Clinical and Health Affairs
Lead Posioning in Minnesota Medicaid Children
By Erik W. Zabel, Ph.D., M.P.H., and Susan Castellano
Abstract
This article presents data on blood-lead testing in Minnesota children from 1999 through 2003. The number of Minnesota children younger than 6 years of age who were tested for blood lead increased from approximately 38,000 in 1999 to 61,000 in 2003. The rate of blood-lead testing in 9- to 30-month-old children enrolled in Medical Assistance and MinnesotaCare, the state’s Medicaid programs for children, increased from 17% to 29% between 1999 and 2003. The rate of elevated blood-lead levels (10 µg/dL or greater) in all tested children declined from 6% in 1999 to 2.7% in 2003. However, the rate of elevated blood-lead levels in children enrolled in the 2 public programs in 2003 (3.4%) was 2-fold higher than that in children who were not on Medical Assistance or MinnesotaCare (1.5%). The percentage of all children with elevated blood-lead levels who were retested within 3 months increased from 39% in 1999 to 50% in 2003.
Lead poisoning is one of the primary environmental health concerns for children, yet it can be prevented entirely by eliminating exposure to lead. Fortunately, children’s blood-lead levels have been greatly reduced since the 1970s mainly because of the phase-out of leaded gasoline and the 1978 ban on lead in residential paint. However, old lead paint and contaminated soil remain major sources of lead poisoning for some children.1-4 Toddlers are at the greatest risk because their nervous systems are developing, they absorb lead more easily because of their nutritional needs combined with lead’s ability to substitute for calcium in enzymatic processes, and they spend time on or near the floor and put things in their mouths, making them more likely to come in contact with lead-contaminated soil, paint dust, or paint chips.5 Additional risk factors include living in a home built before 1950 because of the higher lead concentration in paint, having poor nutrition (diets that are high in fat or low in calcium and iron promote lead absorption), having a family member with a lead-related occupation or hobby, having contact with jewelry that contains lead, and using traditional remedies or consuming imported spices or Mexican candy that contain lead. Children who are minorities, have low socioeconomic status, and live in the inner city are disproportionately affected.6,7
Lead poisoning during early childhood is particularly insidious because it usually has no symptoms, and neurodevelopmental effects such as lower IQ and behavioral problems may not appear until the child begins school. Blood-lead testing is the only way to determine whether a child has lead poisoning. Both the Centers for Disease Control and Prevention (CDC) and the Minnesota Department of Health (MDH) consider a blood-lead concentration of 10 μg/dL or greater to be an elevated blood-lead level.
Children who receive health care through the Medicaid program are at increased risk for lead poisoning. The third National Health and Nutrition Examination Survey (NHANES III) found that Medicaid enrollees accounted for 83% of the U.S. children ages 1 to 5 years who had blood-lead levels of 20 µg/dL or greater.8 Regional and local studies have produced similar results; however, the absolute risk for each geographic area depends on characteristics such as the percentage of old housing stock and the presence of industries that use lead. One study in western Minnesota found a 1.6-fold higher risk of elevated blood-lead levels among children on Medical Assistance (MA) and MinnesotaCare (MNCare), the state’s Medicaid programs for children, than among those who were not enrolled in the programs.9 Because of the increased risk of elevated blood-lead levels in children on Medicaid, federal law requires testing of all children receiving Medicaid Early and Periodic Screening, Diagnostic, and Treatment Program services during well-child visits at 1 and 2 years of age, and any visit thereafter through age 5 if the child has not previously been tested.
Analysis of blood-lead testing and Medicaid enrollment data from 1995 through 1998 showed that only 13% of all children ages 9 to 30 months in the state who were enrolled in MA or MNCare were tested for lead in 1998.10 The study also found that among children ages 9 to 30 months who were on MA or MNCare and were seen for a well-child visit, only 23% were tested for lead in their bloodstream. This represented a decline from the 1995 testing rate of 26%, but it was above the national rate of 19% estimated by NHANES III.8 The NHANES III report also described a 2-fold higher risk of elevated blood-lead levels in Medicaid children. The current analysis sought to determine whether this downward testing trend has continued or improved because of efforts by the MDH, the Minnesota Department of Human Services (DHS), and others to expand lead testing of Medicaid children in the state.
Methods
♦ Data Preparation
Laboratories performing blood-lead analyses on people in Minnesota are required by statute to report results to the Environmental Health Division of the MDH. These reports make up Minnesota’s blood-lead surveillance system. Because the DHS maintains separate files on MA and MNCare enrollment, a matching process was necessary to link children’s enrollment and lead-testing information. Children enrolled in MA or MNCare for at least 1 month between January 1, 1993, and December 31, 2003, were selected from DHS data. Corresponding data on blood-lead tests were selected from the MDH lead surveillance database by choosing children age 6 years and younger who were tested for lead poisoning between January 1, 1999, and December 31, 2003. Age was calculated as the difference between the date of the blood-lead sample draw and the child’s birth date. DHS billing data provided the number of MA and MNCare children seen for well-child visits.
♦ Data Linking
An iterative 8-step matching process similar to that used by Castellano et al. and Gyllstrom et al. was used to link children in the DHS database with children in the MDH lead surveillance database.10,11 The variables used in matching were first name, middle name, last name, gender, and day, month, and year of birth. The first iteration matched on exact first and last name and date of birth. The remaining iterations became progressively broader on matching criteria. Results of all iterations after the first were verified on paper. The information on children who matched was then compared with that in the lead surveillance database to determine which tests occurred when the child was enrolled in MA or MNCare.
♦ Calculation of MA and MNCare Enrollment
Total enrollment of children in MA and MNCare to determine blood-lead testing rates was calculated 2 different ways. The first was simply to total the number of children enrolled for any length of time each year who were of the appropriate age at some point during that year. Because some children who met the age criteria were only enrolled for a short period of time during each year, they had less of a chance of being seen for a provider visit during the period of analysis, leading to an artificially low screening rate. In order to take into account the amount of time children were enrolled in MA or MNCare, enrollment was also calculated as the total number of “child years” per calendar year. The number of child years, or full years of enrollment, was calculated by summing the total number of months of enrollment for all children per calendar year and dividing by 12 months. Therefore, one child year is equivalent to one child being enrolled in MA or MNCare for a full calendar year. This was merely a method of accounting for less enrollment time for some children.
♦ Race and Ethnicity Data
Racial and ethnic information was available for children enrolled in MA and MNCare. The race categories were Asian, Pacific Islander/Native Hawaiian, black or African-American, American Indian/Alaskan Native, and white. Because so few children fell in the Pacific Islander/Native Hawaiian category, that category was folded into one called Asian/Pacific Islander. The children who had a reported Hispanic ethnicity were counted as Hispanic in the analysis. The percent of children with unknown or missing racial/ethnic information increased from 4.7% in 1999 to 8.4% in 2003. Those children were included in a separate category for analysis.
♦ Age Categories
Two age groups were used in the analyses—children younger than 72 months of age and those 9 to 30 months of age. The category “children younger than 72 months” was used because this is the group covered by both CDC and MDH blood-lead screening and treatment guidelines. It is also used when describing elevated blood-lead levels and follow-up testing rates. “Children ages 9 to 30 months” was used because this age range captures children who should be tested according to federal Medicaid requirements (age 1 and 2 years at well-child visits) and reflects the higher risk of elevated blood-lead levels in younger children. These 2 age groups allowed a direct comparison with the previous report covering data from 1995 through 1998.10
Results
♦ Linking Process
The MDH Lead Surveillance Database contained information about 213,663 children who were younger than 6 years of age at the time of their lead test between January 1, 1999, and December 31, 2003. The DHS MA and MNCare enrollment dataset contained information on 307,677 children with birth dates between January 1, 1993, and December 31, 2003. A total of 107,649 children were matched between the 2 databases.
♦ Rate of Testing
The number of children tested for blood lead increased between 1999 and 2003. The rate of lead poisoning testing in children enrolled in MA and MNCare was calculated using 3 different denominator populations: the total number of children enrolled in the public programs, calculated “child years” of enrollment, and the number of children seen for well-child visits. Using all 3 denominators, the percentage of MA- or MNCare-enrolled children who were tested for lead also increased between 1999 and 2003. The percentage of children on MA and MNCare ages 9 to 30 months who were tested for lead rose from 16% in 1999 to 29% in 2003. A similar increase—from 14% to 20%—was seen for all children younger than 72 months. When child years were used to calculate the screening rate, the percentage of children 9 to 30 months old tested in 2003 was 51% versus 28% for all children younger than 72 months. The percentage of children with well-child visits who were tested for lead increased from 1999 to 2003, with a higher percentage of children ages 9 to 30 months tested (37% to 53%) compared with children younger than 72 months (27% to 34%).
♦ Rate of Elevated Blood-Lead Levels
The percentage of tested children who had elevated blood-lead levels dropped throughout the period analyzed. For all children younger than 72 months, it decreased from 6.0% in 1999 to 2.8% in 2003. For children ages 9 to 30 months, it fell from 5.2% to 2.6%. However, the rate of elevated blood-lead levels in children on public health programs remained approximately twice that of children who were not enrolled in such programs (3.5% and 1.9%, respectively, in 2003 for all children younger than 72 months; and 3.3% and 1.7%, respectively, for children 9 to 30 months). Figure 2 describes this trend for all children younger than 72 months.
♦ Rate of 3-Month Follow-up Testing
The percentage of children with elevated blood-lead levels who were retested within the recommended 3 months increased from 39% in 1999 to 50% in 2003 for children who were and were not enrolled in public health insurance programs. It remained slightly higher in children who were not on MA or MNCare. The follow-up rate was slightly higher in children younger than 72 months compared with children 9 to 30 months old. However, the percentage of children retested within 3 months remained low at approximately 50% to 60% for both groups.
♦ Race/Ethnicity Analysis
Based on total enrollment, child years, or well-child visits, the rate of testing increased consistently for children ages 9 to 30 months in all racial and ethnic groups. Blood-lead testing rates in 2003 were highest in the unknown race/ethnicity category (40%), followed by the black or African-American (36%), Asian/Pacific Islander (34%), white (30%), Hispanic (25%), and American Indian/Alaskan Native (24%) groups.
Elevated blood-lead levels declined for all racial groups between 1999 and 2003. The rate of elevated blood-lead levels in 2003 was highest among black or African-American (5.0%), Asian/Pacific Islander (4.0%), Hispanic (5.0%), and American Indian/Alaskan Native children (4.3%). Lower rates were found in the white (2.4%) and unknown (2.9%) categories. Follow-up testing increased somewhat for all racial groups between 1999 and 2003. The rate of 3-month follow-up testing in 2003 was highest in the white (59%), unknown (58%), and Hispanic (55%) groups, with lower rates of follow-up in American Indian/Alaskan Native (49%), black or African-American (35%), and Asian/Pacific Islander (40%) children.
Discussion
The number of children tested for lead in Minnesota who were not enrolled in public health care programs increased between 1999 and 2003. The release of statewide screening guidelines in 2000 and ongoing efforts to increase screening in high-risk populations are likely reasons for this increase. Also, the rate of elevated blood-lead levels in tested children decreased during the period studied. This is likely the result of an increase in testing for all children and ongoing efforts to reduce lead sources. It is important to emphasize the fact that the children whose data is entered in the blood-lead surveillance system do not reflect a random sample of Minnesota children. Testing is performed in response to guidelines and/or at the discretion of the health care provider and parent.
Providers cannot test children who are not seen for routine care. To look at only the children enrolled in public health care programs who are seen for preventative visits, we used DHS billing records to calculate the blood-lead testing rate in children seen for well-child visits. Even though the rate of testing during well-child visits increased from 37% in 1999 to 53% in 2003, the rate remained well below the optimal 100%.
Our analysis also found that children enrolled in MA and MNCare had an approximately 2-fold higher rate of elevated blood-lead levels than children who were not enrolled in those programs. This is very similar to the increase that has been shown in national studies and in a previous analysis of Minnesota blood-lead surveillance data.8,9 Improvements have been made in the blood-lead testing rate of children on MA and MNCare. However, more work needs to be done to ensure testing because of the higher risk of lead poisoning among this population and federal laws requiring all Medicaid children to be tested during both 1- and 2-year well-child visits.
The percentage of children with elevated blood-lead levels who were retested within the recommended 3 months increased from 1999 to 2003. Still, the rate among both children who were and were not enrolled in public health programs remained low, approximately 50% to 60%. Even though the population covered by MA and MNCare tends to be more mobile and less likely to use health care resources, improving the follow-up testing rate should be an important goal of future childhood lead poisoning prevention efforts.
The percentage of children with elevated blood-lead levels dropped for all racial and ethnic groups between 1999 and 2003. This occurred among children ages 9 to 30 months and all children younger than 72 months. However, in 2003 the rate of elevated blood-lead levels remained higher in Asian/Pacific Islander, black or African-American, Hispanic, and American Indian/Alaskan Native children who were tested compared with white children and children with unknown race/ethnicity. This correlates with previous studies that showed a higher risk for lead poisoning in minority populations, which is likely due to living in substandard rental housing and living in urban areas that have higher concentrations of lead in soil.6,7
Testing rates were highest in Asian/Pacific Islander and black or African-American children and lowest among Hispanic and American Indian/Alaskan Native children. The follow-up rate increased for all racial groups between 1999 and 2003. However, the 3-month follow-up rate was highest among whites and the unknown group and lowest in the Asian/Pacific Islander and black or African-American groups.
Limitations
Use of the data in this report to describe the overall Minnesota population is limited. First, surveillance data are not a random sample of Minnesota children. Providers and parents make the decision to test each child, and they have a variety of reasons for doing so. Second, the race and ethnicity data used in this report are for children in public health care programs only. The differences between racial and ethnic groups may not be similar in those children who are not enrolled in such programs. Also, race and ethnicity data were missing or unknown for some children (8% in 2003). This may lead to an inaccurate analysis by race/ethnicity. Finally, there may be errors in matching between databases even though attempts were made to search for nonidentical but similar names and birthdates.
Conclusion
Our analysis found that the rate of blood-lead testing in children on MA and MNCare increased from 1999 to 2003 in contrast to the decline observed between 1995 and 1998. The rate of elevated blood-lead levels continued to decline in both children who were and were not enrolled in public health programs. The rate of follow-up testing in children with elevated blood-lead levels remained low, however, and improving follow-up should be a focus of efforts in childhood lead poisoning prevention. Continuing analysis of blood-lead surveillance data for all children will help track Minnesota’s progress toward elimination of childhood lead poisoning by 2010, which is one of the goals of the Healthy People 2010 federal initiative. MM
For blood lead screening guidelines and more information on lead, visit the MDH Lead Program Web site at www.health.state.mn.us/divs/eh/lead. For guidelines specific to children enrolled in Medicaid, see pages C-19 and C-20 in the Child and Teen Checkups Provider Manual at www.dhs.state.mn.us/main/groups/county_access/documents/pub/
dhs_id_000307.hcsp.
This publication was supported in part by a Cooperative Agreement from the Centers for Disease Control and Prevention (US7/CCU522841-03). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention, the Minnesota Department of Health, or the Minnesota Department of Human Services.
Erik Zabel is principal epidemiologist for the Minnesota Childhood Lead Poisoning Prevention Program at the Minnesota Department of Health. Susan Castellano is maternal and child health assurance manager in the Performance Measurement and Quality Improvement Division at the Minnesota Department of Human Services.
References
1. Centers for Disease Control and Prevention. Update: blood lead levels. MMWR. 1997;46(7):141-6.
2. Lanphear BP, Matte TD, Rogers J, et al. The contribution of lead-contaminated house dust and residential soil to children’s blood lead levels. A pooled analysis of 12 epidemiologic studies. Environ Res. 1998;79:51-68.
3. Lanphear BP, Hornung R, Ho M, Howard CR, Eberle S, Knauf K. Environmental lead exposure during early childhood. J Pediatr. 2002;140:40-7.
4. Mielke H, Reagan P. Soil is an important pathway of human lead exposure. Environ Health Perspect. 1998;106(Suppl. 1):217-29.
5. Lanphear BP, Dietrich K, Auinger P, Cox C. Cognitive deficits associated with blood lead concentrations <10 microg/dl in U.S. children and adolescents. Public Health Rep. 2000;115:521-9.
6. Pirkle JL, Brody DJ, Gunter EW, 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.
7. Pirkle JL, Kaufmann RB, Brody DJ, Hickman T, Gunter EW, Paschal DC. Exposure of the U.S. population to lead, 1991-1994. Environ Health Perspect. 1998;106:745-50.
8. U.S. General Accounting Office. Medicaid: elevated blood lead levels in children. 1998; GAO Publication No. GAO/HEHS-98-78.
9. Zabel EW, Falken MC, Sonnabend M, Alms M, Symonik D. Prevalence of elevated blood-lead levels and evaluation of a lead risk screening questionnaire in rural Minnesota. J Environ Health. 2005;68(2):9-15,36.
10. Castellano SE, Boler MJ, Falken M, Zabel E, Michael A, Virnig D. Elevated blood lead levels in Minnesota and the Medicaid population. Available at: http://www.health.state.mn.us/divs/eh/lead/reports/medicaidleadrpt.pdf. Accessed March 22, 2006.
11. Gyllstrom ME, Jensen JL, Vaughan JN, Castellano SE, Oswald JW. Linking birth certificates with Medicaid data to enhance population health assessment: methodological issues addressed. J Public Health Manag Pract. 2002;8:38-44.