HOME
SHOSHONE NEWS-PRESS, Jan. 17, 2002, p. 6.
Common Sense
Truth behind proposed plan's blood lead levels is dizzying
By Ron Roizen, Ph.D.
SNRC Science Committee
First, a quick word about terminology. Regarding blood lead levels in the population, the percentage of children found to be at or above 10 micrograms per deciliter of blood ("µg/dL") is termed the "exceedance rate." An exceedance rate of 5 percent or greater triggers EPA's call for community-wide remediation at Superfund sites.
In a previous article I suggested that "...it is now credibly arguable that fewer than 5 percent of the Silver Valley's children -- whether inside or outside 'the box' -- have blood-lead levels as high as 10 µg/dL." In other words, I suggested that the current exceedance rate should not trigger the EPA's remediation-requiring standard.
Table 4-1: Blood Lead Levels in 1- to 6-Year-Old Children in the Affected Communities in the Coeur d'Alene Basin, Excluding the Bunker Hill Box
<-----------------1996 to 2000 Data--|--2000_Data_Only------>
Age
(years)No. of children tested Percent of children
10-plus µg/dLPercent of children
15-plus µg/dLNo. of children tested Percent of children
10-plus µg/dLPercent of children
15-plus µg/dL1 40 20.0 5.0 18 16.7 11.1 2 46 26.1 17.4 13 15.4 0 3 52 19.2 7.7 18 11.1 5.6 4 57 12.3 5.3 14 21.4 7.1 5 62 8.1 3.2 14 21.4 0 6 46 6.5 2.2 25 4.0 0
My statement would seem to fly in the face of the EPA's own estimate of childhood blood leads as offered in the recently published proposed plan. The EPA's estimate was summarized in Table 4-1 of the plan, which table (reproduced above) presents blood lead results gathered in yearly community surveys from 1996 to 2000.
The EPA table's column for combined surveys (see shaded column) reports that that more than 5 percent of children aged 1-6 exceeded the EPA's 10 µg/dL standard. From 6-year-olds (who showed the lowest exceedance rate, at 6.5 percent) to 2-year-olds (who showed the highest rate, at 26.1 percent), all age groups exceeded the 5 percent standard. Though it is not specifically reported in this table, the aggregate or overall exceedance rate for children aged 1-6 for these survey data was 14.9 percent. (Incidentally, the figures in the "15-plus µg/dL" columns of the table are an included subset of associated figures in the "10-plus µg/dL" columns.)
The table also presents data from the year-2000 survey separately, the purpose being to demonstrate the durability of exceedance rates into year-2000. "Although there is some variability in the data," the PP's text offered, "the composite data from 1996 to 2000 is comparable overall to the blood lead data collected in 2000 alone" (PP, p. 4-1).
Obviously, the EPA's image of the childhood blood-lead situation as offered in this table is sharply divergent from the one offered in my statement.
Yet, the exceedance rates reported in the Proposed Plan's Table 4-1 must be regarded with caution. Consider just two points:
- The Proposed Plan was published on Oct. 29, 2001, some four days after the results of the most recent, 2001 blood-lead survey were announced to the public in Kellogg. The 2001 survey results differed markedly from both the 1996-1999 compliation and the 2000 survey results. The 2001 results reported an overall exceedance rate of 6.1 percent, now only a single percentage point above the EPA's 5 percent exceedance standard. The new exceedance rate figure was almost 60 percent below the 14.9 percent aggregate rate for combined years, as offered in Table 4-1. The new year-2001 results were devastating for EPA's case according to the logic Table 4-1 had reflected. That table had used year-2000 date to validate the rates generated in past years' surveys. Doing so in effect harbored two crucial implications: first, it showed that EPA recognized the importance of demonstrating the up-to-dateness of its blood-lead rates, and, second, it showed that EPA was quite happy to use the most recent year's survey rates to provide that demonstration. Hence, and by the same token, year-2001's much more favorable results in effect invalidated the EPA's claim of year-to-year stability and also provided a considerably more favorable picture of current blood-lead rates. Moreover, if (as one might guess) EPA's Region X received the findings of the year-2001 blood lead survey some days, weeks, or even months before they were released to the public on Oct. 29, 2001, then Region X's use of year-2000 data rather than year-2001 data in Table 4-1 might hint at the possibility of deliberate misrepresentation.
In short, the proposed plan's Table 4-1 relied on survey data that were known by Region X (1) to be nonrepresentative, (2) to be less recent than the newly available year-2001 data, and, finally, (3) to be inappropriate for projecting population exceedance rates.
- Equally noteworthy is the fact that the proposed plan's Table 4-1 employs nonrepresentative survey data -- data that EPA's own Human Health Risk Assessment (HHRA) report asserts should not be used to describe population exceedance rates. There is a story here and it bears a brief re-telling. The SNRC Science Committee objected to the use of nonrepresentative survey data to select between, and validate, divergent exceedance estimates coming from the EPA's "IEUBK" computer simulation model. The Science Committee objected that the EPA's procedure placed far too much weight on known-to-be-inadequate survey data. The Science Committee also objected that these survey data contained multiple counts of the same individuals -- about one-third of all "observations" comprising the 1996-1999 survey data were contributed by individuals counted more than once in the data. EPA scientists apparently heard these objections and took them to heart because the final draft of the HHRA added caveats report saying that these survey data could not be used to describe blood-lead rates in the general population. New warnings of this kind were added at three different points in the revised and final text. The first and most detailed of these warnings included the following words:
The majority of blood lead data available for the Coeur d’Alene Basin are observational and opportunistic based on voluntary participation in health response programs and were not solicited for experimental or survey purposes. Blood lead levels are not, nor were ever, intended to be randomized.[...]2As a result, the representativeness of this self-selected population to all the children in the Basin is unknown, and the true prevalence of blood lead levels exceeding 10 µg/dl cannot be determined with these data. Because there is not a representative database for blood lead levels, this variable cannot be used to quantitatively assess the risk or probability of exceeding blood lead criteria in the overall population, especially for young children. (HHRA, p. 6-4)
Yet, and astonishingly, these same data -- the very same to which the above warnings applied -- supplied the combined-survey figures presented in the proposed plan's Table 4-1.Hence it is the science committee's contention that Table 4-1's exceedance rates have little value in describing the Basin's present childhood blood-lead situation.
1 It may be mentioned in passing that Table 4-1's label indicating "1996 to 2000 Data" is incorrect and the label should instead read "1996 to 1999 Data." Table 6-4b in the full Human Health Risk Assessment (HHRA) reports exactly the same data though it includes no year-2000 data -- and therefore the Proposed Plan's table combines 1996-1999 and not 1996-2000 data.2An ellipsis appeared at this point in my quotation of HHRA text indicating that words had been omitted from the quotation. Unfortunately, the ellipsis was dropped in the newspaper's rendering of the quotation.
"Common Sense" is a column by members of the Shoshone Natural Resouces Coalition's Science Committee. Responses are invited from those who agree and those who disagree.