Elevated blood lead levels of refugee children in the United States: a systematic review of recent literature (2011–2021)

Literature search

Search strategies were created in partnership with a medical librarian, which incorporated database specific subject headings and keywords related to the concepts of lead poisoning or blood lead levels, refugees or immigrants, and children. Search terms related to lead included: “lead,” “blood lead level,” “lead poisoning,” “Pb” and “BLL.” Search terms related to refugee children included: “refugee,” “asylee,” “migrant,” and “newcomer.” The terms were combined using database specific advanced search syntax; see Supplementary Material for full search strategy. Searches were performed March 2, 2021 on seven bibliographic databases: Ovid MEDLINE, Scopus, Web of Science Core Collection, APA PsycINFO (Ovid), CINAHL (EBSCO), Cochrane Library (Wiley), and Academic Search Premier (EBSCO).

Article selection process

All results were exported to Rayyan, a web-based application for systematic reviews [29]. Duplicates were removed. Two independent reviewers (JSB, JCB) examined the results, iteratively and systematically applying a PICOS criteria for inclusion/exclusion:

P (Population) Refugee children living in the US.

I (Intervention) Any intervention, including none at all.

C (Comparison) To non-refugee populations, or to CDC BLL reference level [9].

O (Outcome) Blood Lead Level (BLL) or Prevalence of Elevated BLL.

S (Study Design) Cross sectional, longitudinal, or other applicable studies measuring BLL.

Initially, both reviewers individually screened the titles of each article for eligibility using outlined inclusion and exclusion criteria. Congruence was quantified and any discordance between reviewers was resolved by another review and discussion by both reviewers together. The screening process was repeated, this time using the abstracts of all remaining articles. Finally, the methods section of each article was independently read in full by both reviewers, to ensure that the remaining articles met the PICOS criteria and were appropriate for inclusion in this review.

Eligibility criteria

Articles were excluded if the study population was not focused on refugee children ≤18 years of age. To capture effects on children, studies were excluded if they did not specifically focus on children. For example, a study that looked at the effects of lead on women of childbearing age (13–25 years) was excluded. Articles were excluded if the study location was not in the US or if the article was not in English. All studies that did not measure BLL or compare BLL to the general population or against the CDC BLL reference value were excluded. Studies were excluded if they did not specifically identify their population as refugees. In order to summarize the most recent data, studies were limited to those published in the previous 10 years, 2011–2021. The reference lists of included articles were scanned for any applicable articles that may have been missed in the retrieval of the original articles. The remaining articles were included in this review.

Study focus

Many of the studies included in this review focused exclusively on BLL in refugee children. However, a few of the studies (30–32) did not specifically study BLL, rather, they looked more broadly at a variety of health outcomes among refugee children, and BLL was only one of many outcomes examined. While these studies did report data on prevalence of BLL, the analysis surrounding lead poisoning was often less in-depth than the studies that focused primarily on lead.

Population

Case definition

Study design

The studies were primarily cross sectional (looking at one selection of samples from a certain time) or longitudinal (retrospectively looking at one participant’s BLLs over multiple time periods). In general longitudinal studies documented an “initial” BLL shortly after the participant’s arrival to the US, and then at one or more BLLs at follow up time points. Yun et al., Kotey et al., Geltman et al., Shakya and Bhatta, and Seagle et al. used data collected from various health departments, refugee agencies or clinics (generally data from the initial domestic refugee exam), and while they often looked at data collected over multiple years, only one data point was analyzed for each participant [30, 32, 35, 37, 41]. Ritchey et al. also collected one cross sectional data point, and was the only study that collected their own original data [33]. The remaining seven studies looked at data with an initial data point plus one or more follow up data points for each individual [31, 34, 36, 38], [39], [40, 42], although not all studies provided detailed results on the follow up data points.

The studies included in this review looked at a variety of covariates in their analyses. Some of the most frequently studied variables included: age, sex or gender, country of origin, year of arrival, anemia. Figure 2 provides a visual of some the most frequently analyzed variables and how they were distributed among the studies. Some of the less frequently reported variables included: vitamin levels, number of children in the family, English proficiency, route of entry, and if treatment was given for elevated BLL.

Figure 2:

Frequently examined variables.

Overall findings

All studies found high prevalence of elevated BLL in refugee children when compared to the general population. Only one study, Sandell et al., reported a prevalence of elevated BLL in refugee children that was <10%. Sandell et al.’s study did not focus solely on lead but examined many health factors in refugee children <18 years. They reported that in both cohorts of their study, one from 2007 to 2009 and another in 2013, that elevated lead levels were present, with 5.6 and 7.8% prevalence in each cohort [31]. Although this was the lowest prevalence among refugee populations in the included studies, it is still an unacceptably high prevalence.

Other studies reported the prevalence of elevated BLL in refuge children to be between 10 and 20%. Kotey et al. found that among refugee children <15 years-old resettled in Kentucky, prevalence of elevated BLL was 11.2% [37]. Pezzi et al. found in their study of refugee children aged 6 months-16 years that the overall prevalence of elevated BLL during initial testing was 19.3% [38], although elevated BLL prevalence declined by arrival year from 24.4% in 2010 to 14.4% in 2014 (p<0.001). In their study, among children with follow-up test results, elevated BLL prevalence was 22.7%: 16.3% had elevated BLLs on both initial and follow up and 6.3% had newly elevated BLL [38]. While there were some newly elevated BLLs, the overall median BLL declined significantly between initial and follow up testing from 8 μg/dL (95% CI 8.0–8.7), to 7.0 μg/dL (95% CI 6.2–7.1) [38]. Eisenberg et al. found in their sample of newly arrived refugee children that 16% had elevated BLL. They also analyzed a follow-up data point, noting that post-resettlement the mean BLL of children who were elevated on arrival had decreased, however, 24 children (6%) had an increase of ≥5 μg/dL, and 7% had a newly elevated BLL [34]. When Lupone et al. looked at refugee children aged 0–16 years who were screened in a Syracuse refugee clinic, they found that 17% of newly arrived refugee children had elevated BLL, with 15.3% between 5.0 and 9 μg/dL, and 1.7% highly elevated at >10 μg/dL [39]. On follow up, 10% had elevated BLL, with 8.3% new exposures. 2.7% had normal BLL on arrival which was elevated to >5 μg/dL at follow up. 29.9% of children were found to have increased BLL from baseline, although not necessarily elevated above the reverence level of 5 μg/dL [39]. Finally, when Seifu et al. studied refugee children aged 6 months-16 years who were seen at the International Family Medicine Clinic, they found that prevalence of elevated BLL was 13% [40]. Notably, of those with elevated BLL only 59% received adequate follow-up until a non-elevated BLL was reached. 13% received follow up testing but still had an elevated BLL, and 28% did not have any follow up test [40].

One study reported prevalence of elevated BLL between 20 and 30%. Shakya and Bhatta found in their population of refugee children <18 in Ohio that 20.2% had BLL ranging from 5-10 μg/dL and 2.1% had BLL >10 μg/dL [41]. For children under 6, prevalence of elevated BLL was 27.1%. Prevalence of elevated BLL declined from 24.4 to 16.9% when time from arrival to screening increased from 30 days to more than 90 days [41].

Some studies reported prevalence of elevated BLL of more than 30% the refugee population they examined. Geltman et al. found in their population of refugee children <7 years resettled in MA that 41.9% of refugee children had BLL ≥5 μg/dL and 7.9% had a BLL ≥10 μg/dL. They reported that mean BLLs declined from 7.58 μg/dL in 2004 to 4.03 μg/dL in 2015 [35]. Ritchey et al. found that in 2009, 72 Burmese refugee children (37%) tested positive with a BLL ≥5 μg/dL and 14 (7.1%) children tested positive for BLL ≥10 μg/dL [33]. They identified 6 new cases (had not been previously identified, who would have gone without case management if it were not for the study), and four who were born in the US [33]. Williams et al. found in their study of refugee children in Rhode Island that BLL ranged from 0-28 μg/dL with an average of 5 μg/dL. While the case definition for elevated BLL was >10 μg/dL in this study, they also reported that nearly 40% of refugee children in the study (100 of 257) were elevated in the 5–9 μg/dL range [42]. 23 children (9%) had at least one sample ≥10 μg/dL and 5 children (2%) had a sample ≥15 μg/dL. Of the 23 children with elevated BLL ≥10 μg/dL, 4 experienced an increase after their initial screening. Notably, two of them had moved from their initial housing placement to a secondary home [42]. 22 children experienced an increase of 2 μg/dL or more between the two screens, about half of these had moved to secondary housing.

One study reported prevalence based on country of origin. Yun et al. studied refugee children <19 years from 4 sites, with 3 out of 4 sites reporting BLLs for children <8 years in their sample. 4.7% had missing values, and children with missing BLL results tended to be younger than those with non-missing results. They found that refugee health profiles were distinct depending on the country of origin, and even between Burmese children who arrived via Thailand vs. those who arrived via Malaysia [30]. They found varying prevalence of BLL, defined as >5 μg/dL, depending on the country of origin: Bhutan 26.8%, Burma (via Thailand) 23.7%, Burma (via Malaysia) 10.5%, DRC 25%, Ethiopia 13.1%, Iraq 19.9%, and Somalia 19.8% [30].

Finally, two studies looked at odds ratio (OR) for elevated BLL rather than prevalence. Raymond et al. looked at refugee and non-refugee children living in the same buildings at the time of sampling, examining residents from one building in Manchester, New Hampshire and another in Providence, Rhode Island. They found that, after controlling for confounders, refugee children in Manchester were more likely to have an elevated BLL than non-refugee children living in the same building (OR 2.09, 95% CI 1.18–3.69) [36]. They did not however, find a significant difference between the two groups in Providence (OR 1.23, 95% CI 0.87–1.75) [36]. Seagle et al. looked at Cuban refugees arriving in Texas, and found that individuals paroled after arrival were less likely to screen positive for elevated BLL than those who obtained refugee status in Cuba (children ≤16 years old, 5.2 vs. 12.3%; adjusted prevalence ratio: 0.42, 0.28–0.63). Of the 1,178 children with valid BLL results, 8% were elevated (and 0.8% elevated ≥10 μg/dL).

Comparisons

Most of the studies reported the prevalence of elevated BLL among refugee children in their study, using ≥5 μg/dL, ≥9 μg/dL, or ≥10 μg/dL as the definition of elevated BLL. While some studies did not provide any direct comparison between refugee groups and the general population [30, 31, 39, 40, 42], many used a comparison to give an idea of the scope of the disparity in elevated BLL faced by refugee children.

Some studies provided a comparison between the general population in their city or state with their sample of refugee children. Ritchey et al. compared prevalence of elevated BLL in Burmese refugee children <6 years living in Indiana with all children <6 that were screened in Indiana in 2008, reporting that the prevalence of elevated BLL was 10.7 times higher in refugee children [33]. Kotey et al. provided a comparison between prevalence of elevated BLL in refugees in their study (11.2%) and overall prevalence in Kentucky (only 0.36%) [37].

Other studies provided a comparison to the national average instead of, or in addition to, the state average. Geltman et al. provided a comparison between prevalence of BLL ≥5 μg/dL in their study population (33%) and the US in 2012 (5.0%) and in Massachusetts (3.0%) [35]. Geltman et al. also provided an additional comparison for their most recent year of data, 2015, in which the prevalence of elevated BLL ≥5 μg/dL in refugee children (29%) was nearly 10 times higher of children <6 in the general population (3%) [35]. Shakya and Bhatta provided a comparison between refugee children <6 in their study population and the US and state prevalence for children <6, observing that elevated BLL prevalence in refugee children <6 was nearly 4–7 times higher than the US national prevalence of 4.7%, and the Ohio state prevalence of 6.7% [41]. Eisenberg et al. compared the initial prevalence of elevated BLL of refugee children in their study (16%) against the prevalence of the general population of children aged 1–5 years in the US (1.4%) [34]. Additionally, Eisenberg calculated the rate ratio for BLL rising to ≥20 μg/dL in the year after arrival as 12.3 (95% CI 6.2–24.5) when compared to other communities considered high-risk in MA [34].

A few studies used other groups to draw comparison against their study population of refugees. Raymond et al. looked at a comparison between refugee and non-refugee children who were living in the same apartment buildings, finding that in Manchester refugee children were more likely to have elevated BLL >10 compared to non-refugee children (OR 2.09, 95% CI 1.18–3.69), although they did not find a significant difference when looking at an apartment in Providence [36]. Pezzi et al. provided a comparison between their study population and the National Health and Nutrition Examination Survey (NHANES) estimate, with the prevalence of elevated BLL for refugee children age 1–5 (23.7%) tenfold higher than the NHANES estimated prevalence of among all 1–5 year-old children (2.3%) [38]. Seagle et al. compared elevated BLL prevalence among Cuban refugees ≤16 years who were granted refugee/parolee status in Cuba (12.3%) vs. those who were granted parolee status upon arrival in the US (5.2%) [32]. Figure 3 provides a graphical representation of studies that provided an overall elevated BLL prevalence of their study population and compared it to a general population such as a national or state average.

Figure 3:

Overall prevalence of elevated BLL in refugee (study) group, and prevalence of elevated BLL in a non-refugee comparison group, as reported by the study authors.

*Same study looked at an apartment building in providence and did not find a difference between the groups in that location.

Identified risk factors/associations

Many of the studies reported variables or ‘risk factors’ related to increased prevalence of elevated BLL within their study populations. Some of the most frequently studied variables were age, sex or gender, place of origin, year of arrival, season of testing, health variables and environmental variables. Williams et al. [42], and Sandell et al. [31] included some variables in their study, but did not report finding associations with elevated BLL prevalence in their studies.

Conclusions

This systematic review found that the literature consistently reported that refugee children are at high risk for elevated BLL. It supports that both premigration factors (lead exposures in home country, refugee camp, or during flight) and postmigration factors (exposures to lead after arrival in the US) can contribute to this disparity. Premigration risk is supported by the findings that BLL decreased from initial testing upon arrival to retesting, or had an inverse relationship between time from arrival to testing. Other studies found differences in prevalence of elevated BLL depending on the country or region of origin, suggesting that location-specific factors premigration or culturally specific factors pre and postmigration may play a role in driving the disparity. Ritchey et al. provided some evidence that specific cultural remedies, both pre and post migration, may be responsible for some disparity [33]. Other studies provided support that post-migration environmental factors, such as age of housing, may impact elevated BLL in refugee children. Findings of newly elevated blood lead levels in follow up tests also support that refugee families face continued lead hazards post-migration.

Limitations

While this review supports that refugee children are at high risk for lead exposure and elevated BLL, limitations exist. First, not all studies used venous sampling for lead tests, and some studies did not specify which type of testing was used. Capillary lead testing is known to have high false-positive rates, which can be related to the process in which the sample was collected [43, 44]. Next, not all studies collected, adjusted for, or reported the same variables. This makes it difficult to compare results and it precludes a more comprehensive understanding of which factors may be driving the observed disparity. Finally, the review is subject to all limitations as discussed by the authors of the included studies, which may include small sample size, lab errors, potential misclassification, or other data errors.

Future research

Future research is needed to better understand the scope and causes of the disparity in blood lead levels among refugee children and how to address it. Studies that look at children born in the US to refugee parents could help elucidate the extent to which post-migration factors contribute to the disparity. Studies that examine environmental variables would be helpful in understanding which environmental factors may be influencing the disparity.

Many refugee families have fled dire circumstances in their home country and may not be expecting to be confronted with additional environmental hazards immediately upon resettlement. Qualitative studies that seek to understand what refugee parents know or have been taught about lead hazards in the US would be useful, as well as qualitative or other studies that seek to understand what methods would be most effective for providing parental education on environmental or residential lead hazards, and how to protect their families.

Practical significance

This review has potentially valuable implications for those involved in refugee care in the US, such as clinicians, public health agencies, resettlement workers and policy makers. Resettlement agencies should provide education to their workers on the importance of assessing the homes that they will place newly arriving refugee into for lead hazards. Lead hazards can be invisible or nearly invisible. As such, in addition to a visual inspection, a thorough environmental assessment of potential housing units is important when seeking safe housing. This may include testing for lead dust with wipes (independently or in collaboration with local health or housing officials), as well as testing of water and exterior soil. As there is no cure or reversal of the effects of lead exposure, primary prevention is critical. Refugee families should be provided timely, comprehensive, and understandable (this may require written translations or verbal dissemination via interpreter) information on lead hazards, their effect of exposure, and how to prevent or minimize exposure if living in a residence known to have lead hazards such as peeling paint or antiquated plumbing.

Clinicians and public health agencies should be aware that refugee children are at heightened risk for exposure to lead, prior to arrival and after resettlement in the US, and should pay particular attention to children arriving from high-risk countries or regions such as Afghanistan (where BLLs were driven up by the belated banning of leaded gasoline, among other factors). Prompt testing upon arrival (per CDC guidelines) is critical to identify cases of elevated BLL, so that treatment can be given if needed, and proper education can be provided on lead, nutrition, and prevention of further exposure. Assessment of potential ongoing non-traditional causes of exposure, such as ethnic remedies, should be made.

Policy makers (local, national, and international) should be aware of the continued scourge of lead on children, who are exposed to it through no fault of their own, and carry the weight of its irreversible impact for the rest of their lives. Policy makers must be aware of the disparities in lead exposure for refugees and other minorities, and give the issue the attention it deserves-as a matter of environmental justice. Lead poisoning in refugee children is a preventable injustice. While providing refugee families (and non-refugee families) education and resources to minimize exposure to lead in their homes is important, true environmental justice calls for elimination of this risk altogether-through policies that extend funding for lead abatement in homes and schools, and replacing lead laterals and other plumbing, and strictly regulating the use of lead in commercial toys and products.

This review has found consistency among studies that demonstrate the existence of a disparity in lead exposure among refugee children, and highlighted some of the potential causes. Health professionals and policy makers are called to work towards eliminating lead exposures in refugee families, and all families.