Artificial Christmas trees: how real are the lead exposure risks?
Exposure to lead has long been recognized as a major public health issue in the United States and other industrialized nations. During the late 1980s and early 1990s, it was discovered that lead would cause permanent and irreversible neurological damage, especially in infants and young children, at far lower exposure levels than previously believed (McMichael et al., 1988; Sciarillo, Alexander, & Farrell, 1992). Although recent data show that baseline blood lead levels have been declining over the last two decades (Mattuck, Beck, Bowers, & Cohen, 2001), lead is still a significant threat to health, particularly in children.
A recent study involving multivariate analysis of 4,853 study subjects ranging in age from 6 to 16 years found reductions in cognitive performance associated with blood lead levels as low as 2.5 micrograms per deciliter ([micro]g/dL) (Lanphear, Dietrich, Auinger, & Cox, 2001). Thus, lead-containing products with even moderate potential to cause human exposure are becoming recognized as having public health significance. One particular study found that even children whose blood lead levels (BLLs) never exceeded the Centers for Disease Control and Prevention (CDC) level of concern (10 [micro]g/dL) could have decreases in IQ of 2.6 points per 10 [micro]g/dL increase in BLLs (Winter, 2001). In the most recent study of the neurological effects of low-level lead exposure, Canfield and co-authors (2003) found a 7.4-point IQ deficit (with a 95 percent confidence interval of 3.2-12.9 points, as measured by the Stanford-Binet Intelligence Scale and the Revised Wechsler Preschool and Primary Scale of Intelligence) as BLLs increased from 1 to 10 [micro]g/dL. This finding strongly suggests that neurological damage may be caused at even lower BLLs. This latest study, which tested 172 children ranging in age from six months to five years and measured nine confounding factors, further underscores the importance of identifying and addressing even relatively small lead exposure concerns.
Several meta-analyses have been done to further investigate the connection between BLLs and IQ deficits. Some researchers (Kaufman, 2001; Pocock, Smith, & Baghurst, 1994) suggest that although there is some evidence that supports the connection, other explanations need considering before definitive conclusions can be drawn on the subject. These researchers fear that recent studies have not adequately allowed for confounding factors and that other shortcomings in the studies may lead to improper conclusions. Other researchers (Needleman & Bellinger, 2001) argue that meta-analysis shows that lead does indeed have a negative impact on IQ, even when multiple variables have been controlled for in different statistical models.
In addition to intellectual effects, studies have connected lead exposure with behavioral and emotional problems, such as aggressive or anxious behavior, in children (Burns, Baghurst, Sawyer, McMichael, & Tong, 1999). Lead may also affect the growth of children, as was documented in a study that measured decreases in height, head circumference, and chest circumference with increasing lead levels (Kafourou et al., 1997).
Within the past 10 years, it has become recognized that polyvinyl chloride (PVC, or vinyl) plastic products often contain relatively large amounts of lead added as stabilizers. In 1995, it was discovered that imported vinyl mini-blinds contained so much lead that the surface dust produced as a result of direct sunlight and heat was resulting in cases of acute lead poisoning of young children who handled and played with them (Norman, 1996; B.C. Lee of U.S. Consumer Products Safety Commission, personal communication to M.F. Toro, July 24, 1996). Studies conducted in 1997 found that several commonly used children's products, such as PVC raincoats, book bags, and beach bags accumulated high levels of lead on surfaces after exposure to sunlight (DiGangi, 1997; Maas, Smith, Patch, & Thornton, 1997).
Artificial Christmas trees made of PVC have become very common in the United States; out of the 76 million family groups in the country, an estimated 50 million households own and use them (Fields & Casper, 2001). Nearly 20 million of the trees in these households are nine years or older (U.S. Bureau of the Census, 2000). Thus, there is a potential for lead exposure from the handling of the trees during assembly, disassembly, and routine usage, as well as from contact with areas underneath the trees. The purpose of this investigation was to begin to determine the potential lead exposure from typical household usage of these products.
Eight 7-foot artificial (PVC) Christmas trees, each from a different major manufacturer, were employed for part of this study. Four of these trees were newly purchased, and the other four had been in residential use for periods ranging from 7 to 17 years. The new trees were sent directly from the manufacturer, and length-of-service documentation for the used trees was provided by Foundation E.A.R.T.H. of St. Louis, Missouri. The manufacturer and tree specifications are summarized in Table 1. A sample of the needles (0.25 grams [g]-0.69 g) was cut from each tree for subsequent lead analysis. Tree needle samples were ashed for five hours at 400[degrees]C before acid/hydrogen peroxide digestion.
Each tree was then assembled in the laboratory by an investigator after a thorough handwashing. Before and after assembly, the subject's hands were wiped with a laboratory wipe to remove all metal/dust material present on the hands. Each wipe was hot-block-digested according to National Institute for Occupational Safety and Health (NIOSH) Method 7082 (NIOSH, 1994) with concentrated nitric acid and 30 percent hydrogen peroxide, and the digestate was analyzed for lead.
Following assembly, a new clean laboratory paper surface 120 centimeters (cm) X 120 cm was placed under each tree. Investigators took wipe samples weekly for four weeks by wiping the entire area with a laboratory wipe. They took control samples by wiping an immediately adjacent treeless laboratory paper surface 120 cm X 120 cm. All wipe samples were digested as described above with nitric acid and hydrogen peroxide. The laboratory where the trees were erected had windows only along a long north-facing wall, so no direct sunlight struck the trees. At the conclusion of the four-week experiments addressing surface dust deposition, each tree was disassembled and placed back into its storage box by a subject, with the hands wiped before and after disassembly (as noted above for the assembly procedure) to determine how much lead was transferred to the hands during disassembly.
All digested wipe samples were analyzed for lead according to Standard Method 3113B (Clesceri, Greenberg, & Eaton, 1998) for electrothermal atomic absorption spectrometry (EAAS). This method basically involves digesting the wipe in a hot concentrated mixture of nitric acid and 30 percent hydrogen peroxide followed by EAAS analysis. Calculations were then made to express the amount of lead on the gloved hands in terms of the total mass (in [micro]g). The lead content in the plastic needles themselves was calculated as micrograms of lead per gram of needle (i.e., [micro]g/g, or ppm), and the lead in the settled dust beneath the trees was expressed as [micro]g/[cm.sup.2] of surface area.
In this experiment, research testing kits were mailed to 127 households that had ordered the kits from Foundation E.A.R.T.H. The availability of the research kit was announced to the public primarily through a Christmas season evening news story carried by 73 NBC news affiliate stations across the United States. Each testing kit contained instructions, a research questionnaire, sample identification labels, a laboratory wipe, one plastic headspace vial, and a pair of laboratory gloves. Individuals were instructed, upon receipt of the kit, to open the plastic vial to have it ready to accept their wipe sample. Next, participants put on the gloves and removed and unfolded the laboratory wipe. A Christmas tree branch section approximately 30 cm in length was selected, and the wipe was carefully wrapped around the branch section. The participants took the wipe sample by applying pressure and pulling the wipe over the entire 30 cm section. After the first wipe sample was completed, the wipe was folded in half so that any dust was on the inside of the fold. A second pass was made with the same wipe; it used a second 30-cm branch section following the same methods. The wipe was folded a second time and placed into the plastic vial, which was then capped. A sample identification label was affixed to the vial, and the kit was mailed back to the authors laboratory, where the wipes were analyzed for lead. Digestion and analysis of the wipes were conducted with the same methodology as in Experiment 1. Calculations were made to express the amount of lead in each wipe in terms of the total mass (in micrograms).
Participants also were asked to fill out a research questionnaire. The questionnaire was used to determine the number and ages of children in the household, child involvement in handling of the Christmas tree, length of tree ownership, manufacturer of the tree, country of origin, approximate age of tree, and location of tree storage in the off-season. The questionnaire also asked what the participant would do with the tree if it was found to have high lead levels. The choices included "discard tree and buy a real tree," "dispose of tree and buy another artificial tree," or "take extra care to avoid lead exposure when setting up and using the tree."
Results and Discussion
Table 1 summarizes the metal concentrations found in the vinyl needles themselves. Two of the used trees exhibited relatively high levels of lead, with the used American Tree sample having high levels of lead. The levels of lead were nondetectable in the other six trees. These results suggest that lead was used more commonly as a PVC stabilizer in the past.
The results of dust wipe samples taken beneath the laboratory-erected trees are summarized in Table 2. Only used-tree specimens CH03 and CH06 exhibited relatively high lead levels in the settled dust, which is consistent with the metal-assay results shown in Table 1. A scenario of actual lead exposure for the sample with the highest lead level, American Tree, can be made under the assumption that young children might crawl and otherwise place their hands on the affected under-tree surface (floor, wrapped presents, etc.) once per week during a four-week Christmas tree season and pick up perhaps 25 percent of the total dust in the 1.49 [m.sup.2] area, for a total of 630 [micro]g. The U.S. Consumer Products Safety Commission (CPSC) has estimated, based on various behavioral studies, that approximately 50 percent of hand-absorbed material will be ingested by a child three years of age or younger (U.S. Consumer Product Safety Commission, 1997), which would entail, in this case, in an approximate acute ingestion of 315 [micro]g. This figure would equate to approximately 0.86 [micro]g/day ingestion spread out over an entire year. California's Proposition 65 requires a warning label if a consumer product results in an average daily lead exposure of 0.5 [micro]g/day or greater. CPSC does not classify a consumer product as hazardous unless it exposes the average user to at least 15 [micro]g/day of lead; neither regulation specifies a limit for acute short-term lead exposures. The dust-wipe data for the Puleo brand tree give an annual exposure estimate of about 0.42 [micro]g/day, just below the Proposition 65 limit, while the remaining products produce estimated daily exposures between 0.02 and 0.16 [micro]g/day. It is important to note that even the six trees with nondetectable lead content produced dust lead levels 4 to 24 times background control levels, which suggests that all of the trees probably contained at least some lead stabilizer. These tests provide only very rough exposure estimates because only a single tree of each brand was tested, but they do suggest at least some lead exposure potential even from new trees.
The mass of lead transferred to subjects' hands during assembly and disassembly of PVC Christmas trees is summarized in Table 3. Again, the results are consistent with the needle lead concentrations shown in Table 1, with trees CH06 and CH03 showing the highest handling transfer levels. Sample CH06 resulted in a total lead transfer from assembly and disassembly of 30.4 [micro]g, which, spread over a four-week Christmas season, translates to a daily exposure of about 1 [micro]g per day and, spread over an entire year, to about 0.1 [micro]g/day. Actual ingestion would most likely be only 10 percent of these amounts for an adult and 50 percent for a young child. Thus, lead exposure from assembly and disassembly would appear to be relatively minor compared with exposure to a child playing around and under a tree.
A total of 127 in-service trees were tested in this part of the experiment, and 42 trees, or 33.1 percent, were observed to have detectable levels of lead in their PVC needles. The lower limit for the analytical method was 1.5 [micro]g of lead. Analysis of the questionnaire data showed that 66.7 percent of the households with detectable levels of lead had children living in the household. Of this number, 47.6 percent had children five years of age or younger, while 23.8 percent had children two years of age or younger in the residence. Many volunteer participants were not able to determine the tree manufacturer or manufacturer location, and thus no conclusions regarding the relationship between the amount of lead in the tree and the trees manufacturer or country of origin can be made from the data.
Of the 47 trees with detectable lead levels, 35.7 percent were stored in the basement, 28.6 percent were stored in the attic, and 14.3 percent were stored in the garage. Christmas tree storage location was considered a relevant questionnaire inquiry; it had been hypothesized that trees stored in locations with higher temperatures might display greater lead levels since heat encourages the breakdown of PVC materials. The data, however, do not support this hypothesis. The attic was assumed to be the hottest storage location, but of the 40 trees that were stored in the attic, only 30 percent had detectable lead levels. Of the 15 trees that were stored in the garage, 40 percent had detectable lead levels. In the basement storage location, which was assumed to be the coolest, 37 percent of 41 trees had detectable lead levels.
Table 4 shows the number of trees in several tree age categories. The mean lead concentration found in kit wipe samples, by tree age category, as well as the percentage of trees in each age category with lead levels that exceeded the detection limit, are shown. Fifteen trees had ages that were listed on the questionnaire as unknown, and results for those trees were not included in the analysis. The data shown in Table 4 suggest that, while the percentage of artificial trees manufactured with at least some lead stabilizer has decreased only modestly, the amount of lead stabilizer used has apparently been reduced to a much larger extent.
The last two columns in Table 4 deal with possible lead exposure from the artificial Christmas trees. Direct mouthing exposure was calculated on the assumption of once-daily 100 percent transfer of lead from a 30-cm tree branch to a child's mouth over a 30-day Christmas season. The scenario that was used for estimating exposure from hand-to-mouth transfer was a once-daily handling of a 30-cm tree branch with a 50 percent transfer rate over a 30-day Christmas season. A young child might likely handle more than one tree branch per day. A typical daily exposure might more commonly involve the handling of three branches per day on a 10-15 year old tree; with this amount of physical contact, the child would be exposed to 6.72 [micro]g/day of lead, or 201.6 [micro]g over the Christmas season.
On average, the data collected from the wipe samples showed that exposure risks are generally relatively low. Some percentage of young children, however, will come into physical contact with their Christmas tree more than once a day, and some children will touch much more than just one tree branch per day. These factors combined suggest that it is possible for a child to be exposed to a far greater amount of lead than the data immediately suggest. For instance, if perhaps the worst-case scenario of child exposure involves a child who mouths 10 tree branches per day on trees that are 20 years old or greater, and in addition handles a total of 300 cm of branch per day, the exposure for this child could be as great as 225.6 [micro]g per day, or 6,768 [micro]g over the 30-day Christmas season (or nearly 20 [micro]g/day averaged over the entire year). When one applies the Food and Drug Administration (FDA, 1993) 0.16 factor for converting [micro]g/day of lead exposure to [micro]g/dL increases in BLL, this value translates to a BLL increase of over 3.0 [micro]g/dL. Extrapolation from the recent work of Canfield and co-authors (2003) produces an estimated IQ deficit for this scenario of about 2.5 points. Although no studies have ever addressed the range of hand-to-mouth contact a young child might have with an artificial Christmas tree, clearly many children will have much higher exposure than was estimated as typical in this study. With an estimated 50 million out of 76 million families owning and using artificial Christmas trees in the United States, and with almost 11.5 million children two years of age and younger living in the United States (U.S. Bureau of the Census, 2000), if even 50 percent of these children lived in a home with an artificial Christmas tree, the most exposed 1 percent of this population (a combination of the most dangerous child behavior and the trees with the highest lead content) would have a much higher exposure level than estimated for the typical child; mathematically, this percentage would translate to about 57,500 children.
The experiments conducted in the first phase of the research do not support the contention that the PVC Christmas trees currently being manufactured represent more than a relatively small lead exposure hazard across the entire population of U.S. children. This conclusion is tempered by the fact that only one tree of each major brand was tested.
In contrast, two of the four older used trees that were tested contained relatively high levels of lead (1,527 and 7,184 [micro]g/g) in the PVC needles. These levels are well in excess of the less than 400 [micro]g/g that CPSC recommends children's products contain, and ingestion calculations indicate that trees of these two types probably expose young children to lead levels at least in the range of California Proposition 65 limits.
While clearly not an acute toxicity threat by themselves, a significant fraction of older artificial trees are probably exposing children and adults to enough lead to be at least a noteworthy public health issue. The experiments described in this paper indicate that it is probably appropriate to caution families--especially families with older PVC Christmas trees, but even families with new ones--to thoroughly wash hands immediately after tree assembly and disassembly, and especially to limit the access of children to areas under erected trees. Direct mouthing contact, frequent branch handling, or both by young children would appear to have the potential for causing lead exposures of considerably greater health significance.
Data collected from the second phase of the authors' research generally confirm that on average, lead exposure from artificial Christmas trees is relatively low. A worst-case scenario, however, would result in very harmful lead exposure. For the safety of all children, it is probably appropriate to request that PVC Christmas tree manufacturers formally commit to banning the use of lead in the PVC formations employed in these products. Until they do so, it would be wise to limit the amount of physical contact that children have with artificial Christmas trees.
TABLE 1 Lead Content of Artificial Christmas Tree Needles ID Type Manufacturer Length of Service Pb (Years) ([micro]g/g) CH 01 New The Christmas House 0 *ND CH 02 New Holiday Tree and Trim 0 ND CH 03 Used Puleo 13 1,527 CH 04 New Christmas Direct 0 ND CH 05 New Tree Classics 0 ND CH 06 Used American Tree 17 7,184 CH 07 Used Hudson Valley Tree 8 ND CH 08 Used General Form Plastic 7 ND Blank ND *ND: less than 25 [micro]g/g. TABLE 2 Lead in Settled Dust Beneath Standing Christmas Trees ID Manufacturer Week 1 ([micro]g/100 [cm.sup.2]) CH 01 The Christmas House 0.127 CH 02 Holiday Tree and Trim 0.0538 CH 03 Puleo 0.424 CH 04 Christmas Direct 0.114 CH 05 Tree Classics 0.298 CH 06 American Tree 2.43 CH 07 Hudson Valley Tree 0.0807 CH 08 General Form Plastic 0.0129 Blank 0.0107 ID Week 2 Week 3 ([micro]g/100 [cm.sup.2]) ([micro]g/100 [cm.sup.2]) CH 01 0.0980 2.66 CH 02 0.0258 0.157 CH 03 1.71 4.98 CH 04 0.207 0.107 CH 05 0.0893 0.157 CH 06 3.20 5.39 CH 07 0.109 0.133 CH 08 0.475 0.0635 Blank 0.0151 0.0258 ID Week 4 Total ([micro]g/100 [cm.sup.2]) ([micro]g/100 [cm.sup.2]) CH 01 0.171 3.06 CH 02 0.539 0.775 CH 03 1.16 8.28 CH 04 0.170 0.596 CH 05 0.114 0.658 CH 06 5.94 17.0 CH 07 0.157 0.479 CH 08 0.151 0.701 Blank 0.0732 0.125 TABLE 3 Lead on Hands After Assembly and Disassembly of Artificial Christmas Trees ID# Assembly Disassembly Blank Wipe Transfer Blank Wipe Transfer ([micro]g) ([micro]g) ([micro]g) ([micro]g) Ch 01 <0.1 0.7 <0.1 0.5 Ch 02 <0.1 0.8 <0.1 0.3 Ch 03 <0.1 6.9 <0.1 2.3 Ch 04 <0.1 0.2 <0.1 0.8 Ch 05 <0.1 0.3 <0.1 0.9 Ch 06 <0.1 15.0 <0.1 15.4 Ch 07 <0.1 4.4 <0.1 1.2 Ch 08 <0.1 0.8 <0.1 0.5 TABLE 4 Mean Lead Mass and Estimated Lead Exposures from In-service Artificial Christmas Trees Age of Trees (Years) 0-5 5-10 10-15 15-20 >20 Number of trees 61 21 16 9 5 in age category Percentage of trees 26.2 23.8 37.5 33.3 100 above detection limit (1.5 [micro]g Pb) Mean lead mass on 1.45 0.91 4.48 2.92 11.28 wipe ([micro]g) Estimated mean total direct 43.5 27.3 134.4 87.6 338.4 mouthing exposure over Christmas season ([micro]g) Estimated mean total 21.8 13.7 67.2 43.8 169.2 hand-to-mouth transfer exposure over Christmas season ([micro]g)
Burns, J.M., Baghurst, P.A., Sawyer, M.G., McMichael, A.J., & Tong, S. (1999). Lifetime low-level exposure to environmental lead and children's emotional and behavioral development at ages 11-13 years: The Port Pirie Cohort Study. American Journal of Epidemiology, 149(8), 740-749.
Canfield, R.L., Henderson, C.R., Cory-Slechta, D.A., Cox, C., Jusko, T.A., & Lanphear, B.P. (2003). Intellectual impairment in children with blood lead concentrations below 10 [micro]g per deciliter. New England Journal of Medicine, 348(16), 1517-1526.
Clesceri, L.S., Greenberg, A.E., & Eaton, A.D. (Eds.). (1998) Standard methods for the examination of water and wastewater (20th ed.). Baltimore: United Book Press, Inc.
DiGangi, J. (1997) Lead and cadmium in vinyl children's products: A Greenpeace expose (Report No. 38). Washington, DC: Greenpeace USA.
Fields, J., & Casper, L.M. (2001) America's families and living arrangements: March 2000 (Current Population Reports, 20-537). Washington, DC: U.S. Bureau of the Census.
Kafourou, A., Touloumi, G., Makropoulos, V., Loutradi, A., Papanagiotou, A., & Hatzakis, A. (1997). Effects of lead on the somatic growth of children. Archives of Environmental Health, 52(5), 377-383.
Kaufman, A.S. (2001) Do low levels of lead produce IQ loss in children? A careful examination of the literature. Archives of Clinical Neuropsychology, 16 (4), 303-341.
Lanphear, B.P., Dietrich, K.N., Auinger, P., & Cox, C. (2000). Cognitive deficits associated with blood lead concentrations <10 [micro]g/dL in U.S. children and adolescents. Public Health Reports, 115(6), 521-529.
Maas, R.P., Smith, R.M., Patch, S.C., & Thornton, L.R. (1997). Assessment of potential lead exposure from children's vinyl products (Tech. Rep. No. 97-043). Ashville, NC: UNC Asheville Environmental Quality Institute, 18.
Mattuck, R.L., Beck, B.D., Bowers, T.S., & Cohen, J.T. (2001). Recent trends in childhood blood lead levels. Archives of Environmental Health, 56(6), 536-541.
McMichael, A.J., Baghurst, P.A., Wigg, N.R., Vimpani, G.V., Robertson, E.F., & Roberts, R.J. (1988). Port Pirie Cohort Study: Environmental exposure to lead and children's abilities at the age of four years. New England Journal of Medicine, 319 (8), 468-475.
Needleman, H.L., & Bellinger, D. (2001). Studies of lead exposure and the developing central nervous system: A reply to Kaufman. Archives of Clinical Neuropsychology, 16(4), 359-374.
National Institute for Occupational Safety and Health (1994). Lead by flame AAS method 7082, i.2. Manual of Analytical Methods (4th ed.). Washington, DC: United States Department of Health and Human Services.
Norman, E. (1996.) Imported, plastic mini-blinds identified as potential source of lead poisoning. North Carolina Department of Environment and Natural Resources. http://www.enr.state.nc.us/newsrels/lead.htm (7 Sept. 2004).
Pocock, S.J., Smith, M., & Baghurst, P. (1994). Environmental lead and children's intelligence: A systematic review of the epidemiological evidence. British Medical Journal, 309(6963), 1189-1197.
Sciarillo, W.G., Alexander, G., & Farrell, K.P. (1992). Lead exposure and child behavior. American Journal of Public Health, 82(10), 1356-1360.
U.S. Bureau of the Census. (2000). Census 2000 Summary. http://factfinder.census.gov (2 June 2004).
U.S. Consumer Product Safety Commission. (1997). CPSC staff report on lead and cadmium in children's polyvinyl chloride (PVC) products. Washington, DC: Author.
U.S. Food and Drug Administration. (1993). Lead in Food, Part II, 58 Federal Register 117 (codified at 21 CFR [section][section] 109, 189).
Winter, M. (2001). Heavy metal: It doesn't take nearly as much lead as the CDC thought to lower a child's IQ. Human Ecology, 29(4), 10-12.
Richard P. Maas, Ph.D.
Steven C. Patch, Ph.D.
Tamara J. Pandolfo
Corresponding Author: Richard P. Maas, Co-director, UNC-Asheville Environmental Quality Institute, CPO #2331, One University Heights, Asheville, NC 28804. E-mail: email@example.com.