Nonlinear association between soil lead and blood lead of children in metropolitan New Orleans, Louisiana: 2000–2005

Available online at www.sciencedirect.com

Science of the Total Environment 388 (2007) 43 – 53 www.elsevier.com/locate/scitotenv

Nonlinear association between soil lead and blood lead of children in metropolitan New Orleans, Louisiana: 2000–2005 ☆ Howard W. Mielke a,⁎,1 , Chris R. Gonzales a , Eric Powell a , Morten Jartun b , Paul W. Mielke Jr. c a

Xavier University of Louisiana, College of Pharmacy, 1 Drexel Drive, New Orleans, Louisiana, 70125, USA b Geological Survey of Norway, NO-7491 Trondheim, Norway c Department of Statistics, Colorado State University, Fort Collins, CO 80523, USA Received 1 June 2007; received in revised form 27 July 2007; accepted 3 August 2007 Available online 19 September 2007

Abstract Metropolitan New Orleans is unique because it has a universal blood lead (BL) screening dataset (n = 55,551) from 2000–2005 spatially coupled with a soil lead (SL) dataset (n = 5467) completed in 2000. We evaluated empirical associations between measurements of SL and BL exposure responses of children in New Orleans by stratifying the databases by Census Tracts and statistically analyzing them with permutation methods. A consistent curvilinear association occurred annually between SL and BL with robust significance (P-values b 10− 23). The mathematical model of the pooled BL datasets for 2000–2005 is: BL = 2.038 + 0.172 × (SL)0.5 (agreement (ℜ) of 0.534, an r2 of 0.528, and a P-value of 1.0 × 10− 211) indicating that chance alone cannot explain the association. Below 100 mg/kg SL children's BL exposure response is steep (1.4 μg/dL per 100 mg/kg), while above 300 mg/kg SL the BL exposure response is gradual (0.32 μg/dL per 100 mg/kg). In 1995, the BL prevalence was 37% ≥ 10 μg/dL for the most vulnerable poor and predominantly African-American children. In the era of universal screening the prevalence of elevated BL is 11.8% ≥ 10 μg/dL for the general population of children. The SL map describes community variations of potential BL exposure. If health effects occur at BL ≥ 2 μg/dL, then 93.5% of the children in New Orleans are at risk. These results reinforce the proposal that prevention of childhood Pb exposure must include SL remediation as demonstrated by a New Orleans pilot project and a proactive Norwegian government program. © 2007 Elsevier B.V. All rights reserved. Keywords: Childhood lead exposure; Urban soil lead remediation; Health disparity; Environmental justice

☆

The views are those of the authors and do not necessarily reflect the positions or policies of the funding agencies or the Louisiana Office of Public Health. ⁎ Corresponding author. Center for Bioenvironmental Research at Tulane and Xavier Universities, 1430 Tulane Avenue SL-3, New Orleans, LA 70112, USA. Tel.: +1 504 988 3889. E-mail address: [email protected] (H.W. Mielke). 1 Present address: Department of Chemistry, Tulane University, New Orleans, LA 70118 USA and Center for Bioenvironmental Research at Tulane and Xavier Universities, 1430 Tulane Avenue SL-3, Tulane University, New Orleans, LA 70112, USA. 0048-9697/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2007.08.012

1. Introduction Previous assessments of the association between lead-contaminated soil and lead exposure have found widely varying relationships between soil lead concentrations (SL) and children's exposure as measured by blood lead (BL) (Reagan and Silbergeld, 1989; Xintaras, 1992). This study describes the empirical association of SL measurements paired with BL exposure responses of

44

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

children living in residential communities of metropolitan New Orleans. For perspective on SL, New Orleans is located on the Delta of the Mississippi River. The sediments from the Mississippi River are the parent materials of the alluvial soils of metropolitan New Orleans, and currently contain only trace amounts of lead (Pb) (median = 5 mg/kg) (Mielke et al., 2000). Despite the clean quality of the parent soils, the city of New Orleans now has ten census tracts (U.S. Census Bureau enumeration districts) with a median SL of ≥ 1000 mg/kg (Mielke et al., 2006a). Urban soils integrate all dust sources of Pb, including deteriorated lead-based paint (or its haphazard removal by power sanding, sand blasting, etc.), leaded gasoline emissions, and incinerator or industrial Pb emissions that have accumulated in the environment (Mielke, 1999, 2005). Soils then are both a sink and a source of Pb dust. In metropolitan New Orleans, two SL surveys have been conducted (Mielke et al., 2005b). The first survey (Survey I) was completed in 1991 and was the basis for the first evaluation of the association between SL and BL (Mielke et al., 1997). The BL data were obtained from the Louisiana Office of Public Health in 1995. The exposure responses of BL to both age of housing (HA) and SL were evaluated (Mielke et al., 1997). In that study the P-value of the association between BL and HA (10− 12) was 12 orders of magnitude larger than the P-value between SL and BL (10− 24) (Mielke et al., 1997). This fact focused our attention on SL as a major factor in children's Pb exposure response as measured by BL. Detailed evaluation of the association between SL and BL of New Orleans children more clearly demonstrated the nonlinear association between SL and BL and the larger implications to society of the observed relationship (Mielke et al., 1999). The mathematical model of the SL and BL relationship was: BL = 3.06 + 0.33 (SL)0.5 (correlation coefficient = 0.69 between modeled BL and observed BL and P = 3.5 × 10− 22) (Mielke et al., 1999). A similar association between SL and BL was found by Johnson and Bretsch (2002) in their study of Syracuse, NY. The purpose of this study is to review the relationship between SL and BL using two new databases completed since the year 2000. The SL data for this study is derived from Survey II of New Orleans completed in 2000 (Mielke, 2002; Mielke et al., 2002, 2005b). In 2000, the Louisiana Department of Health began conducting universal BL screening of children living in metropolitan New Orleans. The present study describes and evaluates the empirical associations between Survey II SL measurements and childhood BL for the years 2000– 2005 within residential communities of metropolitan

New Orleans. Also, the findings from 2000–2005 were compared with those from 1995. 2. Materials and methods Two datasets, SL and BL, were stratified by census tracts and assembled for this study. 2.1. Survey II soil lead (SL) data The SL dataset was assembled from samples collected on the top 2.5 cm of the soil surface within residential neighborhoods of metropolitan New Orleans (Mielke et al., 2005b). Wherever possible, 19 samples per census tract were collected as described previously (Mielke et al., 2005a,b). The soil samples were stratified by 1990 Census Tracts (n = 286). Sampling was conducted by using U.S. Census Bureau maps as a guide (U.S. Census Tracts and Block Number Areas, 1993). Overall, Survey II included 5467 surface samples collected from 286 census tracts (Mielke et al., 2005b). The extraction is based on room temperature leachate methods using 1 M nitric acid (HNO3), a scheme that correlates well with total methods (Mielke et al., 1983; U.S. EPA, 1996). The method has the advantage of more closely resembling physiologic conditions compared with extraction methods based on the use of boiling and concentrated HNO3. The extraction protocol requires mixing 0.4 g of dry and sieved (#10 USGS—2 mm) soils with 1 M HNO3 and agitated at slow speed on an Eberbach shaker for 2 hours at room temperature (∼22 °C). The extract is then centrifuged (10 min at 1600 ×g) and filtered using Fisherbrand P4 paper. The extract is stored in 20 ml polypropylene scintillation vials until analyzed. A Spectro Analytical Instruments CIROS CCD Inductively Coupled Plasma Atomic Emission Spectrometer (ICP-AES) is used to analyze the metals in each sample. The ICP-AES is calibrated with NIST traceable standards, and a laboratory reference, at a rate of 1 per 15 samples, is analyzed during each run. Internal laboratory references included one low SL sample from New Orleans City Park and one high SL sample from the junction of Elysian Fields and Interstate 10 in the inner city of New Orleans. Duplicate extractions are included for every 15 samples. The final SL database is the median result of all samples collected in each census tract of metropolitan New Orleans (Mielke et al., 2005b). 2.2. 2000–2005 blood lead (BL) data The BL dataset of children less than 6 years old was organized by the Section of Environmental

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

45

Fig. 1. Map of SL in metropolitan New Orleans according to the results of Survey II, and the number of times BL was included in each Census Tract during the 6 years of screening. This study indicates that SL is strongly associated with BL exposure response; the darker the shade of red the higher the SL and the higher the corresponding median BL of children living in that community of New Orleans.

Epidemiology and Toxicology and the Louisiana Childhood Lead Poisoning Prevention Program (LALCLPPP), Louisiana Office of Public Health. The LALCLPPP program follows the Centers for Disease Control and Prevention (CDC) protocols for collection, preparation, and analysis of BL results (U.S. CDC, 1991). Each BL was geocoded and matched to the corresponding census tract. If there were less than five children with BL samples in a given census tract for a particular year, then that census tract was excluded. Each year the main variable of the dataset is the yearly number of census tracts with BL results that could be matched (i.e. paired) with SL results. The SL and BL datasets include the following information: median SL concentration for all census tracts matched with the median BL for census tracts with test results of five or more children.

2007). A more detailed discussion of the statistical method follows: The chance corrected agreement measure (ℜ) between n pairs of values, (xi, yi) for i = 1,…, n, is given by

2.3. Data analysis

Specifically δ, the mean absolute difference between xi and yi for i = 1,…,n, is the simple version of the multivariate randomized block permutation (MRBP) statistic where there are only two blocks, and μ is the expected value of δ under the null hypothesis that each of the n! possible orderings of either x1,…,xn or y1,…,yn are equally likely (Mielke and Berry, 2007). As defined,

Data analysis included evaluation of the paired SL and BL data by Census Tract for each year and for pooled data for all 6 years. To avoid making any artificial distributional assumptions, permutation methods are used to evaluate all results (Mielke and Berry,

< ¼ 1  d=l where d ¼ n1

n X

jxi  yi j

i¼1

and l ¼ n2

n X n X i¼1

jxi  yj j:

j¼1

46

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

Table 1 Description of the yearly databases used to determine the exposure response of BL to measurements of SL for the children of metropolitan New Orleans Year

Tracts matched SL and BL Excluded tracts Included tracts SL used BL used BL excluded % ≥10 μg/dL % ≥2 μg/dL 2a 45b 32b 23b 20b 29b 69b 218

1995 158 2000 280 2001 284 2002 286 2003 285 2004 282 2005 263 2000–2005 1680

156 235 252 263 265 253 194 1462

2157 4432 5279 4963 4995 4781 3665 28,115

5640 9756 11,107 11,635 11,846 7823 3384 55,551

65 103 82 66 58 70 156 535

37.0 20.0 14.3 12.0 11.5 10.5 10.7 c 11.8

NA 92.8 93.5 91.4 93.7 91.2 94.1 c 93.5

For 1995, 156 Census Tracts were included with matched SL and BL data. For the 2000–2005 data sets, a total of 286 Census Tracts had SL data available for matching with BL. The universal screening prevalence for BL ≥ 10 μg/dL and ≥2 μg/dL are given respectively for each year. The bottom row lists the median prevalence for 6 years. Notes: a 2 Census Tracts of Survey I lack SL data while in Survey II all Census Tracts had SL data. b Census Tracts with less than 5 BL samples for a given year. c Median Prevalence (%) during 6 years from 2000–2005.

− 1 b ℜ ≤ 1, ℜ = 1 implies perfect agreement, ℜ N 0 implies agreement, and ℜ ≤ 0 implies no agreement. If xi = yi for i = 1,…, n, then perfect agreement exists between xi and yi for i = 1,…,n, δ = 0, and ℜ = 1. Thus the closer ℜ is to 1, the higher the agreement will be. If ℜo is the observed value of ℜ, then the probability that ℜ ≥ ℜo under the null hypothesis is the probability value (P-value) in question. Incidentally, if δo is the observed value of δ, then this P-value is also the probability that δ ≤ δo. Since n! is extremely large for each application of this paper, the exact P-value is not attainable. Also because each exact P-value is very small for each of the applications, the preferred approximate resampling P-value would be uninformative, i.e., 10 − 6 for each case if the number of resamplings is 1,000,000 and δo is assumed to be one of the 1,000,000 resamplings. Thus an approximate Pearson type III P-value based on the exact mean, variance, and skewness of δ under the null hypothesis is Table 2 Metropolitan New Orleans associations between soil lead (SL) and children's blood lead (BL) exposure response for 1995 and 2000–2005 Year

ℜ

r2

P-value

1995 2000 2001 2002 2003 2004 2005 2000–2005

0.536 0.572 0.480 0.577 0.560 0.561 0.447 0.534

0.602 0.618 0.357 0.650 0.606 0.621 0.476 0.528

9.2 × 10− 27 1.5 × 10− 40 4.3 × 10− 36 1.7 × 10− 30 7.5 × 10− 42 4.7 × 10− 39 2.3 × 10− 24 1.0 × 10− 211

Notes: The agreement measure is ℜ, the squared correlation coefficient is r2, and the P-value is given for each year and for the pooled data for all years, 2000–2005.

obtained for each application. As defined, ℜ is simply a standardized version of δ. Also ℜ has an advantage over the Pearson product–moment correlation coefficient, r, since (1) ℜ is a measure of agreement rather than a measure of linearity, and (2) ℜ is a far more stable (robust) value than r since ℜ is based on ordinary Euclidean distances whereas r is based on squared Euclidean distances (Mielke and Berry, 2007). There are a number of complex polynomials that can be used to fit paired SL values with BL values. The simplest expression is in the form x = a + byc where x denotes BL and y denotes SL. 3. Results Figure 1 is a map of SL in metropolitan New Orleans according to the results of Survey II, as well as the number of times each census tract was included in the dataset during the 6 years of BL collection. Fig. 1 shows that during each year, BL screening was conducted throughout most areas of metropolitan New Orleans. All of the data used to evaluate the association between SL and BL are described in Table 1. The same SL data from Survey II, completed in 2000, were used for matching BL for each year of analysis from 2000–2005. In each of the years, some census tracts were excluded because the number of BL samples collected did not meet the minimum criteria of 5 BL samples per census tract (see Fig. 1). Table 1 shows the number of tracts available for matching based on all of the census tracts with BL data during each year, the number of tracts excluded (i.e. census tracts with b5 BL results), the number of included census tracts (i.e. available census tracts minus excluded census tracts), the actual number of SL

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

Fig. 2. The curvilinear association between median soil Pb and median blood Pb exposure response for 1995 compared with 2000–2005. Note that the curves for 2003, 2001, 2002 and 2005 are essentially the same and thus rest on top of each other.

samples and BL samples used for each year, and the actual number of BL excluded. Overall, 535 BL results (about 1%) were excluded. The smallest number of census tracts for a single year occurred in 2005 because on 29 August, the blood lead screening program was Table 3 The predicted median BL exposure response associated with median SL of children living in metropolitan New Orleans from the mathematical model BL = 2.038 + 0.172 × (SL)0.5 Median SL (mg/kg)

Predicted median BL (μg/dL)

5 10 20 25 50 80 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800

2.4 2.6 2.8 2.9 3.3 3.6 3.8 4.5 5.0 5.5 5.9 6.3 6.6 6.9 7.2 7.5 7.7 8.0 8.2 8.5 8.7 8.9 9.1 9.3

47

curtailed after the catastrophic flooding of New Orleans by the storm surge from Hurricane Katrina. In addition to the results for 1995, the final dataset includes 55,551 BL samples collected in 286 census tracts between 2000 and 2005. Table 1 also gives the prevalence (%) of children with BL of ≥ 10 μg/dL and ≥ 2 μg/dL, for 1995 and 2000–2005, respectively. The last line of Table 1 indicates the median prevalence of childhood exposure response during the 6 years of pooled BL data. Table 2 provides the descriptive statistical data for each year, including 1995, of BL sampling. ℜ is relatively large (ranging from 0.447 to 0.577) indicating a robust measurement of agreement between SL and BL of matched data by census tract for each year as well as for the pooled data. The corresponding P-value for each year indicates that the fit between SL and BL is highly significant and not due to chance. The nonlinear relationships between median SL and median BL stratified by census tracts of metropolitan New Orleans are presented in Fig. 2. For comparison purposes, Table 2 lists the results of the 1995 BL dataset and SL dataset in New Orleans (Mielke et al., 1999). In Fig. 2, the 1995 dataset indicated more extreme BL exposure results than observed during 2000–2005. Table 3 lists the median SL associated with the predicted median BL of children living in residential census tracts. In metropolitan New Orleans the median SL ranges from 6.2 mg/kg to 1789 mg/kg. The lowest median BL is 2.4 indicating that in addition to SL, other

Fig. 3. A scatter plot of the pooled SL and BL data for 2000–2005 (n = 1680). The model of the association between median SL and median BL exposure response is: BL = 2.038 + 0.172 × SL)0.5. Despite the apparently large range of scatter, the overall association between SL and BL is extremely strong.

48

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

sources of Pb play a role in exposure. While soils are a sink and also a source of Pb dust from multiple sources, other sources exist such as food, water, toys, Pb-based paint, etc. Not having measures of a large number of Pb sources is a limitation of this study. SL is however, a common source for most children living in metropolitan New Orleans. The pooled results of 2000–2005 are given in Fig. 3 along with a scatter plot of all of the data used for the evaluation. Least sum of absolute deviations regression is used to model the data for all matched cases for the pooled dataset of 2000–2005. The mathematical model of the relationship between SL and BL is best described by the following formula: BL = 2.038 + 0.172 × (SL)0.5. The measure of agreement (ℜ) is 0.534, the squared correlation coefficient (r2) is 0.528 and the P-value is 1.0 × 10− 211 indicating that chance is extremely unlikely to explain the association found in this study. 4. Discussion 4.1. Historical perspective on products containing lead Processes that account for accumulated Pb in urban soils include at least the two high product volume materials that contained lead: lead-based paint and tetraethyl lead (TEL) additives to gasoline (Clark et al., 1991). Each product contained at least 6 million metric tons of Pb or a combined total of at least 12 million metric tons of Pb in the U.S.A. The first high product volume material, lead-based paint, generally receives most of the attention. The peak use of lead-based paint occurred in the 1920s and sharply declined by the 1940s; it was banned from household use in 1978. When lead-based paint deteriorates or is removed by power sanding or sandblasting the lead accumulates in soils; most of the lead-based paint appears to remain intact as a coating on residential buildings (Mielke et al., 2001). Because of the product history of lead-based paint, age of housing (HA) is often used as a surrogate for Pb in the environment (Jacobs et al., 2002; Sutton et al., 1995; Sargent et al., 1995). However, Pb dust did not accumulate in soil in a pattern that is congruent with HA alone. Studies conducted in Minnesota indicated relatively low conformity between HA and BL (Mielke et al., 1989). The second high product volume substance was tetraethyl lead (TEL). TEL was introduced in the 1920s as an additive for enhancing octane of gasoline. The use of this product was controversial from the beginning but assurances were given that the benefits outweighed the health concerns (Lewis, 1985; Rosner and Markowitz,

1985; Nriagu, 1990). TEL use increased slowly until the 1940s. By the 1940s Pb from TEL exceeded the amount of Pb in paint products. In the years following the 1940s TEL use rose steeply and peaked in the late 1960s and early 1970s. Its use declined first to protect the catalytic converter and later for concerns about health with the rapid phase down in 1986 (Airborne Lead Reduction Act, 1984). TEL was directly associated with the traffic volume and dispersed into the environment in the form of small dust particles. The Pb loading by millions of tons of Pb dust particles resulted in environmental SL accumulations in residential communities. SL conforms to traffic volume during the era of the use of TEL in gasoline. This connection explains why SL is significantly lower in older, but low traffic flow communities of North Minneapolis, and higher in newer but high traffic flow communities of South Minneapolis and the observation that SL in the oldest communities with the low traffic, such as the small town of Rochester, Minnesota contains only a fraction of the SL than the communities of South Minneapolis with high traffic (Mielke et al., 1989). These early studies also indicated that children's BL corresponded with SL in the same community and were precedence for the current study (Mielke et al., 1989). Because of the enormity of the amount of Pb dust generated daily, banning TEL from gasoline, unlike banning Pb in paint, resulted in an immediate decline of blood Pb (Brody et al., 1994; Pirkle et al., 1994). However, despite the banning of TEL in gasoline for highway usage in the U.S., and lead-based paint for household uses, the legacy of the use of these two high product volume consumer materials remains accessible as fine particles of Pb dust that accumulated in the soils, especially within inner-city urban environments (Mielke, 1999; Mielke et al., 2007). 4.2. Comparison between 2000–2005 and 1995 datasets and health disparities There is a major difference (Tables 1 and 2, and Fig. 2) between the curve for the SL and BL first described for the 1995 data (Mielke et al., 1997, 1999) compared with the curves for the 2000–2005 datasets. The 1995 dataset reflects the children who obtained their health care from public health clinics and the BL data was biased toward the poor and African-American portion of the population and represents the most lead-exposed children of society (Mielke et al., 1997, 1999). The era of universal screening, the years 2000–2005, resulted in obtaining BL data from the general population of New Orleans children throughout the city as illustrated in Fig. 1. Thus,

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

the 2000–2005 dataset is from a range of socio-economic groups of children. The exceptionally high BL results in 1995 therefore reflect the health disparity observed in the African-American group of children compared with the general population of children observed in 2000–2005. Improved understanding is advancing regarding factors contributing to higher Pb exposure in the AfricanAmerican population than observed in the general New Orleans population. One factor is environmental; as a group African-American children of New Orleans live in the most lead contaminated areas of the city (Mielke et al., 1999; Mielke, 1999). Another factor is physiological; African-American children generally have higher serum vitamin D concentrations due to skin pigment (Kemp et al., 2006). Serum vitamin D increases the gastrointestinal absorption of Pb and the movement of Pb from bone stores resulting in a higher disposition for BL increases among African-American children than for children with less skin pigment and lower serum vitamin D (Kemp et al., 2006). 4.3. Blood lead and health effects Recent advances in understanding the health effects from Pb exposures as low as 2 μg/dL are being demonstrated. First, IQ deficits from Pb exposure are being recognized as nonlinear, and exposure less than the U.S. guideline (b 10 μg/dL) results in a larger decline in IQ (7.4 points per 10 μg/dL) than blood Pb over the entire range of exposures (4.6 I.Q. points per 10 μg/dL) (Canfield et al., 2003). A non-linear dose–response of BL and intelligence quotient was also found in a pooled dataset of seven international prospective studies (Rothenberg and Rothenberg, 2005). In New Orleans the neurotoxic concern was indicated by 4th grade students whose achievement scores were directly associated with the amount of Pb (and other metals) accumulated in the soils of each school district (Mielke et al., 2005a). The amount of Pb accumulated in preKatrina New Orleans affected children's health and welfare. Other research demonstrates that BL is significantly associated with both myocardial infarction and stroke mortality, at BL levels N2 μg/dL (Menke et al., 2006). In addition, research demonstrates associations between human Pb exposure and a range of chronic diseases and health effects including cataracts, kidney function, osteoporosis and diabetes (Lin et al., 2003; Schaumberg et al., 2004; Campbell et al., 2004; Tsaih et al., 2004). Thus, there are multiple observed health effects from Pb exposure that begin in childhood and extend over the entire human lifespan (Schwartz and Hu, 2007).

49

The nonlinear BL–response to SL below 100 mg/kg combined with the nonlinear health effects of low BL exposure exacerbates the problem for children. In accord with the nonlinear relationships, children are far more sensitive to lower amounts of Pb in soils than is generally recognized. The current Healthy People 2010 goal is no child with a BL N 10 μg/dL by the year 2010 (Healthy People 2010). In pre-Katrina New Orleans, 11.8% of children were excessively Pb exposed. Even more disconcerting is that when the median SL is 400 mg/g (the U.S. SL standard), the median BL is 5.5 μg/dL (see Table 3), and when the median SL is 5 mg/kg, the median BL is 2.4 μg/dL. Both of these exposure levels are N2 μg/dL indicated for neurotoxicity and many diseases states (Canfield et al., 2003; Menke et al., 2006). For 2000–2005 the average BL prevalence ≥ 2 μg/dL for New Orleans children was 92.8%. If health effects research continues to find BL exposures of ≤ 2 μg/dL then the amount of Pb accumulated in cities such as New Orleans support the prediction by the late Clair Patterson (1980) that millions of tons of Pb have virtually made older U.S. cities uninhabitable. Without corrective action, the children returning to New Orleans are likely to return to an environment with the same or, because of lack of Pb-safe practices during most of the renovation, even higher exposure risk than they experienced before the catastrophic flooding of New Orleans. 4.4. Comparison of New Orleans exposure response results to U.S. government agency standards The current U.S. EPA Pb standards for residential soils are 400 mg/kg for bare soils in play areas and 1200 mg/kg for the rest of the property (U.S. EPA, 2001; U.S. HUD, 1999). U.S. agencies have evaluated the association between SL and BL. In general, agency studies rely on a mix of data often derived from smelter and mining sites, as well as limited urban data, to model associations between SL and BL (Reagan and Silbergeld 1989; Xintaras, 1992). One critical feature is that the soil measurements are reported at SL above 250 mg/kg (ATSDR Health Consultation) or even commencing with 400 mg/kg (U.S. EPA IEUBK). ATSDR reports a change of 1.2 μg/dL BL per 1000 mg/kg SL (from 250 to 1250 mg/kg). The EPA uses the Integrated Exposure Uptake Biokinetic Model (U.S. EPA IEUBK). In batch mode for children the lowest SL number given is 400 mg/kg; EPA reports a 7.2 μg/dL increase for a 1000 mg/kg increase from 500 to 1500 mg/kg SL. The empirical results from New Orleans provide a fundamentally different understanding about SL

50

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

measurements and BL exposure response by children. Above a median SL of 300 mg/kg, New Orleans children experience a median BL exposure of 0.32 μg/dL per 100 mg/kg SL. However, at exposures below 100 mg/kg, there is over a 4 fold increase of exposure sensitivity; the median BL response is 1.4 μg/dL per 100 mg/kg. From the empirical perspective of New Orleans children, the nonlinear curve has major implications to health and welfare in U.S. society. The mechanism connecting SL with BL exposure response is most likely the ubiquitous hand-to-mouth and teething activities exhibited by young children everywhere. When SL decreases, hand lead also decreases (Sayre et al., 1974; Thornton et al., 1994; Viverette et al., 1996; Mielke et al., 1996; Nielsen and Kristiansen, 2004). For children at play, the larger the amounts of Pb dust in soils of play areas, the larger the amount of Pb that appears on hands and the greater the potential for children to ingest increased amounts of Pb in this route of exposure. The potential for Pb exposure from the surface layer of the soil is orders of magnitude larger than the potential for exposure from indoor floors (Mielke et al., 2007). Furthermore, when there is drought, and soils are exceptionally dusty, this condition leads to higher Pb dust loading and inhalation of Pb dust, and the BL of children increases; whereas in wetter periods when soils contain more moisture, BL decreases (Laidlaw et al., 2005). Several actions to reduce SL and protect children have been tested (Hilts, 1996; Mielke et al., 1992; Sayre et al., 1974). Because of the importance of the topic of remediation, we briefly describe two actions to prevent childhood exposure to SL, a pilot project in New Orleans and a proactive national program initiated by the Norwegian Parliament. 4.5. Recover New Orleans pilot project In New Orleans a project was conducted whereby Pb contaminated properties (n = 25) were treated with a cover of 15 cm of low SL Mississippi River alluvium obtained at the Bonnet Carré Spillway (BCS) (Mielke et al., 2006a,b). There were several phases of soil collection on the project properties and these included: pre-treatment (median surface soil Pb was 1051 mg/kg, range 5–19,627); after BCS cover the median SL decreased to 6.3 mg/kg (range 3–18); just before Katrina, a soil collection was partially completed and finally, a post-Katrina collection on all 25 properties. Twenty-three of the treated properties were flooded. The objective was to compare Pb changes of the pre-Katrina with the post-Katrina soil collection. After catastrophic

flooding the clean soil remained relatively undisturbed; the soil had small increases of median SL of 12 mg/kg for vacant lots and 6 mg/kg for properties with homes. Processes accounting for Pb increases include Pb-based paint abatement on one property, home construction on the vacant lots, and resuspension and deposition of Pb dust from neighboring properties. New Orleans has a huge reservoir of clean sediment passing by the city at an average rate of about 300 metric tons per minute. As part of the post-Katrina recovery, the combined benefits of Pb-safe paint renovation (Mielke et al., 2001) plus clean soil cover (alone estimated as a one time expense of $225–$290 million) should outweigh the estimated annual $76 million cost of Pb poisoning of children (Mielke et al., 2006a,b). The amount of money (U.S. $300 million) needed to improve the environment to protect children is miniscule compared to the money ($100+ billions) being spent to rebuild and renovate 80% of New Orleans from flood damage after Hurricane Katrina. 4.6. Norway's action plan to remedy soils for small children Clean playground areas intended for use by young children are a high priority in Norway. The Norwegian Parliament has decided to investigate all playgrounds at day care centers, parks and elementary schools in Norway, starting with the 10 largest cities and 5 most important industrial areas in 2007. A total number of 2000 day care centers are to be investigated through 2009. The Geological Survey of Norway (NGU) has studied urban soil pollution in several Norwegian cities (Ottesen and Volden, 1999; Ottesen and Langedal, 2001; Jartun et al., 2003). Traffic emissions, building materials, and city fires have contributed to high levels of different pollutants such as lead (Pb), zinc (Zn), Table 4 The Norwegian soil guidelines for inorganic and organic compounds Pollutant

Action level (mg/kg dw)

Guideline concentrations, most sensitive area use (mg/kg)

Lead (Pb) Arsenic (As) Cadmium (Cd) Mercury (Hg) Nickel (Ni) Benzo(a)pyrene PAH16 PCB7

100 20 10 1 135 0.5 8 0.5

60 2 3 1 50 0.1 2 0.01

The action level is for all areas with residential soil and the guideline concentrations are for play areas intended for children.

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

mercury (Hg), polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), and dioxins in urban soils of the older and central areas of cities. Compared to New Orleans, the contamination of soils in Norwegian cities is considered moderate. However, the possible contamination of sand and soil in sensitive areas such as playgrounds and day care centers may give small children, which often spend their entire day outdoors playing close to the soil, an unnecessary additional exposure to hazardous chemicals. The NGU has recently carried out a specific study of sand and soil contamination in surface materials in day care centers in Oslo, the capital of Norway. The survey was administered by the NGU in close cooperation with city government, the Norwegian Pollution Control Authority, and the Norwegian Institute of Public Health. A total of 6754 soil samples (surface, 0–2 cm) from 699 public and private day care centers were collected and analyzed for the content of heavy metals, PAHs and PCBs. Action levels for different pollutants in soil from day care centers are given in Table 4 and provides perspective on guidelines and activities to protect children in Norwegian cities (Haugland et al., 2005). Norway's guideline of 100 mg/kg for SL is the highest permitted Pb concentration on playgrounds in Oslo, and one forth the U.S. bare SL standard of 400 mg/kg for play areas for children or one twelfth the U.S. SL standard (1200 mg/ kg) for other areas of a yard (U.S. HUD, 1999; U.S. EPA, 2001). Table 4 also lists the Norwegian soil guidelines on sensitive areas intended for children's play for Pb (60 mg/kg) and several other metals and organic compounds. Research about SL and concern for quantities at a fraction of the U.S. standards is ongoing in other countries (e.g. Ljung et al., 2006). The action levels are set by the Norwegian Institute of Public Health in cooperation with the NGU. If the action level concentration of one of the pollutants listed in Table 4 is exceeded in one single sample from a day care center, a plan for remediation/improvement of the soil must be initiated. Action levels may however differ slightly from area to area depending on the potential exposure pathways, such as soil ingestion, dermal contact, homegrown vegetables on the property, and use of groundwater. Thirty-five percent of the 699 day care centers in Oslo needed to carry out an action plan on their properties. All day care centers must also remove CCA-impregnated wood. The city of Oslo has estimated a total cost of about $6 million in 2006 and about $5 million in 2007 to complete these remediation plans. Individual plans must be made for each day care center depending on the topography of the given area. However, the recommendations of proper actions are: a)

51

Remove the polluted soil down to 8–12 inches depth, b) cover with a geotextile (fiber cloth), and c) cover the area with clean materials (sand or soil) from approved distributors. The materials must meet the requirements of most vulnerable area use as set by the Norwegian Pollution Control Authority (see Table 4). 5. Conclusion The robust nonlinear association between SL and BL provides critical information about an additional approach to resolving the national health issue of excessive childhood Pb exposure, and predicts that reducing SL in residential areas will also reduce children's BL. Meeting the U.S. goal of “no children with an elevated blood lead level (10 μg/dL or higher)” will require enlightened understanding and activities for reducing the Pb exposure in New Orleans (Healthy People 2010). Remediation of SL is a means for primary prevention of children's Pb exposure. Efforts to remediate or encapsulate contaminated soils, as piloted in New Orleans and initiated as a proactive program in Norway, should be encouraged to prevent Pb exposure of children. Acknowledgements Funding: ATSDR/MHPF cooperative agreement and HUD Lead Technical Study grant to Xavier University of LA. Assistance with obtaining the Metropolitan New Orleans 2000–2005 blood lead data: Kathleen Aubin, Dianne Dugas, Section of Environmental Epidemiology and Toxicology, Louisiana Office of Public Health, and Ngoc Huynh, Louisiana Childhood Lead Poisoning Prevention Program. Thanks also to Tina Mielke for editorial suggestions. References Airborne Lead Reduction Act of 1984: Hearings on S.2609 before the U.S. Senate, Committee on Environment and Public Works, 98th Cong., 2nd Session. ATSDR Health Consultation. Analysis of risk factors for childhood blood lead levels El Paso, Texas, 1997–2002 El Paso County Metal Survey El Paso, El Paso County, Texas http://www.atsdr. cdc.gov/hac/PHA/elpaso2/elp_p1.html [Accessed March 1, 2007]. Brody DJ, Pirkle JL, Kramer RA, Flegal KM, Matte TD, Gunter EW, et al. Blood lead levels in the U.S. population: phase 1 of the Third National Health and Nutrition Examination Survey (NHANES III, 1988 to 1991). JAMA 1994;272(4):277–83. Campbell JR, Rosier RN, Novotny L, Puzas JE. The association between environmental lead exposure and bone density in children. Environ Health Perspect 2004;112(11):1200–3. Canfield RL, Henderson Jr CR, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP. Intellectual impairment in children with blood lead

52

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53

concentrations below 10 mg per deciliter. N Engl J Med 2003;348: 1517–26. Clark S, Bornschein R, Succop P, Roda S, Pease B. Urban lead exposures of children in Cincinnati, Ohio. Chem Speciat Bioavailab 1991;3 (3/4):163–71. Haugland T, Ottesen RT, Volden T, Jartun M. Jordforurensning i OBYbarnehager innenfor Ring 2 (Urban soil pollution in central city day care centers, Oslo, Norway). NGU Rep 2005;064:128 [in Norwegian]. Healthy People 2010. http://www.healthypeople.gov/Document/HTML/ Volume1/08Environmental.htm#_Toc490564710 [Accessed 5 March 2007]. Hilts SR. A cooperative approach to risk management in an active lead/ zinc smelter community. Environ Geochem Health 1996;18:17–24. Jacobs DE, Clickner RP, Zhou JY, Viet SM, Marker DA, Rogers JW, et al. The prevalence of lead-based paint hazards in U.S. housing. Environ Health Perspect 2002;110:A599–606. Jartun M, Ottesen RT, Steinnes E. Urban soil pollution and the playfields of small children. J Phys IV Fr 2003;107:671–4. Johnson DL, Bretsch JK. Soil lead and children's blood lead levels in Syracuse, NY, USA. Environ Geochem Health 2002;24(4):373–85. Kemp FW, Neti PVSV, Howell RW, Wenger P, Louria DB, Bogden JD. Elevated blood lead concentrations and vitamin D deficiency in winter and summer in young, urban children. Environ Health Perspect 2006. doi:10.1289/ehp.9389 [Online 18 December 2006. http://dx.doi.org/, accessed 28 February 2007]. Laidlaw MAS, Mielke HW, Filippelli GM, Johnson DL, Gonzalez CR. Seasonality and children's blood lead levels: developing a predictive model using climatic variables and blood lead data from Indianapolis, Indiana, Syracuse, New York and New Orleans, Louisiana (USA). Environ Health Perspect 2005;113(6):793–800. Lewis J. Lead poisoning: a historical perspective EPA journal — May 1985; 1985. [http://www.epa.gov/history/topics/perspect/lead.htm, Accessed 2 February 2007]. Lin JL, Lin-Tan DT, Hsu KH, Yu CC. Environmental lead exposure and progression of chronic renal diseases in patients without diabetes. NEJM 2003;348:277–86. Ljung K, Selinus O, Otabbong E, Berglund M. Metal and arsenic distribution in soil particle sizes relevant to soil ingestion by children. Appl Geochem 2006;21(9):1613–24. Menke A, Muntner P, Batuman V, Silbergeld EK, Guallar E. Blood lead below 0.48 mmol/L (10 mg/dL) and mortality among US adults. Circulation 2006;114:1388–94. Mielke HW. Lead in the inner-cities. Am Sci 1999;87:62–73. Mielke HW. Research ethics in pediatric environmental health: lessons from lead. Neurotoxicol Teratol 2002;24(4):467–9. Mielke HW. Lead's toxic urban legacy and children's health. Geotimes 2005;50(5):22–6. Mielke PW, Berry KJ. Permutation methods: A distance function approach. 2nd Edition. Springer Verlag; 2007. Mielke HW, Anderson JC, Berry KJ, Mielke PW, Chaney RL. Lead concentrations in inner city soils as a factor in the child lead problem. Am J Public Health 1983;73:1366–9. Mielke HW, Adams JL, Reagan PL, Mielke PW. Soil-dust lead and childhood lead exposure as a function of city size and community traffic flow: the case for lead abatement in Minnesota. Environ Geochem Health 1989;9:253–71 (suppl). Mielke HW, Adams JE, Huff B, Pepersack J, Reagan PL, Stoppel D, et al. Dust control as a means of reducing inner-city childhood Pb exposure. Trace Subst Environ Health 1992;25:121–8. Mielke HW, Viverette LR, Brisco M, Scott L, Gonzales C, Schaefer J. Surface dust on hands as a method for assessing and preventing

childhood lead exposure. In: Johnson BL, Xintaras C, Andrews Jr JS, editors. Proceedings of the 2nd international conference on hazardous waste: Impact on human and ecological health. Princeton, NJ: Princeton Scientific Publishing; 1996. p. 151–5. Mielke HW, Dugas D, Mielke PW, Smith KS, Smith SL, Gonzales CR. Associations between lead dust contaminated soil and childhood blood lead: a case study of urban New Orleans and rural Lafourche Parish, Louisiana, USA. Environ Health Perspect 1997;105(9): 950–4. Mielke HW, Smith MK, Gonzales CR, Mielke Jr PW. The urban environment and children's health: Soils as an integrator of lead, zinc and cadmium in New Orleans, Louisiana, U.S.A. Environ Res 1999;80(2):117–29. Mielke HW, Gonzales CR, Smith MK, Mielke Jr PW. Quantities and associations of lead, zinc, cadmium, manganese, chromium, nickel, vanadium, and copper in fresh Mississippi delta alluvium and New Orleans alluvial soils. Sci Total Environ 2000;246:249–59. Mielke HW, Powell E, Shah A, Gonzales C, Mielke PW. Multiple metal contamination from house paints: consequences of power sanding and paint scraping in New Orleans. Environ Health Perspect 2001;109:973–8. Mielke HW, Gonzales CR, Powell ET, Shah A, Mielke Jr PW. Natural and anthropogenic processes that concentrate Mn in rural and urban environments of the lower Mississippi River Delta. Environ Res 2002;90(2):157–68. Mielke HW, Berry KJ, Mielke PW, Powell ET, Gonzales CR. Multiple metal accumulation as a factor in learning achievement within various New Orleans communities. Environ Res 2005a;97(1):67–75. Mielke HW, Gonzales C, Powell E, Mielke Jr PW. Changes of multiple metal accumulation (MMA) in New Orleans soil: preliminary evaluation of differences between Survey I (1992) and Survey II (2000). Int J Environ Res Public Health 2005b;2(2):84–90. Mielke HW, Powell ET, Gonzales CR, Mielke Jr PW, Ottesen RT, Langedal M. New Orleans soil lead (Pb) cleanup using Mississippi River alluvium: need, feasibility and cost. Environ Sci Technol 2006a;40(08):2784–9. Mielke HW, Powell ET, Gonzales CR, Mielke PW. Hurricane Katrina´ s impact on New Orleans soils treated with low lead Mississippi river alluvium. Environ Sci Technol 2006b;40:7623–8. Mielke HW, Powell ET, Gonzales CR, Mielke Jr PW. Potential lead on play surfaces: evaluation of the “PLOPS” sampler as a new tool for primary lead prevention. Environ Res 2007;103:154–9. Nielsen JB, Kristiansen J. Remediation of soil from lead-contaminated kindergartens reduces the amount of lead adhering to children's hands. J Expo Anal Environ Epidemiol 2004;2004:1–7. Nriagu JO. The rise and fall of leaded gasoline. Sci Total Environ 1990;92:1-28. Ottesen RT, Langedal M. Urban geochemistry in Trondheim, Norway. NGU Bull 2001;438:63–9. Ottesen RT, Volden T. Jordforurensning i Bergen Urban soil pollution in Bergen. NGU Rep 1999;99.022:27 [in Norwegian]. Patterson CC. An alternative perspective—lead pollution in the human environment: origin, extent, and significance. lead in the human environment. National Research Council, National Academy of Sciences. Nation Academy Press, Washington, D.C. 1980; pp 265–349. Pirkle JL, Brody DJ, Gunter EW, Kramer RA, Paschal DC, Flegal KM, et al. The decline in blood lead levels in the United States: the National Health and Nutrition Examination Surveys (NHANES). JAMA 1994;272:284–91. Reagan PL, Silbergeld EK. Establishing a health based standard for lead in residential soils. In: Hemphill, Cothern, editors. Trace

H.W. Mielke et al. / Science of the Total Environment 388 (2007) 43–53 substances in environmental health, vol. 23. Environ Geochem Health; 1989. p. 199–238. [(Suppl) to Volume 12, (1990)]. Rosner D, Markowitz G. A ‘gift of God’?: the public health controversy over leaded gasoline during the 1920s. Am J Public Health 1985;75(4):344–52. Rothenberg SJ, Rothenberg JC. Testing the dose–response specification in epidemiology: Public health and policy consequences for lead. Environ Health Perspect 2005;113(9):1190–5. Sargent JD, Brown MJ, Freeman JL, Bailey A, Goodman D, Freeman Jr DH. Childhood lead poisoning in Massachusetts communities: its association with sociodemographic and housing characteristics. Am J Public Health 1995;85:528–34. Sayre JW, Charney E, Vostal J, Pless IB. House and hand dust as a potential source of childhood lead exposure. Am J Dis Child 1974;127:167–70. Schaumberg DA, Mendes F, Balaram M, Dana MR, Sparrow D, Hu H. Accumulated lead exposure and risk of age-related cataract in men. JAMA 2004;292:2750–4. Schwartz BS, Hu H. Adult lead exposure: time for change. 1st article of a mini-monograph, “lead exposure and health effects in adults: evidence, management, and implications for policy”. Environ Health Perspect 2007;115(31):451–92. Sutton PM, Athanasoulis M, Flessel P, Guirguis G, Haan M, Schlag R, et al. Lead levels in the household environment of children in three high-risk communities in California. Environ Res 1995;68:45–57. Thornton I, Watt JM, Davies DJA, Hunt A, Cotter-Howells J, Johnson DL. Lead contamination of UK dusts and soils and implications for childhood exposure: an overview of the work of the Environmental Geochemistry Research Group, Imperial College, London, England 1981–1992. Environ Geochem Health 1994;16(3/4):113–22. Tsaih S-W, Korrick S, Schwartz J, Amarasiriwardena C, Aro A, Sparrow D, et al. Lead, diabetes, hypertension, and renal function: The normative aging study. Environ Health Perspect 2004;112(11): 1178–82.

53

U.S. CDC. Preventing lead poisoning in young children. Atlanta, GA: Centers for Disease Control and Prevention; 1991. U.S. Census Tracts and Block Numbering Areas. New Orleans, LA MSA, Table 32 selected structural characteristics of housing units: 1990; and summary tape file 3A, Louisiana 040, Lafourche Parish 050, Census tracts 140. Washington, DC: U.S. Department of Commerce, Economics and Statistics Administration, Bureau of Census; 1993. p. 565–615. U.S. EPA (U.S. Environmental Protection Agency). Lead; identification of dangerous levels of lead; final rule. 40CFR745. Fed Regist 2001;66:6763–5. U.S. EPA IEUBK Batch mode for SL and BL of children http://www. epa.gov/superfund/lead/ieubkfaq.htm#depth1 [Accessed March 1, 2007]. U.S. EPA. Urban soil lead abatement demonstration project. Volume 1: EPA Integrated Report. EPA/600/P-93/001aF. Research Triangle Park, NC: U.S. Environmental Protection Agency, 1996. U.S. HUD. (U.S. Housing and Urban Development Agency). Requirements for notification, evaluation and reduction of leadbased paint hazards in federally owned residential property and housing receiving federal assistance. Fed Regist 1999;64(178): 50140–231. Viverette L, Mielke HW, Brisco M, Dixon A, Schaefer J, Pierre K. Environmental health in minority and other underserved populations: benign methods for identifying lead hazards at day care centres of New Orleans. Environ Geochem Health 1996;18:41–5. Xintaras C. Impact of lead-contaminated soil on public health U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Agency for Toxic Substances and Disease Registry 1992; 05/01/1992. http://wonder.cdc.gov/wonder/ prevguid/p0000015/p0000015.asp [Accessed on March 1, 2007].