Abstract
Background:
Despite declining lead exposure among U.S. children, childhood blood lead level (BLL) undertesting and elevation remains a public health issue. This study explores the impacts of maternal, infant, and neighborhood characteristics on the receipt of lead testing and having elevated BLLs (EBLLs) among children under age two.
Methods:
Pennsylvania infants born in 2015 and 2016 were followed to 24 months. Birth certificate data were linked to 2015 through 2018 blood lead surveillance data and neighborhood data on household income, poverty, and the burden of houses built before 1970. Generalized linear mixed models were used to examine the individual and neighborhood characteristics independently and/or interactively affecting the likelihood of lead testing and of having EBLLs.
Results:
A total of 48.6% of children were tested for BLLs, and 2.6% of them had confirmed EBLLs. The likelihood of lead testing and of having EBLLs among non-Hispanic black children was respectively 7% and 18% higher than white children. Children born to mothers with the lowest educational attainment (<high school), with self-payment as a payment source for delivery, and without WIC enrollment were at higher risk of undertesting. Children living in neighborhoods of the lowest quartile of household income and the highest quartile of poverty and old housing were more likely to have EBLLs. Different neighborhood characteristics modified the associations between some individual factors (such as race/ethnicity, payment source for delivery, and WIC enrollment) and the odds of undertesting and of having EBLLs.
Conclusion:
This cohort analysis provides more accurate estimates of lead screening rates and the percentages of EBLLs than cross-sectional analysis. Some maternal and infant demographics significantly impact the risk of undertesting and of having EBLLs, and some of the effects vary across different neighborhood characteristics. These findings can help lead prevention programs to target screening and treatment resources to children with specific characteristics.
Keywords: birth cohort, blood lead test, elevated blood lead level, maternal and infant demographics, neighborhood characteristics
Introduction
Blood lead levels (BLLs) for U.S. children have declined dramatically over the past several decades.1 However, childhood exposure to lead and elevated blood lead levels (EBLLs) remains the important causes of various health problems, including decreased intelligence quotient, damaged nervous system, developmental delays, and neurobehavioral deficits.2-6 The Centers for Disease Control and Prevention (CDC) updated its recommendations on the blood lead reference value to 5 μg/dL, used to identify children with EBLLs in 2012.7 It is estimated that roughly 500 000 U.S. children under 2 years of age are still at risk of lead poisoning.8 Therefore, state and local health departments need to identify children with EBLLs as early as possible so that they can receive follow-up cares as needed.
Disparities in lead exposure and the burden of lead poisoning persist disproportionately among specific population groups, such as racial and ethnic minorities, parents with relatively low educational attainment, and people who participated in Medicaid or enrolled in The Special Supplemental Nutrition Program for Women, Infants, and Children (WIC).9-13 In addition, children from deprived neighborhoods (higher proportions of homes built before 1970, lower household income, and higher levels of poverty) were associated with increased BLLs.13-17 A recent study indicated the black-white racial gap of BLLs was exacerbated among children living in neighborhoods with higher socioeconomic positions.13 However, limited information is available on how neighborhood characteristics modify associations between demographic characteristics and the likelihood of having EBLLs.
Childhood blood lead screening and follow-up monitoring and care provide information that forms the basis for planning, executing, and evaluation of lead poisoning prevention policies and programs. The Centers for Medicare and Medicaid Services (CMS) requires all children enrolled in Medicaid to receive blood lead screening tests at ages 1 year and 2 years18; however, a large proportion of uninsured or privately insured newborns are not screened. A previous study showed that approximately 57% of children who participated in Medicaid in 9 U.S. states did not receive lead testing by 2 years of age.19 Another study showed that 50% or fewer children in New York, New Jersey, Pennsylvania, and Michigan were screened for BLLs before 6 years of age, and Pennsylvania children had the lowest screening rate and the highest burden of EBLLs.20 Only a few studies have examined independent effects of selected characteristics such as age, race/ethnicity, parental educational attainment, Medicaid enrollment, and neighborhood characteristics on the likelihood of receipt of lead testing.10,11 It is unclear whether community-level characteristics interact with individual risk factors on the likelihood of being tested for BLLs. More studies that combine individual and neighborhood characteristics data are needed to better depict this association.
This study uses a cohort analytic design in which newborns of Pennsylvania resident mothers were followed up to 2 years of age to estimate the rates of lead testing and the percentages of having EBLLs by maternal, infant, and neighborhood characteristics and to evaluate the independent impacts of selected characteristics on the odds of receiving lead testing and of having confirmed EBLLs. The interaction effects between maternal and infant demographics and neighborhood characteristics were also evaluated.
Methods
Data Source
Newborns born to Pennsylvania resident mothers in 2015 and 2016 were followed up to their second birthday and these birth cohorts’ vital statistics data were obtained from birth certificates. Demographic information on maternal and infant characteristics was obtained from the birth certificate and categorized as follows: gender (male or female), race/ethnicity (Hispanic, non-Hispanic white, non-Hispanic black, non-Hispanic Asian, or other), maternal educational attainment (< high school: less than high school graduate; high school/some college: high school graduates or had attended some college but had not received a college degree; ≥ college: college degree or higher; or other), principal source of payment for delivery (private insurance, Medicaid, self-payment, or other), maternal smoking (yes or no: mothers reported cigarette smoking or no cigarette smoking during the 3 months before pregnancy or during pregnancy; or unknown), WIC enrollment (yes or no: mothers participated or did not participate in WIC program; or unknown), maternal infection (yes: maternal infections, including gonorrhea, syphilis, herpes simplex virus, chlamydia, tocolysis, or external cephalic version, were present or treated during pregnancy; no: no maternal infection was present or treated during pregnancy), and maternal risk factors (yes or no: mother had or did not have risk factors, including pre-pregnancy diabetes, gestational diabetes, pre-pregnancy hypertension, gestational hypertension, previous pre-term birth, previous poor pregnancy outcomes, vaginal bleeding, pregnancy resulted from infertility treatment, or previous cesarean, during pregnancy). For children’s neighborhood characteristics, census tract-level information on median household income (household income), the percentage of families and people whose income in the past 12 months is below the poverty level (poverty), and the percentage of housing units built before 1970 (old housing) were obtained from the U.S. Census Bureau 2012 to 2016 American Community Survey 5-Year Estimates.21 Census tracts were ranked based on the percentage of each neighborhood characteristic and were assigned to a quartile for each neighborhood characteristics respectively. Census tract-level neighborhood characteristic data were linked to birth certificate data based on each child’s maternal residential address which was geocoded using ArcGIS (ArcGIS Desktop: Release 10.4.1. Redlands, CA: Esri, 2016).
The Pennsylvania Department of Health (DOH) requires all health care service providers to report all blood lead test results from both venous and capillary specimens for persons under 16 years of age, and most of the reports are submitted electronically through the Pennsylvania National Electronic Disease Surveillance System (PA-NEDSS). All reported data for children who had at least 1 blood lead test from 2015 to 2018, including those collected for screening, confirmation, or follow-up purposes were included in the analyses. In accordance with CDC’s current definition of an EBLL, the Pennsylvania DOH uses a single capillary or venous lead test at or above the reference value of 5 µg/dL to identify children with EBLLs. A confirmed EBLL is defined as a venous lead test ≥5 μg/dL, or 2 capillary lead tests ≥5 μg/dL drawn within 84 days of each other. An unconfirmed EBLL is defined as a capillary lead test ≥5 μg/dL with no other blood test done in the next 84 days.
Data Linkage
Deterministic linkage was used to compare several demographic identifiers (first name, last name, date of birth, gender, and zip code of the residence) across birth certificate data and blood lead surveillance data, and constructed a series of linking steps, starting with the most restrictive criteria to determine whether record pairs agree on all identifiers. If a record did not meet the first round of matching criteria, it was passed to the subsequent linking step for further comparison based on a match on partial identifiers. In situations where full or partial identifiers were incorrect, we matched based on the comparison of encrypted identifiers. For example, the first name and last name that sounded similar but had different spellings were converted by the Soundex coding system. First and last name and birth month and date were also assessed as being potentially transposed during matching.
A simple random sampling method was used to select a subset of the matched records after each step for manual review and validation. Some matched records that failed to be validated by the manual review were put back into the linkage process for subsequent comparison. After completing the linkage process, if a child whose birth certificate data linked to multiple lead test results in the same linking step, we only retained 1 matched record which was linked to the first of multiple lead test results. If a child whose birth certificate data linked to multiple lead test results in different linking steps, we only retained 1 matched record which was linked in an earlier (more restrictive) linking step. Additionally, we manually reviewed a child’s multiple lead test records which were linked to different records in the birth certificate 1 by 1 and only retained 1 of them with optimal validity and reliability.
Birth certitificate data of 278 807 children born to Pennsylvania resident mothers in 2015 and 2016 were used to link to 284 755 children’s blood lead test results archived in PA-NEDSS. After completing the linkage process and the manual review for validation, a total of 149 264 children’s birth certificate data were successfully matched with their blood lead test results.
Statistical Analysis
Descriptive analyses were conducted to explore how the percentages of children tested for BLLs before 12 or 24 months of age and the percentages of tested children with unconfirmed or confirmed EBLLs vary by maternal and infant demographics and by neighborhood characteristics among the 2015 birth cohort and the 2016 birth cohort separately. Separate generalized linear mixed models (GLMMs) were constructed to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) to assess the independent relationships between each potential risk factor with the likelihood of lead testing and of having confirmed EBLLs after adjusting for the random effects of the census tract in the models. Besides all independent variables, two-way interactions between maternal and infant demographics and neighborhood characteristics were incorporated into the model 1 by 1 to explore if the impacts of individual factors on 2 outcomes of interest vary by different levels of neighborhood characteristics. All analyses were performed using SAS software version 9.4 (SAS Institute, Cary, North Carolina, USA).
Results
Overall, 48.3% of children born in 2015 received a blood lead test before 2 years of age, and this percentage increased to 49.0% in the 2016 birth cohort. Non-Hispanic black children had the highest rate of lead testing (63.4% and 63.0% in the 2015 and 2016 birth cohort, respectively), and non-Hispanic white children had the lowest rate (44.1% and 45.3% in the 2015 and 2016 birth cohort, respectively) before 2 years of age. Considering maternal educational attainment, the rate of lead testing was highest among children born to mothers with “high school/some college” educational level. Considering the principal source of payment for delivery, the rate of lead testing was the lowest among children born to mothers with self-payment. Children who enrolled in WIC, whose mothers smoked either before pregnancy or during pregnancy, and whose mothers had infections during pregnancy had higher rates of lead testing. Children who lived in neighborhoods of higher quartiles of poverty and old housing also had higher rates of lead testing (Table 1).
Table 1.
Number and Percentage of Children Tested for BLLs before 2 Years of Age by Maternal and Infant Demographics and Neighborhood Characteristics, 2015 and 2016 Pennsylvania Birth Cohorts.
| 2015 birth cohort | 2016 birth cohort | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total | BLL test < 1 year | BLL test < 2 years | Total | BLL test < 1 year | BLL test < 2 years | |||||
| Na | N | %b | N | %b | Na | N | %b | N | %b | |
| Overall | 137 246 | 37 428 | 27.3 | 66 233 | 48.3 | 135 641 | 37 914 | 28.0 | 66 505 | 49.0 |
| Maternal and infant demographics | ||||||||||
| Sex | ||||||||||
| Female | 67 169 | 18 245 | 27.2 | 32 263 | 48.0 | 65 969 | 18 515 | 28.1 | 32 441 | 49.2 |
| Male | 70 076 | 19 183 | 27.4 | 33 970 | 48.5 | 69 667 | 19 399 | 27.9 | 34 064 | 48.9 |
| Race | ||||||||||
| Hispanic | 14 748 | 3753 | 25.5 | 7822 | 53.0 | 15 110 | 3835 | 25.4 | 7873 | 52.1 |
| Non-Hispanic Asian | 5118 | 1418 | 27.7 | 2633 | 51.5 | 4990 | 1376 | 27.6 | 2523 | 50.6 |
| Non-Hispanic black | 18 073 | 5813 | 32.2 | 11 450 | 63.4 | 17 730 | 5712 | 32.2 | 11 164 | 63.0 |
| Non-Hispanic white | 92 069 | 24 375 | 26.5 | 40 613 | 44.1 | 90 363 | 24 710 | 27.4 | 40 948 | 45.3 |
| Otherc | 7238 | 2069 | 28.6 | 3715 | 51.3 | 7448 | 2281 | 30.6 | 3997 | 53.7 |
| Maternal educational attainment | ||||||||||
| <High school | 17 483 | 4057 | 23.2 | 7772 | 44.5 | 16 661 | 3760 | 22.6 | 7195 | 43.2 |
| High school/some college | 58 822 | 18 111 | 30.8 | 31 849 | 54.1 | 57 583 | 17 616 | 30.6 | 31 044 | 53.9 |
| ≥College | 60 072 | 15 062 | 25.1 | 26 268 | 43.7 | 60 546 | 16 337 | 27.0 | 27874 | 46.0 |
| Otherd | 869 | 198 | 22.8 | 344 | 39.6 | 851 | 201 | 23.6 | 392 | 46.1 |
| Payment source for delivery | ||||||||||
| Private insurance | 79 599 | 20 151 | 25.3 | 35 306 | 44.4 | 77 273 | 20 540 | 26.6 | 35 597 | 46.1 |
| Medicaid | 44 605 | 14 763 | 33.1 | 26 627 | 59.7 | 43 972 | 14 456 | 32.9 | 25 991 | 59.1 |
| Self-payment | 6419 | 608 | 9.5 | 1093 | 17.0 | 6162 | 455 | 7.4 | 829 | 13.5 |
| Othere | 6623 | 1906 | 28.8 | 3207 | 48.4 | 8234 | 2463 | 29.9 | 4088 | 49.7 |
| WIC enrollment | ||||||||||
| Yes | 49 725 | 17 278 | 34.8 | 30 525 | 61.4 | 47 264 | 16 197 | 34.3 | 28 565 | 60.4 |
| No | 84 477 | 19 412 | 23.0 | 34 344 | 40.7 | 85 408 | 20 977 | 24.6 | 36 586 | 42.8 |
| Unknown | 3044 | 738 | 24.2 | 1364 | 44.8 | 2969 | 740 | 24.9 | 1354 | 45.6 |
| Maternal smoking | ||||||||||
| Yes | 23 490 | 7342 | 31.3 | 12 610 | 53.7 | 21 592 | 6717 | 31.1 | 11 662 | 54.0 |
| No | 111 858 | 29 610 | 26.5 | 52 741 | 47.2 | 112 486 | 30 743 | 27.3 | 54 035 | 48.0 |
| Unknown | 1898 | 476 | 25.1 | 882 | 46.5 | 1563 | 454 | 29.1 | 808 | 51.7 |
| Maternal infection | ||||||||||
| Yes | 7760 | 2442 | 31.5 | 4438 | 57.2 | 7740 | 2427 | 31.4 | 4367 | 56.4 |
| No | 129 486 | 34 986 | 27.0 | 61 795 | 47.7 | 127 901 | 35 487 | 27.8 | 62 138 | 48.6 |
| Maternal risk factor | ||||||||||
| Yes | 47 500 | 12 594 | 26.5 | 22 596 | 47.6 | 48 510 | 13 233 | 27.3 | 23 570 | 48.6 |
| No | 89 746 | 24 834 | 27.7 | 43 637 | 48.6 | 87 131 | 24 681 | 28.3 | 42 935 | 49.3 |
| Neighborhood characteristics | ||||||||||
| Quartile of household income | ||||||||||
| 1st | 37 743 | 11 982 | 31.8 | 22 607 | 59.9 | 36 522 | 11 733 | 32.8 | 21 798 | 59.7 |
| 2nd | 31 718 | 9617 | 30.3 | 15 844 | 50.0 | 31 849 | 9849 | 30.2 | 16 134 | 50.7 |
| 3rd | 34 948 | 8568 | 24.5 | 14 720 | 42.1 | 34 234 | 8203 | 25.0 | 14 323 | 41.8 |
| 4th | 32 809 | 7251 | 22.1 | 13 042 | 39.8 | 33 030 | 8129 | 22.0 | 14 247 | 43.1 |
| Quartile of poverty | ||||||||||
| 1st | 31 300 | 7288 | 23.3 | 12 869 | 41.1 | 30 366 | 7640 | 25.2 | 13 258 | 43.7 |
| 2nd | 34 055 | 8641 | 25.4 | 14 368 | 42.2 | 33 086 | 8329 | 25.2 | 14 126 | 42.7 |
| 3rd | 31 649 | 8882 | 28.1 | 14 946 | 47.2 | 32 797 | 9447 | 28.8 | 15 672 | 47.8 |
| 4th | 40 232 | 12 611 | 31.4 | 24 041 | 59.8 | 39 387 | 12 498 | 31.7 | 23 446 | 59.5 |
| Quartile of old housing | ||||||||||
| 1st | 35 406 | 6788 | 19.2 | 12 013 | 33.9 | 35 824 | 6994 | 19.5 | 12 659 | 35.3 |
| 2nd | 32 468 | 8598 | 26.5 | 14 479 | 44.6 | 31 764 | 8415 | 26.5 | 14 316 | 45.1 |
| 3rd | 32 299 | 9735 | 30.1 | 17 274 | 53.5 | 32 382 | 9990 | 30.9 | 17 584 | 54.3 |
| 4th | 37 073 | 12 307 | 33.2 | 22 467 | 60.6 | 35 670 | 12 515 | 35.1 | 21 946 | 61.5 |
Abbreviation: BLLs, blood lead levels.
Total number of children born in 2015 and 2016 by maternal and infant demographics and neighborhood characteristics.
The percentage of children born in 2015 and 2016 with a blood lead test before the age of 1 and of 2 years by maternal and infant demographics and neighborhood characteristics.
Other race includes all other races, unknown or missing race.
Other maternal educational attainment includes unknown or missing maternal educational attainment.
Other principal source of payment for delivery includes unknown or missing principal source of payment for delivery.
The percentage of confirmed EBLL among children tested for BLLs was 2.8% in the 2015 birth cohort and 2.5% in the 2016 birth cohort. In terms of racial disparities, non-Hispanic black children had the highest percentage of having EBLLs (4.4% and 4.4% in the 2015 and 2016 birth cohort, respectively), while non-Hispanic white children had the lowest percentage (2.1% and 1.9% in the 2015 and 2016 birth cohort, respectively). By maternal educational attainment, children born to mothers with “<high school” education level had the highest percentage of having EBLLs. Considering the principal source of payment for delivery, the percentage of having EBLLs was the highest among children born to mothers with self-payment and lowest among children born to mothers with private insurance. Children who enrolled in WIC, whose mothers smoked either before pregnancy or during pregnancy, and whose mothers had infections during pregnancy had higher percentages of having EBLLs. Children who lived in neighborhoods of lower quartiles of household income and higher quartiles of poverty and old housing had higher percentages of having EBLLs (Table 2).
Table 2.
Number and Percentage of EBLLs Among Children Tested for BLLs Before 2 Years of Age by Maternal and Infant Demographics and Neighborhood Characteristics, 2015 and 2016 Pennsylvania Birth Cohorts.
| 2015 birth cohort | 2016 birth cohort | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Tested children | Unconfirmed EBLL | Confirmed EBLL | Tested children | Unconfirmed EBLL | Confirmed EBLL | |||||
| Na | N | %b | N | %b | Na | N | %b | N | %b | |
| Overall | 66 233 | 1044 | 1.6 | 1826 | 2.8 | 66 505 | 828 | 1.3 | 1675 | 2.5 |
| Maternal and infant demographics | ||||||||||
| Sex | ||||||||||
| Female | 32 263 | 496 | 1.5 | 868 | 2.7 | 32 441 | 392 | 1.2 | 813 | 2.5 |
| Male | 33 970 | 548 | 1.6 | 958 | 2.8 | 34 064 | 436 | 1.3 | 862 | 2.5 |
| Race | ||||||||||
| Hispanic | 7822 | 172 | 2.2 | 279 | 3.6 | 7873 | 121 | 1.5 | 238 | 3.0 |
| Non-Hispanic Asian | 2633 | 53 | 2.0 | 88 | 3.3 | 2523 | 47 | 1.9 | 89 | 3.5 |
| Non-Hispanic black | 11 450 | 223 | 2.0 | 505 | 4.4 | 11 164 | 151 | 1.4 | 488 | 4.4 |
| Non-Hispanic white | 40 613 | 554 | 1.4 | 870 | 2.1 | 40 948 | 479 | 1.2 | 777 | 1.9 |
| Otherc | 3715 | 42 | 1.1 | 84 | 2.3 | 3997 | 30 | 0.8 | 83 | 2.1 |
| Maternal educational attainment | ||||||||||
| <High school | 7772 | 247 | 3.2 | 355 | 4.6 | 7195 | 192 | 2.7 | 349 | 4.9 |
| High school/some college | 31 849 | 594 | 1.9 | 971 | 3.1 | 31 044 | 445 | 1.4 | 870 | 2.8 |
| ≥College | 26 268 | 199 | 0.8 | 484 | 1.8 | 27 874 | 185 | 0.7 | 436 | 1.6 |
| Otherd | 344 | 4 | 1.2 | 16 | 4.7 | 392 | 6 | 1.5 | 20 | 5.1 |
| Payment source for delivery | ||||||||||
| Private insurance | 35 306 | 355 | 1.0 | 693 | 2.0 | 35 597 | 272 | 0.8 | 623 | 1.8 |
| Medicaid | 26 627 | 609 | 2.3 | 979 | 3.7 | 25 991 | 468 | 1.8 | 919 | 3.5 |
| Self-payment | 1093 | 25 | 2.3 | 60 | 5.5 | 829 | 24 | 2.9 | 35 | 4.2 |
| Othere | 3207 | 55 | 1.7 | 94 | 2.9 | 4088 | 64 | 1.6 | 98 | 2.4 |
| WIC enrollment | ||||||||||
| Yes | 30 525 | 649 | 2.1 | 1003 | 3.3 | 28 565 | 493 | 1.7 | 911 | 3.2 |
| No | 34 344 | 376 | 1.1 | 782 | 2.3 | 36 586 | 322 | 0.9 | 737 | 2.0 |
| Unknown | 1364 | 19 | 1.4 | 41 | 3.0 | 1354 | 13 | 1.0 | 27 | 2.0 |
| Maternal smoking | ||||||||||
| Yes | 12 610 | 297 | 2.4 | 373 | 3.0 | 11 662 | 234 | 2.0 | 358 | 3.1 |
| No | 52 741 | 732 | 1.4 | 1418 | 2.7 | 54 035 | 586 | 1.1 | 1285 | 2.4 |
| Unknown | 882 | 15 | 1.7 | 35 | 4.0 | 808 | 8 | 1.0 | 32 | 4.0 |
| Maternal infection | ||||||||||
| Yes | 4438 | 98 | 2.2 | 140 | 3.2 | 4367 | 74 | 1.7 | 114 | 2.6 |
| No | 61 795 | 946 | 1.5 | 1686 | 2.7 | 62 138 | 754 | 1.2 | 1561 | 2.5 |
| Maternal risk factor | ||||||||||
| Yes | 22 596 | 366 | 1.6 | 612 | 2.7 | 23 570 | 272 | 1.2 | 645 | 2.7 |
| No | 43637 | 678 | 1.6 | 1214 | 2.8 | 42 935 | 556 | 1.3 | 1030 | 2.4 |
| Neighborhood characteristics | ||||||||||
| Quartile of household income | ||||||||||
| 1st | 22 607 | 556 | 2.5 | 1010 | 4.5 | 21 798 | 396 | 1.8 | 924 | 4.2 |
| 2nd | 15 844 | 239 | 1.5 | 345 | 2.2 | 16 134 | 211 | 1.3 | 324 | 2.0 |
| 3rd | 14 720 | 169 | 1.2 | 285 | 1.9 | 14 323 | 152 | 1.1 | 270 | 1.9 |
| 4th | 13 042 | 80 | 0.6 | 186 | 1.4 | 14 247 | 69 | 0.5 | 157 | 1.1 |
| Quartile of poverty | ||||||||||
| 1st | 12 869 | 112 | 0.9 | 183 | 1.4 | 13 258 | 85 | 0.6 | 175 | 1.3 |
| 2nd | 14 368 | 162 | 1.1 | 277 | 1.9 | 14 126 | 139 | 1.0 | 211 | 1.5 |
| 3rd | 14 946 | 194 | 1.3 | 326 | 2.2 | 15 672 | 198 | 1.3 | 348 | 2.2 |
| 4th | 24 041 | 576 | 2.4 | 1040 | 4.3 | 23 446 | 406 | 1.7 | 941 | 4.0 |
| Quartile of old housing | ||||||||||
| 1st | 12 013 | 102 | 0.9 | 196 | 1.6 | 12 659 | 86 | 0.7 | 188 | 1.5 |
| 2nd | 14 479 | 170 | 1.2 | 246 | 1.7 | 14 316 | 156 | 1.1 | 230 | 1.6 |
| 3rd | 17 274 | 271 | 1.6 | 478 | 2.8 | 17 584 | 246 | 1.4 | 413 | 2.4 |
| 4th | 22 467 | 501 | 2.2 | 906 | 4.0 | 21 946 | 340 | 1.6 | 844 | 3.9 |
Abbreviation: EBLLs, elevated blood lead levels.
Total number of children born in 2015 and 2016 with a blood lead test before the age of 2 years by maternal and infant demographics and neighborhood characteristics.
The percentage of tested children under the age of 2 years who had unconfirmed or confirmed EBLLs by maternal and infant demographics and neighborhood characteristics.
Other race includes all other races, unknown or missing race.
Other maternal educational attainment includes unknown or missing maternal educational attainment.
Other principal source of payment for delivery includes unknown or missing principal source of payment for delivery.
The adjusted ORs and 95% CIs were estimated from the GLMMs to identify significant independent factors of the likelihood of receipt of lead testing and of having confirmed EBLLs (Table 3). Non-Hispanic black children had 7% higher odds of receipt of lead testing (adjusted OR = 1.07, 95% CI: 1.04, 1.11) and 18% higher odds of having EBLLs (adjusted OR = 1.18, 95% CI: 1.06, 1.31) as compared with non-Hispanic white children. Children born to mothers with “<high school” educational level had 15% lower odds of receipt of lead testing (adjusted OR = 0.85, 95% CI: 0.82, 0.88) and 75% higher odds of having EBLLs (adjusted OR = 1.75, 95% CI: 1.55, 1.98) compared with those with “≥ college” education level. Compared with children born to mothers with private insurance as the payment source for delivery, children born to mothers with self-payment had 69% lower odds of receipt of lead testing (adjusted OR = 0.31, 95% CI: 0.29, 0.32) and 89% higher odds of having EBLLs (adjusted OR = 1.89, 95% CI: 1.51, 2.36), and the odds of both outcomes were higher among children born to mothers with Medicaid. Children with WIC enrollment were more likely to receive lead testing and were less likely to have EBLLs than children without WIC enrollment. The adjusted ORs of receipt of lead testing increased in a stepwise fashion for higher quartiles of old housing, reaching 1.97 (95% CI: 1.84, 2.10) for the highest quartile. The adjusted ORs of having EBLLs decreased in a stepwise fashion for higher quartiles of household income, reaching 0.58 (95% CI: 0.47, 0.72) for the highest quartile. Additionally, the odds of having EBLLs was 29% (adjusted OR = 1.29, 95% CI: 1.05, 1.58) and 44% (adjusted OR = 1.44, 95% CI: 1.24, 1.68) higher for children living in neighborhoods of the highest quartile of poverty and old housing, respectively.
Table 3.
Adjusted Odds Ratios (95% Confidence Intervals) for Associations between Selected Characteristics and Receipt of Blood Lead Testing and Having a Confirmed EBLL Among Children Under 2 Years of Age, 2015−2016 Pennsylvania Birth Cohort.
| Blood lead testing | Confirmed EBLL | |||
|---|---|---|---|---|
| Maternal and infant demographics | ||||
| Sex | ||||
| Female | 1.00 | 1.00 | ||
| Male | 0.99a | (0.97, 1.01)b | 0.96 | (0.90, 1.03) |
| Race | ||||
| Non-Hispanic white | 1.00 | 1.00 | ||
| Hispanic | 0.99 | (0.96, 1.02) | 0.85 | (0.75, 0.96) |
| Non-Hispanic Asian | 0.95 | (0.90, 0.99) | 1.16 | (0.98, 1.36) |
| Non-Hispanic black | 1.07 | (1.04, 1.11) | 1.18 | (1.06, 1.31) |
| Otherc | 1.14 | (1.10, 1.19) | 0.90 | (0.76, 1.07) |
| Maternal educational attainment | ||||
| ≥College | 1.00 | 1.00 | ||
| <High school | 0.85 | (0.82, 0.88) | 1.75 | (1.55, 1.98) |
| High school/some college | 1.09 | (1.07, 1.11) | 1.23 | (1.12, 1.36) |
| Otherd | 0.62 | (0.56, 0.69) | 1.98 | (1.39, 2.83) |
| Payment source for delivery | ||||
| Private insurance | 1.00 | 1.00 | ||
| Medicaid | 1.19 | (1.17, 1.22) | 1.25 | (1.15, 1.37) |
| Self-payment | 0.31 | (0.29, 0.32) | 1.89 | (1.51, 2.36) |
| Othere | 1.01 | (0.97, 1.05) | 1.07 | (0.91, 1.25) |
| WIC enrollment | ||||
| No | 1.00 | 1.00 | ||
| Yes | 1.75 | (1.72, 1.79) | 0.89 | (0.82, 0.96) |
| Unknown | 1.05 | (0.99, 1.11) | 0.86 | (0.66, 1.10) |
| Maternal smoking | ||||
| No | 1.00 | 1.00 | ||
| Yes | 1.07 | (1.04, 1.09) | 1.06 | (0.96, 1.16) |
| Unknown | 0.90 | (0.83, 0.97) | 1.16 | (0.90, 1.49) |
| Maternal infection | ||||
| No | 1.00 | 1.00 | ||
| Yes | 1.10 | (1.06, 1.14) | 0.87 | (0.77, 1.00) |
| Maternal risk factor | ||||
| No | 1.00 | 1.00 | ||
| Yes | 0.96 | (0.94, 0.98) | 1.01 | (0.94, 1.08) |
| Neighborhood characteristics | ||||
| Quartiles of household income | ||||
| 1st | 1.00 | 1.00 | ||
| 2nd | 1.00 | (0.95, 1.05) | 0.71 | (0.62, 0.81) |
| 3rd | 0.97 | (0.90, 1.03) | 0.76 | (0.64, 0.91) |
| 4th | 1.02 | (0.94, 1.10) | 0.58 | (0.47, 0.72) |
| Quartiles of poverty | ||||
| 1st | 1.00 | 1.00 | ||
| 2nd | 0.97 | (0.93, 1.02) | 1.08 | (0.92, 1.26) |
| 3rd | 0.99 | (0.94, 1.05) | 1.15 | (0.96, 1.36) |
| 4th | 1.05 | (0.98, 1.13) | 1.29 | (1.05, 1.58) |
| Quartiles of old housing | ||||
| 1st | 1.00 | 1.00 | ||
| 2nd | 1.39 | (1.31, 1.46) | 0.87 | (0.74, 1.02) |
| 3rd | 1.66 | (1.57, 1.77) | 1.11 | (0.95, 1.29) |
| 4th | 1.97 | (1.84, 2.10) | 1.44 | (1.24, 1.68) |
Abbreviation: EBLL, elevated blood lead level.
Adjusted odds ratio.
95% confidence interval.
Other race includes all other races, unknown or missing race.
Other maternal educational attainment includes unknown or missing maternal educational attainment.
Other principal source of payment for delivery includes unknown or missing principal source of payment for delivery.
Interaction terms were added to the GLMM model 1 by 1 to assess if the impacts of individual factors varied at different levels of neighborhood characteristics on the 2 outcomes of interest. Compared with non-Hispanic white children, Hispanic and non-Hispanic Asian children had lower odds of receipt of lead testing in relatively deprived neighborhoods (the first and second quartiles of household income and the third and fourth quartiles of poverty and old housing), although some comparisons were not statistically significant. Compared with children born to mothers with “≥college” education level, children born to mothers with “<high school” educational level were generally less likely to receive lead testing except those in the least deprived neighborhoods (the fourth quartile of household income and the first quartile of poverty and old housing). Compared with children born to mothers with private insurance as the payment source for delivery, children born to mothers with Medicaid had significantly higher odds of receipt of lead testing except in the most deprived neighborhoods, while the odds for children born to mothers with self-payment were significantly lower and fluctuated widely by different levels of neighborhood characteristics. Children with WIC enrollment had significantly higher odds of receipt of lead testing in each quartile of neighborhood characteristics, and the odds gradually decreased in more deprived neighborhoods (Table 4).
Table 4.
Adjusted Odds Ratios (95% Confidence Intervals) for Associations between Maternal and Infant Demographics and Receipt of Blood Lead Testing Among Children Under 2 Years of Age, by Quartiles of Each Neighborhood Characteristics, 2015−2016 Pennsylvania Birth Cohort.
| Quartiles of household income | Quartiles of poverty | Quartiles of old housing | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1st | 2nd | 3rd | 4th | 1st | 2nd | 3rd | 4th | 1st | 2nd | 3rd | 4th | |
| Race | ||||||||||||
| Non-Hispanic white | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Hispanic | 0.94a (0.9, 0.99)b | 0.93 (0.87, 0.99) | 1.10 (1.03, 1.18) | 1.09 (1.00, 1.18) | 1.08 (1.00, 1.17) | 1.05 (0.98, 1.13) | 1.00 (0.93, 1.07) | 0.95 (0.91, 1.00) | 1.20 (1.12, 1.29) | 1.07 (0.99, 1.16) | 0.96 (0.90, 1.01) | 0.88 (0.83, 0.93) |
| Non-Hispanic Asian | 0.91 (0.85, 0.98) | 0.90 (0.82, 0.98) | 1.06 (0.96, 1.16) | 0.95 (0.86, 1.06) | 0.99 (0.89, 1.10) | 1.01 (0.92, 1.12) | 0.93 (0.85, 1.02) | 0.91 (0.85, 0.98) | 1.03 (0.93, 1.15) | 1.01 (0.91, 1.12) | 0.87 (0.80, 0.95) | 0.91 (0.84, 0.98) |
| Non-Hispanic black | 1.07 (1.02, 1.12) | 1.02 (0.96, 1.10) | 1.09 (1.01, 1.18) | 1.06 (0.96, 1.17) | 1.05 (0.95, 1.16) | 0.99 (0.91, 1.08) | 1.08 (1.01, 1.17) | 1.08 (1.03, 1.14) | 1.21 (1.11, 1.33) | 1.17 (1.08, 1.28) | 0.99 (0.93, 1.05) | 1.03 (0.98, 1.08) |
| Otherc | 1.21 (1.11, 1.31) | 1.27 (1.17, 1.39) | 1.12 (1.08, 1.21) | 1.03 (0.97, 1.10) | 1.03 (0.96, 1.10) | 1.09 (1.01, 1.18) | 1.19 (1.09, 1.30) | 1.27 (1.18, 1.37) | 1.05 (0.97, 1.12) | 1.12 (1.02, 1.22) | 1.13 (1.05, 1.23) | 1.25 (1.16, 1.35) |
| Maternal educational attainment | ||||||||||||
| ≥College | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| <High school | 0.84 (0.80, 0.88) | 0.77 (0.73, 0.82) | 0.78 (0.73, 0.84) | 1.02 (0.93, 1.13) | 1.05 (0.96, 1.15) | 0.81 (0.76, 0.87) | 0.78 (0.74, 0.83) | 0.80 (0.77, 0.84) | 0.97 (0.91, 1.05) | 0.77 (0.73, 0.82) | 0.90 (0.85, 0.95) | 0.71 (0.68, 0.75) |
| High school/some college | 0.99 (0.96, 1.04) | 1.06 (1.03, 1.11) | 1.12 (1.08, 1.17) | 1.18 (1.14, .123) | 1.14 (1.10, 1.19) | 1.14 (1.09, 1.18) | 1.12 (1.08, 1.16) | 0.97 (0.93, 1.01) | 1.34 (1.29, 1.39) | 1.18 (1.13, 1.22) | 1.01 (0.97, 1.05) | 0.87 (0.83, 0.90) |
| Otherd | 0.60 (0.52, 0.70) | 0.66 (0.53, 0.84) | 0.54 (0.42, 0.70) | 0.61 (0.46, 0.81) | 0.73 (0.55, 0.99) | 0.65 (0.50, 0.85) | 0.55 (0.43, 0.69) | 0.58 (0.50, 0.67) | 0.63 (0.47, 0.84) | 0.77 (0.59, 0.99) | 0.60 (0.49, 0.74) | 0.51 (0.44, 0.60) |
| Payment source for delivery | ||||||||||||
| Private insurance | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Medicaid | 1.01 (0.97, 1.05) | 1.23 (1.19, 1.28) | 1.37 (1.32, 1.43) | 1.47 (1.39, 1.55) | 1.41 (1.34, 1.49) | 1.37 (1.31, 1.43) | 1.31 (1.26, 1.36) | 0.99 (0.96, 1.03) | 1.68 (1.61, 1.76) | 1.37 (1.32, 1.43) | 1.14 (1.09, 1.18) | 0.91 (0.88, 0.95) |
| Self-payment | 0.61 (0.55, 0.68) | 0.22 (0.19, 0.24) | 0.20 (0.18, 0.23) | 0.33 (0.28, 0.38) | 0.29 (0.25, 0.33) | 0.22 (0.19, 0.24) | 0.20 (0.18, 0.22) | 0.56 (0.51, 0.62) | 0.23 (0.20, 0.26) | 0.16 (0.14, 0.18) | 0.57 (0.50, 0.65) | 0.56 (0.50, 0.63) |
| Othere | 0.99 (0.92, 1.06) | 0.99 (0.92, 1.07) | 1.02 (0.95, 1.11) | 0.97 (0.89, 1.06) | 0.96 (0.87, 1.04) | 1.02 (0.94, 1.10) | 1.01 (0.94, 1.09) | 0.98 (0.92, 1.05) | 1.11 (1.03, 1.21) | 1.05 (0.97, 1.14) | 1.05 (0.97, 1.13) | 0.83 (0.77, 0.89) |
| WIC enrollment | ||||||||||||
| No | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Yes | 1.53 (1.48, 1.59) | 1.85 (1.79, 1.92) | 1.90 (1.82, 1.97) | 1.96 (1.86, 2.07) | 1.91 (1.81, 2.00) | 1.98 (1.90, 2.06) | 1.90 (1.83, 1.98) | 1.50 (1.45, 1.55) | 2.25 (2.16, 2.35) | 2.05 (1.67, 2.13) | 1.65 (1.58, 1.71) | 1.41 (1.36, 1.46) |
| Unknown | 1.03 (0.93, 1.15) | 1.20 (1.06, 1.35 | 0.93 (0.83, 1.05) | 1.02 (0.92, 1.14) | 0.95 (0.85, 1.07) | 1.05 (0.94, 1.18) | 1.19 (1.06, 1.34) | 0.99 (0.89, 1.09) | 1.05 (0.94, 1.17) | 1.19 (1.06, 1.34) | 0.93 (0.83, 1.04) | 0.99 (0.89, 1.09) |
| Maternal smoking | ||||||||||||
| No | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Yes | 0.90 (0.86, 0.93) | 1.11 (1.07, 1.16) | 1.23 (1.18, 1.29) | 1.19 (1.12, 1.26) | 1.19 (1.12, 1.26) | 1.23 (1.17, 1.29) | 1.14 (1.09, 1.19) | 0.89 (0.85, 0.92) | 1.37 (1.30, 1.44) | 1.22 (1.17, 1.28) | 1.02 (0.97, 1.06) | 0.83 (0.80, 0.87) |
| Unknown | 0.84 (0.75, 0.94) | 0.90 (0.75, 1.07) | 0.90 (0.76, 1.07) | 1.02 (0.86, 1.23) | 1.01 (0.84, 1.23) | 0.98 (0.82, 1.16) | 0.87 (0.73, 1.05) | 0.83 (0.75, 0.92) | 0.95 (0.81, 1.12) | 1.14 (0.96, 1.36) | 0.85 (0.73, 0.99) | 0.79 (0.70, 0.89) |
| Maternal infection | ||||||||||||
| No | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Yes | 1.01 (0.96, 1.07) | 1.13 (1.05, 1.22) | 1.22 (1.13, 1.32) | 1.17 (1.06, 1.28) | 1.19 (1.08, 1.30) | 1.25 (1.15, 1.35) | 1.16 (1.08, 1.25) | 0.99 (0.94, 1.04) | 1.32 (1.22, 1.44) | 1.19 (1.10, 1.28) | 1.02 (0.96, 1.09) | 1.02 (0.96, 1.08) |
Adjusted odds ratio.
95% confidence interval.
Other race includes all other races, unknown or missing race.
Other maternal educational attainment includes unknown or missing maternal educational attainment.
Other principal source of payment for delivery includes unknown or missing principal source of payment for delivery.
Compared with non-Hispanic white children, Hispanic and non-Hispanic Asian children had significantly higher odds of having EBLLs in the least poor neighborhoods, while non-Hispanic black children had significantly higher odds in the lowest household income and the poorest neighborhoods. Compared with children born to mothers with private insurance as the payment source for delivery, the odds of having EBLLs for children born to mothers with Medicaid or self-payment were significantly higher and fluctuated widely by different levels of neighborhood poverty. Children with WIC enrollment were less likely to have EBLLs only in neighborhoods of the highest quartile of old housing (Table 5).
Table 5.
Adjusted Odds Ratios (95% Confidence Intervals) for Associations between Maternal and Infant Demographics and Having a Confirmed EBLL Among Children Under 2 Years of Age, by Quartiles of Each Neighborhood Characteristics, 2015−2016 Pennsylvania Birth Cohort.
| Quartiles of household income | Quartiles of poverty | Quartiles of old housing | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1st | 2nd | 3rd | 4th | 1st | 2nd | 3rd | 4th | 1st | 2nd | 3rd | 4th | |
| Race | ||||||||||||
| Non-Hispanic white | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| Hispanic | 0.82a (0.7, 0.96)b | 0.73 (0.54, 0.98) | 1.17 (0.86, 1.57) | 0.94 (0.59, 1.50) | 1.46 (1.01, 2.15) | 0.62 (0.39, 0.96) | 0.81 (0.60, 1.11) | 0.83 (0.71, 0.96) | ||||
| Non-Hispanic Asian | 1.06 (0.85, 1.32) | 1.15 (0.80, 1.65) | 1.33 (0.89, 2.00) | 1.64 (0.98, 2.73) | 2.55 (1.67, 3.89) | 0.90 (0.54, 1.50) | 0.98 (0.65, 1.48) | 1.09 (0.88, 1.36) | ||||
| Non-Hispanic black | 1.25 (1.10, 1.43) | 0.75 (0.57, 0.98) | 0.95 (0.68, 1.33) | 1.32 (0.83, 2.10) | 1.35 (0.88, 2.09) | 0.66 (0.43, 1.02) | 1.04 (0.81, 1.35) | 1.21 (1.06, 1.38) | ||||
| Otherc | 0.75 (0.57, 0.98) | 0.82 (0.57, 1.19) | 1.24 (0.87, 1.76) | 1.12 (0.75, 1.66) | 1.32 (0.89, 1.96) | 0.78 (0.50, 1.23) | 1.23 (0.89, 1.72) | 0.73 (0.56, 0.94) | ||||
| Payment source for delivery | ||||||||||||
| Private insurance | 1.00 | 1.00 | 1.00 | 1.00 | ||||||||
| Medicaid | 1.51 (1.17, 1.94) | 1.15 (0.94, 1.41) | 1.13 (0.95, 1.34) | 1.31 (1.17, 1.48) | ||||||||
| Self-payment | 2.84 (1.43, 5.66) | 1.26 (0.62, 2.59) | 2.98 (1.98, 4.48) | 1.62 (1.18, 2.20) | ||||||||
| Otherd | 1.12 (0.66, 1.90) | 0.78 (0.48, 1.26) | 0.65 (0.42, 1.00) | 1.29 (1.05, 1.58) | ||||||||
| WIC enrollment | ||||||||||||
| No | 1.00 | 1.00 | 1.00 | 1.00 | ||||||||
| Yes | 0.83 (0.66, 1.04) | 0.91 (0.75, 1.10) | 1.11 (0.96, 1.29) | 0.80 (0.72, 0.89) | ||||||||
| Unknown | 1.50 (0.83, 2.72) | 1.04 (0.56, 1.92) | 0.97 (0.57, 1.65) | 0.64 (0.44, 0.93) | ||||||||
| Maternal smoking | ||||||||||||
| No | 1.00 | 1.00 | 1.00 | 1.00 | ||||||||
| Yes | 1.34 (1.05, 1.71) | 1.00 (0.81, 1.25) | 1.25 (1.07, 1.47) | 0.90 (0.79, 1.04) | ||||||||
| Unknown | 0.80 (0.25, 2.45) | 1.20 (0.56, 2.57) | 1.57 (0.97, 2.53) | 1.04 (0.74, 1.46) | ||||||||
Abbreviation: EBLL, elevated blood lead level.
Adjusted odds ratio.
95% confidence interval.
Other race includes all other races, unknown or missing race.
Other principal source of payment for delivery includes unknown or missing principal source of payment for delivery.
Discussion
This cohort study indicated that approximately 49% of newborns tested for BLLs before 2 years of age which was much higher than the screening rate (29%) reported in the previous Pennsylvania childhood lead surveillance annual report based on 1 calendar year.22 This cohort analysis, using birth certificate data linked to blood lead test data and neighborhood characteristics data, enables us to more accurately estimate the rate of receipt of lead testing and the proportion of children with EBLLs by maternal and infant demographics and neighborhood characteristics. It provides more accurate estimates than the cross-sectional study design which included blood test results reported on a calendar year view and did not include children who had been tested in the previous year or will be tested in the following year. Pennsylvania does not mandate a statewide universal screening which may result in a lower lead screening rate when compared with Philadelphia (76.5% in children under the age of 24 months) and New York state (76.7% in children under the age of 18 months) who have such mandates in place.23,24 Estimated percentages of receipt of lead testing and of having EBLLs, which were higher among children with specific demographics and in deprived neighborhoods, may reflect a true increased risk of lead exposure or more robust and targeted lead testing among that specific group of children. We found that the rate of lead testing was relatively low among children who were non-Hispanic whites, who were born to mothers with the lowest or highest educational attainment, whose payment source for delivery was non-Medicaid (private insurance or self-payment), who didn’t enroll in WIC and who lived in less deprived neighborhoods. Moreover, the percentage of having EBLLs was relatively high among children who were racial and ethnic minorities (especially non-Hispanic black), who were born to mothers with the lowest educational attainment, whose payment source for delivery was Medicaid, who enrolled in WIC, and who lived in more deprived neighborhoods. The above-mentioned results were consistent with previous findings.9-11,15-16
To the best of our knowledge, this is the first population-based study using GLMMs to investigate independent and interaction effects of selected maternal and infant demographics and neighborhood characteristics on the likelihood of receipt of lead testing and of having EBLLs. We found that being non-Hispanic black, having mothers with higher educational attainment, paying for delivery by Medicaid, enrolling in WIC, and living in neighborhoods with higher burdens of old housing were associated with higher odds of receipt of lead testing. Being non-Hispanic black, having mothers with lower educational attainment, paying for delivery by Medicaid or self-payment, and living in the poorest and the oldest neighborhoods were significant risk factors for EBLLs. These results are consistent with findings from previous studies.9-11,15-17,25,26 Besides these well-known risk factors, being non-Hispanic Asian and paying for delivery by self-payment were associated with undertesting of lead and having EBLLs. Parental linguistic and cultural barriers may affect Asian children’s ability to gain access to appropriate and timely health care services. Children without Medicaid or private insurance may have difficulties finding a primary care provider due to the out of pocket costs.
Furthermore, we found that the odds of receipt of lead testing and of having confirmed EBLLs related to disparities in some maternal and infant demographics vary by different levels of neighborhood characteristics. Being Hispanic, having mothers with high school/some college educational level, paying for delivery by Medicaid, and enrolling in WIC were associated with a higher likelihood of receipt of testing in the least deprived neighborhoods, but these positive relationships diminished and even reversed in more deprived neighborhoods.
Additionally, non-Hispanic black children had higher odds of having EBLLs in the most economically disadvantaged neighborhoods compared to non-Hispanic white children, but this significant racial gap was nonexistent in less economically disadvantaged neighborhoods. Our results here are different from findings from the previous study conducted by Moody et al. in the Detroit metropolitan area.13 Their findings showed that the black-white racial gap in blood lead levels was the narrowest for children living in the neighborhoods of the lowest socioeconomic position (SEP), and the gap exacerbated with increasing levels of neighborhood SEP. It is important to note that the outcomes of interest, classification of neighborhood characteristics, and data analysis methods were different in the 2 studies. More studies are needed to address whether the racial differences seen regarding the risk of having EBLLs exacerbates or narrows with increasing levels of neighborhood socioeconomic characteristics.
This study had several limitations. First, the underreporting of blood lead test results by health care service providers, unmatched blood lead test records, and children born to Pennsylvania resident mothers and moved out of state before the receipt of lead testing may have contributed to the underestimate of the screening rate. Secondly, the inherent limitations of accuracy errors in deterministic linkage would introduce bias into analyses, even though we conducted manual validation reviews on matching data to minimize these errors. Finally, because Pennsylvania does not have a statewide universal lead screening mandate for children, it is important to note that the results presented in this study should be interpreted with knowledge of local childhood lead screening related policies. This limits the generalizability of our findings to other areas of the country.
In summary, certain maternal and infant demographics and neighborhood socioeconomic characteristics are significantly associated with undertesting of childhood blood lead and with higher risk of having EBLLs. Therefore, proactive and effective lead screening to identify potentially exposed children is essential. Our findings can not only be used to guide targeted efforts in planning prevention programs but also guide health provider decisions on priorities regarding which children should receive a follow-up test within the recommended time period and treatment if necessary.
Acknowledgments
The author(s) acknowledge the Childhood Lead Poisoning Prevention Program within the Pennsylvania Department of Health Child and Adult Health Services for its assistance in the collection of blood lead test data.
Footnotes
Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Yeh-Hsin Chen 
https://orcid.org/0000-0003-3011-7704
References
- 1. Jones RL, Homa DM, Meyer PA, et al. Trends in blood lead levels and blood lead testing among US children aged 1 to 5 years, 1988–2004. Pediatrics. 2009;123:e376-385. doi: 10.1542/peds.2007-3608 [DOI] [PubMed] [Google Scholar]
- 2. Bellinger DC, Stiles KM, Needleman HL. Low-level lead exposure, intelligence and academic achievement: a long-term follow-up study. Pediatrics. 1992;90:855-861. [PubMed] [Google Scholar]
- 3. McMichael AJ, Baghurst PA, Wigg NR, Vimpani GV, Robertson EF, Roberts RJ. Port Pirie cohort study: environmental exposure to lead and children’s abilities at the age of four years. N Engl J Med. 1988;319:468-475. doi: 10.1056/NEJM198808253190803 [DOI] [PubMed] [Google Scholar]
- 4. Canfield RL, Henderson CR, Jr, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP. Intellectual impairment in children with blood lead concentrations below 10 microg per deciliter. N Engl J Med. 2003;348:1517-1526. doi: 10.1056/NEJMoa022848 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Rummo JH, Routh DK, Rummo NJ, Brown JF. Behavioral and neurological effects of symptomatic and asymptomatic lead exposure in children. Arch Environ Health. 1979;34:120-124. doi: 10.1080/00039896.1979.10667381 [DOI] [PubMed] [Google Scholar]
- 6. Naranjo VI, Hendricks M, Jones KS. Lead toxicity in children: an unremitting public health problem. Pediatr Neurol. 2020;113:51-55. doi: 10.1016/j.pediatrneurol.2020.08.005 [DOI] [PubMed] [Google Scholar]
- 7. Centers for Disease Control and Prevention (CDC). CDC response to advisory committee on childhood lead poisoning prevention recommendations in “low level lead exposure harms children: a renewed call for primary prevention.” Department of Health and Human Services. 2012. Accessed February 1, 2021. https://stacks.cdc.gov/view/cdc/37586 [Google Scholar]
- 8. Centers for Disease Control and Prevention (CDC). Blood lead levels in children aged 1-5 years - United States, 1999-2010. MMWR Morb Mortal Wkly Rep. 2013;62:245-248. [PMC free article] [PubMed] [Google Scholar]
- 9. Aoki Y, Brody DJ. WIC participation and blood lead levels among children 1-5 years: 2007–2014. Environ Health Perspect. 2018;126:067011. doi: 10.1289/EHP2384 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kemper AR, Cohn LM, Fant KE, Dombkowski KJ. Blood lead testing among Medicaid-enrolled children in Michigan. Arch Pediatr Adolesc Med. 2005;159:646-650. doi: 10.1001/archpedi.159.7.646 [DOI] [PubMed] [Google Scholar]
- 11. Kaufmann RB, Clouse TL, Olson DR, Matte TD. Elevated blood lead levels and blood lead screening among US children aged one to five years: 1988-1994. Pediatrics. 2000;106:E79. doi: 10.1542/peds.106.6.e79 [DOI] [PubMed] [Google Scholar]
- 12. Carrel M, Zahrieh D, Young SG, et al. High prevalence of elevated blood lead levels in both rural and urban Iowa newborns: spatial patterns and area-level covariates. PLoS One. 2017;12:e0177930. doi: 10.1371/journal.pone.0177930 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Moody H, Darden JT, Pigozzi BW. The racial gap in childhood blood lead levels related to socioeconomic position of residence in metropolitan Detroit. Sociol Race Ethn. 2016;2: 200-218. [Google Scholar]
- 14. Lanphear BP, Byrd RS, Auinger P, Schaffer SJ. Community characteristics associated with elevated blood lead levels in children. Pediatrics. 1998;101:264-271. doi: 10.1542/peds.101.2.264 [DOI] [PubMed] [Google Scholar]
- 15. Vivier PM, Hauptman M, Weitzen SH, Bell S, Quilliam DN, Logan JR. The important health impact of where a child lives: neighborhood characteristics and the burden of lead poisoning. Matern Child Health J. 2011;15:1195-1202. doi: 10.1007/s10995-010-0692-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Sargent JD, Brown MJ, Freeman JL, Bailey A, Goodman D, Freeman DH., Jr. Childhood lead poisoning in Massachusetts communities: its association with sociodemographic and housing characteristics. Am J Public Health. 1995;85:528-534. doi: 10.2105/ajph.85.4.528 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Moody HA, Darden JT, Pigozzi BW. The relationship of neighborhood socioeconomic differences and racial residential segregation to childhood blood lead levels in metropolitan Detroit. J Urban Health. 2016;93:820-839. doi: 10.1007/s11524-016-0071-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Centers for Medicare and Medicaid Services. Coverage of blood lead testing for children enrolled in Medicaid and the Children’s Health Insurance Program. CMCS Informational Bulletin. 2016. Accessed February 1, 2021. https://www.medicaid.gov/federal-policy-guidance/downloads/cib113016.pdf
- 19. Daniel RL. Most medicaid children in nine states are not receiving all required preventive screening services. Department of Health and Human Services. Office of Inspector General. 2010. Accessed February 1, 2021.https://oig.hhs.gov/oei/reports/oei-05-08-00520.pdf [Google Scholar]
- 20. Neuwirth LS. Resurgent lead poisoning and renewed public attention towards environmental social justice issues: a review of current efforts and call to revitalize primary and secondary lead poisoning prevention for pregnant women, lactating mothers, and children within the U.S. Int J Occup Environ Health. 2018;24(3-4):86-100. doi: 10.1080/10773525.2018.1507291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. United States Census Bureau. 2012–2016 American Community Survey (ACS) 5-year estimates. 2017. Accessed June 15, 2020. https://www.census.gov/acs/www/data/data-tables-and-tools/data-profiles/
- 22. Pennsylvania Department of Health. 2016 childhood lead surveillance annual report. Childhood Lead Poisoning Prevention Program. 2018. Accessed February 1, 2021. https://www.health.pa.gov/topics/Documents/Environmental%20Health/2016%20Childhood%20Lead%20Surveillance%20Annual%20Report.pdf [Google Scholar]
- 23. Philadelphia Department of Public Health. Philadelphia childhood lead poisoning surveillance report 2017. Lead and Healthy Homes Program. 2018. Accessed February 1, 2021. https://www.phila.gov/media/20190319101844/Lead-Surveillance-2017_9.7.2018-final.pdf [Google Scholar]
- 24. New York Department of Health. Percentage of children born in 2013 with a lead screening - aged 0–8 months and 9–17 months, 2013. New York State Community Health Indicator Reports. 2018. Accessed February 1, 2021. https://webbi1.health.ny.gov/SASStoredProcess/guest?_program=/EBI/PHIG/apps/chir_dashboard/chir_dashboard&p=it&ind_id=Cg25 [Google Scholar]
- 25. Vaidyanathan A, Staley F, Shire J, et al. Screening for lead poisoning: a geospatial approach to determine testing of children in at-risk neighborhoods. J Pediatr. 2009;154:409-14. [DOI] [PubMed] [Google Scholar]
- 26. Hore P, Ahmed MS, Sedlar S, Saper RB, Nagin D, Clark N. Blood lead levels and potential risk factors for lead exposures among south Asians in New York City. J Immigr Minor Health. 2017;19:1322-1329. [DOI] [PMC free article] [PubMed] [Google Scholar]