Factors Associated with Lead Exposure in Oaxaca, Mexico

www.nature.com/jea

Factors associated with lead exposure in Oaxaca, Mexico

HERNA´ NDEZ-SERRATO MA. ISIDRA, MENDOZA-ALVARADO LAURA R, ROJAS-MARTI´ NEZ ROSALBA, GONZA´ LEZ-GARZA CARLOS, JENNIFER MEGHAN HULME AND OLAIZ-FERNA´ NDEZ GUSTAVO

Instituto Nacional de Salud Pu´blica, Cuernavaca, Me´xico

Lead intoxication risks were studied in a community of ceramic folk art workers in Oaxaca, a southern state of Mexico, where the manufacture of low- temperature lead ceramic ware is a family tradition and often the only source of income. Variables such as household characteristics, occupation, andlead exposure risk factors were explored. Study participants’ mean blood lead concentration was 43.8 mg/dl (range ¼ 8.4–99.6 mg/dl), which is over the WHO guideline of 40 mg/dl for removing workers from exposure and is the concentration over which renal damage is accelerated. Best predictors for high blood lead concentrations by multivariable regression analysis were: occupation (Po0.0001), gender (P ¼ 0.0002), and the use of glazed stoneware (Po0.0001). This model explained 18% of blood lead variation among the study group. Exposure appears to be primarily associated with antiquated pottery manufacturing techniques and the high degree of contamination prevailing at the production sites, which in most cases are their living quarters. This consequently affects the lead levels of the entire community.

Journal of Exposure Analysis and Environmental Epidemiology (2003) 13, 341–347. doi:10.1038/sj.jea.7500282

Keywords: blood lead, ceramic glaze, pottery

Introduction Mexico over the last two decades after the enforcement of stricter regulations for gasoline and other products (Romieu Exposure to lead can be deleterious to most body systems, et al., 1994). absorbed primarily through respiratory and digestive routes There has long been documentation in Mexico of human and interfering with cellular function and metabolism. The exposure to this pollutant before the time of intensive relationship between exposure and blood lead levels con- industry and gasoline emissions, which has since been traced stitutes a dynamic process in which the blood lead to lead-glazed ceramics, now considered to be the major concentrations are the product of recent exposure, excretion, contributor to increased lead body burdens in urban and its balance in tissues. Elevated body lead concentrations populations (Herna´ ndez-Avila et al., 1991). Lead-glazed produce harmful effects on hematopoietic, hepatic, renal, pottery is used in abundance here for food preparation and reproductive, and gastrointestinal systems (Silbergeld, 1990; storage,andwhilethereisnoofficialinformationonthe Landrigan 1991, 1990). number of users, several studies have indicated that 30% of The sources for lead exposure are generally determined by the urban population in Mexico uses glazed pottery for atmospheric factors, common practices, personal habits, and cooking and/or storage of foods (Herna´ ndez-Avila et al., occupational hazards. Lead is typically released into the 1991; Romieu et al., 1994). Studies show that blood lead environment from lead smelters, emissions from diverse concentrations in families who use these utensils are between manufacturing plants, such as battery production, paint, and 30% and 40% higher than those of families with similar the combustion of leaded gasoline (Coyer and Galvao Luiz, profiles who do not. As lead is more soluble at low pH, acidic 1989; Criterios de salud ambiental 3 Plomo, 1979). For this Mexican staples such as chili, tomatoes, and lemon juice are reason, the highest environmental lead concentrations are much more likely to be contaminated. Exposure is likely most often in highly populated areas. (Flores et al., 1989; much greater among rural populations who, for its afford- Romieu et al., 1992). These lead sources have declined in ability, also serve meals and beverages in glazed pottery (Flores et al., 1989; Magdalena Rojas-Lo´ pez et al., 1994). More important is the heavy occupational exposure asso- 1. Address all correspondence to: Dr. R Rojas-Martı´ nez, Instituto ciated with the production of these ceramics, which was Nacional de Salud Pu´ blica, Calle 7a cda de Fray Pedro de Gante, #50 initially revealed in the late 1960s in a series of studies of Col. Seccio´ n 16, Delg. Tlalpan C.P 14080, Me´ xico, D.F. Tel.: +52-55-56-55-08-45. ceramic folk art workers in small communities in Mexico E-mail: [email protected] (Montoya Cabrera et al., 1978; Garcia Leal et al., 1979). Received 10 October 2001; accepted 18 April 2003 Five million Mexicans make ceramic folk art, and workers in Isidra et al. Factors associated with lead exposure various areas of the country continue to be dependant on sources of lead exposure and its association with blood lead lead-based glazes in their preparation. Workers often live in concentrations in adult community members in Santa Maria small settlements where sanitary services are minimal and Atzompa, allowing for later follow-up work on its physio- both the ceramic work and furnace are located in the home. logical manifestations. Women and children are also at risk as they most often paint, finish, and glaze the pottery with lead oxides, or ‘‘greta’’ (containing 80% lead oxide and 20% water), to produce the Materials and methods desired glazed finish. The risk for inhalation or ingestion of lead oxide begins during the preparation of the enamel and A cross-sectional study design was utilized with a sample of throughout its application, when the skin of the potter is in 553 men and women over the age of 15 years. Sample size direct contact with the greta. Furthermore, the baking was also determined according to the sample required for the process that sets the enamel releases significant amount of prevalence study on kidney damage, using pilot work in the atmospheric lead, with required temperatures as high as area that revealed an expected prevalence of kidney damage 9001C. Some studies have suggested that this may be the in at least 10% of the population (unpublished). A 95% source of increased lead exposure in children who simply live confidence level and 80% precision were applied, along with 2 2 in areas occupied by many ceramic workshops (Garcia Leal the expected frequency, to the formula: N ¼ Z1Àa/2 (1ÀP)/E 2 et al., 1979). The double-edged sword lies in that much more P,whereN ¼ sample size, Z1Àa/2 ¼ confidence level, 2 lead remains in the glaze of this traditional pottery when it is P ¼ frequency of phenomenon, and E ¼ desired precision hardened with these low-temperature methods (Romieu et al., (Lwanga and Lemeshow, 1996). 1994). One out of three homes in Santa Maria Atzompa, to a The repercussions of this exposure have presented total of 185 homes, were randomly selected based on the themselves in several studies revealing subclinical toxicity in population census that reported an average three residents ceramic workers: Studies in 1970s reported that ceramic within the target age group per home. Data were collected by workers in Tonala, Jalisco, had average blood lead means of direct personal interviews conducted by public concentrations between 30 and 40 mg/dl, and in some groups health nurses, who first obtained signed informed consent concentrations reached 85 mg/dl (Garcia de Alba et al., 1979; and provided information on study objectives, procedures, Molina Ballesteros et al., 1980). Another study held in a and implications to participants. A structured questionnaire community in the State of Michoacan reported elevated examining demographic, environmental, occupational, and blood lead concentrations in a population of folk ceramic food-related variables was given to each participant. In the workers, particularly in individuals under 16 years of age. same visit, a blood sample was obtained from each These levels were also significantly associated with the participant by means of venopuncture using a vacutainer presence of a furnace in the home and the usage of lead- containing anticoagulant EDTA K3 (ethylene diamine glazed ceramics for food storage. Here it was concluded that teraacetic acid). Blood lead levels are considered the best the occupational exposure to lead was greater among the indicators for the degree of actual exposure (Silbergeld, subjects involved in the hand painting stage, typically women 1990). These data and sample collection took place over 20 and children (Olaiz et al., 1997). days in August of 1996 and samples were refrigerated at 41C Many rural populations in Mexico are totally or partially until analysis. Lead concentration was determined through economically dependent upon the ceramic industry, in which atomic absorption spectrometry with graphite furnace with a lead salts are commonly used and have been since the arrival detection limit of 1 mg/dl; the Perkin-Elmer model 2100 was of the Spanish 400 years ago. Santa Maria Atzompa, a rural used, with autosampler and recorder model AS-70. Labora- low-income community located 10 km outside of Oaxaca tory work was carried out at the American British Cowdray City, has been of particular interest in this context as the Hospital (ABC) in Mexico City. We validate these measure- region has long produced a unique style of pottery with a ments by considering two independent control methods: the highly glazed, green-tint that is country-renowned. The first consists of the laboratory’s certification program for majority (63%) of its inhabitants are potters who work with competency in measuring blood lead levels run by the Center especially high levels of lead oxide to produce its renowned for Disease Control (CDC), for which it holds a strong green-tint. There has been great resistance to changing to performance record, and the second consists of the use of low lead-free enamel and public health officials have been very and high standards for every 20 lab tests performed. skeptical regarding the health consequences of lead exposure. Univariate analysis was performed on blood lead con- Pilot studies suggesting significant lead exposure in the area centration by extracting its distribution and corroborating hadpromptedresearchontheeffectofleadandkidney extreme values, revealing a normal distribution damage in adults. As research indicates that kidney damage (kurtosis ¼ 0.9004, skewness ¼ 0.4222, median ¼ 43.2, and from lead exposure initiates at an average age of 15, children mean ¼ 43.9), and thus eliminating the need to perform were excluded from the study in order to identify important logarithmic transformation in the analysis.

342 Journal of Exposure Analysis and Environmental Epidemiology (2003) 13(5) Factors associated with lead exposure Isidra et al.

Bivariate analysis sorted occupation groups into two Table 1 . Sociodemographic characteristics in 413 adults from Santa Maria categories, grouping potters into one group and those with Atzompa, Oaxaca. professions unrelated to lead exposure into another, revealing the variables’ association with blood lead concentrations. Characteristic Men N=156 Women N=257 Stratiﬁed analysis was used to determine the possible No. % No. % confounding variables as well as the interaction between Age (years) variables. Finally, a multiple regression analysis was 16–25 38 24.4 76 29.7 performed using the SAS statistical package version 7.0. 26–35 32 20.5 66 25.8 Statistical signiﬁcance was assigned at a P-valuer0.05. 36–45 37 23.7 52 20.3 46–55 21 13.5 23 9.0 Z56 28 17.9 39 15.2

Occupation Results Potter 68 43.6 149 58.0 House-person FF 80 31.3 Data were collected from the 413 (74.6%) subjects who Farmer 30 19.2 FF responded out of the 553 originally selected, the nonresponse Student 11 7.0 14 5.4 Merchant 5 3.2 6 2.3 was associated with heavy migration and the resistance to Employee 42 26.9 8 3.1 blood testing. Of the respondents, a majority were women (62.2%), 56% were between ages 26 and 55 years and a large Education proportion (52.5%) were involved in pottery production. Illiterate 19 12.2 53 20.6 The control group of those working outside of pottery Elementary 79 50.6 141 54.9 ZSecondary 58 37.2 63 24.5 included 19.4% house persons, 7.3% farmers, 6.1% Socialized medical coverage students, and the remaining 12% in a variety of jobs No 122 78.2 207 80.5 unrelated to lead exposure. Only 53.3% had completed Yes 34 21.8 50 19.5 elementary education, and a poor economic situation was a reflected in that 79.6% reported no access to medical services Ingest alcohol No 92 59.4 218 85.5 (Table 1). There were significant lifestyle differences among Yes 63 40.6 37 14.5 gender: males consumed three times more alcohol, and Smokerb smoked eight times more cigarettes than females, important No 104 66.7 246 96.1 factors to consider in assessing kidney damage and lead Yes 52 33.3 10 3.9 exposure. aNoinformationonthreecases. The mean blood lead concentration of the entire study bNo information on one case. group was 43.8 mg/dl, the median at 43.2 mg/dl, with concentrations ranging from 8.4 to 99.6 mg/dl. Figure 1 As we look at the overall increased risk of having high illustrates the significant difference between potters and blood lead levels, Table 3 shows that high blood lead levels nonpotters, although both groups display a normal distribu- were most strongly associated with occupation as workers tion of blood lead concentrations. were 3.4 times more likely to have levels over 40 mg/dl Blood lead concentrations were significantly higher overall (Table 3). The sole consideration of the use of lead-glazed in men (46.7 versus 42.2 mg/dl in women, P ¼ 0.001), use of ceramics for food storage or cooking revealed a 2.6-fold risk lead-glazed ceramics for food storage or cooking (45.6 versus of blood lead concentrations greater than 40 mg/dl (95% 36.2 mg/dl, Po0.0001), cigarette smokers (48.3 versus CI ¼ 1.6–4.3), while cigarette smoking was correlated with 43.0 mg/dl, Po0.0052), and occupations related to lead about 2.5 times the risk of exceeding this parameter (95% exposure (48.2 mg/dl, Po0.0001). It is important to note that CI ¼ 1.344–4.508). even those without occupational exposure risks had very high Among the ceramic workers, lead levels were much higher lead levels: for example, in farmers the average blood lead in cases where a ceramic furnace was present in the home concentration was 43.1 mg/dl, in students 38.6 mg/dl, and in (OR 3.20, 95% CI ¼ 1.4194–7.2143) and in subjects housewives 37.2 mg/dl (Table 2). involved in the baking process (OR 2.21, 95% Viewing Table 2 we see that potters with indoor furnaces CI ¼ 1.1930–4.1222) (Table 3). had concentrations greater than those with outdoor stoves Among nonpotters, we find a highly significant difference (49.2 versus 41.5 mg/dl, P ¼ 0.0032), while those potters in lead blood levels (t ¼ 4.39, P ¼ o0.0001) in those who use directly involved in the baking process suffered from higher ceramics for food cooking and storing and those who do not. concentrations at 50 mg/dl. Those who were exposed to both However, this 12.8% of the population who report not using hazards made up 48% of all potters, who had an average or making lead-glazed pottery still reveal a relatively high concentration of 51 mg/dl. average lead concentration of 32.6 mg/dl.

Journal of Exposure Analysis and Environmental Epidemiology (2003) 13(5) 343 Isidra et al. Factors associated with lead exposure

70 Several different multivariable regression models were

60 Non-Potter performed by varying the variables selected as potential risk Potter factors or confounders. The ﬁnal model yielded the following 50 as the best predictors for blood lead: gender (P ¼ 0.0001), the 40 use of glazed stoneware (Po0.0001), and occupation (P 0.0001). The model explained 18% of the variation in Frequency 30 o blood lead concentration among the study population 20 (F ¼ 31.19 Po0.0001) (Table 4). 10

0 0-10 10.1-20.0 20.1-30.0 30.1-40.0 40.1-50.0 50.1-60.0 60.1-70.0 mayor de 70.0 Discussion Pb blood concentration µg/dl. Figure 1. Lead concentrations in individuals of Santa Maria Lead levels in ceramic workers have been reportedly on the Atzompa, Oaxaca (N=413). decline through the late 1970s and 1980s, but this study

Ta bl e 2 . Distribution of blood lead concentration associated with exposure factors.

Variable Blood lead level

No. Mean (mg/dl) Standard deviation Range Test statistics

Lowest Highest

General population (n=413) Gender Women 257 42.18 13.34 8.4 88.6 t=3.25* Men 156 46.69 14.16 17.4 99.6 P=0.0012 Glazed pottery used for cooking Yes 338 45.60 13.61 8.4 99.6 t=À5.53* No 75 36.17 12.07 9.9 59.8 Po0.0001 House painted on the exterior within the past yeara Yes 71 42.04 15.13 13.8 89.2 t=1.27* No 340 44.33 13.47 8.4 99.6 P=0.2019 Smokerb Yes 62 48.39 14.13 23.6 99.6 t=À2.81* No 350 43.07 13.64 8.4 94.4 P=0.0052 Occupation Pottery 217 48.24 12.97 15.6 99.6 Housewife 80 37.22 12.97 8.4 88.6 f=9.95** Farmer 30 43.15 13.05 24.2 79.0 P=0.0001 Student 25 38.67 11.87 15.8 55.2 Merchant 11 33.18 10.08 18.2 48.6 Employee 50 41.08 14.02 13.8 89.2

Pottery (n=217) Molding pottery piecesa Yes 172 47.3 12.44 15.6 82.8 t=2.42* No 43 52.6 14.10 27.0 99.6 P=0.0163 Applying enamela Yes 83 48.6 12.47 20.0 72.8 t=À0.27* No 132 48.2 13.25 15.6 99.6 P=0.7905 Baking processa Yes 114 50.7 12.85 15.8 82.8 t=À2.95* No 101 45.6 12.53 27.0 99.6 P=0.0035 Presence of furnace out-doors Yes 189 49.2 12.96 26.6 70.0 t=À2.99* No 28 41.5 11.13 15.6 99.6 P=0.0032

aNo information on two cases. bNoinformationononecase. * t-test. **ANOVA.

344 Journal of Exposure Analysis and Environmental Epidemiology (2003) 13(5) Factors associated with lead exposure Isidra et al.

Ta bl e 3 . Odd ratios and for selected variables based on likelihood of having levels over 40 mg/dl.

Variables Category OR 95% CI Adjusted

OR 95% CI

General population (n=413) Glazed pottery used for cooking Yes 2.58 1.566 4.279 2.74* 1.644 4.582 Smoker Yes 2.46 1.344 4.508 2.07* 1.050 4.098 Occupation Potter 3.42 2.286 5.133 4.167* 2.695 6.443

Only pottery (n=216) Apply enamel Yes 1.36 0.7211 2.5715 1.32** 0.7025 2.5177 Baking process Yes 2.21 1.1930 4.1222 1.81** 0.917 3.576 Presence of furnace in-doors Yes 3.20 1.4194 7.2143 2.73** 1.1831 6.3052

*Adjusted by age and gender. **Adjusted by gender.

Ta bl e 4 . Predictors for lead blood concentration. example, found that women who reported frequently preparing, storing, and/or serving food in lead-glazed Multivariate regression ceramics had mean blood levels of 12.7 mg/dl (SD ¼ 9.0), Variables Coefficient P 95% CI whereas women who used LGC only occasionally had a mean blood lead level of 7.1 mg/dl (SD ¼ 4.5) (Herna´ ndez- Intercept 31.64680 o0.0001 28.5304 34.74056 Avila et al., 1991). While clearly unhealthy, these levels do Gender 5.61539 0.0001 3.08909 8.14170 o not approach those found in this community where the Glazed 6.72498 o0.0001 3.50085 9.94912 pottery used average concentration was 43.8 mg/dl. for cooking We observed high lead exposure even in those residents Occupation 8.79044 o0.0001 6.27783 11.30305 who do not use or make pottery, and an unusual normal R2=0.18 (F=31.19 Po0.0001). distribution in both groups in contrast with most other studies that report lognormal distributions. Also, the model based on occupation, gender, and the use of glazed stoneware reveals a community where risk factors for exposure are still only served as a predictor for 18% of the lead levels. Both of very much in existence, as are the resulting high blood lead these results reveal a near-universal contamination within the levels (Garcia de Alba et al., 1979; Molina et al., 1982; Olaiz population, and can be explained as we consider the social et al., 1997). Lead concentrations were very high among both context of this community that both produces and uses pottery and nonpottery workers, and 91.7% of the group pottery with very high lead concentrations. The fumes displayed blood lead levels that exceed the 25 mg/dl maximum produced in the baking process are important contaminants, for biological safety recommended by the Center for Disease and all community members will likely be exposed to their Control and Prevention (CDC, 1996). More disquieting is neighbors’ living environments in the daily exchanges so that the World Health Organization considers blood lead common in Mexican communities, as well as the resulting concentrations greater than 40 mg/dl to be the guideline for external contamination to some degree. While external removing workers from exposure, and this study revealed contamination was not measured in our study, this trend that 59.3% of this population have levels above this limit, hasbeenobservedinchildrenlivinginaceramic-dependent with an average lead blood concentration of 43.8 mg/dl. community, who were exposed indirectly from playing in (WHO, 1989). contaminated soil in addition to direct exposure from the Our results are supported by past studies that concluded ovens and the greta (Montoya and Hernandez, 1981; Molina that the use of lead-oxide-based glaze and the antiquated et al., 1982). The potency of fumes produced in the baking methods used in processing pottery poses a true threat to the process is highlighted again in the fact that men exhibited the well-being of all community members. We have also highest blood lead concentrations, as found in past studies concluded that the use of glazed stoneware for cooking and (Olaiz et al., 1997). Their role is most often in warming the food storage is a significant predictor for elevated blood lead stove and monitoring the baking process, while women have concentrations in Mexico (Flores et al., 1989; Herna´ ndez- more direct contact with the enamel in its application to the Avila et al., 1991; Olaiz et al., 1997). ceramic. It is also worth noting that while the manual However, blood lead levels were much higher relative to painting of enamel and involvement in the baking process past studies looking exclusively at lead-glazed ceramic usage significantly affected lead blood levels, only the presence of in urban communities. Past work in Mexico City, for an oven in the home was a significant contributor to the

Journal of Exposure Analysis and Environmental Epidemiology (2003) 13(5) 345 Isidra et al. Factors associated with lead exposure prediction model. Another factor that dilutes the prediction ceramics in the kitchen, as a simple change such as assigning model is the sharing and exchange of meals common in acidic foods to nonceramic dishware would influence lead Mexican culture, exposing those who do not identify concentrations in the Mexican population. High-risk groups themselves as cooking with lead-glazed ceramics at home. must also remain in focus for their continual exposure as A third possible compounding factor is the community’s lifelong members of the community. It is being increasingly proximity to Oaxaca City, where industrial and vehicular acknowledged that kidney damage is not only associated with lead sources should be considered. dosage, but also with the duration of exposure to lead It is interesting that lead levels were about 10 mg/dl higher sources. Prolonged or repeated exposure is thought to in potters who smoke, and about 5 mg/dl higher in the general increase clinical manifestations, most often occurring with population. Lead is usually considered to be only an blood lead concentrations greater than 40 mg/dl (Nolan and occasional contaminant in cigarettes when lead arsenate is Shaikn, 1992). The monitoring of chronic conditions in the used as an insecticide in tobacco growing, with levels on the population of Santa Maria Atzompa is an ongoing project. decline in line with new regulations (Watanabe et al., 1987; Nuwayhid et al., 2001). These results suggest either that they are a more important source of lead contamination than References typically reported or that smoking in contaminated environ- CDC. Adult blood lead epidemiology and surveillance-United States, ments, or with lead-contaminated hands, significantly second quarter, 1996. MMWR 1996: 45: 919–920. elevates inhalation and absorption of lead. Coyer O. and Galvao Luiz A.C. Serie de Vigilancia 8 PLOMO. Centro Mexico, as the sixth largest lead-producing country in the Panamericano de Ecologia Humana y Salud, 1989: pp. 19–24. world, has introduced strong measures to minimize potential Criterios de salud ambiental 3 Plomo. Centro Panamericano de Ecologia sources of lead exposure (Romieu et al., 1994; Romieu et al., Humana y Salud, 1979: pp. 66–84. Diario Oficial Mexicano, tomo CDLXXXII, No. 10, NOM.010-SSA1- 1997). The implementation of surveillance systems in 1991 1993. Gobierno Constitucional de los Estados Unidos Mexicanos, has been highly effective in reducing air lead concentrations, Me´ xico City, 1993. falling in Mexico from an annual average of 1.95 mg/m3 in Driscoll W., Mushak P., Garfias J., Rothenberg S.J., Reducing lead 1988 to 0.28 mg/m3 in 1994 alongside plummeting blood lead in gasoline. Mexico experience. Environ Sci Technol 1992: 26: concentrations in the capital’s residents (Driscoll et al., 1992; 1702–1705. Flores L.E., Alarco´ n C.J., Bobadilla J.L., et al. Factores asociados a los Herna´ ndez-Avila and Plazualos-Rendon, 1995). More niveles de plomo en sangre en residentes de la ciudad de Me´ xico. recent attempts to find multilevel solutions to lead exposure Salud Pu´blica de Mex 1989: 31(5): 625–633. led to the establishment of regulations through the Health Garcı´ a Leal R., Gonza´ lez Ramı´ rez D., Molina Ballesteros G.X. Ministry on numerous products, including the lead content Contaminacio´ n con plomo en el medio rural. In: Intoxicacion por of high-temperature glazed ceramic ware and glassware (2.5– plomo. Subdireccion General Medica, JMSS, Mexico. 1979: 129–142. Garcı´ a de Alba J.E., Molina Ballesteros G, Cuento Sa´ nchez L., et al. 7.0 ppm) (Diario Oficial Mexicano, 1993). These guidelines Niveles de plomo en alfareros. Higiene 1979: 29(3): 129–139. have run alongside population-directed educational pro- Herna´ ndez-Avila M., and Plazualos-Rendon E. (Eds). El plomo: Un grams on pottery within the Health Sector, but it is obvious problema de salud pu´ blica en Me´ xico. Intoxicacio´ n por plomo en that a more integral approach is needed in a nonregulated Me´ xico: Prevencio´ n y Control. In: Perspectiva en Salud Publica 21. industry where livelihoods and tradition are at stake. Instituto Nacional de Salud Publica, Cuernavaca, Me´ xico, 1995. pp. 13–14. Feasible, affordable alternatives must be considered. For Herna´ ndez-Avila M., Romieu I., Rı´ os C., et al. Lead-glazed ceramics as example, updating clay ovens to gas or high-temperature major determinants of blood lead levels in Mexican women. Environ ovens that produce less exhaust, or moving the furnaces out Health Perspect 1991: 94: 117–120. of the house altogether would be greatly effective, but are not Landrigan P.J. Current issues in the epidemiology and toxicology of financially feasible options for many community members. A occupational lead exposure. Environ Health Perspect 1990: 89: 61–66. new boron-based alternative to the traditional leaded ‘‘greta’’ Landrigan P.J. Strategies for epidemiological studies of lead in bone in was introduced in Atzompa, but did not provide the same occupationally exposure populations. Environ Health Perspect 1991: quality in color that the community is renowned for. 91: 81–86. Moreover, they have also been using lead-oxide glaze for Lwanga S.K., Lemeshow S. Sample size determination in health studies. hundreds of years and thus the connection between the World Health Organization, Geneva, 1991. Magdalena Rojas-Lo´ pez, Carlos Santos-Burgoa, Camilo Rı´ os, et al. enamel and illness seems abstract. Studies revealing astound- Use of lead-glazed ceramics is the main factor associated to high lead ing lead concentrations such as this might be convincing in blood levels in two Mexican rural communities. J Toxicol Environ enough to motivate changes in the methods employed with Health 1994: 42(1): 45–52. the support of the health sector and community leaders, Molina G., Zun˜ iga M., Ca´ rdenas A., et al. Concentracio´ ndeplomoen allowing the community to slowly reshape its economic base sangre de los nin˜ os de familias alfareras. Bol Sanit Panam 1982: 92: 33–40. on boron-based alternatives or in another niche. In the Molina Ballesteros G., Zun˜ iga-Charles M., Ca´ rdenas A., et al. meantime, more integrated educational approaches should be Alteraciones psicolo´ gicas en nin˜ os expuestos a ambientales dome´ s- taken to raise basic awareness in the use of lead-glazed ticos ricos en plomo. Bol Sanit Panam 1983: 94(3): 239–247.

346 Journal of Exposure Analysis and Environmental Epidemiology (2003) 13(5) Factors associated with lead exposure Isidra et al.

Molina Ballesteros G., Zun˜ iga-Charles M., Garcia J., et al. Exposicio´ na America and the Caribbean. Environ Health Perspect 1997: plomo en una poblacio´ ndealfareros.Arch Invest Me´d 1980: 11: 105: 398–405. 147–155. Romieu I., Palazuelos E., Herna´ ndez-Avila M., et al. Sources of lead Montoya Cabrera M., Moise´ s S., and Barquet R. Saturnismo por loza exposure in Mexico City. Environ Health Perspect 1994: 102: de barro vidriado. Rev Med IMSS 1978: 4: 249–257. 384–389. Montoya M., Herna´ ndez A. Intoxicacio´ n por plomo. Origen de la Romieu I., Palazuelos E., Meneses F., et al. Vehicular trafﬁc as a exposicio´ n al metal en tres familias. Rev Med IMSS 1981: 19: determinant of blood-lead leves in children: a pilot study in Mexico 337–339. City. Arch Environ Health 1992: 47(4): 246–249. Nolan C.V., Shaikn Z.A. Lead nephrotoxicity and associated disorder: Ruiz Sandoval S.G., Envenenamiento lento por plomo en los habitantes biochemical mechanisms. Toxicology 1992: 73: 127–146. de Oaxaca. Gac Med Mex 1878: 13: 393–403. Nuwayhid I., McPhaul K., Bu-Khuzam R., Duh S.H., Christenson Silbergeld E.K. Implications of new data on lead toxicity for managing R.H., Keogh J.P. Determinants of elevated blood lead levels among and preventing exposure. Environ Health Perspect 1990: 89: 49–59. working men in Greater Beirut. J Med Liban 2001: 49(3): 132–139. Watanabe T., Kasahara M., Nakatsuka H., Ikeda M. Cadmium and Olaiz G., Rojas R., Fortoul T.I., et al. High blood lead levels in ceramic lead contents of cigarettes produced in various areas of the world. Sci Folk Art Workes in Michoacan, Me´ xico. Arch Environ Health 1997: Total Environ 1987: 66: 29–37. 52(1): 51–55. WHO. WHO Environmental Health Criteria Series No. 85: Lead, Romieu I., Lacasna M., McConnell R., and the Lead Research Group Geneva. Environmental Aspects. 1989: 106 pp. ISBN 92 4 1542853 of the Pan-American Health Organization. Lead Exposure in Latin

Journal of Exposure Analysis and Environmental Epidemiology (2003) 13(5) 347