Persistent Organic Pollutants: How did they get there?!
In an earlier post I wrote about polybrominated diphenyl ethers (PBDEs), which are among the list of chemical substances that are classified as persistent organic pollutants (POPs). POPs are chemicals that have three main characteristics: 1) they are stable compounds, enabling them to persist in the environment; 2) they are lipid (fat) soluble, which combined with their stability, enables them to accumulate up the food chain; 3) they have the ability to act as endocrine (hormone) disruptors [1]. New studies continue to discover the presence of POPs in environments where they have never been produced or even used before, indicating their ability to be transported over long-ranges. These characteristics, along with the increasing evidence of adverse health outcomes associated with exposure, have sparked international discussion about the need to urgently reduce and eliminate the production of these chemicals [2].
Dioxins and Furans: Source and Health Impacts
Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) are another type of POPs that were studied in the e-waste literature I reviewed. There is a recurrent theme in my postings about the dangers associated with primitive e-waste recycling techniques used in poorer countries where a vast majority of the world's e-waste is processed. In the case of PCDD/Fs, their affiliation with e-waste depends on these crude recycling processes. Plastics made from polyvinyl chloride (26% of the plastic found in e-waste by volume), once processed through uncontrolled open burning, can generate PCDD/Fs [3].
Dioxins and furans can enter the body via inhalation, ingestion and skin absorption. Exposure to PCDD/Fs at high levels can lead to chloracne (severe skin disease), darkening of the skin, and altered liver function. Long-term exposure can lead to damage of the immune, nervous and endocrine systems and impaired reproductive function [3,5]. Specifically, dioxins are classified as Group 1 "known human carcinogens" according to the World Health Organization's (WHO) International Agency for Research on Cancer [4]. These carcinogenic effects have only been observed with high dose exposures; there is insufficient evidence to prove that low-level exposure to dioxins and furans can cause cancer [5].
The Tolerable Monthly Intake for dioxins and furans is 70 picograms per kilogram of body weight (pg/kg), as outlined by the WHO. This is the amount that can be ingested per month over a lifetime without inducing substantial health risk [5].
What's been reported in the E-waste literature?
A number of studies have quantified the level of PCDD/Fs in air, soil, dust, sediment, freshwater, fish, and cow milk samples, in a number of regions where e-waste recycling has taken place [6]. My interests lied in those studies that quantified human internal exposure to these hazardous agents, using biological markers such as human breast milk, placenta, hair, blood and urine. I did review one study that estimated daily human exposure using many environmental samples (soil, dust, and surface samples). This study was an interesting starting point for my research in this area because I kept these estimates in mind when looking at the studies that measured internal exposure using biological markers. The estimated daily intake of PCDD/Fs via soil/dust ingestion and dermal exposure, according to this study, was two times higher for people who are exposed to e-waste recycling facilities in Taizhou (2.3 and 0.363 pg/kg/day for children and adults, respectively), compared to people who are exposed to chemical industrial sites (0.021 and 0.0053 pg/kg/day for children and adults, respectively) in various areas also in Eastern China [7]. Note that these estimates did not include a number of other sources of PCDD/Fs exposure such as through food, water, breast milk etc.
From here I turned to another study that was a health risk assessment (systematic calculation of risk) of dioxins and furans, using samples of human milk, placenta and hair from residents also in the Taizhou region*. The 10 study participants were all women of child-bearing age who had been either exposed to e-waste recycling activities as residents in Taizhou, or unexposed residents in a neighbouring city (245 km away from Taizhou) for at least two years, and who had given birth at either of two study sites between August and December 2005. Each biological sample provided a different picture of the overall exposure. PCDD/Fs in hair samples indicated accumulation and atmospheric deposition on the hair surface; consequently, it is difficult to distinguish between internal and external contamination [8]. Analysis of the placenta is a good indication of prenatal exposure, and breast milk samples reflects maternal body burden and the postnatal transfer of PCDD/Fs to infants [9].
The results showed that there were significantly different PCDD/F concentrations in the placenta and hair samples after comparing the exposed group to the unexposed group. The total concentrations found in breast milk samples from the exposed group were two times higher than the unexposed group, but this difference was not found to be statistically significant (p>0.05). Background pollution, dietary habits and personal characteristics were all important factors influencing the total concentrations of PCDD/Fs [9]. All human milk samples from the Taizhou group, and 80% of the samples taken from the women in the reference group, exceeded the European Union's maximum permitted levels in milk (3 pg WHO-TEQ/g lipid) [10]. The health risk assessment for infants estimated that the daily intake of PCDD/Fs via breast milk was 102.98 +/- 67.65 pg TEQ/kg body weight/day in the exposed group and 45.83 +/- 36.22 pg TEQ/kg body weight/day in the unexposed. Both of these values exceed the WHO tolerable daily intake (1-4 pg TEQ/kg body weight/day) [11]. High intake exceeding the toxicological limit during breast feeding is concerning because of the long length of time that it takes for the toxins to be removed from the body (7 to 12 years), faster and greater absorption in infants and children, and the immature body defenses in infants [9].
A third study was found that measured PCDD/F concentrations in hair samples from 64 randomly selected male workers at e-waste factories in eastern China. The study did not have it's own comparison group, and was therefore not looked at in great detail. The results from this study similarly found vastly greater concentrations of PCDD/Fs in hair samples analyzed compared to other known contaminated areas, and compared to the concentrations found in healthy study participants in Japan [12].
Tying things together
At the beginning of this post I mentioned that one of the properties of POPs is that they can travel over long distances. I was kind of surprised to have never really found a comprehensive article reviewing whose health is being affected by e-waste (big and small, currently or in the future). I read a bunch of media articles and blog posts talking about how as North Americans, we don't really experience the major effects of the international shipment and mismanagement of e-waste. I think this is a huge misconception, because we are all impacted by e-waste, even if we aren't the ones working in the family-owned workshops. If not in the present, then most likely in the future, we will all be impacted by e-waste if innovative and stricter policies aren't implemented.
Hopefully this is motivation for those decision makers that aren't motivated by the devastating results presented above, and in previous posts, about the human health impacts already occurring.
* Taizhou has become one of the main receivers of e-waste in China, in recent years. It receives an estimated 2.2 million tonnes of e-waste on an annual basis, and the industry employs an estimated forty thousand people. Most of the recycling that takes place in this region involves open burning, acid baths, and manual disassembly of e-waste [9].
Sunday, December 13, 2009
Saturday, December 12, 2009
DNA Damage and Chromosomal Aberrations: Threats to Current and Future Generations
Environmental and human health concerns related to electronic waste were first raised by the international community more than a decade ago [1]. After noticing the rapidly expanding mountains of e-waste in places like Guiyu, in the Guangdong Province of China, governments have been persuaded to take action to alleviate the negative impacts of e-waste. For example, in Jinghai County of Tianjin, in Northern China, the local government has built an "environmental protection industry park" to minimize environmental pollution [2]; however, researchers have noted that many residents have not been using the park in order to reduce costs [3]. The manual disassembly and illegal burning of e-waste continues to occur [4]. These primitive practices not only create an enormous amount of environmental pollution, but also disseminate genotoxic agents that threaten the health of current and future generations living in the local environment.
What are Genotoxins?
Genotoxins are agents that damage the genetic material in cells. Furthermore, they are toxins that have been found to be mutagenic or carcinogenic, meaning they are capable of causing genetic mutations or the development of cancer [5]. The genotoxins associated with e-waste include: metals such as chromium, beryllium, and cadmium; chlorinated dioxins and furans formed from the burning of plastics; and, flame retardants such as polybrominated diphenyl ethers [6].
What is DNA damage? And a chromosomal aberration?
DNA damage is an umbrella term used to define all types of physical abnormalities in DNA. Typically, these abnormalities are detected and fixed by the body, and in fact are often occurring naturally in the body [3,7]. DNA damage is particularly problematic when it leads to DNA mutations: changes to the DNA sequence [8].
A chromosome is an organized structure of DNA; therefore, a chromosomal aberration is just a change in the normal structure or number of chromosomes within a cell [9].
Changes to a cell's DNA or chromosomes can lead to a number of pathologies including genetic disorders, infertility, spontaneous abortions, elevated cancer risk and premature aging [3,5].
Does exposure to e-waste processing lead to DNA damage and chromosomal aberrations?
Three studies were found that looked at DNA damage associated with exposure to e-waste processing. One of the studies was already discussed in my post about polybrominated diphenyl ethers (PBDE) [10]. Of the remaining two, one measured DNA damage associated with chromium exposure from e-wastes [7], and the other focused on measuring elevations in risk of DNA damage and chromosomal aberrations attributable to e-wastes [3]. Measuring the damage to genetic information in cells holds great public health importance because it enables the investigation of toxicities that may accumulate in the human body over many years and essentially go undetected until pathologies develop. These markers can be used to emphasize the urgent need for public health interventions in order to mitigate exposure and potentially avoid serious health outcomes. Not only are there cumulative health risks for the current populations exposed, but with damages to the genetic information of these individuals there are also great healths risks for future generations (their offspring)[3].
All three studies did find an elevated risk of DNA damage; hence, the discovery that exposure to e-waste pollutants produced by the dumping and recycling of e-wastes may be mutagenic. The PBDE study estimated that the odds of developing genetic damage after having a history of working with e-waste in village in southeast China was 38.85 times the odds for someone of similar characteristics, who didn't have a history of working with e-waste (range, 1.11 to 1358.71; p=0.044) [10]. Similarly, in the study looking at chromium exposure, umbilical cord blood chromium levels (UCBCL) in neonates was significantly positively correlated with the time their mothers spent roaming in e-waste recycling sites during pregnancy. Furthermore, UCBCLs in neonates was discovered to have a significant positive correlation with DNA damage of lymphocytes (white blood cells) [7]. This confirms the hypothesized risk of DNA damage beginning early in the life of new generations, as a result of e-waste pollutants.
The third paper I found was exceptionally well done. Both the exposure and outcome was well measured, the methods were transparent and validated, and the author's suggestions were insightful and well supported. This study looked at multiple types of chromosomal aberrations and DNA damage. Their analysis detected a significant difference in DNA damage among the randomly selected permanent residents of three villages with numerous e-waste disposal sites (exposed, 171 villagers), compared to randomly selected permanent residents from neighbouring towns without any e-waste disposal sites (unexposed, 30 villagers). Additionally, the total chromosomal aberration rate in the exposed group was significantly higher than in the unexposed group (5.50% versus 0.28%, respectively; p<0.0001)[3].
Bottom Line
The unsafe handling and recycling of e-waste can lead to human exposure to genotoxic substances, which has been shown to lead to genetic damage. These findings are extremely worrisome, and emphasize the need for stricter control measures to eliminate outdated recycling practices and the illegal importation of e-waste in order to protect current and future generations from the immediate and long-term effects of pollutants from e-waste. Fundamental changes are needed in the approach to e-waste management, and they need to come quickly.
What are Genotoxins?
Genotoxins are agents that damage the genetic material in cells. Furthermore, they are toxins that have been found to be mutagenic or carcinogenic, meaning they are capable of causing genetic mutations or the development of cancer [5]. The genotoxins associated with e-waste include: metals such as chromium, beryllium, and cadmium; chlorinated dioxins and furans formed from the burning of plastics; and, flame retardants such as polybrominated diphenyl ethers [6].
What is DNA damage? And a chromosomal aberration?
DNA damage is an umbrella term used to define all types of physical abnormalities in DNA. Typically, these abnormalities are detected and fixed by the body, and in fact are often occurring naturally in the body [3,7]. DNA damage is particularly problematic when it leads to DNA mutations: changes to the DNA sequence [8].
A chromosome is an organized structure of DNA; therefore, a chromosomal aberration is just a change in the normal structure or number of chromosomes within a cell [9].
Changes to a cell's DNA or chromosomes can lead to a number of pathologies including genetic disorders, infertility, spontaneous abortions, elevated cancer risk and premature aging [3,5].
Does exposure to e-waste processing lead to DNA damage and chromosomal aberrations?
Three studies were found that looked at DNA damage associated with exposure to e-waste processing. One of the studies was already discussed in my post about polybrominated diphenyl ethers (PBDE) [10]. Of the remaining two, one measured DNA damage associated with chromium exposure from e-wastes [7], and the other focused on measuring elevations in risk of DNA damage and chromosomal aberrations attributable to e-wastes [3]. Measuring the damage to genetic information in cells holds great public health importance because it enables the investigation of toxicities that may accumulate in the human body over many years and essentially go undetected until pathologies develop. These markers can be used to emphasize the urgent need for public health interventions in order to mitigate exposure and potentially avoid serious health outcomes. Not only are there cumulative health risks for the current populations exposed, but with damages to the genetic information of these individuals there are also great healths risks for future generations (their offspring)[3].
All three studies did find an elevated risk of DNA damage; hence, the discovery that exposure to e-waste pollutants produced by the dumping and recycling of e-wastes may be mutagenic. The PBDE study estimated that the odds of developing genetic damage after having a history of working with e-waste in village in southeast China was 38.85 times the odds for someone of similar characteristics, who didn't have a history of working with e-waste (range, 1.11 to 1358.71; p=0.044) [10]. Similarly, in the study looking at chromium exposure, umbilical cord blood chromium levels (UCBCL) in neonates was significantly positively correlated with the time their mothers spent roaming in e-waste recycling sites during pregnancy. Furthermore, UCBCLs in neonates was discovered to have a significant positive correlation with DNA damage of lymphocytes (white blood cells) [7]. This confirms the hypothesized risk of DNA damage beginning early in the life of new generations, as a result of e-waste pollutants.
The third paper I found was exceptionally well done. Both the exposure and outcome was well measured, the methods were transparent and validated, and the author's suggestions were insightful and well supported. This study looked at multiple types of chromosomal aberrations and DNA damage. Their analysis detected a significant difference in DNA damage among the randomly selected permanent residents of three villages with numerous e-waste disposal sites (exposed, 171 villagers), compared to randomly selected permanent residents from neighbouring towns without any e-waste disposal sites (unexposed, 30 villagers). Additionally, the total chromosomal aberration rate in the exposed group was significantly higher than in the unexposed group (5.50% versus 0.28%, respectively; p<0.0001)[3].
Bottom Line
The unsafe handling and recycling of e-waste can lead to human exposure to genotoxic substances, which has been shown to lead to genetic damage. These findings are extremely worrisome, and emphasize the need for stricter control measures to eliminate outdated recycling practices and the illegal importation of e-waste in order to protect current and future generations from the immediate and long-term effects of pollutants from e-waste. Fundamental changes are needed in the approach to e-waste management, and they need to come quickly.
Friday, December 11, 2009
Continuing on the Topic of Heavy Metals Exposure
I wish I could post a few pictures that I came across in one of the articles I read on e-waste recycling sites in Bangalore, India. If you have access to articles in Chemosphere, here are the details. Figure 1 in this journal article includes a spread of pictures of the facilities and procedures used by these e-waste recyclers. Included were photos of child and adult workers using mercury and corrosive acids to extract valuable metals. No protective equipment is worn, bare hands are shown carrying containers filled with corrosives, and an indoor open fire pit is photographed with an individual standing over the fire, burning e-waste with mercury. I've read about these crude practices, but the photos are very powerful illustrations.
Continuing on the topic of heavy metal exposure, this post will discuss internal mercury and cadmium exposure levels as reported in three studies. These studies looked at potential exposure as a result of either occupational exposure to elements in e-waste or environmental exposure due to proximity to e-waste sites. A comparison group was used in all three studies in order to associate elevated mercury and cadmium levels with e-waste recycling activities.
You can learn more about which electronic devices contain mercury and cadmium, as well as an overview of the known human health risks of exposure, in my post entitled What Are You Leaving Behind?
Mercury Exposure
I could only find one study that measured mercury exposure among e-waste workers and neighboring community members, and the quality of the study was quite poor. Hair samples were used as the biological measure to assess mercury exposure, which can be a good measure of the variation in mercury intake over a long period of time, if the hair is analyzed in segments [1]. This study, to my understanding, did not analyze the hair in segments.
The study I'm referring to was conducted in Bangalore, India, and is the same article I mentioned above (containing the photos) [2]. The aim of the study was to identify exposure to a number of different metals, mercury being one of them. Only males were looked at in this study: eleven exposed and eight control participants. As a results of the small sample size, a number of factors that may have influenced the findings were either not measured or not able to be considered in the analysis (i.e. age, diet, smoker status and years of employment). Overall, I found the methods section to be poor, as it lacked a detailed description of how the hair samples were taken and analyzed.
The results of the study did not indicate a statistically significant difference between the male workers in two e-waste recycling sites in Bangalore (means=0.4 and 0.1 micrograms per gram), compared to men from Chennai, India, the control group (mean=0.19 micrograms per gram). In fact, the concentrations reported were below the averages found in other parts of India (mean= 0.8 micrograms per gram) [3], and well below the levels deemed to be normal and acceptable in Canada (less than 6 micrograms per gram) [1]. The authors suggest that this may be due to very little consumption of fish by the residents in this area, since fish consumption is seen as a major source of mercury exposure in humans [1,2].
The study also measured mercury content in soil samples from a crude backyard e-waste recycling site in Bangalore (mean=1.8 micrograms per gram; range, 0.09 to 59 micrograms per gram), and from a control site in Bangalore (<0.05 micrograms per gram). These levels are also well below the Canadian guidelines for mercury content in soil: agricultural and residential areas must be less than 6.6 micrograms per gram, and industrial areas must be less than 50 micrograms per gram [1].
There is clearly a gap in the current literature describing human mercury exposure associated with e-waste recycling. Since this is the first and only study, to my knowledge, that has looked at mercury, further investigation is needed.
Cadmium Exposure
My literature search discovered three studies that looked at cadmium exposure and e-waste disassembly. Two of the three measured concentrations in hair samples, and one measured it in blood. All three studies measured internal exposure to additional heavy metals, and consequently, have already been discussed in the sections on lead exposure (Zheng et al., 2008) and mercury exposure (above).
As already mentioned, Canada doesn't have a guidance value for safe blood cadmium levels (BCLs) [4]. According to the literature, the threshold BCL deemed to be a risk for toxicity is five micrograms per litre [5,6]. In the study by Zheng et al., BCLs were measured in children ages one to seven from Guiyu, a well known area where primitive family-run recycling is practiced by nearly 60 to 80% of families [6]. In the Guiyu group, the mean BCL was 1.58 micrograms per litre (range, 0 to 9.2), which was statistically significantly different from the reference group (from Chendian) mean BCL of 0.97 micrograms per litre (range, 0 to 3.49). Other studies have found high levels of cadmium in Guiyu, in samples of dust, water, and river sediments, which further supports the finding of elevated BCLs in this population of children, by comparison to those in the reference group [7]. Furthermore, factors related to e-waste recycling showed a statistically significant correlation with BCLs. In Guiyu, e-waste materials and process residues are dumped in workshops, yards, roadsides, open fires, irrigation canals, ponds, and rivers. As a result, house dwelling location, father's occupation, and child playing areas that were close to e-waste increased one's risk of exposure to cadmium, and therefore, were all factors found to be associated with the elevated BCLs. Since smoking is a major source of non-occupational exposure to cadmium, secondary smoke exposure was assessed, but surprisingly was not found to be significantly correlated with BCLs. Even though the mean BCLs were found to be below the threshold of five micrograms per litre, recent studies have suggested that the tolerable cadmium limit be lowered since kidney damage was observed at lower concentrations [8]. To further put these values into perspective, the Canadian Health Measures Survey from 2007/08 detected BCLs of 0.15 micrograms per litre in youth ages 6 to 19 [9].
In the other two studies that were looked at, a statistically significant difference in cadmium exposure was also observed, which was believed to be attributable to primitive e-waste recycling. The first study compared cadmium concentrations in hair among male e-waste workers in Bangalore, India to a control group from Chennai, India [2]; and, the second study compared cadmium concentrations in hair among residents in Taizhou, in southeastern China where extensive e-waste recycling takes place in family-run workshops, to two industrialized cities within the same province (Ningbo and Shaoxing) [10]. Once again, the quality of both studies was lacking, and because there currently is no standardized protocol for hair analysis, a comparison of the results from the two studies was not made [10].
A paper that I had come across in my literature search was referred to in the Taizhou study, which looked at cadmium levels in locally grown rice. The researchers calculated that Taizhou rice would alone contribute to 67% of the World Health Organization's tolerable daily intake for cadmium [11]. Given this, it was not surprising to see that the Taizhou participants showed approximately twice as high levels of cadmium than the control participants [10]. Other sources of cadmium exposure that are unrelated to e-waste, such as smoker status, food and drinking water levels, were not measured in this study. However, their methods for determining that the high levels of cadmium content in hair were likely due to e-waste activities were quite sophisticated (Pearson's correlation and principal component analysis).
Continuing on the topic of heavy metal exposure, this post will discuss internal mercury and cadmium exposure levels as reported in three studies. These studies looked at potential exposure as a result of either occupational exposure to elements in e-waste or environmental exposure due to proximity to e-waste sites. A comparison group was used in all three studies in order to associate elevated mercury and cadmium levels with e-waste recycling activities.
You can learn more about which electronic devices contain mercury and cadmium, as well as an overview of the known human health risks of exposure, in my post entitled What Are You Leaving Behind?
Mercury Exposure
I could only find one study that measured mercury exposure among e-waste workers and neighboring community members, and the quality of the study was quite poor. Hair samples were used as the biological measure to assess mercury exposure, which can be a good measure of the variation in mercury intake over a long period of time, if the hair is analyzed in segments [1]. This study, to my understanding, did not analyze the hair in segments.
The study I'm referring to was conducted in Bangalore, India, and is the same article I mentioned above (containing the photos) [2]. The aim of the study was to identify exposure to a number of different metals, mercury being one of them. Only males were looked at in this study: eleven exposed and eight control participants. As a results of the small sample size, a number of factors that may have influenced the findings were either not measured or not able to be considered in the analysis (i.e. age, diet, smoker status and years of employment). Overall, I found the methods section to be poor, as it lacked a detailed description of how the hair samples were taken and analyzed.
The results of the study did not indicate a statistically significant difference between the male workers in two e-waste recycling sites in Bangalore (means=0.4 and 0.1 micrograms per gram), compared to men from Chennai, India, the control group (mean=0.19 micrograms per gram). In fact, the concentrations reported were below the averages found in other parts of India (mean= 0.8 micrograms per gram) [3], and well below the levels deemed to be normal and acceptable in Canada (less than 6 micrograms per gram) [1]. The authors suggest that this may be due to very little consumption of fish by the residents in this area, since fish consumption is seen as a major source of mercury exposure in humans [1,2].
The study also measured mercury content in soil samples from a crude backyard e-waste recycling site in Bangalore (mean=1.8 micrograms per gram; range, 0.09 to 59 micrograms per gram), and from a control site in Bangalore (<0.05 micrograms per gram). These levels are also well below the Canadian guidelines for mercury content in soil: agricultural and residential areas must be less than 6.6 micrograms per gram, and industrial areas must be less than 50 micrograms per gram [1].
There is clearly a gap in the current literature describing human mercury exposure associated with e-waste recycling. Since this is the first and only study, to my knowledge, that has looked at mercury, further investigation is needed.
Cadmium Exposure
My literature search discovered three studies that looked at cadmium exposure and e-waste disassembly. Two of the three measured concentrations in hair samples, and one measured it in blood. All three studies measured internal exposure to additional heavy metals, and consequently, have already been discussed in the sections on lead exposure (Zheng et al., 2008) and mercury exposure (above).
As already mentioned, Canada doesn't have a guidance value for safe blood cadmium levels (BCLs) [4]. According to the literature, the threshold BCL deemed to be a risk for toxicity is five micrograms per litre [5,6]. In the study by Zheng et al., BCLs were measured in children ages one to seven from Guiyu, a well known area where primitive family-run recycling is practiced by nearly 60 to 80% of families [6]. In the Guiyu group, the mean BCL was 1.58 micrograms per litre (range, 0 to 9.2), which was statistically significantly different from the reference group (from Chendian) mean BCL of 0.97 micrograms per litre (range, 0 to 3.49). Other studies have found high levels of cadmium in Guiyu, in samples of dust, water, and river sediments, which further supports the finding of elevated BCLs in this population of children, by comparison to those in the reference group [7]. Furthermore, factors related to e-waste recycling showed a statistically significant correlation with BCLs. In Guiyu, e-waste materials and process residues are dumped in workshops, yards, roadsides, open fires, irrigation canals, ponds, and rivers. As a result, house dwelling location, father's occupation, and child playing areas that were close to e-waste increased one's risk of exposure to cadmium, and therefore, were all factors found to be associated with the elevated BCLs. Since smoking is a major source of non-occupational exposure to cadmium, secondary smoke exposure was assessed, but surprisingly was not found to be significantly correlated with BCLs. Even though the mean BCLs were found to be below the threshold of five micrograms per litre, recent studies have suggested that the tolerable cadmium limit be lowered since kidney damage was observed at lower concentrations [8]. To further put these values into perspective, the Canadian Health Measures Survey from 2007/08 detected BCLs of 0.15 micrograms per litre in youth ages 6 to 19 [9].
In the other two studies that were looked at, a statistically significant difference in cadmium exposure was also observed, which was believed to be attributable to primitive e-waste recycling. The first study compared cadmium concentrations in hair among male e-waste workers in Bangalore, India to a control group from Chennai, India [2]; and, the second study compared cadmium concentrations in hair among residents in Taizhou, in southeastern China where extensive e-waste recycling takes place in family-run workshops, to two industrialized cities within the same province (Ningbo and Shaoxing) [10]. Once again, the quality of both studies was lacking, and because there currently is no standardized protocol for hair analysis, a comparison of the results from the two studies was not made [10].
A paper that I had come across in my literature search was referred to in the Taizhou study, which looked at cadmium levels in locally grown rice. The researchers calculated that Taizhou rice would alone contribute to 67% of the World Health Organization's tolerable daily intake for cadmium [11]. Given this, it was not surprising to see that the Taizhou participants showed approximately twice as high levels of cadmium than the control participants [10]. Other sources of cadmium exposure that are unrelated to e-waste, such as smoker status, food and drinking water levels, were not measured in this study. However, their methods for determining that the high levels of cadmium content in hair were likely due to e-waste activities were quite sophisticated (Pearson's correlation and principal component analysis).
Thursday, December 10, 2009
Heavy Metals: Alarming Results
Guiyu, a region in the Chinese province of Guangdong, together with New Delhi in India, is one of the most popular destinations for e-waste [1,2]. Within 52 squared kilometers, Guiyu alone has been estimated to accommodate millions of tonnes of domestic and overseas e-waste per year [3]. Several studies have detected soaring levels of toxic heavy metals in Guiyu, in samples of dust, soil, river sediment, surface water, and groundwater [1,2,4,5,6,7]. Levels well above Canadian environmental standards.
In a previous post I outlined the human health risks of exposure to lead, mercury, and cadmium. At the end of each description I provided a threshold or guidance value set out by the Canadian government, which essentially is the exposure value deemed to be "safe" based on past research. In my literature review I came across a number of studies looking at these particularly toxic heavy metals, and describing the exposure levels among residents of communities where e-waste disassembly is occurring. I found the results of these studies astounding! And you will eventually come to understand why, after reading this post. Given the extensive amount of research that already describes the human health impacts of exposure to these metals, a lot of the studies did not measure health outcomes (i.e. cancer, behavioural changes, kidney damage, etc.). The studies I chose to look at were those that measured internal exposure (human hair, blood, umbilical cord blood and meconium samples).
Lead Exposure
I have created and included a table outlining the blood lead levels in the three studies that measured it. As you look at the levels reported, recall what was mentioned in my previously posting about lead: no lead exposure levels are deemed safe in children, and the current Canadian threshold for treatment is 10 micrograms per decilitre in blood.
In a previous post I outlined the human health risks of exposure to lead, mercury, and cadmium. At the end of each description I provided a threshold or guidance value set out by the Canadian government, which essentially is the exposure value deemed to be "safe" based on past research. In my literature review I came across a number of studies looking at these particularly toxic heavy metals, and describing the exposure levels among residents of communities where e-waste disassembly is occurring. I found the results of these studies astounding! And you will eventually come to understand why, after reading this post. Given the extensive amount of research that already describes the human health impacts of exposure to these metals, a lot of the studies did not measure health outcomes (i.e. cancer, behavioural changes, kidney damage, etc.). The studies I chose to look at were those that measured internal exposure (human hair, blood, umbilical cord blood and meconium samples).
Lead Exposure
I came across four studies measuring blood lead levels in children, and one study looking at blood lead levels in adults. Unfortunately, one of the studies in children [8], as well as the study of adult exposure [9], could not be retrieved due to a lack of availability. I attempted to retrieve the articles from the main authors but was unsuccessful. In addition, I also found two studies that measured lead content in hair samples. One was conducted in China[10] and the other was conducted in India (one of the very few published studies I looked at that was not conducted in China)[11]. Human scalp hair is a convenient and less invasive biological measure that can reflect long-term lead exposure. However, for this posting I decided to focus on the three studies that looked at blood lead levels, since blood samples are the most commonly used, and also deemed to be the most accurate assessment of human lead exposure [10,12].
I have created and included a table outlining the blood lead levels in the three studies that measured it. As you look at the levels reported, recall what was mentioned in my previously posting about lead: no lead exposure levels are deemed safe in children, and the current Canadian threshold for treatment is 10 micrograms per decilitre in blood.
Click to enlarge!
As is apparent in the table, blood lead levels were strikingly high in the study populations observed overall, but even more so in the children living in Guiyu, where e-wastes are handled. In the Li et al. study, neonatal exposure to lead was highly correlated with parental activities in e-waste recycling. Lead content in meconium was an important measure in this study as it is believed to reflect the accumulation of lead during pregnancy when mothers have elevated levels of lead circulating in their blood [13]. This was important because maternal blood lead levels were not measured. With regards to the two other studies (Huo et al. & Zheng et al.), the control group participants were selected from the same region, Chendian, and essentially the same study design was used, only the studies were conducted two years apart (2004 versus 2006, respectively). In both studies, residence in Guiyu was a risk factor for elevated blood lead levels. In the later study, Zheng et al., additional risk factors were identified: paternal employment in e-waste recycling (OR=4.61; p=0.003), amount of time spent playing outside near the road everyday (OR=1.73; p=0.02), frequency that the child would suck their fingers (OR=2.85; p=0.009), and age (OR=1.72; p=0.033). Even though the study participants in both studies were different, they came from the same population; therefore, it is interesting to compare the mean lead levels and percent of participants with lead levels above 10 micrograms per decilitre in each study. It appears that the blood lead levels are decreasing, although I'm not sure whether this difference is statistically significant. The Chinese government has implemented control measures to reduce pollution produced from primitive e-waste recycling. Such measures include stricter control of the importation of e-waste, and health education campaigns about the dangers and means of preventing lead poisoning [14,15].
Mercury & Cadmium Exposure
In my next post I'll continue to unveil the results from studies on mercury and cadmium exposure.
Wednesday, December 9, 2009
What Are You Leaving Behind?
When you move on to the latest technology, what happens to the items you leave behind?
Well, it gets added to the 4,750 tonnes of lead, 4.5 tonnes of cadmium, and 1.1 tonnes of mercury contained in personal electronics that are disposed in Canada each year, according to Environment Canada [1]. These figures are grossly underestimated.
Slightly more than half of the metals that are found in a typical desktop computer include copper, aluminum, lead, gold, zinc, nickel, tin, silver, and iron, while the remaining portion is composed of platinum, palladium, mercury, cobalt, antimony, arsenic, barium, beryllium, cadmium, chromium, selenium and gallium [2]. One of the reasons we don't want old electronics to end up in landfills is because toxic metals such lead and mercury can leach into the water and soil, and eventually circulate throughout the food chain, and possibly end up in our bodies [2]. While equipment is intact, these heavy metals don't pose a risk to human health, but when electronics are discarded and/or recycled in uncontrolled environments, the hazardous components are released into the environment posing great amounts of risk [2,3]. Despite the fact that most heavy metals are toxic and bioaccumulative at low concentrations, the main heavy metals investigated in the e-waste disposal literature that I found were lead, beryllium, cadmium and mercury.
Health Risks of Lead Exposure
Lead is one of the most commonly used heavy metals -- it is used in both computer and television screens, and in the solder used to anchor various circuit board components. Toxicity tests of laptops, VCRs, printers and remote-control devices have been conducted in the US, and found that a substantial proportion of these electronics exceeded the US safety standards for lead [3]. The main reason for having a product safety standard for lead is because its deleterious effects on human health have been widely studied, and are well known to be quite serious. According to Health Canada, short term exposure to high levels of lead can cause vomiting, diarrhea, convulsions, coma or even death. The main areas of the body affected by lead are the brain, kidney, and nervous system [4]. Once exposed to lead, it can remain in your body for years in bone or circulating through the blood stream [5]. Children are particularly susceptible to lead at even lower levels of exposure, due to increased absorption. The harms noted in children include impacts on intellectual development, behaviour, size and hearing. During pregnancy, lead can also cross the placenta and affect the unborn child. Studies have shown that female workers who are exposed to high levels of lead have more miscarriages and stillbirths [4,5]. In Canada, corrective action is taken when patients present with blood lead levels exceeding 10 micrograms per decilitre [4]. Recent studies have shown that blood lead levels lower than 10 micrograms per decilitre were associated with reduced IQ scores and academic skills [6]; therefore, no level of exposure has been deemed safe.
Health Risks of Beryllium Exposure
Beryllium is sometimes used in circuit boards as an electrical connector and/or to insulate microprocessors [2]. When improperly handled during disposal or recycling, beryllium dust can be released, which is known to cause severe lung disease and lung cancer [6,7]. Exposure thresholds have not been accurately set for beryllium in Canada, according to a recent risk assessment for generic e-waste processing facilities in Canada [8]. Interestingly, this Canadian risk assessment found exposure levels to both lead and beryllium to be above the occupational exposure limits outlined by the ACGIH. The current occupational exposure limit in Ontario is 0.002 micrograms per meter cubed (time–weighted average exposure value) [9].
Health Risks of Cadmium Exposure
The predominant use of cadmium is in rechargeable batteries. In addition to this, cadmium can be found in plastics, cadmium plated steel, solders, and TV picture tubes [10]. Cadmium toxicity can lead to kidney, bone, and pulmonary damage. There are three modes of exposure: dermal, pulmonary (lungs), and gastrointestinal (mouth); cadmium cannot cross the placenta. The main organ for long-term cadmium accumulation is the kidney, hence its toxic effects on the kidney with life-time exposure [11]. Acute toxicity due to cadmium exposure can lead to nausea, vomiting, weakness, shortness of breath, lung edema (fluid in the lungs) and possibly death [11,12]. Chronic exposure has been linked to kidney damage, bone mineral density loss and hypertension [13]. Additionally, the International Agency for Research on Cancer classifies cadmium as carcinogenic, with exposure primarily linked to lung cancer [14]. Currently, there is no Canadian blood cadmium guidance value for the general population; although, according to the preliminary results from the new Canadian Health Measures Survey, the geometric mean blood cadmium in Canadians aged 6 to 79 was 0.35 micrograms per litre [13].
Health Risks of Mercury Exposure
An estimated 22 percent of the mercury used world-wide each year goes into electrical and electronic equipment, including batteries, flat-panel display screens, and switches [15]. Even though very small amounts of mercury are used in these products, very small levels of mercury exposure are known to cause damage to the brain, spinal cord, kidneys, liver and a developing fetus. The human health risks of mercury exposure have been recognized for quite some time, and consequently is a well researched area. To date, mercury exposure is understood to have neurological, renal (kidney), cardiovascular and immunological impacts. In extreme cases, long-term exposure can lead to coma or death. Neurodevelopmental problems in children can also develop as a result of mercury exposure while in the womb [16,17]. Recent studies have noted adverse health events occurring at even lower mercury exposure levels [16]. The Health Canada guidance value for total blood mercury concentrations is 20 micrograms per litre for adults (a threshold value has not been set for children) [17].
Well, it gets added to the 4,750 tonnes of lead, 4.5 tonnes of cadmium, and 1.1 tonnes of mercury contained in personal electronics that are disposed in Canada each year, according to Environment Canada [1]. These figures are grossly underestimated.
Slightly more than half of the metals that are found in a typical desktop computer include copper, aluminum, lead, gold, zinc, nickel, tin, silver, and iron, while the remaining portion is composed of platinum, palladium, mercury, cobalt, antimony, arsenic, barium, beryllium, cadmium, chromium, selenium and gallium [2]. One of the reasons we don't want old electronics to end up in landfills is because toxic metals such lead and mercury can leach into the water and soil, and eventually circulate throughout the food chain, and possibly end up in our bodies [2]. While equipment is intact, these heavy metals don't pose a risk to human health, but when electronics are discarded and/or recycled in uncontrolled environments, the hazardous components are released into the environment posing great amounts of risk [2,3]. Despite the fact that most heavy metals are toxic and bioaccumulative at low concentrations, the main heavy metals investigated in the e-waste disposal literature that I found were lead, beryllium, cadmium and mercury.
Health Risks of Lead Exposure
Lead is one of the most commonly used heavy metals -- it is used in both computer and television screens, and in the solder used to anchor various circuit board components. Toxicity tests of laptops, VCRs, printers and remote-control devices have been conducted in the US, and found that a substantial proportion of these electronics exceeded the US safety standards for lead [3]. The main reason for having a product safety standard for lead is because its deleterious effects on human health have been widely studied, and are well known to be quite serious. According to Health Canada, short term exposure to high levels of lead can cause vomiting, diarrhea, convulsions, coma or even death. The main areas of the body affected by lead are the brain, kidney, and nervous system [4]. Once exposed to lead, it can remain in your body for years in bone or circulating through the blood stream [5]. Children are particularly susceptible to lead at even lower levels of exposure, due to increased absorption. The harms noted in children include impacts on intellectual development, behaviour, size and hearing. During pregnancy, lead can also cross the placenta and affect the unborn child. Studies have shown that female workers who are exposed to high levels of lead have more miscarriages and stillbirths [4,5]. In Canada, corrective action is taken when patients present with blood lead levels exceeding 10 micrograms per decilitre [4]. Recent studies have shown that blood lead levels lower than 10 micrograms per decilitre were associated with reduced IQ scores and academic skills [6]; therefore, no level of exposure has been deemed safe.
Health Risks of Beryllium Exposure
Beryllium is sometimes used in circuit boards as an electrical connector and/or to insulate microprocessors [2]. When improperly handled during disposal or recycling, beryllium dust can be released, which is known to cause severe lung disease and lung cancer [6,7]. Exposure thresholds have not been accurately set for beryllium in Canada, according to a recent risk assessment for generic e-waste processing facilities in Canada [8]. Interestingly, this Canadian risk assessment found exposure levels to both lead and beryllium to be above the occupational exposure limits outlined by the ACGIH. The current occupational exposure limit in Ontario is 0.002 micrograms per meter cubed (time–weighted average exposure value) [9].
Health Risks of Cadmium Exposure
The predominant use of cadmium is in rechargeable batteries. In addition to this, cadmium can be found in plastics, cadmium plated steel, solders, and TV picture tubes [10]. Cadmium toxicity can lead to kidney, bone, and pulmonary damage. There are three modes of exposure: dermal, pulmonary (lungs), and gastrointestinal (mouth); cadmium cannot cross the placenta. The main organ for long-term cadmium accumulation is the kidney, hence its toxic effects on the kidney with life-time exposure [11]. Acute toxicity due to cadmium exposure can lead to nausea, vomiting, weakness, shortness of breath, lung edema (fluid in the lungs) and possibly death [11,12]. Chronic exposure has been linked to kidney damage, bone mineral density loss and hypertension [13]. Additionally, the International Agency for Research on Cancer classifies cadmium as carcinogenic, with exposure primarily linked to lung cancer [14]. Currently, there is no Canadian blood cadmium guidance value for the general population; although, according to the preliminary results from the new Canadian Health Measures Survey, the geometric mean blood cadmium in Canadians aged 6 to 79 was 0.35 micrograms per litre [13].
Health Risks of Mercury Exposure
An estimated 22 percent of the mercury used world-wide each year goes into electrical and electronic equipment, including batteries, flat-panel display screens, and switches [15]. Even though very small amounts of mercury are used in these products, very small levels of mercury exposure are known to cause damage to the brain, spinal cord, kidneys, liver and a developing fetus. The human health risks of mercury exposure have been recognized for quite some time, and consequently is a well researched area. To date, mercury exposure is understood to have neurological, renal (kidney), cardiovascular and immunological impacts. In extreme cases, long-term exposure can lead to coma or death. Neurodevelopmental problems in children can also develop as a result of mercury exposure while in the womb [16,17]. Recent studies have noted adverse health events occurring at even lower mercury exposure levels [16]. The Health Canada guidance value for total blood mercury concentrations is 20 micrograms per litre for adults (a threshold value has not been set for children) [17].
Flame Retardants: Here, There... Everywhere
"We're definitely eating them and probably inhaling a small amount" said Dr. Arnold Schecter, professor of environmental sciences at the University of Texas Health Centre[1].
What Dr. Schecter is referring to are the unseen but widely found polybrominated diphenyl ethers (PBDEs) that are almost guaranteed to be present in any home or area where high-tech electronics like TVs, computers or cell phones have been in use [1,2]. PBDEs are synthetic chemical compounds that are used as flame retardants (chemicals that are added to polymers to prevent fires) in upholstery furniture, carpet backing textiles, foam, plastics, and electrical and electronic equipment [1,2]. Great concern has been expressed by a number of researchers about the escalating volume of PBDEs that are persistent in the environment, and their potential to disrupt endocrine function, neurodevelopment and increase ones risk of cancer [3,4]. It was incredible learning about the various environmental media that PBDEs have been detected in: air, water, sediment, soil and biota from literally across the globe [1,3,4,5]. PBDEs have been detected in human adipose tissue, blood, and breast milk, and repeat measures have proven that these concentrations are clearly increasing over time [1,2,5].
PBDE Exposure and E-Waste
PBDE Exposure and E-Waste
In my literature search I came across two studies that looked at both human internal exposure to PBDEs associated with e-waste disassembly, and two health outcomes thought to be related to exposure: increased cancer incidence and altered thyroid function. Both studies conclude with a plea for more research on the human health impacts of exposure to PBDEs, since there is an apparent gap in the present scientific literature. A number of in vitro and in vivo studies have been conducted, however, few studies have involved humans.
In the study that looked at cancer and PBDE exposure, the researchers assessed the level of internal exposure to PBDEs among a convenience sample of cancer patients living around an e-waste disassembly site [6]. After an increasing incidence of cancer (such as liver and lung cancer) was noted among residents surrounding the e-waste disassembly site, the researchers decided to look at the body burden of PBDEs (through kidney, lung and liver tissue samples) among local cancer patients who visited a surgical ward. Unfortunately, the study did not have a comparison group; therefore the association between cancer and PBDE exposure in e-waste dumping grounds could not be measured. The study could only conclude that the PBDE internal exposure levels for the study participants (174.1-182.3 ng/g lipid) were notably higher than those reported among the European population, but comparable to those reported in the USA population (based on previous exposure studies). The study also identified the main congeners of PBDEs in the tissue samples (PBDE47, PBDE28 and PBDE209), which was consistent with previous measures of food, air particles, e-waste residues and soil samples collected from the disassembly site in an earlier study. These findings suggest that the congeners can enter the body through three main routes of exposure: inhalation, ingestion and skin penetration. Clearly, a more in-depth investigation of the potential association between cancer and PBDE exposure is needed. After a quick literature search of PBDEs and cancer on MEDLINE I could only find one case control study of non-Hodgkin lymphoma published in 1998 [7].
The second study I looked at was a cross-sectional study of residents in two villages: a village close to an e-waste recycling site (exposed group) and a village located 50 km away from the e-waste site (control group) [8]. Due to PBDEs’ structural similarity to thyroid hormone and polychlorinated biphenyls (PCBs), researchers have been investigating their ability to cause thyroid hormone disruption and DNA damage (can lead to cancers and developmental disorders) [8,9,10] . In the following study, the researchers measured serum levels of PBDEs as an indicator of internal exposure, serum levels of thyroid stimulating hormone (TSH) as an indicator of thyroid function (host response), as well as urinary 8-hydroxydeoxyguanosine (8-OHdG) and frequencies of the cytokinesis-block micronucleus (CBMN) as indicators of the presence of DNA damage. Not only was the median concentration of total PBDEs in the atmosphere of the e-waste recycling site 47 times higher than the concentration in the control site (7149 vs. 150 pg/m3, respectively), but also the median serum PBDEs level for the exposed participants was more than twice that in the controls (382 vs. 158 ng/g lipid, respectively). These median serum levels in both villages were notably higher than those previously reported from studies in Spain (median, 12 ng/g lipid)[11], New Zealand (median, 7.17 ng/g lipid) [12], and Japan (median, 2.89 ng/g lipid) [13].
The results also showed that the levels of serum TSH among the participants exposed to e-wastes were significantly increased (p<0.01). A dose response trend was noted between the levels of PBDE exposure and TSH levels; however, the study did not account for other pollutants within the e-waste dismantling site that may have impacted thyroid hormone homeostasis. The study did account for some factors that can affect the balance of TSH in the human body (ie. BMI, age and sex), however, a number of endogenous and exogenous factors were not measured, reducing the strength of these findings.
With respect to DNA damage, there was no evidence of excess oxidative DNA damage (8-OHdG) that may have been caused by PBDE exposure. The researchers suspect that this may have been because the study was underpowered. The CBMN assay did however find that having a history of working with e-wastes increased ones risk for micronucleated binucleated cells (MNed BNC) 28-fold. Current evidence suggests that these MNed BNC can lead to cancer (14,15). Although two in vitro studies were cited as connecting PBDEs to an increased frequency of MNed BNC, exposure to other genotoxic agents found in e-waste but not included in the study, must be accounted for in future studies.
The study findings do suggest that PBDE exposure may indeed interfere with the thyroid hormone system and cause genetic damage, but further studies are needed. The alarmingly high level of exposure in both villages in Zhenjiang emphasizes the need for enhanced surveillance and control of the occupational and environmental exposure to PBDEs, particularly in areas where e-waste recycling is taking place. Even though human health effects of PBDEs exposure are only starting to be increasingly explored, a number of countries and organizations are recommending the mitigation of environmental exposure. The US Environment Protection Agency has already classified decaBDEs (one of three mixtures of PBDEs) as possible human carcinogens [16], and two PBDE congeners have already been banned in the European Union since 2004 [17].
What is Canada doing?
- As of December 2006, Environment Canada and Health Canada have prepared screening assessments on several PBDEs.
o Health Canada’s screening assessment found human PBDE exposure estimates to be well below the levels believed to cause health effects in laboratory animals.
- The Government of Canada now prohibits the production, use and importation of specific PBDEs.
- Health Canada has implemented ongoing food monitoring and will continue to conduct research on PBDEs [2].
Wednesday, December 2, 2009
Climate Change, Mixed Messaging & Superfreakonomics
A really important comment was brought up this week in my environmental epidemiology lecture on climate change. I wanted to speak to this point a bit more after coming across a series of writings about a recently published book called Superfreakonomics by Steve Levitt and Stephen Dubner.
The comment in class was about the difficulty of sorting through the mixed messaging surrounding climate change, and the role that health promoters need to play in order to address this issue at the individual level. This was a really important point to bring up because it is difficult to know what our role as individuals is when we speak about climate change. This point is particularly important not only because lifestyle changes need to be made, but also because political will needs to be generated.
I came across a prime example of this mixed messaging after watching a CBC interview of Steve Levitt. One of the chapters in his new book is about climate change and why he thinks that the current climate change efforts are flawed. This peaked my interest, and so I tried to track down his chapter online. Instead I came across an interesting series of blog posts between the author and a climate change economist. Thankfully, the chapter doesn't refute the current scientific consensus about climate change, but rather argues that the "solution" isn't in changing human contributions to the problem (adaptation strategies). I don't fully understand what aerosol geo-engineering is, but the authors present this as a harmless and cheap quick fix for global warming. Skeptics have highlighted that the authors ignored readily available literature detailing the severe risks associated with aerosol geo-engineering and reasons why emission reduction is the dominant strategy chosen. Even though I have not read the chapter myself, I remain quite uncomfortable with the thought that if this book was even remotely as successful as the first Freakonomics, 4 million people could potentially be mislead by the message in this chapter.
The comment in class was about the difficulty of sorting through the mixed messaging surrounding climate change, and the role that health promoters need to play in order to address this issue at the individual level. This was a really important point to bring up because it is difficult to know what our role as individuals is when we speak about climate change. This point is particularly important not only because lifestyle changes need to be made, but also because political will needs to be generated.
I came across a prime example of this mixed messaging after watching a CBC interview of Steve Levitt. One of the chapters in his new book is about climate change and why he thinks that the current climate change efforts are flawed. This peaked my interest, and so I tried to track down his chapter online. Instead I came across an interesting series of blog posts between the author and a climate change economist. Thankfully, the chapter doesn't refute the current scientific consensus about climate change, but rather argues that the "solution" isn't in changing human contributions to the problem (adaptation strategies). I don't fully understand what aerosol geo-engineering is, but the authors present this as a harmless and cheap quick fix for global warming. Skeptics have highlighted that the authors ignored readily available literature detailing the severe risks associated with aerosol geo-engineering and reasons why emission reduction is the dominant strategy chosen. Even though I have not read the chapter myself, I remain quite uncomfortable with the thought that if this book was even remotely as successful as the first Freakonomics, 4 million people could potentially be mislead by the message in this chapter.
Subscribe to:
Posts (Atom)