Mattress Manufacturing: Bowtie Dermal Risk Assessment Model

Several sources indicate that dermal exposure to harmful substances in an occupational setting is a significant problem both globally and in the U.S. The authors examined the latest detailed data available from the Bureau of Labor Statistics (BLS). Polyurethane mattress manufacturing requires the use of toxic substances like Toluene diisocyanate (TDI). This presents challenges from an occupational health perspective. Dermal exposure to toxic chemicals is sometimes difficult to assess due to a lack of exposure limits. The purpose of this study is to estimate the risk of TDI, tetrahydrofuran (THF), and boric acid exposures to the mattress manufacturing operators. Field sampling methods were considered, and a new bowtie dermal risk assessment model was developed.

KEY WORDS: dermal exposure, risk assessment, bowtie dermal risk assessment model, field sampling

Polyurethane foam was created in 1937 by Otto Bayer, based on a reaction between a polyester diol and a diisocyanate. During World War II, polyurethane was used for the insulation of refrigerators and aircrafts. This polymer was cheaper and easier to shape, and it had great potential to be used in many practical applications (Gama et al., 2018). Mattress polyurethane is formed by mixing a polyol with Toluene diisocyanate (TDI) in the presence of suitable catalysts and additives.

The purpose of this research project is to evaluate TDI exposures utilizing field methods and colorimetric wipe samplers (direct-read wipes for surface chemical detection). The project was initiated due to skin and respiratory irritation complaints from mattress manufacturing workers in Columbia, South America. The authors wanted to answer the following research questions: Are colorimetric methods suitable for TDI risk assessment during mattress manufacturing? Is the bowtie dermal risk assessment model applicable to such a process?

The scope of the study was to estimate the risk of TDI, tetrahydrofuran (THF), and boric acid exposure to the mattress manufacturing operators. At the time of the sampling, there were no American Industrial Hygiene Association (AIHA) accredited laboratories in South America. TDI samples are usually collected on glass fiber filters (GFFs) or impingers (or a combination of the two) based on methods developed by the Occupational Safety and Health Administration (OSHA) and the National Institute for Occupational Safety and Health (NIOSH). However, due to restrictions in sampler storage and transportation before and after sampling, in addition to the limited ability to ship the collected samples in cold conditions within 24 hours, the authors had to consider field sampling methods. Therefore, TDI colorimetric tubes and direct-read wipes for surface TDI contamination were acquired. For this project, the authors concentrated on surface sampling and detector tubes to estimate the risk of dermal exposure.

Dermal exposure to harmful substances in an occupational setting is a significant problem globally. According to OSHA (2017), the number of cases and the rate of skin disease in the U.S. exceeds recordable respiratory illnesses. According to the latest detailed data available from the Bureau of Labor Statistics (BLS) in 2017, 24,800 recordable skin diseases or disorders were reported at a rate of 2.2 diseases or disorders per 10,000 employees, compared to 14,900 respiratory illnesses with a rate of 1.3 illnesses per 10,000 employees.

In 2015, 28,300 recordable skin diseases or disorders were reported by the BLS at a rate of 2.6 diseases or disorders per 10,000 employees, compared to 17,200 respiratory illnesses with a rate of 1.5 illnesses per 10,000 employees.

In 2013, 33,600 recordable skin diseases or disorders were reported by the BLS at a rate of 3.2 diseases or disorders per 10,000 employees, compared to 19,600 respiratory illnesses with a rate of 1.8 illnesses per 10,000 employees.

The authors summarized the number and ratios of recordable skin diseases or disorders against respiratory illnesses in the U.S. The results are presented in Figs. 1 and 2.

The trend, presented in Fig. 2, is obvious. There are approximately 1.7 recordable skin diseases or disorders for every respiratory illness. A significant number of chemicals are readily absorbed through the skin. These chemicals can cause undesirable health effects and/or contribute to the dose absorbed by inhalation of the chemical from the air (OSHA, 2017).

In many cases, absorption of chemicals through the skin can occur without being observed by the affected employee. As indicated in the latest BLS trends, the skin has a more significant route of exposure than the respiratory system. This especially applies to non-volatile chemicals. Such chemicals can remain on an employee’s clothes or work surfaces for long periods of time. It is virtually impossible to determine the number of occupational illnesses caused by skin absorption of chemicals. However, it can be estimated that the risk of skin diseases or disorders represent a significant health risk for employees and organizations.

2.1 Dermal Risk Assessment

There are a number of dermal risk assessment models available. However, one of the most user-friendly models is the Dermal Risk Assessment Model (DRAM)™  (AIHA, 2022). This tool provides a systematic screening evaluation of the relative risks of dermal exposure to material and may be especially useful for the purposes of prioritizing additional analysis for specific materials or scenarios. It runs only in Microsoft Excel (with macro-enabled mode) but requires no other software. What is unique about this tool is the fact that it includes a deterministic (single value inputs) option and a Monte Carlo simulation (distributions of input values) option.

The tool uses information about the nature of the dermal toxicity and categorical choices for exposure factors such as dermal contact area, contact frequency, dermal retention time, dermal concentration/loading, and dermal penetration potential. The factors are used in an algorithm to estimate the risk and plot it on the risk grid. The authors and developers of the tool are Jennifer Sahmel, Daniel Drolet, and Susan Arnold.

One of the limitations of the model is that it only allows for single compound/substance dermal risk assessment. The tool does not address the risk summation or additive/synergistic effects of multiple substance exposures. To address the limitations, the authors developed the Bowtie DRAM. This model is based on the well-established simplified model for assessing and managing workplace risk factors, as shown in Fig. 3.

To further develop the DRAM, the authors selected a real-case scenario to present a practical application of the newly developed method.

2.2 Practical Application of the Bowtie Dermal Risk Assessment Tool

A small company that produces foam mattresses imports a variety of chemicals used in the production process. Three chemicals were selected to demonstrate the applicability of the Bowtie Risk Assessment methodology. For instance, polymeric polyols are generally used to produce other polymers. They react with isocyanates to make polyurethanes used in mattress manufacturing. Furthermore, tetrahydrofuran and boric acid are also added to the mix. The process is shown in Fig. 4.

The production process requires the mixing of chemicals and polymeric polyols, as shown in Exhibit 1.

The mixing operator used nitrile gloves and work clothing for skin protection and a full-face respirator for respiratory protection. An industrial hygienist collected air samples from nine locations. Corresponding surface wipe samples were also collected from the same locations. The results from screening for airborne concentrations and surface contamination of TDI are presented in Table 1.

Toluene diisocyanate 0.02 detector tubes are specialized and sophisticated tubes that require 25 strokes. The standard deviation is ±30 %. Before the measurement, the lower reagent ampoule was broken and the liquid was transferred to the indication layer, resulting in the color changing to yellow. Next, the upper reagent ampoule was broken, and the liquid was transferred to the indication layer so that it returned to a white color. After performing 25 pump strokes, the sampling professional had to wait for 15 minutes before evaluating the indication. After that, the discoloration was compared to the color comparison tube, as shown in Table 1.

Direct-read wipes for surface TDI detection provided on-the-spot detection of trace chemicals on surfaces. The highly sensitive wipes offer low minimum detectable limits with few, if any, known interferences. On-the-spot testing is faster and less expensive than sending samples to a lab. Although the screening methods used in this case are no replacement for the evaluation of airborne concentrations for compliance purposes, the results provided valuable information on TDI concentrations in air at specific times and locations. Moreover, surface sampling demonstrated the advantages of using the direct-reading technique to evaluate on-the-spot presence of TDI, thus preventing cross-contamination and improving procedures for the removal of the chemical from affected areas. The design enables the safety professional involved in sampling to clearly see color changes regardless of surface dirt.

For this case study, the researchers focused preliminary on the dermal exposure. The readers are encouraged to identify other occupational risks, but the authors wanted to point out the applicability of the Bowtie DRAM. The following chemicals were addressed in the study: toluene diisocyanate (TDI), tetrahydrofuran (THF), and boric acid.

Composite surface samples were collected at key locations along work areas to evaluate the risk of exposure to employees who may be required to work in specific locations for long periods and potentially come into contact with exposed surfaces. In addition, area sampling was used to identify any potential “hot spots” and areas that may require special control measures. Not surprisingly, the surfaces close to the mixing area showed the highest concentration of TDI, as shown in Exhibit 2.

Composite dermal surface samples were collected from the hands and lower arms of the operators, as shown in Exhibit 3.

Bello et al. (2007) concluded that “Integrated animal and human research is needed to better understand the role of skin exposure in human isocyanate asthma and to improve diagnosis and prevention. In spite of substantial research needs, sufficient evidence already exists to justify greater emphasis on the potential risks of isocyanate skin exposure.” Again, in this case study, the authors addressed only the skin exposure risks.

The second substance of concern was THF. Contact with THF can severely irritate and burn the skin and eyes possibly causing eye damage. Repeated contact with the skin can cause dryness, cracking, and a rash (New Jersey Department of Health and Senior Services, 2004). The third substance, boric acid, can cause irritation, redness, and pain.

The authors proposed a new best practice model to assess the dermal exposure risks. This new model addresses the multiple chemical risks associated with the current state of mattress production. This model can be easily applied to other scenarios where multiple chemical exposures exist. The approach starts with risk identification. Using a modified OSHA Job Hazard Analysis form is shown in Fig. 5.

Next, the latest DRAM^TM was used to assess the risk involved. All three substances produced similar dermal risk assessment results under the conditions, as shown in Fig. 6.

All three substances were estimated to fall in the “Medium Risk” category. The results were transferred to the qualitative job risk assessment form, as shown in Fig. 7.

The readers will notice that most of the risk assessment methods are linear. In a way, most of the current methods don’t address the risk summation or additive/synergistic effects of multiple chemical exposures. Therefore, the Bowtie DRAM was developed. The risk levels are derived from the DRAM tool. The estimated risk levels hyperlinked to the Bowtie DRAM are shown in Fig. 8.

The readers will notice that all three substances are estimated as “Medium Risk,” but the total risk is estimated as a “High” risk category.  That is due to the possibility of TFH skin contact causing skin cracking and a rash. In such cases, TDI and boric acid can cause severe skin diseases, disorders, or even enter the bloodstream. Obviously, such dermal risk will not be considered an acceptable level of risk. Therefore, the risk has to be reduced using the Hierarchy of Risk Treatment (HoRT), as shown in Fig. 9.

If the toxic chemicals used in the manufacturing process are substituted with less toxic substances, risk re-assessment shall be performed, as shown in Fig. 10. In this case, it can be suggested to substitute TDI, THF, and boric acid, currently used with organic cotton, wool, and organic latex.

Substitution with less toxic substances is not always possible. In addition, the Occupational and Environmental Health and Safety (OEHS) professionals have to consider new risks introduced to the process. For instance, eco-friendly mattresses may have to be reinforced with carbon nanotubes. This introduces new risks that will have to be re-assessed.

This research project clearly demonstrates that field sampling methods can be used to quickly estimate the risk of toxic substances exposures. However, due to the sample size and the uniqueness of the case study, the presented methodology can be applied in a different manufacturing environment to further validate the Bowtie DRAM. This methodology saves time and shipping, and sometimes, it is the only available option. In addition, the newly developed Bowtie DRAM allows for risk summation estimation.

4.1 Benefits

• OESH Risk Reduction
• Reduced turnover rate
• Fewer legal issues
• Improved company reputation

The Bowtie Dermal Risk Assessment Model is a useful tool for risk summation estimation. It can be used as a supplement to the DRAM tools. HoRT ranks the effectiveness of various risk reduction interventions in order, starting with avoidance, elimination, substitution, minimization, simplification, engineering, warning, and then administrative and personal protection controls.

It is likely that combinations of different DRAM tools will be needed to properly communicate the dermal exposure risks.

AIHA, Dermal Risk Assessment Model (DRAM), Falls Church, VA, 2022.

Bello, D., Herrick, C.A., Smith, T.J., Woskie, S.R., Streicher, R.P., Cullen, M.R., Liu, Y., and Redlich, C.A.,  Skin Exposure to Isocyanates: Reasons for Concern, Environmental Health Perspectives, vol. 115(3), pp. 328–335, 2007. DOI: https://doi.org/10.1289/ehp.9557

Gama, N.V., Ferreira, A., and Barros-Timmons, A.,  Polyurethane Foams: Past, Present, and Future, Materials, from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6213201/, 2018.

New Jersey Department of Health and Senior Services,  Hazardous Substance Fact Sheet: Tetrahydrofuran, from: https://nj.gov/health/eoh/rtkweb/documents/fs/1823.pdf, 2004.

Occupational Health and Safety Administration,  Dermal Exposure, from: https://www.osha.gov/dermal-exposure#:~:text=In%202018*%2C%2025%2C000%20recordable%20skin,1.7%20illnesses%20per%2010%2C000%20employees, 2017.

Popov, G., Bowtie Dermal Risk Assessment Model, (2024).

Popov, G.,  A Model for Assessing and Managing Workplace Ergonomics Risk Factors, Shift, from https://shiftr2p.com/guides/a-model-for-assessing-and-managing-workplace-ergonomics-risk-factors.

Table SNR07, Illness Cases by Category of Illness – Rates, Counts, and Percent – Industry Division – 2013, U.S. Bureau of Labor Statistics, from https://www.bls.gov/iif/nonfatal-injuries-and-illnesses-tables/soii-summary-historical/ostb3968.pdf, 2013.

Table SNR07, Illness Cases by Category of Illness – Rates, Counts, and Percent – Industry Division – 2015, U.S. Bureau of Labor Statistics, 2015.

Table SNR07, Nonfatal Occupational Illnesses by Major Industry Sector and Category of Illness, U.S. Bureau of Labor Statistics, from https://www.bls.gov/…/nonfatal…illnesses-tables/…/supplemental-table-7-20… www.bls.gov › iif › supplemental-table-7-2017-national, 2017.

Table SNR07, Nonfatal Occupational Illnesses by Major Industry Sector and Category of Illness,U.S. Bureau of Labor Statistics,  from https://www.bls.gov/…/nonfatal…illnesses-tables/…/supplemental-table-7-20… www.bls.gov › iif › supplemental-table-7-2017-national, 2017.

U.S. Department of Labor, Dermal Exposure – Overview, Occupational Safety and Health Administration, from https://www.osha.gov/dermal-exposure#:~:text=In%202018*%2C%2025%2C000%20recordable%20skin,1.7%20illnesses%20per%2010%2C000%20employees, 2018.