Approaches to Assessing Risks from Chemicals Used in Toys − Rakyat Biologi

In the absence of population based measures of risks associated with chemical exposures from toys, alternative approaches to assessing or predicting risks are needed, particularly in the area of non-acute health risks. Ideally, we would know the precise chemical content of toys, the full spectrum of toxicities associated with all chemicals used in toys, the bioavailability of potentially hazardous chemicals within toys, the level of exposure from intended play and reasonably anticipated abuse, and the number of children with unsafe exposures would all be objectively measurable. There are many barriers to gathering such complete information, yet all of these categories of information are relevant and necessary if risk is to be quantitatively defined.

Most often the chemical content of toys is not readily available. Commercial toy manufacturers are likely to know relevant information on chemical hazards from safety data sheets transmitted from suppliers. It cannot be assumed, however, that toy manufacturers know the exact chemical content of all of the components of their products. Toys may be made from a number of pre-manufactured parts, such as fabric, metal pieces (springs, rods), electronic motors, beads, stuffing, etc. for which the precise chemical content may not be readily available, and chemical manufacturers often supply chemicals as a mixture that meets certain performance specifications. Absent strict specifications about chemical content, there also may be batch to batch variation in chemical mixtures used for toys. Additionally, the chemicals used in toys may change rapidly in response to market forces, or may be protected as proprietary information. The chemicals used in toys made by smaller manufacturers, at home, or in cottage industries may be less well controlled and/or undocumented.

Even if chemical content is consistent and recorded, toxicological information about chemicals used in toys may not be complete. The Forum IV information paper noted that complete basic Screening Information Data Sets (SIDS) were not available for over 80% of the 5000 high production volume (HPV) chemicals in current use.[i] Since that time, more data are becoming available on HPV chemicals. (see SIDEBAR) SIDS data, however, are for screening purposes only, and not intended to directly explore risks to children from early life-stage exposures. For non-HPV chemicals fewer toxicity data are available. Thus, there remain few data directly related to understanding the potential for early life toxicity. The SIDEBAR describes some efforts to address this knowledge gap. A wide range of potential chemical harms pertain to early life exposures, and protocols to test for some hazards are still underdevelopment. Both the dose and the timing of exposures are important in evaluating children's chemical risk. New work evaluating low-dose effects of chemical exposures on cell-to-cell signaling, endocrine sensitive pathways, and functional developmental differentiation have added to concerns about how to evaluate chemical toxicities in infants, children and adolescents. Thus, even if all chemical content were known and fully disclosed, without complete, life-stage specific, toxicological data it may be difficult to ensure chemical safety.

The bioavailability of a chemical in the context of toy safety refers to the ability of the chemical to be released from a product or toy and absorbed into a child's body via the gastrointestinal tract, the lungs, or the skin and mucus membranes. If a chemical is not bioavailable, even if it has some toxicity in pure form, it may not represent a health hazard. Both the physical design of a toy and the chemical composition can affect the bioavailability of a specific chemical. For example, a toy that contains a liquid can be problematic if the structure is not engineered to ensure that the liquid remains locked in place. A plastic toy may contain plasticizers, softeners or stabilizers to reduce fragility and breakage, but these modifying chemicals may be capable of leaching from the structural material, making them bioavailable to a child rubbing or mouthing the object.

It is well established that bioavailability does not correlate simply with chemical content; i.e. the mere presence of a chemical in a toy does not translate into exposure.[ii] Methods for estimating a child's level of exposure to chemicals from playing with toys are under active development.[iii] A variety of approaches are being explored, but none has emerged as the "gold standard." Because children interact with toys differently at different ages and developmental stages, exposure models must consider a range of behaviors including both intended use and reasonably anticipated misuse of the toy. Children interact and experiment with objects in their environment and will mouth, throw, hit, scrape, scuff, bend and break toys just to learn what will happen. They may attempt to eat small toys or toy fragments or insert such small pieces into nose, ears or other body orifices. They will cuddle, rub, or sniff surfaces of toys. Different children have different sensory preferences, and the behavior of a given child will change from day to day. Finally, while certain behaviors tend to cluster at certain chronological ages, (e.g. mouthing behaviors peak between 6 and 36 months), there is a considerable range of normal behaviors which extend well beyond the "average" or "median" age. This range is further enlarged when considering children with developmental delay or disability, mental retardation, emotional or behavior problems, or other mental health conditions.

Some of the approaches used to estimate children's exposures have involved short-term direct observation of children at play to assess behaviors such as mouthing of objects and hands, analysis of video recordings of children a play, and parental logs of specified behaviors over several days.[i]^,[ii],[iii] In some cases, adults have been used to simulate play in a variety of settings. In vitro, and mechanical approaches have been used to simulate chewing of toys and measure chemical leaching rates.[iv] All of this work has been done in industrialized, western countries, raising questions of cultural bias in the data collected, and its relevance to the rest of the world. On the other hand, these approaches can be enhanced and adjusted statistically to aid in accounting for various sources and levels of uncertainty.

For any model to be useful, it must be validated, a process which involves testing predictions for accuracy and reproducibility against empiric data from groups of representative individuals. None of the available exposure models for assessing likely or predicted exposures from toys has been validated on large numbers of children, across cultures and regions, or with matched biomonitoring and toy chemical content sampling. All of these models are based on small sample sizes and short duration of observations. For ethical reasons, it will never be possible to validate the accuracy of exposure model predictions using direct biomonitoring of children exposed to toys of known chemical content. Validated in vitro studies, and validated and controlled studies of adults will always need to be extrapolated to the special case of children and some uncertainty is unavoidable. In vitro leaching methods were developed in Europe in the late 1990s as a proposed basis for a regulatory approach to limit the risks from exposure to phthalates in teething toys, although the conclusion that these methods had been validated for this purpose received some criticism.[v]^,[vi]

Assessing chemical risks to children from exposure to toys requires considering information on the nature of the chemical itself, the actual use of the product and the characteristics of the population or populations of concern in order to establish an expected range of effects. (Figure 1) By convention, chemical risk assessment is a four part process involving 1) identifying a specific chemical hazard, 2) determining the amount or dose that causes damage, 3) estimating levels of exposure in the population at risk, and 4) generating a risk estimate usually expressed as the probability of adverse health effects from a range of specific exposures to the populations of concern. For non-cancer causing chemicals, it is generally assumed that there is a threshold effect, i.e. a level of exposure below which no harm occurs. There is no threshold assumed for carcinogens. In general, the likelihood of cancer occurring is considered proportional to the dose, and exposures are usually averaged over a 70 year lifetime regardless of when exposures occur. Regulations are then set in the case of carcinogens to prevent excess cancers above a level set by authorities, and in the case of non-carcinogens to hold exposures below the theoretical threshold level causing disease. When assessing chemical risks for children, however, additional issues must be considered.[vii] The traditional emphasis on dose may be inadequate without also considering the timing of the exposure with respect to critical windows of development. For example, recent data suggest that the relative importance of early life-stage exposures to later disease, both cancer and non-cancer, may be disproportionately large for some chemicals.[viii] This may require additional "weighting" of chemical exposures during childhood, rather than a simple averaging. In some cases there is an additional uncertainty factor added for children. The assumption that non-cancer adverse health outcomes exhibit threshold behavior is called into question when toxicities associated with early lifestage exposures are considered. An example of this is the failure to demonstrate a true threshold to the developmental neurotoxicity of lead poisoning in children.[ix] These examples illustrate the complexities involved in attempting to assess and mitigate risks, particularly non-acute risks, to children from chemicals in toys. When considering the toxic potential of a chemical used in a toy, it should be within the context of the benefits of using that same chemical, as well as the suite of alternatives to that use, i.e., changes in toy design and the use of alternatives.

[i] Reed KJ, Jimenez M, Freeman NCG, Lioy PJ. Quantification of children's hand and mouthing activities through videotaping methodology. J Exposure Analysis Environ Epidemiol 1999:9;513-520.

[ii] Koneman WH. Phthalate release from soft PVC baby toys. Report from the DUtch Consensus Group. RIVM report 613320 002. September 1998.

[iii] Greene M. Report to CPSC. Mouthing Times of Young Children from Observational Data. Page 16-48. June 2002. Available at http://www.cpsc.gov/LIBRARY/FOIA/FOIA02/brief/Fiveyearpt4.pdf

Accessed 24 May 2006

[iv] Bouma K, Schael DJ. Migration of phthalates from PVC toys into saliva stimulant by dynamic extraction. Food Additive Contaminants 2002:19;602-612.

[v] Santillo, D., Johnston, P. & Singhofen, A. (1999). Critique of the validation studies conducted to date of in vitro methods for determination of leaching rates of phthalates from PVC toys (conducted by TNO and LGC), and of the in vivo study underlying the validation of the Dutch methodology (as conducted by RIVM). Submitted to the EU Scientific Committee for Toxicity, Ecotoxicity and Environment. Greenpeace Research Laboratories Technical Note 02/99, September 1999. Available at http://www.greenpeace.to/publications_pdf/CSTEE%202001%20comments.PDF Accessed 6 June 2006

[vi]Santillo, D., Johnston, P. & Stringer, R. (2001). Comments on the opinion expressed by the CSTEE regarding the report "Validation of methodologies for the release of diisononylphthalate (DINP) in saliva simulant from toys (2001 EUR 19826 EN)". Greenpeace Research Laboratories Technical Note 09/2001, July 2001: 4 pp. Available at http://www.greenpeace.to/publications_pdf/cstee%20critique.PDF Accessed 6 June 6, 2006

[vii] IFCS FCS Working Group Chaired by Hungary. Protecting Children from Harmful Chemical Exposures; Chemical Safety and Children's Health. IFCS/FORUM-IV/11 INF 7 October 2003. Available at http://www.who.int/ifcs/documents/forums/forum4/en/11inf_en.pdf Accessed 13 May 2006

[viii] IFCS FCS Working Group Chaired by Hungary. Protecting Children from Harmful Chemical Exposures; Chemical Safety and Children's Health. IFCS/FORUM-IV/11 INF 7 October 2003. Available at http://www.who.int/ifcs/documents/forums/forum4/en/11inf_en.pdf Accessed 13 May 2006

[ix] 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(16);1517-1526.

[i]IFCS FCS Working Group Chaired by Hungary. Protecting Children from Harmful Chemical Exposures; Chemical Safety and Children's Health. IFCS/FORUM-IV/11 INF 7 October 2003. Pg 6-7. Available at http://www.who.int/ifcs/documents/forums/forum4/en/11inf_en.pdf Accessed 13 May 2006

[ii] Stringer R, Labunska I, Santillo D, Johnson P, Siddorn J, Stephensen A. Concentration of phthalate esters and identification of these additives in PVC children's toys. Environmental Sci Pollution Res 2000:7(1);27-36.

[iii] Hubal EAC, Sheldon LS, Burke JM, McCurdy TR, Berry MR, Rigas ML, Zartarian VG, Freeman NCG. Children's exposure assessment: A review of factors influencing children's exposure, and the data available to characterize and assess that exposure. Environ Health Perspect 2000:108;475-486.