Children encounter the toxic metal cadmium frequently in early life. It is present in tobacco smoke, a variety of foods, air pollution, and consumer products like cheap jewellery, and some plastics. Cadmium is a human carcinogen (Group 1)  and is toxic to kidneys, skeletal and respiratory systems, and neurodevelopment [2–5]. Early life exposures to cadmium are of particular concern [3,6–9]. Globally, hundreds of millions of people are exposed to elevated cadmium beginning in early life  (Figure 1).
Figure 1. Graphical abstract for this study. Abstracts in Persian and Chinese can be found in file S1 in the Online Supplementary Document.
Low- and middle-income countries (LMICs) face a heavy burden of cadmium contamination due to industrialization [11–13]. Developing countries experience extensive uncontrolled cadmium release into the environment [14,15]. Food crops irrigated with cadmium-contaminated water are major sources of exposure . Rice (Oryza sativa) is a staple crop for over half of the world’s population and provides more than 20% of the calories consumed globally . It is also a known high-accumulator of cadmium from the environment, making it a major source of cadmium in LMICs [10,18–20]. Undernutrition and micronutrient deficiency has been found to increase cadmium uptake in pregnant mothers in LMICs [21–23]. Cadmium is commonly used in cheap jewellery and toys sold in LMICs such as Nigeria , Cambodia , and China , either made locally or imported. A study from China reported that cadmium exposure from food contributes substantially to stroke and coronary heart disease burden .
E-waste also contains cadmium (eg, nickel-cadmium batteries) and is transported in large quantities – often illegally – from high-income countries (HICs) to LMICs . Major destinations for e-waste are Africa, Southeast Asia, Central America, and South America, where some components can be recycled . Due to inadequate e-waste management infrastructure, hazardous e-waste materials frequently end up in landfills, open-air burning, or open dumping sites near residential areas . Due to its toxicity, bioavailability, and soil concentration at e-waste dump sites, cadmium is considered a high environmental risk, especially for children [28,30]. At an informal e-waste recycling site in Lagos State, Nigeria, soil cadmium levels exceeded Dutch soil guideline values , along with elevated levels of antimony, chromium, copper, lead, manganese, nickel, and zinc.
Cadmium exposure is a global problem. Factories refining non-ferrous metals or recycling cadmium-containing scrap and e-waste, and waste incinerators (especially of cadmium-containing batteries and plastics) are sources of cadmium pollution [31–33]. Airborne particles containing cadmium can travel thousands of kilometres from their source . Mining and drainage from mines (cadmium is a by-product of zinc mining) and waste disposal contaminates bodies of water. Foods grown in environments contaminated by cadmium from industrial emissions and runoff can also accumulate cadmium, such as grains, vegetables, meat, and seafood . Tobacco leaves also naturally accumulate relatively high concentrations of cadmium . Sewage sludge, manure, and some fertilizers may leach cadmium into agricultural soil . Contaminated roadside soils are frequently used for growing food crops . In many areas, cadmium accumulates in soil more quickly than it is removed, resulting in a gradual increase in cadmium in soil and crops .
Due to the widespread and increasing nature of cadmium pollution, it is critical to identify methods to prevent and reduce exposures in early life. At a policy level, a number of interventions have aimed to reduce cadmium exposure through both remediation and prevention . In the best-known example, hundreds of victims – primarily women – developed Itai-itai disease from cadmium pollution emanating from the Mitsui Kamioka mine in Toyama Prefecture, Japan, over several decades until the 1960s . The victims eventually succeeded with extensive advocacy and legal action, and decontamination measures were implemented which significantly remediated the pollution . Such extensive – and expensive – clean-up of cadmium pollution is exceedingly rare. In many cases, local and national governments in LMICs have called for restoring degraded environments  and (occasionally) stopping agricultural production in contaminated soils or watersheds . The European Union set maximum levels for cadmium in foodstuffs in 2006 and decreased them in 2021 . Targeted legislation, improved remediation, and more stringent controls through the 1998 Aarhus Protocol have contributed to significant progress in reducing cadmium emissions in high-income countries since 1990. Between 2005 and 2019, overall cadmium emissions in the European Union declined by 33% , although a few countries saw a small increase. The Convention on Long-Range Transboundary Air Pollution addresses cadmium emissions . The WHO Framework Convention on Tobacco Control  aims to reduce tobacco smoke in indoor workplaces, public transport, and public places. While this convention could potentially reduce cadmium exposures, its success has been uneven and limited, especially in low-income countries, where smoking has increased .
Still, the United Nations Environment Programme notes that “these existing efforts are likely still inadequate to eliminate or minimise cadmium exposures from anthropogenic sources globally as a whole” . Weak enforcement of quality control regulations and increased demand are, for example, important contributors to cadmium exposures from jewellery and toys in LMICs .
This review aims to identify and summarize the range of interventions implemented in LMICs for pregnancy, infancy, and childhood and assess their effectiveness for improving human health with the intention to inform policy development for reducing children’s exposure to toxic environmental chemicals.
Theoretical model of how the interventions work
Environmental pollution from toxicants such as cadmium is increasing. Cadmium is a carcinogen and toxic to multiple organ systems, and early life is a vulnerable window for exposure and toxicity. Human exposures occur through food grown on contaminated land, air pollution, and unsafe consumer products. Due to rapid industrialization, LMICs are facing growing public health risks due to cadmium. Given the widespread nature of cadmium pollution, interventions have been tested to reduce early life exposures in a wide range of settings. Interventions typically take a prevention approach (preventing human exposure to cadmium), a remediation approach (cleaning up existing cadmium from the environment), or a treatment approach (medical or nutritional treatments that attempt to reduce the impact of cadmium in the body). Any of these approaches could in theory promote human health, but in practice, their success will depend on effectiveness, feasibility, cost, penetration, and social inequalities. Large-scale clean-up of existing cadmium pollution and reduction of future pollution will require identification of realistic and cost-effective, context-specific intervention strategies. Interventions that address early life exposures may be the most effective at reducing disease burden and promoting life-long health.
This systematic review aimed to collect and assess research on programs, policies, and other interventions that aimed to reduce or prevent exposure to the toxic metal cadmium in pregnancy and childhood, specifically in LMICs. We assessed whether these interventions were effective at reducing or preventing cadmium exposures in early life. In addition, we assessed whether the interventions were associated with improved human health outcomes. Our methods were described in the protocol established before the review and registered with PROSPERO (CRD42021235435).
The literature search was done on November 10, 2020, using the following databases: Scopus, Web of Science, PubMed, Global Health Medicus, and Greenfile. The search string was developed with guidance from a research librarian with expertise in systematic reviews. It used a combination of Medical Subject Headings (MeSH) and free-text keywords and was adapted by one reviewer to fit each database. Prior to conducting a full search, test searches were done to refine the search string with help from the second reviewer and the research librarian. Once the search strings had been developed, a full search was completed. The search was unrestricted by language and publication date. Reference lists of included articles were then hand-searched for additional relevant articles to screen. Search strings and numbers of results per database are provided in the Online Supplementary Document (S1).
Duplicates were removed first in Endnote (version X9.2) and then Rayyan . Two double-blinded reviewers screened the titles and abstract for eligibility and labelled them with “include”, “exclude” or “maybe”. Any discrepancies were resolved by discussion between the two reviewers. After reaching consensus on eligible articles, the two reviewers performed double-blind full-text screening for inclusion based on the PICO criteria (Table 1). Full-text articles that were not found online were requested through the university library. If an article could not be found there either, it was excluded.
Table 1. Systematic review PICO criteria for inclusion
To be included in this review, studies needed to be primary research that included an evaluation of cadmium chemical body burden in study participants (human biomonitoring) and/or in child- or pregnancy-relevant environments or products (environmental biomonitoring), within the context of an intervention. The populations of interest were children, infants, neonates, and pregnant women in LMICs. Interventions could include policies, programs, environmental clean-up/mitigation, rehabilitation, counselling, parenting programs, nutrition programs, clinical research, health education, or other relevant interventions at the household, community, or policy levels. Only primary research was included; secondary analyses such as reviews were excluded. Inclusion and exclusion of studies are reported based on PRISMA guidelines .
Bibliography for all included articles is provided in the Online Supplementary Document (S1). Included articles were extracted by one reviewer into a template tailored for this review; all extractions were checked by the second reviewer. The template included study characteristics, type of intervention, population studied, measurement of cadmium (either human or environmental biomonitoring), outcomes of interest, and comparators. For studies that did not report all data needed for data analysis, we requested the data from the corresponding author by email. Studies were grouped into either human or environmental interventions because the study methods and outcomes were categorically different and were better suited to different analysis approaches. The template used for data extraction is provided in the Online Supplementary Document (S2).
Study quality and risk of bias were assessed using a template tailored to this review, based on modified GRADE criteria , in the categories of risk of bias, inconsistency, indirectness, imprecision, and publication bias. Included articles were all scored for quality by one reviewer; all scores were checked by a second reviewer. Table 2 and Table 3 display quality assessments for all included studies with heat colouring indicating quality (lower scores lighter, better scores darker). Quality of each included study was scored as no (0 points), partially (0.5), or yes (1 point) for each question (Box 1).
Table 2. Interventions for humans to reduce early life exposure to cadmium in LMICs*
LMIC – low- and middle-income country
*Human health outcome(s) colour coding explained: green for significant potential to improve human health, yellow for inconclusive/mixed effect, blue for neutral effect, black for detrimental effect.
Table 3. Environmental remediation interventions to reduce early life exposure to cadmium in low- and middle-income countries*
LMIC – low- and middle-income country
*Potential for human health impact colour coding explained: green for significant potential to improve human health, yellow for inconclusive/mixed effect, blue for neutral effect, black for detrimental effect.
†Cost information given where available.
Box 1. Modified GRADE criteria  used to assess study quality.
- Is the study design a randomized trial?
- Does the analysis include assessment of dose-response effects? (i.e., dose/magnitude of intervention linked to magnitude of health response/effect)
- Did the analysis include at least 3 of the most relevant confounders?
- Was there blinding of participants and personnel (i.e., no potential for performance bias)?
- Was an objective outcome used?
- Less than 20% participant dropout from enrollment to analysis? (i.e., no potential attrition bias)
- Were all relevant data reported for the outcome of interest (i.e., no potential selective reporting)? (no potential reporting bias)
- No other biases reported? (i.e., no potential of other bias)
- Did the intervention end as scheduled (i.e., not stopped early)?
- Do the authors provide confidence intervals for effect sizes?
- Was the direction of effect consistent across participants?
- Was the included outcome a direct outcome (i.e. not a surrogate/proxy outcome)?
- Was the outcome timeframe sufficient?
- Were the conclusions based on direct comparisons?
- Study included at least 10 participants in each group?
- Was there no evidence of serious harm associated with treatment?
- There was no industry influence on the study? (i.e., author affiliations or funding sources)
- Scoring: No = 0 points
- Partially = 0.5 point
- Yes = 1 point
Data synthesis and analysis
Synthesis without meta-analysis (SWiM, also known as narrative synthesis)  was used for analysis of the data extracted through the systematic review. The SWiM approach was adopted due to the wide heterogeneity of study designs and materials used in the reviewed studies (eg, diverse cadmium sources; human body burden vs environmental biomonitoring; wide variety in study quality). SWiM allowed for the best use of the available data and for capturing the complexity for this research question.
Because few studies reported data suitable for meta-analysis and few were of high quality, the review authors opted to present relevant information in table format using the units provided by the original studies. There was a lack of similar-enough data to calculate standardized effect sizes or create a standardized metric for this review.
Articles were categorized as either human interventions (Table 2) or environmental interventions (Table 3) and analysed separately due to large methodological differences. In analysing the findings, reviewers considered both the study’s quantitative results the study authors’ overall interpretation of their own intervention’s success and limitations. Reviewers assessed these for direction of effect, so the synthesis primarily used a vote counting approach based on the reported direction of effects. For each study, the intervention was categorized in Table 2 and Table 3 based on our assessment of the reported results and study authors’ interpretations as
- Green: positive. This means the results indicated significant potential to improve human health and that all reported findings and author interpretations pointed in the same direction towards better health.
- Yellow: inconclusive/mixed. This means that some of the results indicated a potential for better health outcomes while other results were either neutral or suggested a potential for worse health;
- Blue: neutral. This means that the results indicated neither the potential for better nor worse health; or
- Black: detrimental. This means that the reported findings indicate that the intervention was associated with worse health outcomes.
This review explored heterogeneity in the reported effects using tables and visual elements. Extracted data on the following factors was presented and compared: intervention characteristics, participant age (for human biomonitoring studies), time to follow-up, biomonitoring methods, cost, study quality, and human health outcome (for human biomonitoring studies) or estimated human health impact (for environmental studies). Tables are organized by intervention type. For human studies, results are presented separately for children and women, as these were two distinct target populations for interventions. Table 2 and Table 3 include information on all studies, regardless of study quality score. However, we emphasize the findings with highest study quality scores: within each sub-category of intervention from the tables (eg, human intervention to indoor air filter), results from the studies with highest quality are also described briefly in the text.
We systematically reviewed five databases (Scopus, Web of Science, Global Health Medicus, Greenfile, and PubMed) and found 4098 articles, 26 of which were relevant for this review (Figure 2 and Figure 3). The review found a mix of studies of interventions for humans and for environmental remediation, four of which examined policy-level changes [41,72–74].
Figure 2. Map indicating study locations for included studies. China and Iran were the countries with most included studies.
Figure 3. PRISMA flowchart for inclusion in systematic review.
Studies of interventions for humans (Table 2) focused on nutritional supplements (n = 7), medicine/clinical care (n = 2), or rice cooking education (n = 1). These studies generally involved human biomonitoring (eg, urine, blood, breastmilk, hair, or faeces) in children, pregnant or lactating women, or women of childbearing age. Shafiei et al.  did not include biomonitoring, but assessed change in diet. Most studies recruited a typical community population; a few examined subgroups, namely children with physician-diagnosed asthma developmental disorders and nutritional deficiency , autism , or children from middle and lower socioeconomic status . Notably, two studies [52,53] evaluated interventions on both children and pregnant women, and the results are therefore listed on separate rows in Table 2. Indoor air cleaners demonstrated effectiveness at reducing indoor concentrations of cadmium when deployed for two trimesters during pregnancy , and showed mixed effectiveness when deployed over a shorter period (two weeks) during childhood [65,66]. Nutritional and medical interventions ranged from provision of fresh jujube fruit (Ziziphus jujuba Mill.)  or probiotic yogurt (containing Lactobacillus rhamnosus) during pregnancy/postpartum , to chelation therapy in children ; studies were too small and dissimilar to compare here and warrant more extensive clinical investigation.
Most studies investigated environmental remediation interventions (Table 3), including for cooking. The most common was agricultural soil remediation (n = 6), followed by environmental regulations (n = 4), indoor air filtration (n = 3), rice cooking method (n = 3), and cookware and microwaving (n = 2). In general, these studies used environmental biomonitoring and indirect measures for estimating human health impact (eg, estimated daily intake, hazard quotient); Barn et al.  used human biomonitoring, but is included in Table 3 with other air filtration studies for consistency. Remediation of agricultural soil using biochar reduced the estimated children’s daily intake of cadmium ; application of rice husk waste to soil also lowered the child health risk index linked to tomato consumption . Other types of organic waste were less effective or ineffective at immobilizing cadmium and (in one case) even enhanced uptake of cadmium into food plants . Rice cooking in either high or low water did not show conclusive results for reducing cadmium content , but pre-rinsing five times and soaking of rice was associated in one study with lower hazard quotient for children and adults .
We reviewed only studies from LMICs. The 26 studies were conducted in 21 different countries (Figure 2), with one study  accounting for 10 of those. China (n = 8), Iran (n = 5), Pakistan (n = 3), and Russia (n = 2) were the only countries where more than one study had been conducted, with studies in China and Iran accounting for 50% of the studies.
Of the 26 studies, 14 showed significant potential to improve human health, nine showed inconclusive/mixed effect, and three showed neutral effect. The majority (n = 7) of the human intervention studies indicated associations between the intervention and improved health, while two studies showed a neutral effect. From the environmental studies, seven showed that the intervention improved health, but most environmental interventions (n = 9) showed an inconclusive/mixed effect, while one demonstrated a neutral effect on health outcomes. No studies indicated only detrimental effects of the intervention.
Six studies provided cost information for interventions, and one estimated financial benefit of the intervention. Of these studies, HEPA and activated carbon air filter (n = 3) and soil amendment (n = 2) had more than one. The cost of one HEPA and activated carbon air filter was in the range of US$100-500, with a replaceable filter which costs US$50-100 that must changed every six months. The cost of soil amendment depended on the materials used, with one study  reporting US$0.08-0.69 per kilogram of sewage sludge biochar produced and another  reporting US$3885 per hectare for biochar, phosphate materials and compost raw materials. Cost information was also retrieved from Weidenhammer et al.  where the cost for Xylan® coating of cookware was found to be US$1.00 for three pots, with a cost of equipment ranging from US$10 000 to US$20 000. Only the study by Luzhetsky et al.  estimated a financial benefit of the intervention, reporting prevented annual GDP losses of 13246 rubles per person (approximately one week of average 2018 wages) with technologies for treating physical development disorders.
Of the 26 studies included, 23 investigated other metals or pollutants in addition to cadmium. The most common contaminants included were lead (n = 19), arsenic (n = 14), zinc (n = 10), chromium (n = 9), copper (n = 8), nickel (n = 7), and manganese (n = 7). Additionally, iron, aluminum, cobalt, mercury, antimony, vanadium, molybdenum, and titanium were reported in more than one study. Outside of metals, particulate matter smaller than 2.5 µm (PM2.5) were assessed in two studies focused primarily on air pollution.
The average study quality for all studies included was 11 (scale of 0 to 17 best), based on a modified GRADE scale  (Box 1). Two of the human intervention studies were of average quality, while seven were below average; ten of the environmental studies were of average quality, while seven were below average. Five studies randomized the participants to treatment or control group [50,53,55,64,65], five reported a dose-response effect [58,59,62,64,70], and only three studies provided confidence intervals for effect sizes [50,64,73]. Three studies blinded the personnel and participants [65,66,69]. Overall, Barn et al.  was rated as the study with highest overall quality, with a score of 14.5. Data used in study quality assessment is provided in the Online Supplementary Document (S3).
Robustness of the synthesis
These studies included substantial heterogeneity, which provided limited evidence behind any single type of intervention. Where possible, similar data points were identified and extracted from all studies and presented in a transparent way via tables and text. This review attempted to reduce bias by using a modified GRADE quality scale. Given the widespread nature of cadmium contamination, interventions targeting different exposure pathways (such as food vs air pollution) will by nature have different expected levels of effectiveness, which also make them difficult to compare. Moreover, social inequalities in health (such as access to nutritional foods, health care, and education) will impact the context surrounding each intervention and are beyond the scope of this preliminary review. Nevertheless, taken together, these interventions studies underline the challenges of addressing cadmium as a public health problem in early life, and the limited tools available to solve it.
This systematic review identified a range of interventions aimed at reducing early life exposures to cadmium in LMICs. A total of 26 studies from 21 countries were included, targeting children, pregnant and postpartum women, and women of child-bearing age. All in all, these studies provided some evidence of effectiveness, but none appeared to offer a realistic solution for cadmium pollution at scale. Agricultural and food preparation interventions were relatively frequent, particularly related to rice, though deploying the agricultural interventions at scale would require substantial initial investment. Air filtration studies indicated some effectiveness in reducing indoor cadmium exposures during pregnancy. Four studies used long-term follow-up design to examine children’s cadmium exposure risks related to smelting, mining, and industrial pollution [41,72–74], none of which eliminated the cadmium pollution. The knowledge gaps identified by this review can be addressed by higher-quality studies and innovative interventions. Except for one educational program , all studies took a remediation approach, which may have contributed to the overall limited effectiveness.
Children are more exposed to cadmium than the general population, as are vegetarians, smokers, and people living in highly contaminated communities . Cadmium has drawn comparisons to lead in terms of widespread exposures and multi-system toxicity. Like lead, pesticides, and other environmental toxicants, cadmium has many entry points into the human exposome during pregnancy, infancy, and childhood. Like for other environmental toxicants, efforts to mitigate children’s exposures to cadmium have failed to sufficiently protect children’s health.
In this review, intervention designs ranged widely in their attempt to reduce human exposure to cadmium. This underlines the need to prevent uptake into the human body directly (eg, remove cadmium from foods at home, outcompete nutritionally, filter indoor air) and indirectly (eg, close smelters, remove cadmium from agricultural soils). Prevention of new exposures to environmental toxicants must be prioritized along with large-scale investment in remediation, especially in the Global South. At the same time, it is important to avoid “regrettable remediation” whereby the remediation efforts cause more damage than the substance itself . Intervention research is useful for understanding the cost-benefit trade-offs and, in the case of environmental pollution, helping guide evidence-driven, effective allocation of limited resources to protect health.
LMICs face substantially different pollution mixtures than HICs – such as worse air pollution, garbage burning, and fewer regulations on chemicals in consumer products. What’s more, the chemical cocktails children are exposed to likely exert larger toxic effects since they mix with other inequities that affect health and brain development, such as poverty, low-quality housing, malnutrition, and social stressors. Twenty-three of the studies reviewed here collected data on multiple metals or other pollutants, most commonly lead (n = 19), arsenic (n = 14), zinc (n = 10). However, none reported results from mixture analyses with cadmium. Yet the toxicity of the neurotoxicant lead appears to increase in the presence of high levels of other metals such as cadmium . Many environmental interventions target routes of exposure, rather than single toxicants. A better understanding of how interventions affect exposure to chemical mixtures – and their potentially synergistic effects – would therefore be valuable and better tailored to children’s real-world exposures.
Foodborne cadmium was estimated to account for 12 224 illnesses, 2064 deaths, and 70 513 DALYs in 2015 , making it a significant contributor to global health hazards. Rice, a staple crop for billions around the world, is a source of exposure to toxic metals. Extensive research has examined how health risks from rice are influenced by different rinsing, soaking, and cooking methods (eg, high water, low water, high-pressure cooking, and microwave) . While the studies reviewed here did show some evidence for removal of cadmium risks related to specific types of rice preparation, types of rice and study methods differed. Cadmium-related risks from rice consumption should be a focus of additional research and action to remediate contaminated rice paddies. Additional studies have explored alternate cooking methods (eg, parboiling husked rice) [80,81] and potential agricultural interventions , but did not include human exposure estimates and were therefore not included in this review.
In Iran, cadmium contamination has been identified in staple foods including cereal, legumes, canned tuna fish, vegetables, fruit juice, and egg; approximately 75% of rice samples had cadmium levels higher than the national maximum value (0.06 mg/kg) . Between 2013  and 2016 , cadmium and lead levels in human breastmilk increased in Isfahan, a large industrial city in Iran, indicating increasing pollution.
A meta-analysis of 343 studies reported lower concentrations of cadmium in organic foods, particularly cereals (eg, wheat), compared to conventional crops . While a full assessment of organic farming practices was beyond the scope of this review, six studies did investigate soil remediation interventions. Use of biochar reduced estimated children’s daily intake of cadmium. The environmental impact of burning plant waste to produce biochar was not taken into consideration in these studies, however, and may complicate the picture.
Certain probiotics may offer an opportunity to prevent uptake of metals such as cadmium into cells in the digestive tract; however research into such technologies is still in its infancy . Bisanz et al  did not find evidence for a protective effect in Tanzania of provision of locally produced probiotic yogurt containing Lactobacillus rhamnosus, either during pregnancy or childhood.
Only one included study took a prevention approach. Shafiei et al.  tested an educational intervention on knowledge of rice contamination. While the characteristics of the three sample groups were not well-balanced, both intervention groups showed an increase in consumption of local rice (described as uncontaminated) and decrease in consumption of foreign rice. Health education and parenting programs have been examined for other toxicants such as lead , bisphenol A , and various consumer materials containing hazardous chemicals , with mixed results . Education around children’s environmental health is an important area for additional prevention efforts and study.
Currently, no effective means for reducing cadmium absorption following inhalation have been reported, and no treatments other than supportive care and avoidance of additional risk are presently known for reducing latent effects on lung function . Chelation has been studied as an intervention for children with known exposure to cadmium and specific diagnoses, such as autism . Research on chelation treatment following cadmium exposure is continuing , but currently, some forms of chelation may be useful while others are ineffective, likely due to the rapid uptake of cadmium into tissue . Some chelators may even worsen cadmium toxicity. Prevention is always considered better than treatment.
Tobacco use is therefore a large source of exposure globally; for heavy smokers, daily intake of cadmium from smoking may exceed that from food. Due to extensive research on smoking and tobacco cessation (eg, ), this study did not focus on smoking-related cadmium exposures. In general, the included studies showed moderate to low quality due in several cases to lack of comparison groups, non-randomized designs, little detail provided to support results (eg, no confidence intervals reported), small sample sizes, and short follow-up durations. This review relies on intervention effectiveness reported by each of the studies included. For these reasons, in addition to the wide variety of interventions studied, meta-analysis was not possible, and effect sizes are therefore not standardized or possible to compare directly here. While this reduces this study’s ability to synthetize the effectiveness of any one specific type of analysis, it does provide a more comprehensive summary of the types of intervention research that have been implemented and the limitations in the existing evidence base.
Cadmium remains a threat to human health worldwide. Children living in rapidly industrializing LMICs face additional risks for exposure to cadmium and countless other environmental toxicants . Costs of cadmium remediation will likely require extensive investment in the short term, given the costs associated with clean-up interventions. Prevention of exposure in early life remains a stronger approach than remediation, and they must be prioritized together to promote lifelong health and well-being.
We thank Nataliia Korotkova (NTNU) for valuable assistance with Russian translation; librarian Magnus Rom Jensen (NTNU) for guidance in developing the search strategy; and Zohreh Jalili PhD and Shanshan Xu for translating the abstract into Persian and Chinese, respectively. We thank authors of the studies reviewed here who responded to our queries and provided additional information where requested.
Data availability: The full dataset supporting the reported results is available in the Online Supplementary Document extraction datafile (S2) and quality assessment datafile (S3).
Ethics: No ethical approval was required for this systematic review.