A New Slant on Pediatrics’ Report of Falling Elevated Blood Lead
Levels
Lelia M. Coyne, PhD, MST, NE-Certified Lead-based Paint Risk Assessor
Independent Researcher
April 2009
The recent Associated Press article by Lindsay Tanner1 (Based on a
recent Pediatrics article by Jones, et al. March 10, 20092) has left some
of us scratching our heads. This study reports and interprets demographic
trends in testing and the national prevalence of elevated blood lead
levels (EBLLs). Data were from National Health and Nutrition Examination
Survey (NHANES)' "nationally representative samples" surveyed since 1988.
In the article three statements are made which call for serious
consideration both of their meaning as applied to any specific region, and
also the implications of this meaning with respect to national lead poison
prevention policy.
Two of the statements appear in the Abstract.
“The prevalence of elevated blood levels, greater than or equal to 10
micrograms/dL among children decreased from 8.6% in 1988-1991 to 1.4% in
1999-2004, which is an 84% decline.”
“Blood lead testing of Medicaid-enrolled children increased to 41.9%
from 19.2% in 1988-1991.
The first sentence in the discussion states that:
“Blood lead levels (BLLs) in US children continue to decrease most
likely as a result of an intense coordinated effort to control or
eliminate lead sources in children’s environments by government officials,
health care and social service providers, and the communities most at
risk.”
The results reported are substantially inapplicable to conditions I
know to exist, not just in my local community, but also my entire state.
For instance, in no part of Nebraska (NE), do the reporting
statistics parallel the 41.9% stated in the Pediatrics article to describe
the BLL testing level of Medicaid-enrolled children nationally. At its
high point in 2005, NE, statewide, tested 14.6% of its children. Half of
these tests were performed in one of our 93 counties. Six of these
counties reported no screenings at all in 2005, and in another twenty-
nine, less than 5% of children six and under were screened. The Medicaid
status of these children was not, nor is now, systematically reported to
the State. Examining the process of reporting of BLL levels here, it is
not even clear how this could be practically retrospectively determined.
Furthermore, in NE, because of the broad distribution of pre-1950 housing,
it would be legitimate to question whether EBLLs are, in actuality,
clustered among Medicaid children.
Pediatrics’ attributes the improvement to “an intense coordinated
effort to control or reduce exposures to lead in children’s
environments….” This statement controverts strong divergence of emphases
by public health units on lead poison prevention programs even between the
two most populous counties in this state. In one, the Pediatrics’
attribution is accurate to an awe-inspiring degree as the result of
combined efforts between an EPA Superfund cleanup of industrially
contaminated soil, and broadly based community attention to multicultural
outreach, education about prevention of exposure risks, and remediation of
old housing and public schools. The other stoutly maintains lack of a
significant incidence of lead poisoning, in spite of a similarly large
stock of pre-1950 housing and unabated public school facilities. Yet,
both have mirrored the stunning national decrease in the prevalence of
EBLLs (albeit considerably larger in the one relative to the other).
The Pediatrics’ attribution for the extended national decline in
EBLLs cannot be reconciled with disparate NE policies. I therefore
searched for an over-arching explanation for the broadly observed decline
in EBLL prevalence that might amalgamate exposure prevalences in our
“polyglot” society.
An EPA Document3 was compiled which supported the need for a
dramatically lowered acceptable standard for airborne lead. A lowered
standard was recently adopted, in response. Figure 2.8 on page 2-81 may
provide an explanation for numerous disparities that is worthy of more
general consideration. Airborne Pb emissions dropped by 98% between 1970
and 2003. However, the figure caption points out that 5% of this decrease
occurred between 1993 and 2002, well after the complete phase-out of
leaded fuel, which occurred in 1986.
Is it possible that during the period of the NHANES surveys and the
NE surveillance data collection that decreasing national airborne Pb
emissions have remained the dominating factor in lowering BL levels?
If so, rather than being in the midst of “an intense coordinated
effort to control or eliminate lead sources in children’s environments by
government officials, health care and social service providers, and the
communities most at risk” can we, as a nation, not just NE, be still in
very early stages of a widely neglected effort to control exposures of our
children to multitudinous sources of lead? As a society, have we been
falsely cheered by results from a study designed neither to quantify the
degrees of lead poisoning that are being suffered regionally, nor to
evaluate comprehensively the extent of local effectiveness in addressing
specific regional risk factors? Is it the exceptional, rather than the
typical community that is paying serious attention to minimizing the
economic and social consequences of early childhood lead poisoning?
Examination of several other recently published articles4,5,6, which
conduct additional analyses just of the most recent of the NHANES studies,
complicated my consideration of these questions, rather than clarifying
it. Although these articles delineate a number of risk factors, the data
analysis is very complex and hard for all but a very few experts to use.
This is because inferential statistical modeling procedures are employed,
which, fundamentally, are of a different nature from statistical data
presentation methods familiar to most scientists and public health
officials.
The NHANES nationally representative sample found the national
prevalence of BLLs elevated above 10 μg/dL to be 1.4%. In many
specific regions, overall prevalences are acknowledged to be much higher.
How extended can these specific regions be? What is an expected
range of variability about the national figures? How dissimilar do the
local demographic factors have to be from those described by the
“nationally representative sample” to signal existence of a specific local
risk factor, more importantly, to warn that the national figure does not
likely reflect the regional prevalence? How extensive must the testing
coverage be to give a reliably representative picture?
It is difficult even to absorb the meaning of the caution that
“estimates can be generalized only to the US population." Respecting
this, it remains problematic to grasp how these studies can then have
applicable local descriptive or predictive value, and to discern what this
might be.
The Pediatrics article states:
“Since 1997, the Center for Disease Control (CDC) has recommended
that states develop plans to target testing to children at high risk.”
Other Federal Agencies also have long advised us to design and apply
strategies targeted for specific high-risk populations in our own locales.
All of the recent articles restate this directive.
I note that, in spite of the reiterated caution that "nationally
representative samples do not identify or characterize local risks,"
public officials at all levels cite the NHANES national EBLL prevalence as
if it were regionally applicable, even in the absence of such programs.
How many communities are aware of their overall prevalence of EBLLs?
Of those who are, how many have explored the uniformity of this prevalence
throughout their community, and sought to identify vulnerable clusters
within it? How many have tried to correlate EBLL prevalence in their own
community with their prevalences of academic and behavioral problems that
are well proven to be associated with even lower levels of lead exposures?
We have control to alter (and responsibility to do so) local exposure
sources by our community efforts, but in the absence of local surveillance
data we can remain oblivious even of their existence, and certainly are
unable to quantify their direct consequences on the intellectual,
behavioral, and physical health of our citizenry, because the NHANES
surveys do not supply such data. NHANES models, even if their
implications are correctly interpreted, offer only generic guidance to
parents and concerned citizens about specific types of high-risk sources
that may exist in local communities, and the relative risks that they
impose.
Even the national prevalence of 1.4% remains unacceptably high and
this, very possibly, is a serious underestimate of the actual national
prevalence, because many likely scenarios for exposure are not included
explicitly in the survey design. More specific guidance is needed to
inform design of local strategies to reduce lead exposure and to stimulate
us to push for their adoption. To acquire this guidance, we need a means
to identify and quantify risk factors specific to our own localities, via
their measured contribution on local lead exposure prevalences.
Such evidence is provided only via comprehensive, publicly reported
testing. Current surveillance data are most sparse, and results are
essentially inaccessible to any but professional public health
specialists. Most BLL sampling now being done is at the discretion of
private providers, and, for most part, is timed in compliance with rigid
age-based screening intervals. These are prescribed by diverse, loosely
coordinated organizations such as CDC, Medicaid, the American Academy of
Pediatrics, or State Early Childhood Lead Poison Prevention Programs
(ECLPPPs). Their advisories vary widely. Parent compliance with any of
them is subject to widely varying individual accessibility to and
affordability of health care. Physician compliance varies with their
personal diligence actually to provide recommended testing, and then
reliably to report test results to a common database that includes a
modicum of demographic information. Few physicians include responses to
lead exposure questionnaires as a standard component of medical history on
patient intake, or draw BLL samples when a patient complains of vague
symptoms after a household remodeling project, if the activity were even
to be mentioned.
Age-based testing is more apt to capture ongoing chronic, rather than
acute exposures, and thus can be far removed in time and conditions from
the onset of actual exposure of the child, such as from an unsafely
conducted home improvement project or a move. Thus, timely warnings, or
evidence of harm may or may not be provided to primary caregivers and the
broader health community in a way useful to promoting immediately needed
changes of behavior, even among tested children.
Perhaps it is time to get a truly descriptive profile of the
incidence of EBLLs. To do this would require mandated universal screening
over at least the first 7 1/2 years of life using reevaluated and
uniformly applied screening criteria that are designed to capture the bulk
both of chronic and acute exposures. Minimally, EBLL data summaries by
zip code need to be reported to the public. Both parents and concerned
citizens deserve to know how many of our small ones are being subjected to
lead poisoning, when and where it is being acquired, and what they need to
see done to eliminate it. In the lack of surveillance data, continued
NHANES surveys will remain overly restrictive mirrors of the societal
price of lead poisoning.
Lead poisoning in the 21st century remains a weighty problem!
1 Tanner, Lindsay. “Study: Fewer kids have high lead levels. Lincoln
Journal Star, Monday March 2, 2009 (based on a recent Pediatrics article
by Jones Robert L, et al.2).
2. Jones, Robert L., Homa, David M., et al. Trends in blood lead levels
and blood lead testing among U.S. children aged 1 to 5 years, 1988 2004.
Pediatrics [serial online]. 2009;123:e376-e385. Available from: American
Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village,
Illinois, 60007. Accessed March 10, 2009.
2. U.S. Environmental Protection Agency (EPA). “Air Quality Criteria
– Lead” October 2006. Available from: National Center for Environmental
Assessment Research, Triangle Park, NC 27111.
3. Levin Ronnie, Brown, Mary Jean et al., Lead exposures in U.S.
Children, 2008: implications for prevention. Environmental Health
Perspectives. 2008;116(10): 1285-1292.
4. Gaitens, Joanna M., Dixon, Sherry L., et al., Exposure of U.S.
children to residential dust lead, 1999-2004 I. Environmental Health
Perspectives [serial online]. March 2009;117(3):461-467.
5. Dixon, Sherry L, Gaitens, Joanna M., et al., Exposure of U.S.
children to residential dust lead, 1999-2004 II. Environmental Health
Perspectives [serial online]. 2009;117(10): 468-474).
Conflict of Interest:
None declared
eLetters: ![[Read eLetters]](/icons/misc/sectionDOWN.gif){kind=icon}
