US children experience lead poisoning, which has detrimental health effects and significant individual and societal costs. This study aimed to describe the sociodemographic and hospitalization characteristics of children hospitalized for lead poisoning and assess the proportion of inpatients who received blood lead testing and appropriate follow-up testing before hospitalization.
2015–2021 hospital discharge data were linked to lead surveillance data for Pennsylvania children aged 0 to 5 years. Demographics, hospitalization characteristics, and lead testing data from children with a primary diagnosis code of lead poisoning were used. The number of hospitalizations, associated hospital length of stay and charges, and the proportion of inpatients who received lead testing and follow-up testing after identification of high blood lead levels before hospitalization were analyzed by selected characteristics.
During the study period, there were 93 children hospitalized for lead poisoning, incurring 443 inpatient days and approximately 6 million dollars in inpatient charges. Of these inpatients, 69.9% were males, 36.6% were non-Hispanic Black, 67.7% were aged 0 to 2 years, 14% had repeated admissions, and 88.2% of admissions were paid by Medicaid. In addition, 20.4% did not have lead testing, and 34.4% had appropriate follow-up testing before hospitalization. Non-Hispanic whites and children with developmental, behavioral, and emotional disorders had relatively low proportions of having appropriate follow-up testing.
Severe childhood lead poisoning is a significant health and economic burden, especially among children with Medicaid. Further improvements in blood lead testing and follow-up testing can help prevent childhood lead poisoning.
There is indisputable evidence that exposure to lead, even at low levels, is associated with persistent adverse health effects in children.1–4 In response, in 2012, the Centers for Disease Control and Prevention (CDC) lowered the blood lead reference value (BLRV) to 5 µg/dL from 10 ug/dL and further lowered it to 3.5 µg/dL in response to the Lead Exposure Prevention and Advisory Committee recommendation made in 2021.5,6 Young children under 6 years of age are particularly susceptible to lead exposure because of such behaviors as crawling and/or playing on the ground, mouthing objects that increase contact with lead-contaminated dust and soil, and their rapidly developing nervous systems are more sensitive to the impacts of lead.7
The societal cost of lead poisoning in the United States is significant and multifaceted: an estimated $5.9 million in medical care costs, $50.9 billion annually in lost economic productivity, and an estimated $50 000 of lifetime social cost for every 1 unit (ug/dL) increase in a child’s blood lead level (BLL).8,9 Previous research did cost-benefit analysis and revealed that there would be substantial returns to investing in lead poisoning prevention efforts, and blood lead screening is one of the most cost-effective strategies, especially in populations at high risk of lead exposure.10,11 Therefore, understanding the cost of hospitalizations for lead poisoning in different population groups can provide useful information for the further research on economic analysis of various lead poisoning prevention and intervention policies.
As evidenced by the significant reduction in the average BLLs in the United States over the past couple of decades, particularly among children, because of the concerted efforts at the public health and individual levels,12 including improvement in blood lead testing, attention has been increasingly directed to less severe lead exposure. Although the cases of severe lead poisoning are relatively rare, severe lead poisoning can cause multiple organ damage and even death.13 A recent research estimated that less than 550 US children under 18 year of age were hospitalized for lead poisoning between 2006 and 2014.14 Most previous studies on lead poisoning hospitalization were case reports; however, no systematic analysis of childhood lead exposures that required hospitalization has been conducted. This study attempts to fill this knowledge gap.
In this study, we linked hospital discharge data and blood lead surveillance data to demonstrate the characteristics of children hospitalized for lead poisoning under 6 years of age and associated hospital charges in Pennsylvania. Furthermore, we assessed whether blood lead testing had been performed before hospitalization and if follow-up testing was conducted appropriately within the CDC’s recommended schedule after identification of a high BLL.
Methods
Data Sources
Hospital discharge data between 2015 and 2021 from the Pennsylvania Health Care Cost Containment Council and 2011 to 2021 lead testing data from the Pennsylvania Department of Health National Electronic Disease Surveillance System (PA-NEDSS) were used. The Pennsylvania Health Care Cost Containment Council is an independent state agency formed under Pennsylvania statute (Act 89 of 1986, as amended by Act 15 of 2020) with the authority to collect inpatient hospital discharge and ambulatory or outpatient procedure records from hospitals and freestanding ambulatory surgery centers in Pennsylvania. Pennsylvania state regulation requires all clinical laboratories to report all blood lead test results, regardless of the BLLs, on venous and capillary specimens for persons under 16 years of age to the PA-NEDSS.
Study Population
We matched all 2015 to 2021 hospital discharge records for children under 6 years of age to the 2011 to 2021 PA-NEDSS lead testing data using both deterministic and probabilistic matching methods using the patient's name, home address, date of birth, sex, social security number, and insurance information. Lead testing data between 2011 and 2021 were used to ensure we have complete lead testing records for each patient before their hospital admissions for lead poisoning. A patient’s hospitalization for lead poisoning was defined as having either the International Classification of Diseases, Ninth Revision (ICD-9) code of 984* or the Tenth Revision (ICD-10) code of T56.0* in the primary or any secondary principal diagnosis codes. Only patients with a primary diagnosis code of lead poisoning were used for this analysis to ensure they were hospitalized primarily for lead poisoning. Patients with a secondary diagnosis of lead poisoning were included in the sensitivity analysis but were not reported in this evaluation. Children with pervasive and specific developmental disorders (ICD-10 codes of F80* to F89* or ICD-9 codes of 299*) or behavioral and emotional disorders (ICD-10 codes of F90* to F98* or ICD-9 codes of 313*) in any of the 17 secondary discharge diagnoses were further identified, given their increased risk for lead exposure as a result of higher prevalence of pica behavior among these children.15
Each inpatient’s demographics (age, sex, and race and ethnicity); total hospital charges provided by the hospital; primary payment types for hospital charges; developmental, behavioral, and emotional disorders; and residence at diagnosis was collected from the lead testing dataset or hospital discharge dataset. Racial disparities have been noted in the risk of lead poisoning; particularly, African American and Hispanic populations are at higher risk of lead exposure and disproportionately affected by lead poisoning. Our race and ethnicity data included children who self-reported as non-Hispanic Black (hereafter, Black), non-Hispanic white (hereafter, white), Hispanic, and other individuals. In analysis, we categorized children with other, unknown, or missing race and ethnicity as other or unknown race and ethnicity.
Blood Lead Testing and Follow-up Testing Before Hospitalization
We calculated the days between each blood lead specimen collection date and the indexed hospital admission date for each patient. If 1 of the patient’s blood lead specimen collection dates occurred 7 days or more before the indexed hospital admission date, the patient was identified as having blood lead testing for routine screening for lead exposure before hospitalization. An inpatient’s blood lead test done within 6 days or fewer before the admission date was considered as a diagnostic test for lead poisoning rather than a routine screening test for lead exposure. Children with only diagnostic lead testing within 6 days or fewer before the admission date were considered as not having blood lead testing before hospitalization.
For each patient, we indexed the reporting date of each confirmed blood lead test result that was equal to or greater than BLRV before hospitalization and then calculated the days between that date and the subsequent blood lead testing date and the days between that date and the patient’s hospital admission date. We used the CDC’s recommended schedules for follow-up lead testing to categorize whether a child had an appropriate follow-up test after identification of a confirmed BLL ≥ BLRV.16 A patient was considered as having appropriate follow-up testing before hospitalization if the patient received a follow-up blood lead test or was admitted to the hospital for medical evaluation for lead exposure within 92 days after a confirmed BLL of 5 to 19 µg/dL or within 31 days for a confirmed BLL of 20 to 44 µg/dL. To allow for time to schedule a testing appointment or arrange the admission, we defined a patient who received a follow-up blood lead test or was admitted to the hospital for medical evaluation for lead exposure within 2 days after a confirmed BLL of 45 µg/dL or greater as having appropriate follow-up testing. If the patient had at least 1 incident of a confirmed BLL ≥ BLRV, with no follow-up lead test or a follow-up lead test beyond the CDC’s recommended timeframe or if the patient with a concerning BLL was not admitted to the hospital within the timeframe as described above, the patient was categorized as not having appropriate follow-up testing before hospitalization.
Statistical Analysis
We used descriptive statistics to show the number of patients hospitalized for lead poisoning; length of inpatient stay; hospital charges by year of diagnosis; demographic characteristics; primary payment type; maximum BLL before admission; developmental, behavioral, and emotional disorders; and county of residence. We performed a linear regression analysis to test if there is a statistically significant linear trend for the number of lead poisoned inpatients over years 2015 to 2021. We presented the number and proportion of lead poisoned inpatients who have received blood lead testing and appropriate follow-up testing before hospitalization by demographic characteristics, maximum BLL before admission, and developmental, behavioral, and emotional disorders. All statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC).
Results
Ninety three children under 6 years of age were hospitalized with lead poisoning as the primary discharge diagnosis between 2015 and 2021, incurring 443 inpatient days and approximately 6 million dollars for total hospital charges. Thirteen children (14%) had readmissions for lead poisoning and no deaths were noted. Eighty two inpatients (88.2%) had their hospital charges covered by Medicaid insurance, at a total cost of $5.7 million in charges. The number of hospitalized patients for lead poisoning increased from 4 in 2015 to 23 in 2021, with a statistically significant linear trend (P = .0036; coefficient of determination [R2] = 0.84) (Fig 1). Almost 70% of the inpatients were males. Most hospitalizations occurred in children aged 2 years (N = 34) and 1 year (N = 26). Three infants were hospitalized for lead poisoning during the study period. Black children had the largest number of hospitalizations (N = 34), followed by white children (N = 25) and Hispanic children (N = 22) (Table 1).
Number of lead poisoning hospitalized patients by year with the trend line of linear regression between 2015 and 2021.
Number of lead poisoning hospitalized patients by year with the trend line of linear regression between 2015 and 2021.
Number of Lead Poisoning Hospitalized Patients, Length of Stays, and Inpatient Charges by Sociodemographic and Hospitalization Characteristics for Children Under 6 Years of Age, Pennsylvania 2015 to 2021
. | Hospitalized Patients (N) . | Total Length of Stays, Day . | Average Length of Stays Per Patient . | Total Inpatient Charges, ($) . | Average Inpatient Charges Per Patient, ($) . |
---|---|---|---|---|---|
Year | |||||
2015 | 4 | 14 | 3.5 | 165 316 | 41 329 |
2016 | 11 | 41 | 3.7 | 289 516 | 26 320 |
2017 | 12 | 56 | 4.7 | 533 768 | 44 481 |
2018 | 17 | 75 | 4.4 | 984 030 | 57 884 |
2019 | 15 | 42 | 2.8 | 387 552 | 25 837 |
2020 | 16 | 78 | 4.9 | 2 089 193 | 130 575 |
2021 | 23 | 137 | 6.0 | 1 542 087 | 67 047 |
Sex | |||||
Male | 65 | 313 | 4.8 | 4 647 267 | 71 496 |
Female | 28 | 130 | 4.6 | 1 344 193 | 48 007 |
Age at admission, y | |||||
<1 | 3 | 12 | 4.0 | 96 599 | 32 200 |
1 | 26 | 107 | 4.1 | 1 684 752 | 64 798 |
2 | 34 | 121 | 3.6 | 1 455 734 | 42 816 |
3 | 19 | 113 | 5.9 | 1 712 980 | 90 157 |
4 | 6 | 30 | 5.0 | 311 338 | 51 890 |
5 | 10 | 60 | 6.0 | 730 058 | 73 006 |
Race and ethnicity | |||||
Non-Hispanic white | 25 | 93 | 3.7 | 972 132 | 38 885 |
Non-Hispanic Black | 34 | 186 | 5.5 | 2 254 603 | 66 312 |
Hispanic | 22 | 102 | 4.6 | 1 491 198 | 67 782 |
Other or unknowna | 12 | 62 | 5.2 | 1 273 526 | 106 127 |
Primary payment type | |||||
Commercial | 9 | 28 | 3.1 | 211 938 | 23 549 |
Medicaid | 82 | 404 | 4.9 | 5 701 472 | 69 530 |
Self-pay | 2 | 10 | 5.0 | 74 540 | 37 270 |
Other or unknown | 1 | 1 | 1.0 | 3510 | 3510 |
Number of admission(s) | |||||
1 | 80 | 288 | 3.6 | 3 586 338 | 44 829 |
2 | 8 | 58 | 7.3 | 765 763 | 95 720 |
3 | 4 | 74 | 18.5 | 704 042 | 176 011 |
4 | 1 | 23 | 23.0 | 935 317 | 935 317 |
Maximum BLL before admission | |||||
<45 | 11 | 27 | 2.5 | 270 030 | 24 548 |
45–59.9 | 35 | 118 | 3.4 | 813 731 | 23 249 |
60–79.9 | 27 | 132 | 4.9 | 1 292 442 | 47 868 |
80–99.9 | 12 | 82 | 6.8 | 2 439 783 | 203 315 |
≥100 | 8 | 84 | 10.5 | 1 175 474 | 146 934 |
Developmental, behavioral and emotional disorders | |||||
Yes | 44 | 236 | 5.4 | 3 362 667 | 76 424 |
No | 53 | 207 | 3.9 | 2 628 793 | 49 600 |
County of residence | |||||
Philadelphia | 32 | 204 | 6.4 | 3 502 225 | 109 445 |
Allegheny | 7 | 19 | 2.7 | 106 435 | 15 205 |
Berks | 7 | 19 | 2.7 | 161 985 | 23 141 |
Delaware | 6 | 16 | 2.7 | 201 226 | 33 538 |
Lackawanna | 6 | 35 | 5.8 | 466 144 | 77 691 |
Lancaster | 6 | 52 | 8.7 | 447 790 | 74 632 |
Other | 23 | 82 | 3.6 | 969 767 | 42 164 |
. | Hospitalized Patients (N) . | Total Length of Stays, Day . | Average Length of Stays Per Patient . | Total Inpatient Charges, ($) . | Average Inpatient Charges Per Patient, ($) . |
---|---|---|---|---|---|
Year | |||||
2015 | 4 | 14 | 3.5 | 165 316 | 41 329 |
2016 | 11 | 41 | 3.7 | 289 516 | 26 320 |
2017 | 12 | 56 | 4.7 | 533 768 | 44 481 |
2018 | 17 | 75 | 4.4 | 984 030 | 57 884 |
2019 | 15 | 42 | 2.8 | 387 552 | 25 837 |
2020 | 16 | 78 | 4.9 | 2 089 193 | 130 575 |
2021 | 23 | 137 | 6.0 | 1 542 087 | 67 047 |
Sex | |||||
Male | 65 | 313 | 4.8 | 4 647 267 | 71 496 |
Female | 28 | 130 | 4.6 | 1 344 193 | 48 007 |
Age at admission, y | |||||
<1 | 3 | 12 | 4.0 | 96 599 | 32 200 |
1 | 26 | 107 | 4.1 | 1 684 752 | 64 798 |
2 | 34 | 121 | 3.6 | 1 455 734 | 42 816 |
3 | 19 | 113 | 5.9 | 1 712 980 | 90 157 |
4 | 6 | 30 | 5.0 | 311 338 | 51 890 |
5 | 10 | 60 | 6.0 | 730 058 | 73 006 |
Race and ethnicity | |||||
Non-Hispanic white | 25 | 93 | 3.7 | 972 132 | 38 885 |
Non-Hispanic Black | 34 | 186 | 5.5 | 2 254 603 | 66 312 |
Hispanic | 22 | 102 | 4.6 | 1 491 198 | 67 782 |
Other or unknowna | 12 | 62 | 5.2 | 1 273 526 | 106 127 |
Primary payment type | |||||
Commercial | 9 | 28 | 3.1 | 211 938 | 23 549 |
Medicaid | 82 | 404 | 4.9 | 5 701 472 | 69 530 |
Self-pay | 2 | 10 | 5.0 | 74 540 | 37 270 |
Other or unknown | 1 | 1 | 1.0 | 3510 | 3510 |
Number of admission(s) | |||||
1 | 80 | 288 | 3.6 | 3 586 338 | 44 829 |
2 | 8 | 58 | 7.3 | 765 763 | 95 720 |
3 | 4 | 74 | 18.5 | 704 042 | 176 011 |
4 | 1 | 23 | 23.0 | 935 317 | 935 317 |
Maximum BLL before admission | |||||
<45 | 11 | 27 | 2.5 | 270 030 | 24 548 |
45–59.9 | 35 | 118 | 3.4 | 813 731 | 23 249 |
60–79.9 | 27 | 132 | 4.9 | 1 292 442 | 47 868 |
80–99.9 | 12 | 82 | 6.8 | 2 439 783 | 203 315 |
≥100 | 8 | 84 | 10.5 | 1 175 474 | 146 934 |
Developmental, behavioral and emotional disorders | |||||
Yes | 44 | 236 | 5.4 | 3 362 667 | 76 424 |
No | 53 | 207 | 3.9 | 2 628 793 | 49 600 |
County of residence | |||||
Philadelphia | 32 | 204 | 6.4 | 3 502 225 | 109 445 |
Allegheny | 7 | 19 | 2.7 | 106 435 | 15 205 |
Berks | 7 | 19 | 2.7 | 161 985 | 23 141 |
Delaware | 6 | 16 | 2.7 | 201 226 | 33 538 |
Lackawanna | 6 | 35 | 5.8 | 466 144 | 77 691 |
Lancaster | 6 | 52 | 8.7 | 447 790 | 74 632 |
Other | 23 | 82 | 3.6 | 969 767 | 42 164 |
a Other or unknown race and ethnicity includes all other, unknown, or missing races and ethnicities.
Eighty two and forty seven inpatients (88.2% and 50.5%) had a maximum lead test result of ≥45 and ≥60 µg/dL before hospitalization, respectively. Forty four inpatients (47.3%) had diagnosis codes indicative of developmental, behavioral, and emotional disorders. Most children resided in populous counties in Pennsylvania. However, some counties had disproportionally larger numbers of patients hospitalized for lead poisoning. For example, approximately one-third of the hospitalized children lived in Philadelphia County, which comprised only 15% of the state’s population of children under 6 years of age. In contrast, Allegheny County comprised 9.5% of the state’s population of children under 6 years of age and accounted for 7.5% of all hospitalized children (Table 1).
Nineteen inpatients (20.4%) did not have any blood lead test for regular screening for lead exposure before hospitalization. The lead testing rates were similar in male (20%) and female (21.4%) patients. Among different race and ethnicity groups, 26.5% of Black children did not have any lead test before hospitalization, followed by Hispanic (22.7%) and white children (12%). Children with developmental, behavioral, and emotional disorders had a higher lead testing rate before hospitalization (88.6%) than those without such disorders (71.4%) (Table 2).
Number and Proportion of Lead Poisoning Hospitalized Patients Having a Blood Lead Test Before Hospital Admission by Demographic and Hospitalization Characteristics
. | Blood Lead Testing . | |||
---|---|---|---|---|
Yes . | No . | |||
. | N . | % . | N . | % . |
Total | 74 | 79.6 | 19 | 20.4 |
Sex | ||||
Male | 52 | 80 | 13 | 20 |
Female | 22 | 78.6 | 6 | 21.4 |
Race and ethnicity | ||||
Non-Hispanic white | 22 | 88 | 3 | 12 |
Non-Hispanic Black | 25 | 73.5 | 9 | 26.5 |
Hispanic | 17 | 77.3 | 5 | 22.7 |
Other or unknowna | 10 | 83.3 | 2 | 16.7 |
Maximum BLL before admission, µg/dL | ||||
<45 | 9 | 81.8 | 2 | 18.2 |
≥45 | 65 | 78.3 | 17 | 21.7 |
Developmental, behavioral, and emotional disorders | ||||
Yes | 39 | 88.6 | 5 | 11.4 |
No | 35 | 71.4 | 14 | 28.6 |
. | Blood Lead Testing . | |||
---|---|---|---|---|
Yes . | No . | |||
. | N . | % . | N . | % . |
Total | 74 | 79.6 | 19 | 20.4 |
Sex | ||||
Male | 52 | 80 | 13 | 20 |
Female | 22 | 78.6 | 6 | 21.4 |
Race and ethnicity | ||||
Non-Hispanic white | 22 | 88 | 3 | 12 |
Non-Hispanic Black | 25 | 73.5 | 9 | 26.5 |
Hispanic | 17 | 77.3 | 5 | 22.7 |
Other or unknowna | 10 | 83.3 | 2 | 16.7 |
Maximum BLL before admission, µg/dL | ||||
<45 | 9 | 81.8 | 2 | 18.2 |
≥45 | 65 | 78.3 | 17 | 21.7 |
Developmental, behavioral, and emotional disorders | ||||
Yes | 39 | 88.6 | 5 | 11.4 |
No | 35 | 71.4 | 14 | 28.6 |
a Other or unknown race and ethnicity includes all other, unknown, or missing races and ethnicities.
Of inpatients, 65.6% did not have appropriate follow-up testing after identification of a confirmed BLL ≥ BLRV. The proportion of inpatients with appropriate follow-up testing was slightly lower in the females (60.7%) than that in the males (67.7%). Among different race and ethnicity groups, 80% of white children did not receive appropriate follow-up testing before hospitalization, followed by Black (61.8%) and Hispanic children (59.1%). Almost 70% of the inpatients with a maximum BLL ≥ 45 µg/dL did not have appropriate follow-up testing before hospitalization as compared with 36.4% of the inpatients with a maximum BLL < 45 µg/dL. Although inpatients with developmental, behavioral, and emotional disorders were more likely to have lead testing before hospitalization, only 27.3% of these inpatients had appropriate follow-up testing, compared with children without such disorders (40.8%) (Table 3).
Number and Proportion of Lead Poisoning Hospitalized Patients Having the Appropriate Follow-up Before Hospital Admission by Demographic and Hospitalization Characteristics
. | Appropriate Follow-up Testing . | |||
---|---|---|---|---|
Yes . | No . | |||
. | N . | % . | N . | % . |
Total | 32 | 34.4 | 61 | 65.6 |
Sex | ||||
Male | 21 | 32.3 | 44 | 67.7 |
Female | 11 | 39.3 | 17 | 60.7 |
Race and ethnicity | ||||
Non-Hispanic white | 5 | 20 | 20 | 80 |
Non-Hispanic Black | 13 | 38.2 | 21 | 61.8 |
Hispanic | 9 | 40.9 | 13 | 59.1 |
Other or unknowna | 5 | 41.7 | 7 | 58.3 |
Maximum BLL before admission, µg/dL | ||||
<45 | 7 | 63.6 | 4 | 36.4 |
≥45 | 25 | 30.5 | 57 | 69.5 |
Developmental, behavioral, and emotional disorders | ||||
Yes | 12 | 27.3 | 32 | 72.7 |
No | 20 | 40.8 | 29 | 59.2 |
. | Appropriate Follow-up Testing . | |||
---|---|---|---|---|
Yes . | No . | |||
. | N . | % . | N . | % . |
Total | 32 | 34.4 | 61 | 65.6 |
Sex | ||||
Male | 21 | 32.3 | 44 | 67.7 |
Female | 11 | 39.3 | 17 | 60.7 |
Race and ethnicity | ||||
Non-Hispanic white | 5 | 20 | 20 | 80 |
Non-Hispanic Black | 13 | 38.2 | 21 | 61.8 |
Hispanic | 9 | 40.9 | 13 | 59.1 |
Other or unknowna | 5 | 41.7 | 7 | 58.3 |
Maximum BLL before admission, µg/dL | ||||
<45 | 7 | 63.6 | 4 | 36.4 |
≥45 | 25 | 30.5 | 57 | 69.5 |
Developmental, behavioral, and emotional disorders | ||||
Yes | 12 | 27.3 | 32 | 72.7 |
No | 20 | 40.8 | 29 | 59.2 |
a Other or unknown race and ethnicity includes all other, unknown, or missing races and ethnicities.
Discussion
Our analysis shows that the number of hospitalizations for lead poisoning increased from 2015 to 2021 in Pennsylvania. The financial burden associated with lead poisoning hospitalization was high, with almost 6 million dollars in hospital charges for the 93 inpatients, and almost 90% of total charges were paid by Medicaid. Of inpatients, 14% were readmitted for lead poisoning, meaning they were continuously or repeatedly exposed to lead after discharge from the hospital. To reduce the possibility of readmissions for lead poisoning, prompt follow-up care and management after discharge, including home assessment and remediation, regular blood lead testing, and healthcare coordination, are needed.14
In this study, the number of lead poisoning hospitalization in male children was 2.3-fold higher than that in female children, and almost two-thirds of cases occurred in toddlers (1 and 2 years). These disparities are probably because of increased lead exposure caused by increased hand-to-mouth activity and exploratory behavior in males and toddlers. Among all lead poisoned inpatients, children with Medicaid accounted for an overwhelming proportion, meaning economically disadvantaged children were at higher risk of lead poisoning. This provides strong evidence to support mandatory blood lead testing policy that requires children enrolled in Medicaid be tested for lead exposure before 6 year of age. Of the lead poisoned inpatients, 36.6% and 26.9% were Black and white children, respectively. As compared with the background population of children from which the sample originated (12.8% and 63% of the state population were Black and white children, respectively), this suggests Black children bear a disproportionate burden of lead poisoning. In addition, our analysis visually shows geographic variations in lead poisoning hospitalization at the county level in Pennsylvania.
Given that the hospital charge for lead poisoning hospitalization mainly depends on the number of hospitalized patients and hospital length of stay, it is noteworthy that total hospital charges of lead poisoning hospitalization increased over years, and the charges were relatively high for children who were male, toddlers, Black, enrolled in Medicaid, and had higher levels of maximum BLLs. To minimize the above-mentioned disparities in lead poisoning hospitalization and associated healthcare costs, implementing targeted screening and intervention programs, prioritizing environmental remediation in minority neighborhoods, raising awareness for lead poisoning risks through targeted outreach and education, and equitable access to healthcare are necessary.
Because most lead poisoning is insidious and only presents with symptoms when blood lead accumulates to a high level and has already impacted the body systems to produce severe symptoms and disease. Therefore, regular blood lead testing and appropriate follow-up testing after identification of a high BLL are recommended by CDC as an essential prevention strategy for children at risk for lead exposure. A previous study indicated that only 42% of pediatricians and 24% of family practitioners reported they screened the majority of the children by 2 years of age, and about 60% of all respondents reported not providing lead exposure education to half their patients.17 In another survey published in 2017, about half of the physicians said they did not follow the CDC's recommended screening test guidelines.18 Previous studies have shown that low blood lead screening and follow-up testing rates persist disproportionately among racial and ethnic minorities and people with certain sociodemographic characteristics, enrolled in Medicaid, and living in disadvantaged neighborhoods.19–23 In addition, substantial geographic variations presently exist in the childhood blood lead screening and follow-up testing rates within a city or state.20,24
We found that almost 80% of lead poisoning hospitalized patients received a blood lead test before hospitalization, whereas only approximately one-third of inpatients had appropriate follow-up testing after identification of a confirmed high BLL. Further, we found the racial disparity in lead testing rates and follow-up testing rates among lead poisoned inpatients. These findings mean missed intervention opportunities before hospitalization for lead poisoning and imply that some barriers can impede inpatients’ blood lead testing and appropriate follow-up testing. Our available data do not allow us to identify any barriers and further research is needed to identify specific barriers; however, it has been known that these barriers might be caused by parents’ lack of awareness, limited access to healthcare, healthcare cost, lack of screening and follow-up testing protocols, healthcare providers’ knowledge gaps, and inadequate public health efforts.25,26 Addressing these barriers requires a comprehensive and coordinated approach involving public health initiatives, healthcare provider and parent education, community outreach, and targeted lead exposure screening and intervention policy.
According to our analysis findings of the hospital charges, we suggested that increased investments in targeted lead prevention programs, such as improvements in the conduction of blood lead testing and follow-up testing among the high-risk populations, would be a cost-effective strategy to protect children from lead hazards and generate more benefits for the healthcare system and patient’s family besides healthcare cost savings.
This study has several limitations. First, the coronavirus disease 2019 pandemic may impact the numbers of lead poisoning hospitalization after 2020 because of the disruptions in healthcare access, lead exposure changes with longer stay-at-home time, or delayed lead testing; however, we need more years of data to dig into this issue. Second, given the limited availability of relevant information in data used in analysis, we were unable to explore and determine some factors responsible for not having blood lead testing or appropriate follow-up testing. Third, children’s blood lead tests done outside of Pennsylvania before hospitalization and a small portion of point-of-care blood lead tests that were not reported to PA-NEDSS result in incomplete lead testing data, which possibly makes authors unable to capture the true blood lead testing and follow-up testing rates. In this study, there were 11 inpatients with maximum BLLs < 45 µg/dL before hospitalization, and 6 of them with maximum BLLs < 25 µg/dL. These are BLL values that usually do not require hospitalization. We analyzed these children’s secondary diagnosis codes and found some of them had other medical conditions, such as anemia and neurologic disorders. It means that even if we had selected only patients with lead poisoning as the primary diagnosis code, misclassification may occur. Finally, some lead poisoned inpatients had a secondary diagnosis code of developmental, behavioral, and emotional disorders. It is not possible, in this analysis, to determine whether such disorders were related to previous lead exposure or the disorders resulted in behaviors that put these children at increased risk of lead poisoning.
Conclusions
Although the public health focus has been changed to lower lead exposure and its chronic health effects in children, severe lead poisoning is still a significant public health issue, with detrimental health consequences and enormous economic burdens. Therefore, further collaboration and engagement of multiple stakeholders are essential to address childhood lead poisoning and improve high-risk populations’ compliance with the recommendations for blood lead testing and follow-up testing.
Acknowledgments
We thank the Childhood Lead Poisoning Prevention Program within the Pennsylvania Department of Health Child and Adult Health Services for its assistance in collecting blood lead test data.
Dr Chen contributed to study conceptualization and design, analyzed and interpreted the data, drafted the manuscript, and edited and revised the final manuscript; Dr Ma conceptualized and designed the study, conducted initial analyses and data interpretation, drafted the initial manuscript, and critically reviewed and revised the manuscript; Ms Davis coordinated data collection and reviewed and revised the manuscript; Dr Bogen provided input in data interpretation and critically reviewed and revised the manuscript for important intellectual content; Dr Watkins provided specific expertise in data analysis and interpretation and critically reviewed and revised the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
- CDC
Centers for Disease Control and Prevention
- BLL
blood lead level
- BLRV
blood lead reference value
- ICD-9
International Classification of Diseases, Ninth Revision
- ICD-10
International Classification of Diseases, Tenth Revision
- PA-NEDSS
Pennsylvania Department of Health National Electronic Disease Surveillance System
References
Competing Interests
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest relevant to this article to disclose.
Comments