|
Environmental Health Perspectives
Online, Volume 112, Number 5, April 2004
Lead poisoning causes significant public health problems, including
learning disabilities, behavioral problems,
and even death. According to the Centers for Disease Control and
Prevention, blood lead levels (BLLs) above
10 micrograms per deciliter impair learning and behavior in
young children. Early detection and treatment
can limit and sometimes reverse the health problems associated
with elevated BLLs.
In 2002, 92,000 children were tested for elevated BLL in Michigan,
yet just 4.4% showed a BLL greater than
10 micrograms per deciliter. To avoid unnecessary testing, the
Centers for Disease Control and Prevention
funded researchers at Michigan State University (MSU) in East Lansing
and the Michigan Department of Community
Health (MDCH) in Lansing to create an improved screening tool. The
result is the MSU-MDCH BLL Test Program, an
online questionnaire, available at
http://midata.msu.edu/bll/,
that effectively predicts which children are likely to be at
increased risk for lead poisoning. Only those at risk undergo a blood
test, thereby saving laboratory costs.
In Michigan, Medicaid pays $11 for a blood lead test, and private
estimates range up to $30 per test.
Although relatively inexpensive,
thousands of children are tested for lead poisoning yearly, and public
health agencies often pay the bills. The new
website improves cost- effectiveness by identifying those at risk. "It
provides the most bang for limited bucks," says Stan Kaplowitz,
a professor of sociology at MSU and head of
the team.
Previous screening methods to evaluate lead poisoning have
traditionally relied on zip code and/or Medicaid
information. The Michigan survey is far more comprehensive and
asks, for instance, about the child's current
and previous addresses, frequency of pacifier use (which
slightly reduces risk, Kaplowitz says, perhaps because if a child has
a pacifier in his mouth, he's not putting other possibly lead-
contaminated objects there), and exposure to
peeling paint. The statistical calculations also incorporate detailed
information obtained from census block
groups, small geographic units that cover 1,000-4,000 people living in
a neighborhood. Characteristics of census
block groups include the proportion of houses built before 1950,
rental properties, percentage of people who did not graduate
from high school, and proximity to emissions
from industry and roadways. In addition, the evaluation takes into
account personal factors, such as family
income, whether water is obtained through lead pipes, and whether
siblings in the same home have ever been known to have an elevated
BBL.
The website "implements fairly complicated mathematical equations to
predict lead risk based on where a child
lives, characteristics of the neighborhood, and individual
information," says Kaplowitz. The researchers
used BLL data collected over four years from 90,000
one-year-old Michigan children to map correlations
between BLL and census block data, family history, and other
factors to predict who is at risk. Answers to the
online questionnaire are analyzed statistically and compared to
these correlations. The combined information determines
the probability, based on responses to the questionnaire, that a
child's BLL is sufficiently elevated to
warrant testing.
Kaplowitz and his colleagues designed the MSU-MDCH questionnaire to be
answered by personnel in medical offices,
but anyone can submit answers and receive a recommendation within
seconds (although the questionnaire currently can process only
Michigan addresses). The online questionnaire, the only one of its
type in the nation, provides a model for other states, and the
method may be adapted to fit similar data for other locations.
|
