Occupational scenarios and exposure assessment to formaldehyde: A systematic review

Abstract The objectives of the systematic review were to: identify the work sectors at risk for exposure to formaldehyde; investigate the procedures applied to assess occupational exposure; evaluate the reported exposure levels among the different settings. An electronic search of Pubmed, Scopus, Web of Science and ToxNet was carried out for collecting all the articles on the investigated issue published from January 1, 2004 to September 30, 2019. Forty‐three papers were included in the review, and evidenced a great number of occupational scenarios at risk for formaldehyde exposure. All the included studies collected data on formaldehyde exposure levels by a similar approach: environmental and personal sampling followed by chromatographic analyses. Results ranged from not detectable values until to some mg m−3 of airborne formaldehyde. The riskiest occupational settings for formaldehyde exposure were the gross anatomy and pathology laboratories, the hairdressing salons and some specific productive settings, such as wooden furniture factories, dairy facilities and fish hatcheries. Notice that formaldehyde, a well‐known carcinogen, was recovered in air at levels higher than outdoor in almost all the studied scenarios/activities; thus, when formaldehyde cannot be removed or substituted, targeted strategies for exposure elimination or mitigation must be adopted.

products containing and releasing formaldehyde (insulating materials, resins, glues, chipboard, plywood, fabrics, etc), 4 while others are related to activities involving combustion processes, tobacco and e-cigarettes active and passive smoking and cooking (especially frying). 3,5 Formaldehyde is a well-known occupational carcinogen and a recognized sensory irritant compound, especially for sensitive individuals, 6,7 present in many different working scenarios. 8 Indeed, formaldehyde is widely used in numerous production processes and sanitary applications due to its chemical-physical characteristics and broad spectrum microbicide activity. 9,10 The International Agency for the Research on Cancer (IARC) has identified three main occupational scenarios where workers may be exposed to formaldehyde at air concentrations significantly higher than the indoor and outdoor background levels: (i) the production of formaldehyde and/or its solutions; (ii) the production of products containing formaldehyde or during their use and (iii) the combustion of products generating formaldehyde. 8 Thus, workers in industrial production processes (resins, plastics, semi-finished wood products, furnishing accessories and textiles), 11,12 professionals of gross anatomy and pathology laboratories, veterinarians, embalmers, 9,10,[13][14][15] breeders, 16 carpenters, industrial launderers, 17 firefighters, beauticians and printingrooms workers [18][19][20] are the categories at higher risk of exposure to formaldehyde.
In this regard, a robust scientific evidence has highlighted over the years several acute and chronic adverse health effects deriving from such exposure. [21][22][23][24][25] Moreover, after a revision of the scientific literature, IARC in 2004 has classified formaldehyde as group I carcinogen with sufficient evidence for nasopharyngeal carcinoma 26 and, afterward, also for leukaemia. 8 Then, given the evidences, in 2011 the listing status of formaldehyde was changed also from "reasonably anticipated to be a human carcinogen" to "known to be a human carcinogen based on sufficient evidence of carcinogenicity" in the Twelfth Annual Report on Carcinogens of the National Toxicology Program (NTP). 27 More recently, the European Commission (EC) has reclassified formaldehyde to carcinogenic category 1B (may cause cancer by inhalation) and mutagen category 2 (suspected of causing genetic defects). 28 Following this reclassification, the EC Regulation No. 1272/2008 29 on classification, labelling and packaging of substances was amended, and the hazard classification of formaldehyde labelling was modified, as shown in Table 1.
The classification of formaldehyde as a carcinogen has led to the need to re-evaluate the risk management systems for potentially exposed workers, as implemented in the various occupational settings. In particular, the typical chemical risk assessment had to move toward a carcinogenic one. This implied the obligation to first evaluate the replacement of formaldehyde with other non-carcinogenic substances or, if not possible due to technical reasons (often due to cost-benefit constraints), to mitigate any exposure.
The need to carry out accurate occupational risk assessments for formaldehyde has therefore led to the improvement of sampling and analysis methods. In particular, occupational exposure is usually evaluated by active or passive sampling carried out in fixed positions (environmental sampling) and/or through personal samplers. As regards analysis methods, the airborne formaldehyde can currently be  24 and, consequently, the year from which the analytical methods used must surely be reliable and sensitive. In particular, specific objectives were to (i) identify the work sectors at risk of exposure to formaldehyde; (ii) investigate the managing procedures used to assess occupational exposure; (iii) evaluate the reported exposure levels among the different settings. The search terms "(occupational OR workplace OR professional) AND exposure AND formaldehyde" were used. The results obtained by the three different researchers were merged by EndNote X9 software and then all duplicates were removed.

| Inclusion/Exclusion criteria
We included all the original studies, published in Italian or English in the fixed period, reporting experimental data obtained directly by the authors on occupational exposure to formaldehyde. In vitro and animal studies as well as all kind of reviews, reports, monographs,

Practical Implications
• The findings of this systematic review provide an overall picture of the worldwide occupational scenarios at potential exposure to formaldehyde and trace evidences for targeted prevention and mitigation actions.
• Personal samplings show higher levels of airborne formaldehyde than environmental ones. Thus, both monitoring modalities should be performed for an accurate risk assessment.
book chapters and conference acts were excluded. Then, the title and the abstract of the included articles were independently reviewed by three investigators (V.C., D.M. and R.N.P.). Articles which did not fall within the inclusion criteria were excluded during this phase.
In the following phase, the full text of the remaining potentially eligible papers was independently examined by the same three investigators for final decision on their inclusion or not in the review.
During this multi-step exclusion process, any disagreement in the decision on the examined studies was discussed until consensus was reached among the three investigators. All the process was supervised by other two different investigators (C.P. and M.V.).

| Study quality and evaluation
Study quality was independently assessed by three investigators (V.C., D.M. and R.N.P.) using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Analytical Cross-Sectional Studies. 31 This tool was used to evaluate the risk of bias through a checklist of eight questions, which include sample selection (two questions), exposure evaluation (two questions), confounding factors (two questions), outcomes and appropriate statistical analysis (two questions). The possible answers for each question were "yes," "no," "unclear" and "not applicable." According to a previously bias assessment that has been already described, 32 if the answer "yes" was ≥50% of all questions, the evaluated paper was considered with low risk of bias; on the contrary, if the answer "no" was ≥50%, the risk of bias was high.
Finally, if the answer "unclear" was ≥50%, the risk of bias was considered uncertain. Studies that presented high or uncertain risk of bias were excluded from the qualitative synthesis of the present review.
All the studies included in the review were synthetized according to year of publication, country, occupational setting, exposure assessment methodology, and main results. In total, we recovered 1581 studies from all searched databases (584 from PubMed, 889 from Scopus, 108 from ToxNet) and, after removing the 461 duplicates, 1120 articles remained. Out of the remaining 1120 papers, 1051 were excluded after review of their titles and abstracts. Thus, the full-text of 69 papers were searched and evaluated considering the inclusion/exclusion criteria and the quality assessment. After the evaluation of the full-text, 18 articles were excluded for the following reasons: 2 articles evaluated the occupational exposure to other chemicals, 2 did not assess occupational exposure, estimated the exposure to formaldehyde based on data reported in national/institutional databases, reported measured performed before the year 2004, 1 just reported data on biological monitoring, 1 was a proceeding. 6,[10][11][12]18,[33][34][35][36][37][38][39][40][41][42][43][44][45] In Figure  Finally, 8 articles were excluded because at high risk of bias. [46][47][48][49][50][51][52][53] At the end of these steps, 43 articles, all with a low risk of bias, were included in the systematic review. In particular, all the papers All the included articles were grouped according to the studied occupational scenarios as follows: healthcare and research (Table 2), esthetic and wellness (Table 3), industrial (Table 4), fire fighters' and other settings (Table 5).

| Main characteristics of the studies involving healthcare and research settings
In Table 2    active environmental and personal sampling with 2-hydroxymethyl piperidine cartridge and GC-MS, 69 active environmental monitoring with a formaldehyde colorimetric/photoelectric sensor. 73 In one study the exposure was also evaluated by the use of biological monitoring. 65 The levels of environmental formaldehydes ranged from not detectable to more than 4 mg m −3 , with the highest levels recovered in hair saloons. 66,67 F I G U R E 1 PRISMA flow diagram of the literature search

| Main characteristics of the studies involving industrial settings
In Table 4

| Main characteristics of the studies involving fire fighters' and other settings
In Table 5 are reported the studies carried out in fire fighters' and other settings and a miscellanea of other occupational scenarios.
As shown in Table 5

| DISCUSS ION
The present systematic review was conducted to recover scientific evidences on occupational exposure to formaldehyde, in order to define the occupational settings at risk of exposure and the procedures applied to assess exposure levels. These two aims are even more relevant since formaldehyde was recognized as a carcinogen, making mandatory to carry out workers' health surveillance profiled on exposure data.  Besides, it is also demonstrated that formaldehyde exposure can occur not only during the handling of formaldehyde and formaldehyde-treated materials, but also through inappropriate storage of this substance or treated materials and through an ineffective local exhaust ventilation system. 61 The second main recognized sector include esthetic and wellness settings, such as beauty, hair or nail salons and spa. [64][65][66][67][68][69][70][71][72][73][74] In particular, hair dressing activities exposed to higher levels of formaldehyde respect to nail and beauty salons and spas. Probably hair dressers use routinely specific products containing formaldehyde, consciously or because not clearly reported in the label. 68,95 Likewise, detectable levels of formaldehyde were found also in nail and beauty salons due to its presence in care products used in these settings.
As regard to the industrial scenarios, 75-84 airborne formaldehyde is generally present at low levels (micrograms), except where it is directly released from resins 76 or used for its biocidal properties in dairy facilities 76,81 and fish hatcheries. 83 In addition to the well-known exposure profile of firefighters, [85][86][87] other investigated scenarios are very diversified, both in terms of settings (school campuses, laundries, copy centers etc) [88][89][90][91][92][93][94] and in terms of formaldehyde levels (from not detectable to some mg m −3 ).   Finally, given the differences of the studies included in this review in terms of exposed populations and procedures used for assessing airborne formaldehyde levels, we decided to review and summarize the results of the selected studies rather than to carry out a formal meta-analysis. Thus, statistical heterogeneity and publication bias were not assessed.

| CON CLUS IONS
The results of the present review demonstrate that there is a great number of diversified occupational scenarios at risk for formaldehyde exposure. Nevertheless, the monitoring approaches are very similar each other and based on environmental and personal sampling followed by chromatographic analyses, thus allowing data comparison.
The settings at higher levels of airborne formaldehyde resulted the gross anatomy and pathology laboratories, the hairdressing salons and some specific productive settings such as wooden furniture factories, dairy facilities and fish hatcheries. However, it is important to highlight that in almost all the studied scenarios/activities, formaldehyde was recovered in air at levels higher than outdoor. Considering that formaldehyde is a well-known carcinogen, targeted strategies for exposure elimination or mitigation (when formaldehyde cannot be removed or substituted) must be adopted.

ACK N OWLED G EM ENTS
This work was carried out within the BRiC-2018 research projects fi-

CO N FLI C T O F I NTE R E S T S
None of the authors declares any conflict of interest.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/ina.12949.