurrence of cancer inSan Carlos and the exposure to chemicals from oil wells is supported by thefollowing facts:

1. The high rate of cancer found.The risk of developing certain types of cancer in this population is highenough to suspect the presence of some factor which is contributing to theabnormal elevation.

2. Length of residenceThe extended time of residence of patients in the area of study implies apossible environmental carcinogenicd because of the long latency periodrequired by most known carcinogens. This fact is supported by the findingthat more than half of the cancers were diagnosed in the last 3 years.

3. Length of exposure The association between incidence of cancer and oil contamination issupported by the long history of exposure to oil toxins which the people ofSan Carlos have suffered. There have been many complaints from residentsabout oil spills during these twenty years of oil development8,11.

4. The carcinogenic effects of the chemicals.

d Carcinogenic: any biological element capable of triggering the process of cancer formation.

15

It is well known that crude oil and toxic waste from oil stations and wells arehighly carcinogenic8. A study conducted on workers in oil fields14 showed anexcess of leukemia, and numerous studies conducted on residents nearpetrochemistry industries have demonstrated an excess in the rate of cancerincidence and cancer mortality15,16,17. However, no study to date hadinvestigated the relationship between cancer incidence and the residence inareas of oil exploitation.

5. Absence of other risk factors.The affected population lacks one of the most common risk factors forcancer: tobacco. As this is a rural population, it also maintains a healthydiet based mainly on the consumption of rice, yuca, plátano, meat (poultry,beef, pork) and occasionally fish, which lowers the risk of cancer. Inaddition, the population of Quito has also been subjected to the risk ofcancer from urban contamination, which would further increase thedifference between both populations if we were to eliminate this risk factor.

Furthermore, there are no other industries in the area, aside from the oilindustry, which could release cancer-causing toxins.

6. The types of cancers found.The majority of cancers found can be explained by the type of exposure thatthe population has been subjected to. The routes of exposure to toxinsreleased by oil activity mainly include the water and the air. Exposure byway of the air includes migration of the gases from the burning of petroleumand crude oil in wells and stations as well as emission from pools of wasteand oil roads. The water route includes contamination of surface water andof underground wells, which causes the contamination of water useddomestically by the people. The contamination of the air, water, or landcould also affect the consumption of agricultural products and meat.

Some of the possible limitations of this study include:1. The inexactitude of the census of San Carlos, which could cause an

imprecise calculation of the expected cancer rate. However, the estimatedcensus of San Carlos was high in order to avoid this type of problem.

2. The reduced number of cases of this grouping of cancers does not permitstatistically more convincing analysis that could put in evidence othertypes of significant differences, which is reflected in the high confidenceintervals

16

3. The possibility of having a greater number of cancer cases in San Carloswhich have not been diagnosed because access to adequate medical centersfor the residents is difficult.

4. The migration of the population could cause a different number ofobserved cases since individuals with cancer may have moved to otherregions. However, if this had occurred, the number of cancers in the SanCarlos would be even higher.

5. As with other studies on health and the environment, there is theproblem of socioeconomic factors which vary between the areascompared, altering the association between cancer and variable exposure.However, it was not possible to obtain adequate socioeconomic data onthe two populations.

5. Some final recommendations

1. More in-depth studies of cohorts and/or case controls are necessary toconfirm the association and obtain more precise information on the riskfactors which could be producing the high rate of cancers found in thispopulation.

2. A more extensive study in the area in order to adequately detect all of thepossible sources of contamination, eliminate them and establish anadequate monitoring system in the area. As long as the oilcontamination persists, the health of this and similar populations willcontinue to be seriously threatened.

3. Due to the grave nature of cancer, it would be useful to establish anepidemiological monitoring system for the illness which would permit abetter understanding of its distribution and its risk factors with the goalof establishing adequate prevention programs.

17

APPENDIX 1

What is known about the impact of oil on health?

There are two principal sources of information that can be used to evaluatethe health risks of any population exposed to chemicals. The first are studiesof toxicity using laboratory animals and the second, studies in humanpopulations.18

Numerous studies have established that the exposure of animals and humansto oil or its components can result in acute health effects (such as skinproblems) or even deathly diseases (such as cancer)8. In this appendix wepresent a bibliographic review of the effects of oil contamination on animalsand humans.

1. Studies on animals

1.1 Laboratory animals

Toxic effects. Studies in birds have shown that ingestion of oil reduces redblood cells as well as white blood cells in primary19 lymphatic organs. Theadministration of crude oil has also resulted in functional changes in hepaticcells of rats20 and inhibition of testicular development in salmon21. Theeffects of crude oil produced from the burning of petroleum in Kuwaitresulted in diminished survival and growth of the marine fish Menidiaberyllina.22

Cancer. Numerous studies report skin tumors in rats following theapplication of crude oil23-24.

Effects on reproduction. The oral administration of crude oil to pregnant ratsreduced the weight and length of the fetus and multiple exposureconsiderably diminished the mother’s weight. Several studies have alsodemonstrated the pronounced effect of crude oil on the reproductivecapacity of birds after its application to the surface of the egg or after oraladministration27-30.

1.2 Wild animals

18

Fish. Different studies conducted in marine areas have shown the presence ofcrude oil in different species of fish following oil spills31,32. The implicationof these contaminants for the ecosystems and for the population dependenton them is still not clear.

Birds. There is sufficient evidence that oil spills are responsible for massivedeaths of marine birds33,34 and the occurrence of hemolytic anemia has beenobserved in ducks after the ingestion of crude oil35.

Mammals. Significant differences in weight and in blood hemoglobin levelsoccurred in otters inhabiting the contaminated and non-contaminated areasof Prince William Sound (Alaska) after the ExxonValdez oil spill in 1989.An increase in the mortality of seals has also been reported following the oilspills in the North Sea36.

In areas of oil production, the proximity of cattle to the drilling andproduction sites often results in poisoning of the animals from ingestion ofcrude oil, salt water, heavy metals, and caustic chemicals. The most commoncause of illness or death following the exposure to oil particles is pneumoniafrom breathing it in, which can cause a chronic deterioration of health, withdeath occurring after several days or weeks37,38.

2. Studies on humans

Oil production can negatively infringe upon the health of the population inevery phase of its cycle. In this section, due to its implications for the areaof study, its impact on health is described in relation to the initial seismicstudies, drilling, production, and finally transport.

In addition, oil exposure is not limited to the areas close to thecontamination. When oil contaminates the environment, the heaviestparticles tend to be deposited in the sediment from where they can repeatedlycontaminate water sources or be consumed by organisms that can enter thehuman food chain. Lighter oil components can evaporate within hours andbe deposited by air or water at great distances from where they wereproduced8.

Oil or its components can enter into contact with the human body in threeways:• absorption through the skin

19

• ingestion of food and water• inhalation of oil and its gases

The inhabitants of oil production areas of the Oriente are faced withpotential exposure by any of these three routes.

2.1 Exploration phase

We have not found literature about the health effects during this phase.However, in the context of Ecuador, exploration workers have reported lowsalaries and poor conditions of life and work; illnesses related to the work aswell as skin diseases and gastrointestinal diseases are frequent39.

Yellow fever was contracted by many workers (one died) due to penetrationinto the jungle without adequate protection with the vaccine (personalobservation).

Many workers are Indians and their entrance into this world of migratorywork has severed traditional family life and sparked epidemics of influenza,malaria, hepatitis, and venereal diseases in their communities (Amunárriz,personal communication).

2.2 Drilling / Production phase

Activities associated with this phase produce a great variety of contaminantsin the soil, the water, and the air. The communities that live close to the oilwells have a greater probability of suffering exposure to chemicals and toxinswhen they breathe, use water for drinking, bathing, or cooking, and eat foodthat has been in contact with the toxins.

The contaminants of crude oil can be deposited in the earth or ingestedthrough aquatic organisms in quantities that can have adverse effects onhealth and increase rates of malnutrition, especially in children andfishermen, when contaminated fish or their products enter the food chain40.

Sathiakumar et al.14 conducted an epidemiological study on oil and gas fieldworkers and found an association between their work and acute myeloidleukemia.

20

The drilling and production phases also carry the risk of accidents41.Furthermore, the oil industry exposes workers to high levels of noise fromdrills, compressors, generators, etc.42

Currently, three groups of chemical exposure merit a more detailedexplanation.

2.2.1. Crude oil

Crude oil has been defined as a complex mixture, composed primarily ofparaphenol, aromatic hydrocarbons, and particles of other elementsincluding various metals. Among the identified components of crude oil,the aromatic hydrocarbons of toxicological interest are alka-benzols (mainlytoluene and xylene) and the polynuclear aromatic hydrocarbons43.

The effects of acute exposure to crude oil on humans are usually transientand of short duration unless the concentration of the components isunusually high. Such exposures irritate the skin, cause itchiness or irritationof the eyes from accidental contact or exposure to the vapors, and can causenausea, vertigo, headache, or dizziness from prolonged or repeated exposureto low concentrations of its volatile components44. Inhalation of mineraloils can cause lipoid pneumonia and death45.

Of particular concern is exposure to benzene, toluene and xylene. Highconcentrations of benzene cause neurotoxic symptoms, and a prolongedexposure to toxic levels can cause lesions of the bone marrow46. The primaryeffects of toluene and xylene are on the central nervous system. There is notsufficient data on the incidence of cancer following human exposure to thesetoxins. There is also no evidence of birth defects due to benzene, toluene, orxylene.44

Due to its elevated carcinogenic effects and long persistance in theenvironment, benzopyrene seems to be the most important risk factor andmost relevant indicator of crude oil contamination. Adequate data on whichto base a quantitative evaluation on the risk of cancer following the ingestionof HAP are only available for BaP47,48.

There is little information available about the oral toxicity of the HAPs,especially after exposure of long duration.

21

2.2.2. Other chemical exposure

The oil companies in Ecuador have not informed the public of the chemicaldata related to waste from oil drilling. Data from the U.S. show that thewaste from drilling can contain significant quantities of a wide variety ofcontaminants such as antimony, arsenic, cadmium, chrome, lead,magnesium, zinc, benzene and other hydrocarbons as well as toxic levels ofsodium and chloride2.

The heavy metals which pose a threat to health from oil exposure throughdrinking water are mercury and cadmium. Exposure to mercury occursthrough contaminated surface water or underground water or from fishconsumption49. Similarly, the components of alka-mercury are liposolubleand volatile; therefore they carry the risk of being absorbed by the skin orinhaled when bathing in contaminated waters50. The most commonsymptoms from mercury poisoning are disordered thinking, difficultywalking and talking, tunnel vision, and difficulty chewing and swallowing51.The toxicological implications of exposure to low concentrations of mercuryare still not well understood.

Cadmium accumulates throughout one’s life. Environmental exposure tocadmium can occur, as in the case of mercury, through water or foodconsumption. Individuals in Japan who consumed rice contaminated withcadmium developed chronic cadmium poisoning and a reduced lifeexpectancy52. Elevated consumption of cadmium produces nausea, vomiting,abdominal pain, diarrhea, and renal disease. Recent studies have suggestedan increase in lung cancer mortality in workers exposed to cadmium54.Other studies have also indicated an association between the level ofcadmium in drinking water and prostate cancer.55

2.2.3. Air contamination

Air contamination with chemical or solid agents is one of the consequencesof drilling oil. The degree to which the contaminants emitted into the aircontain a risk for the general population depends on various factors; some ofthese include the type and quantity of compounds released, the condition ofthe atmosphere, the number of people exposed and their susceptibility.

The burning of oil and gas contaminates the air with oxides of nitrogen,sulfur, and carbon (CO), as well as with heavy metals , hydrocarbons and

22

diverse particles of carbon. Many of these emissions are toxic, and theoxides of nitrogen can react with solar light to form ozone, an irritant to thehuman respiratory tract.

Although quantitative data on air emissions from oil operations are notavailable in Ecuador – at least to the public- an internal study byPetroecuador found that the levels of contaminated air were alarming2.

Numerous studies have demonstrated the relationship between exposure tothese contaminants and bronco-constrictive effects56, documenting asthmaticresponses in adults as well as in children57. Other studies have alsoassociated an increase in mortality with air contamination58,59.

Investigations of the exposure to ozone and nitrogen dioxide have revealedsymptoms and changes in respiratory mechanisms, bronco-constriction andpulmonary edema60. The clinical effects of acute carbon poisoning vary withthe level of intoxication, from non-specific symptoms (headache, dizziness,fatigue) to death61. Chronic exposure to carbon, at low doses, can affect thecoagulation system increasing the risk of thromboembolism in the heart orbrain62.

Along with these contaminants, the populations exposed to HAP and volatileorganic compounds in the air are at risk for lung cancer63 and adverserespiratory effects64 respectively.

2.3 Transport

The accumulation of water during the construction of roads is inevitable.Mosquitoes carrying malaria proliferate in these surface waters, increasing thespread of the disease. In 1974, more the 50% of the malaria of the BrazilianAmazon was associated with the narrow area of influence of the Trans-Amazon highway65.

Many of the roads of the Oriente are without asphalt and great quantities ofdust from traffic can cause respiratory diseases66. Where the roads arecovered with crude oil, there is often an intense odor, and skin problemsoccur when people walk barefoot.

There also seems to be a strong association between this development and themortality related to vehicles67. On the roads of the Oriente, car accidents are

23

numerous, and are becoming the second highest cause of mortality in smallcities68.

2.4 Oil spills

Epidemiological investigations of the great oil spills have not been frequentlycarried out and have concentrated more on workers involved in the clean-upoperations than on the residents of the areas. These studies have also onlybeen carried out long after the incidents occurred69.

One study conducted a year after the 1989 Exxon Valdez oil spill off thecoast of Alaska suggested an increase in anxiety attacks, post-traumatic stressdisorder, and depressive symptoms in the affected communities after thespill. The populations most at risk for these conditions were women andIndians70.

Another study evaluating the immediate and posterior effects on health afterthe oil spill of the Braer tanker off the coast of Shetland (United Kingdom)was conducted in 1993; anecdotal reports of acute symptoms withoutsignificant differences were found in the exposed population during theinitial phase of study and no differences in the health of the exposedpopulation and the non-contaminated population were found 6 monthslater44,71.

24

APPENDIX 2

Calculation of the standard rate of cancer incidence by age and sex

The standard cancer rate (SCR) is calculated by dividing the number ofobserved cancer cases in the study population by the number of expectedcancers in the same population. To determine the number of expectedcancers, the number of person/years corresponding to the age distribution ofthe study population is multiplied by the cancer rate of the referencepopulation corresponding to the same age distribution.

The age-adjusted calculation of the SCR is illustrated below for tumors of thelarynx.

AGE GROUPOBSERVEDCANCER

STANDARDRATE (100,000)

POPULATION/YEARS

EXPECTEDCANCER

0-4 - 0 940 05-9 - 0 900 0

10-14 - 0 750 015-19 - 0 580 020-24 - 0 470 025-29 - 0 400 030-34 - 0 330 035-39 - 0.4 320 0.0012840-44 - 1.1 230 0.0025345-49 1 0.8 160 0.0012850-54 - 1.5 140 0.0021055-59 - 5.1 90 0.0045960-64 - 6.9 70 0.0048365-69 - 4.2 40 0.0016870-74 - 12.9 30 0.00387>75 - 22.5 50 0.01125

1 0.03341SCR = Observed/Expected = 1/0.03341 = 29.93

25

APPENDIX 3

Estimated distribution of the populationa of San Carlos by age groups.

MALE POPULATION: 550FEMALE POPULATION: 450

AGE GROUP % MALES FEMALES0-4 17 94 775-9 16.4 90 74

10-14 13.5 75 6115-19 10.4 58 4720-24 8.6 47 3925-29 7.2 40 3330-34 5.9 33 2735-39 5.7 32 2640-44 4.2 23 1945-49 2.9 16 1350-54 2.6 14 1255-59 1.6 9 760-64 1.2 7 665-69 0.76 4 370-74 0.58 3 2>75 0.83 5 4

TOTAL 550 450

a The percentage of the population is based on estimates of the National Institute ofStatistics and Census for the rural population of Joya de los Sachas for the year 1998

26

BIBLIOGRAPHY

1. Jochnick C, Normand R, Zaidi S. Rights violations in the EcuadorianAmazon: the human consequences of oil development. Health andHuman Rights 1994; 1(1):82-100.

2. Kimerling J. Amazon crude. Natual Resources Defense Council (ed).Henriksen SS, New York 1991.

3. Unión de Promotores Populares de Salud de la Amazonía Ecuatoriana(UPPSAE). Cultuas bañadas en petróleo: diagnóstico de salud realizadopor promotores. Abya-Yala, Quito 1993.

4. Garcés A. Los colonos de la reserva faunística Cuyabeno: conflictos por eluso de tierras. En Marea negra en la Amazonía: conflictos socioambientalesvinculados a la actividad petrolera en el Ecuador. Abya-Yala, Quito 1995: pp:367-395.

5. Ortiz P, Varea, A. La explotación petrolera en el Ecuador: historia eimpactos socioambientales. En Marea negra en la Amazonía: conflictossocioambientales vinculados a la actividad petrolera en el Ecuador. Abya-Yala,Quito 1995: pp: 71-88.

6. Miami Herald. Eduadorians blame Texaco for leftover oil pollution;February 3, 1998.

7. Kimerling J. Rights, responsabilities, and realities: environmentalprotection law in Ecuador’s Amazon oil fields. Southwestern Journal ofLaw and Trade in the Americas 1995, volume II; number 2; 293-384.

8. Centro de Derechos Económicos y Sociales (CDES). Violaciones dederechos en la Amazonía Ecuatoriana. Hombre y Ambiente 30. Abya-Yala, Quito 1994.

9. Varea A. Falta de transparencia en el manejo de información sobre laauditoría ambiental a la compañia Texaco. En Marea negra en laAmazonía: conflictos socioambientales vinculados a la actividad petrolera en elEcuador. Abya-Yala, Quito 1995: pp: 295-338.

27

10. Frente de Defensa de la Amazonía (FDA). La Texacontaminación en elEcuador, Febrero de 1999. Lago Agrio, Ecuador 1999.

11. Garzón P. Impacto socioambiental de la actividad petrolera: estudio decaso de las comunidades San Carlos y La Prima vera. En Marea negra enla Amazonía: conflictos socioambientales vinculados a la actividad petrolera en elEcuador. Abya-Yala, Quito 1995: pp: 265-294.

12. Instituto Nacional de Estadísticas y Censos (INEC). Cifrando ydescifrando Napo. INEC. 1995.

13. Sociedad de Lucha contra el Cáncer (SOLCA). Cáncer en Quito; anuario1995-1996. Registro Nacional de Tumores, SOLCA, Quito 1998.

14. Sathiakumar N, Delzell E, Cole P, Brill I, Frisch J, Spivey G. A case-control study. American Hournal of Industrial Medicine 1987; 11(6): 615-625.

15. Pan BJ, Hong YJ, Chang GC, Wang MT, Cinkotai FF, Ko YC. Excesscancer mortality among children and adolescents in residential districtspolluted by petrochemical manufacturing plants in Taiwan. Journal ofToxicology and Environmental Health 1994; 43(1): 117-129.

16. Yang CY, Chiu HF, Chiu JF, Kao WY, Tsai SS, Lan SJ. Cancer mortalityand residence near petrochemical industries in Taiwan. Journal ofToxicology and Environmental Halth 1997; 50(3): 265-273.

17. Griffith J, Aldrich T, Drane W. Risk assessment. In Environmentalepidemiology, chapter 10, pp: 212-239.

18. Leighton FA. Clinical, gross, and histological findings in herring gullsand Atlantic puffins that infested Prudhoe Bay crude oil. VetrinaryPathology 1986; 23 (3): 254-263.

19. Khan S, Payne F, Rahimtula AD. Mechanisms of petroleum hydrocarbontoxicity: functional changes in rat liver mitochondria after exposure to aPrudhoe Bay Crude Oil. Toxicology Letters 1986; 32 (1-2): 141-146.

20. Truscott B. Walsh JM, Burton MP, Payne JF, Idler DR. Effect of acuteexposure to crude petroleum on some reproductive hormones in salmon

28

and flounder. Comparative Biochemistry and Physiology. Part C 1983; 75(1): 121-130.

21. Al Yakoob SN, Gundersen D, Curtis L. Effects of the water-solublefraction of partially combusted crude oil from Kuwait’s oil fires (fromDesert Storm) on survival and fires (from Desert Storm) on survival andgrowth of the marine fish Menidia berllina. Ecotoxicology andEnvironmental Safety 1996; 35 (2): 142-149.

22. Clark CR, Walter MK, Ferguson PW, Katchen M. Comparative dermalcarcinogenesis of shale and petroleum-derived distillates. Toxicology andIndustrial Health 1988; 4: 11-22.

23. Wilson JS, Holland LM. Periodic response difference in mouse epidermischronically exposed to crude-oils or BaP: males vs. females. Toxicology1988; 50 (1): 83-94.

24. Schreiner CA. Petroleum and petroleum products: a brief review ofstudies to evaluate reproductive effects. In Advances in modernenvironmental toxicology, vol. III, Assessment of reproductive andteratogenic hazards.

25. Christian MS, Galbraith WM, Voytek P, Mehlman MA (eds). PrincetonScientific Publishers, Princeton NJ 1984; pp: 29-45.

26. Khan S, Irfan M, Rahimtula AD. The hepatoxic potential of a PrudhoeBay crude oil; effet on mouse liver weigh and composition. Toxicology1987; 46:95-105.

27. Hoffman DJ. Embryotoxic and teratogenic effects of petroleumhydrocarbons in mallards (Anas platyrhnchos). Hournal of Toxicologyand Environmental Health 1979; 5: 835-844.

28. Lee Y, O’Brien PJ, Payne JF, Rahimtula AD. Toxicity of petroleum crudeoils and their effect on xenobiotic metabolizing enzyme activities in thechicken embryo in ovo. Environmental Research 1986; 39: 153-163.

29. Walters P, Kahn S, O’Brien PJ, Payne JF, Rahimtula AD. Effectiveness ofa Prudhoe Bay crude oil and its aliphatic, aromatic and heterocyclicfractions in inducing mortality and aryl hydrocarbon hydroxylase inchick embryo in ovo. Archives of Toxicology 1987; 60: 454-459.

29

30. Feuston MH, Mackerer CR, Schreiner CA, Hamilton CE. Systemictoxicity of dermally applied crude oils in rats. Journal of Toxicology andEnvironmental Health 1997; 51(4): 387-399.

31. Lockhart WL, Wagemann R, Tracey B, Sutherland D, Thomas DJ.Presence and implications of chemical contaminants in the freshwaters ofthe Canadian Arctic. Science and Total Environment 1992; 122 (1-2): 165-245.

32. Hellou J, Upshall C, payne JF, Hodson PV. Polycyclic aromaticcompounds in cod (Gadus morhua) from the Northwest Atlantic and St.Lawrence Estuar. Science and Total Environment 1994; 145 (1-2): 71-79.

33. Rebar AH, Lipscomb TP, Harris RK, Ballachey BE. Clinical and clinicallaboratory correlates en sea otters dying unexpectedly in rehabilitationcenters following the Exxon Valdez oil spill. Veterinary Pathology 1995;32 (4): 346-350.

34. Briggs KT, Yoshida SH, Gershwin ME. The influence of petrochemicalsand stress on the immune system of seabirds. Regulatory

35. Yamato O, Goto I, Maede I. Haemolytic anaemia in wild seaducks causedby marine oil pollution. Journal of Wildlife Diseases 1996; 32 (2): 381-384.

36. Jenssen BM. An overview of exposure to, and effects of, petroleum oil andorganochlorine pollution in grey seals (Halichoerus grypus). Science andTotal Environment 1996; 186 (1-2): 109-118.

37. Edwards WC. Toxicology of oil field wastes. Hazards to livestockassociated with the petroleum industry. Veterinary and Clinical NorthAmerican Food and Animal Practice 1989; 5 (2): 363-374.

38. Coppock RW, Mostrom MS, Stair EL, Semalulu SS. Toxicopathology ofoilfield poisoning in cattle: a review. Veterinary and Human Toxicology1996; 38 (1): 36-42.

39. Hoy. Quito, 26 de marzo, 1993.

30

40. Tchounwou PB, AbdelghaniAA, Pramar YV, Heyer LR, Steward CM.Assessment of ptiential health risks associated with ingesting heavy metalsin fish collected frm a hazardous-waste contaminated wetland inLouisiana, USA. Review of Environmental Health 1996; 11 (4): 191-203.

41. McNabb SJ, Ratard RC, Horan JM, Farley TA. Injusries to internationalpetroleum drilling worders, 1988 to 1990. Journal of OccupationalMedicine 1994; 36 (6) 627-630.

42. World Health Organization (WHO) Cmmission on Health andEnvironment. Our planet, our health. World Health Oganization, Geneva1992.

43. IARC. IARC monographs on the evaluation of the carcinogenic risk ofchemicals to man: occupational exposures to petroleum refining; crude oiland major petroleum fuels. Volume 45. IARC, Lyon, 1989.

44. Campbell D, Cox D, Crum J, Foster K, Christie P. Initial effects of thegrounding of the tander Braer on health in Shetland. British MedicalJournal 1993, 307: 1251-1255.

45. Rodriguez MA, Martínez MC, Martínez-Ruiz D, Paz-Giménez M,Menéndez M, Repetto M. Death following crude oil aspiration. Journal ofForensic Science 1991; 36 (4): 1240-1245.

46. McMichael AJ. Carcinogenicity of benzene, toluene and xylene:epidemiological and experimental evidence. IARC Scientific Pbulications1988; 85: 3-18.

47. World Health Organization (WHO). Guidelines for drinking-waterquality. Volume 1, 2nd edition. World Health Organization, Geneva 1984.

48. Attias L, Bucci AR, Maranghi G, Holt S, Marcello I, Zapponi GA. Crudeoil spill in sea water: an assessment of the risk for bathers correlated tobenzo(a)pyrene exposure. Centro European Journal of Public Health1995; 3: 142-145.

49. Boischio AA, Henshel DS. Risk assessment of mercury exposure throughfish consumption by the riverside people in the Madeira Basin, Amazon,1991. Neurotoxicology 1996; 17 (1): 169-175.

31

50. Hu H, Kim NK. Drinking water pollution and human health. In Criticalcondition: human health and the environment. Chivian E, McCally M, Hu H,Haines A (eds). The Massachusetts Institute of Technology, Massachusetts1993.

51. Moreira JC. Threats by heavy metals: human and environmentalcontamination in Brazil. Science and the Total Environment 1996; 188(Suppl 1):S61-71.

52. Nakagawa H, Tabata M, Morikawa Y Senma M, Kitagawa Y, Kawano S,Kido T. High mortality and shortened life-span in patients with itai-itaidisease and subjects with suspected disease. Archives of EnvironmentalHalth 1990; 45 (5): 283-287.

53. Kido T, Nogawa K, Honda R, Tsuritani I, Ishizaki M, Yamada Y,Nakagawa H. The association between renal dysfunction and osteopeniain environmental cadmium-exposed subjects. Environmental Research1990; 51 (1): 71-82.

54. Sorahan T, Lancashire RJ. Lung cancer mortality in a cohort of workersemployed at a cadmium recovery plant in the United States: an analysiswith detailed job histories. Occupational and Environmental Medicine1997; 54 (3): 194-201.

55. Waalkes MP, Rehm S. Cadmium and porstate cancer. Hournal ofToxicology and Environmental Health 1994; 43 (3): 251-56.

56. Linn WS, Avol EL, Peng RC, Shamoo DA, Hackney JD. Replicated dose-response study of sulfur dioxide effects in normal, atopic, and asthmaticvolunteers. American Review of Respiratory Diseases 1987; 136 (5): 1127-1134.

57. Bates DV, Sizto R. The Ontario air pollution study: identification of thecausative agent. Environmental Health Perspectives 1989; 79: 69-72.

58. Salinas M, Vega J. The effect of outdoor air pollution on mortality risk:an ecological study from Santiago, Chile. World Health StatisticsQuarterly 1995; 48 (2): 118-125.

32

59. Pereira LA, Loomis D. Conceicao GM, Braga ALF, Arcas RM, Kishi HS,Singer JM, Bohm GM, Saldiva PH. Association between air pollution andintrauterine mortality in Sao Paulo, Brazil. Environmental HealthPerspecives 1998; 106 (6): 325-329.

60. Hltzman MJ, Cunningham JH, Sheller JR, Irsigler GB, Nadel JA, BousheyHA. Effect of ozone on bronchial reactivity inatropic and nonatropicsubjets. American Review of Respiratory Diseases 1979; 120 (5): 1059-1067.

61. Winter PM, Miller JN. Carbon monoxide poisoning. Journal of theAmerican Medical Association 1976; 236: 1502-1504.

62. Haft JI. Role of blood platelets in coronary artery disease. AmerianJournal of Cardiology 1979; 43: 1197-1206.

63. Mastrangelo G. Fadda E, Marzia V. Polycyclic aromatic hydrocarbons andcancer in man. Environmental Health Perspecives 1996; 104 (11): 1166-1170.

64. Ware JH, Spengler JD, Neas LM, Samet JM, Wagner GR, Coultas D,Ozkaynak K, Schwab M. Respiratory and irritant health effects ofambient volatile organic compounds. The Kanawha County Health Study.American Journal of Epidemiology 1993; 137 (12): 1287-1301.

65. Coimbra CEA. Human factors in the epidemiology of malaria in theBrazilian Amazon. Human Organization 1988; 47 (3): 254-260.

66. Howze EN, Hughes JM. Elevated respirable dust levels along unpavedcoal-haul roads: a rural public health problem. American Review ofRespiratory Diseases 1990; 141 (4): A77.

67. Ezenwa AO. Studies of risks associated with technological development inNigeria. Journal of the Royal Society of Health 1996; 116 (6): 376-380.

68. UNICEF. Situación de las madres y los niños en zonas de grandesproyectos. UNICEF, Quito 1992.

69. Gorman RW, Berardinelli SP, Bender TR. Thealth hazard evaluationreport HETA 89-200 and 89-273-2111, Exxon Valdez Alaska oil spill.

33

Cincinnati Hazard Evaluation and Technical Assistance Branch, NIOSH,US Department of Health and Human Services, 1991.

70. Palinkas LA, Petterson JS, Russell J, Downs MA. Community patterns ofpsychiatric disorders after the Exxon Valdez oil spill. American Journal ofPsychiatry 1993; 150 (10): 1517-1523.

71. Campbell D, Cox D, Crum J, Foster K, Riley A. Later effects ofgrounding of tanker Braer on health in Shetland. British Medical Journal1994, 309: 773-774.