Forensic Evaluation of a Hazardous Household Product

Abstract

Product liability litigation often follows injury to a person, then retention of human factors experts to evaluate causes of the injury. The nature of the investigation may require additional evaluations by other domain experts. This forensic case involved a man who died after using denatured alcohol as a fuel. Forensic analyses indicated that a flame arrestor, plastic fuel container, and adequate warnings most likely would have prevented this fatal incident. In addition, the analyses showed that the product manufacturer failed to guard consumers from the hazards of using the product and failed to warn consumers of the hazards that accompanied use of denatured alcohol.

Keywords

denatured alcohol ethanol flame arrestor Federal Hazardous Substances Act FHSA human factors warnings product liability product design product testing product labels Consumer Product Safety Commission litigation burn accidents ISO standard ANSI standard ANSI Z535.4 ISO 3864-2

A multidisciplinary team not only determined accident causation but also identified the need for improved product design and hazard warnings.

There are many reasons for using human factors/ergonomics (HF/E) technology in product design, which includes product packaging and labeling, such as making a safer product and to prevent litigation and other losses (Cushman & Rosenberg, 1991, pp. 4–5). When use of a product results in injury and litigation, forensic HF/E experts often are retained to evaluate the causes of the injury (Nemire, 2011; Nemire, Cohen & Cohen, 2014). Typical issues to evaluate may include product defect, the reasonableness of behavior of the consumer and the manufacturer, and the adequacy of the product warnings (Deppa & Pollack-Nelson, 2014). A complete evaluation of these issues often involves a multidisciplinary approach and the efforts of experts in other fields.

The current article presents one such multidisciplinary evaluation. The authors – one HF/E consultant, one mechanical engineer, and one chemical and materials engineer – were retained as forensic experts by the plaintiff’s attorney to evaluate the causes of the litigated incident. The forensic analyses described in this article were coordinated by the plaintiff’s attorney, performed individually by each expert, and then shared with the other experts prior to their expert depositions.

Case Scenario

The forensic case involved a man who tried to light the wick of an alcohol lamp filled with denatured alcohol (Figure 1). He did not see a flame. After confirming that the lamp was filled with fuel, he assumed there was no flame because the wick was not yet wet enough with fuel, and he poured more fuel onto the wick. However, there had been a flame; he had failed to notice it. When he poured more fuel onto the wick, the flame from the wick followed the stream of fuel into the container as it was tipped up (clearing the nozzle opening of liquid, creating a flammable vapor path into the container), and caused an explosion that ruptured the bottom of the container and ejected burning fuel onto his body. The man suffered third-degree burns over much of his body, and he later died from his injuries. This scenario is based on many similar cases; the name and details have been changed to protect privacy and confidentiality.

Figure 1.

Example alcohol lamp (left) and can of denatured alcohol (right).

There are many other accounts of people who have been severely burned because they, or someone else, added denatured alcohol to an existing flame that had not been noticed (Nemire, 2014). In addition, there are numerous accounts of injuries resulting from consumers refilling lit firepots with pourable gel fuel, which also burns with an imperceptible flame. Similar incidents have occurred when users of charcoal grills add lighter fluid to charcoal that seems to be taking too long to light, resulting in flash fires and severe burns. Such incidents cause about 25% of burn injuries every year in the United States (Aleccia, 2010).

This article provides details of some of our work to evaluate the causes of injuries and death in this forensic case, which resulted in conclusions, consistent with a hazard control hierarchy (e.g., Christensen, 1986), that the product manufacturer failed to guard consumers from the hazards of using the product and failed to warn consumers of the hazards that accompanied use of denatured alcohol.

Nature of the Hazard

Significant factors in burn incidents involving denatured alcohol are related to the characteristics of the fuel. Denatured alcohol (also known as methylated spirits) is ethanol with additives, such as methanol, to make it toxic and unpleasant to taste. Both ethanol and methanol burn with a natural, pale blue flame that manufacturers refer to as “invisible” because it is difficult to notice under moderate and bright lighting conditions (Equistar, 2003, p. 101).

Ethanol is flammable and has a flash point (the temperature at which the vapor can be ignited) around 15.6°C (60°F), which is right around room temperature. Furthermore, ethanol has a flammable range of 3.3%−19.0% of vapor (Equistar, 2003, p. 59), which means that it can be readily ignitable at a wide range of conditions near room temperature. Consequently, denatured alcohol, which is mostly ethanol, is potentially a bomb given such ease of flammability (Figure 2).

Figure 2.

Plot of vapor concentration versus the temperature of flash point for ethanol.

During a deposition, one defense expert opined that it was likely that denatured alcohol vapors leaking out of the subject can had collected inside the cabinet space where the can had been stored, forming a local buildup of vapor in a closed environment. The defense expert claimed that the vapor then ignited in contact with a heat source and caused the fire. However, the outward bulging of the subject can was indicative of an internal, not external, explosion.

A second defense expert agreed with the plaintiff’s engineering experts that the physical evidence indicated that an ignition occurred inside the container of denatured alcohol, causing overpressure and subsequent rupture of the bottom of the container. Consequently, providing a means to prevent such an internal explosion would have been important to avoid the fatality.

Failure to Guard

If a hazard cannot be designed out of a system, the next best option is to guard users from the hazard (Christensen, 1986). The plaintiff’s medical expert opined that the pattern of the plaintiff’s burn injuries was consistent with holding the can in front of his torso with both hands and with the bottom of the can pointing toward his torso. Given the common, and often unwitting, behavior of pouring fuel onto flames, the manufacturer of the denatured alcohol product should have provided a means to guard from the resultant hazard.

There were at least two ways in which the manufacturer could have guarded against such incidents: provide a flame arrestor at the opening of the container, or manufacture the container from noncorrodible material.

Lack of flame arrestor

Numerous accounts of injuries resulting when consumers refill lit firepots with pourable gel fuel prompted the Consumer Products Safety Commission (CPSC) to ask manufacturers to recall bottles of pourable gel fuel that did not contain a flame arrestor (CPSC, 2011). A flame arrestor works by absorbing heat from a flame traveling through channels designed into the arrestor (Figure 3).

Figure 3.

Example of a charcoal lighter fluid cap without a flame arrestor (left) and with a flame arrestor consisting of three small holes less than 1 mm (.04 in.) in diameter (right).

A flame arrestor placed in the opening of the product container would have effectively guarded users from the hazard of an exploding fuel container (Figure 3; Stevick, Rondinone, & Sagle, 2014). However, containers of denatured alcohol without flame arrestors are still available in retail stores in the United States and other countries.

A series of tests was performed using metal quart containers of denatured alcohol from the subject manufacturer to evaluate the utility of a flame arrestor. A test structure designed specifically for testing quart-size metal containers was used for all testing. It consisted of a pivoting platform to which the container was mounted. A flame source was placed in front of and below the platform. The platform was remotely tilted until the liquid denatured alcohol poured out of the cap opening onto the flame source and then tilted back, exposing the vapor space inside the container to the external flame. (During pouring, the nozzle is typically sealed by the liquid.)

We used newly purchased containers of denatured alcohol with the standard cap (without a flame arrestor) in each of 14 tests. Immediately prior to each test, we measured the approximated vapor concentration in the head space of the container using an oxygen sensor. The amount of denatured alcohol present for each test ranged from 50 ml (1.7 fl oz) to 470 ml (1/2 quart). Test temperatures ranged from 15.6°C to 37.8°C (60°F to 100°F).

In every test, the measured concentration was within the explosive range. In each case, the poured denatured alcohol was ignited by the flame. The ignition propagated back into the container and led to an internal container explosion. The combustion was evidenced by high-velocity ejection of liquid and vapor from the container opening (Figure 4).

Figure 4.

Sequence of images from a video showing the results of testing a container of denatured alcohol with no flame arrestor. The cap with no flame arrestor, but a single opening larger than 2 mm (0.08 in), resulted in internal combustion and explosion within the container. In cases with an intact container, a jetting occurs through the opening, which can cause serious injury if someone is in the line of fire.

In two additional tests, we installed a flame arrestor in the cap opening. Two flame arrestors were used in the testing, one adapted from a RealFlame denatured alcohol container and the other, from a NuFlame denatured alcohol container. In each case, the flame was arrested and prevented combustion inside the container, even with extended direct flame exposure.

During a deposition, a defense expert correctly opined that there were no state or federal codes or standards in the United States that required containers of denatured alcohol to include a flame arrestor. However, if a manufacturer is or should be aware of a hazard associated with using its product, then it is liable for injuries that occur from foreseeable use of the product (Bass, 1986, pp. 35–49). It was foreseeable that someone might pour denatured alcohol onto a wick or fire that had seemed extinguished.

Furthermore, a flame arrestor cap costing less than five cents was available from the same manufacturer that produced the subject container cap. However, the manufacturer failed to provide flame arrestor caps for its denatured alcohol containers. Had the subject container of denatured alcohol included a flame arrestor, it would have prevented a flame from following the stream of fuel into the can, an internal explosion, expulsion of flaming liquid, and the resultant burns and fatality.

Corrodible container

Our flame arrestor testing revealed that pouring denatured alcohol on a flame in the absence of a flame arrestor resulted in explosions of the test containers that ejected burning fuel through the opening of each container rather than through the bottom of the container, as happened in the subject forensic case. Examination of the subject exploded container revealed that it had been severely corroded before the explosion (Figure 5).

Figure 5.

Bottom of exploded subject can showing corrosion.

Compositional analyses of the subject container showed that it was made of low-carbon steel and not stainless steel. The steel sheet end caps were attached by crimping, which formed a deep crevice on the inside of the can (Figure 6). The steel used for the container was susceptible to corrosion without additional coating, and the deep crevice at the bottom of the container was most susceptible to corrosion. Testing showed that the tinning of the crimped bottom edge cap of the subject can was incomplete. Because alcohol (ethanol and methanol) is less dense than water, any water moisture or other water-based contamination will naturally settle into the bottom groove when the can is in its typical upright position.

Figure 6.

Left: Optical microscopy of the cross-section of the bottom of an exemplar can, showing the sidewall, crevice, and the crimped edge. Right: Solder-filled fillet weld that eliminates the chance for crevice corrosion.

The subject container had corrosion marks on the inside walls that suggested that it had been previously opened and used, consistent with deposition testimony, and exposed to moisture from the external atmosphere. Testing showed that water placed into an open exemplar container resulted in significant corrosion after only 3 days’ exposure at room temperature. Crevice corrosion will occur internally at first and greatly weaken the crimped edge, and the end user will have no knowledge of the internal corrosion. It is likely that the explosion inside the subject container ejected burning fuel through the corroded bottom of the container and onto the user because corrosion had severely weakened the bottom seam of the subject can.

Tin coating is a low-cost anticorrosion coating. Full protection requires complete coverage of the exposed steel surface. The manufacturer easily could have mitigated the corrosion weakness by filling the crevice with solder (Figure 6). Alternatively, the manufacturer could have used different materials for constructing the container, such as plastic and aluminum, which would have eliminated water-based corrosion. Had the subject product can been treated with an anticorrosion coating, or been made of noncorrodible material, the resultant burns and fatality most likely would not have occurred.

During a deposition, a defense expert claimed that the subject incident was a result of improper storage of the can, contrary to safety instructions clearly posted on the can label. We evaluate the effectiveness of the instructions to store the can in the next section.

Failure to Warn

If a hazard cannot be designed out of a system, and if users cannot be guarded from the hazard, then the next best option is to warn users of the hazard (Christensen, 1986). Given the nature of the hazards, the least the manufacturer could have done was to warn users.

The invisible flame

As noted earlier, denatured alcohol burns with an invisible flame when viewed under moderate or bright light conditions (Equistar, 2003, p. 101; also see Figure 7 and para. 3 in Figure 8) and is clean burning, meaning that it does not leave contaminants (smoke) when burned (Equistar, 2003, p. 46). However, consumers typically visually check for the presence of a flame. Common experience with flames from fireplaces, campfires, gas ovens, alcohol flames at night or with colorizers, and other sources provide a general experience that one can visually detect the presence of a flame. Consequently, a fuel that burns with an invisible flame would surprise most consumers (Figure 7) and would constitute a hidden hazard (Cohen & Cohen, 2009; Martin, 2000; Nemire, 2014; Williams, Kalsher, & Laughery, 2006).

Figure 7.

The flame from the denatured alcohol lamp is not noticeable and catches the paper on fire.

Figure 8.

Back panel of product can showing directions and safety information and excluding the manufacturer and product name information.

Given that the flame is invisible, it would not be possible for users to comply with the warning to keep the denatured alcohol away from flame: “Danger! Flammable. Keep away from heat, sparks, flame and all other sources of ignition” (Figure 8, para. 6). If one cannot notice the flame, one would not know that he or she was bringing the denatured alcohol close to the flame.

In some countries, such as the United Kingdom, methyl violet dye is added to denatured alcohol to provide a blue color to the liquid in an attempt to prevent accidental consumption (Denatured Alcohol Regulations, 2005). According to preliminary analyses performed by the third author, the dyes added to color the liquid do not change the color of the flame. As a preliminary test, he purchased food coloring from supermarkets and added it to samples of denatured alcohol. Each of five dyes (red, yellow, green, blue, and black) was added to denatured alcohol samples until the alcohol was strongly colored. He then fed the denatured alcohol into a burner similar to that shown in Figure 7 and ignited it. There were no visible flames or colored flames in any of the five samples. The third author suspects that dyes, such as methyl violet dye and food coloring, do not have different-burning flames but, rather, are strictly coloring agents. Additional research may be the topic of follow-up papers.

One of the main purposes of warnings is to inform users about the presence of a hazard of which they may not be aware (Sanders & McCormick, 1993, p. 681). One of the defense experts in this litigation opined that the label on the subject denatured alcohol container provided sufficient warning to users about the hazards of an invisible flame. Consequently, one of the issues in the subject litigation was whether the manufacturer failed to warn users about the hazards and consequences of an invisible flame.

The denatured alcohol container label referred to an invisible flame only in the “HELPFUL TIPS” section (Figure 8, para. 3), stating, “Use caution as flame may be invisible.” This apparent attempt at providing cautionary language was inadequate in many ways. Human factors research has shown that the components of an effective warning include an effective signal word as well as hazard, consequence, and instruction statements (Rogers, Lamson, & Rousseau, 2000; Wogalter, Conzola & Smith-Jackson, 2002). Nemire (2014) noted that the subject product label was inadequate as a warning for a number of reasons, including that the heading “HELPFUL TIPS” failed to indicate a warning, that “Use caution as flame may be invisible” failed to adequately describe the hazard, and that consequence and instruction statements were absent.

If these components of an effective warning had been employed on the subject can of denatured alcohol, and if the warning had been formatted according to guidelines in ANSI Z535.4 (American National Standards Institute [ANSI], 2011), the warning might have been displayed as shown in Figure 9 (similar improved formatting and content also could have been specified by ISO 3864-2; International Organization for Standardization, 2004.) Further design and evaluation (such as described in Frantz, Rhoades, & Lehto, 1999; Robinson, 2009; and Wogalter, Conzola, & Vigilante, 2006) would be needed to determine the effectiveness of this revised warning.

Figure 9.

Example of ANSI Z535.4–formatted warning for flammability and invisible flame, which includes hazard, instruction, and consequence statements. Some of the language in this example warning comes from labels on containers of denatured alcohol from other manufacturers.

We conducted a survey to determine whether consumers were aware of the hazards of an invisible flame when using denatured alcohol, or could become aware by reading the label on the subject can of denatured alcohol (Nemire, 2014). A convenience sample of participants indicated their prior experience with denatured alcohol and listed all hazards they knew or had heard about that were related to using denatured alcohol.

Only 21% of the participants had used denatured alcohol in the past; those respondents listed hazards as flammable (9%) or poisonous (12%). For the 79% who had not used denatured alcohol before, only the hazards of flammable (6%) and poisonous (6%) were listed. None of the participants indicated that the invisible flame could pose a hazard.

Participants then were shown a photograph of the entire back panel of the product (Figure 8) and asked, “Assume you are about to use this product as a fuel in a stove. Please write down any hazards that may be associated with using denatured alcohol as a fuel” (Nemire, 2014, p. 1872). As they looked at the photograph of the product label, most wrote “flammable” (62%), “explosive vapor” (65%), and “poisonous” (76%). However, probably because of label formatting problems (Nemire, 2014; Nemire & Vigilante, 2015), only a few listed “carcinogenic” (12%) and “invisible flame” (12%) as a hazard.

The results showed that 88% of the study participants were not aware of the hazard of an invisible flame even while reading the warning label. These results were consistent with the hypothesized inadequate warning and inadequate warning format (see sidebar). Had the text about the invisible flame included the components of an effective warning (Rogers et al., 2000; Wogalter et al., 2002), and had current ANSI Z535.4 (ANSI, 2011) specifications been used for the warning text, it is likely that users would have noticed and understood that the invisible flame posed a hazard (Nemire, 2014; Nemire & Vigilante, 2015). Consistent with this hypothesis, evaluations of another hazardous household product label presented in the original Federal Hazardous Substances Act (FHSA) format and in a format complying with ANSI Z535.4 showed that participants rated the ANSI format (such as shown in Figure 9) as more likely to be noticed, read, understood, and followed (Nemire, 2016).

The corrodible container

There were no warnings on the subject container label that related to the corrosion of the can, but there were the following instructions (Figure 8, paras. 4 and 5):

Storage: . . . Keep in a cool, dry place . . . Keep container closed when not in use. Do not reuse this container.

Disposal: If container is corroded, rusted or shows signs of deterioration, dispose of product.

The instructions failed to describe the nature of the hazard (an internal explosion that could burst the container and spew flaming fuel on the user) or the consequences of not following the instructions (severe burns; Rogers et al., 2000; Wogalter et al., 2002). Further, because corrosion can start within the container and not be noticed by the consumer, the consumer would not be able to detect corrosion of the container until the container had already been considerably weakened. Consequently, the existing instructions failed to serve as a warning about the nature of the hazard or the consequences of failing to follow the instructions and would not be considered an effective warning. In fact, not a single participant in the Nemire (2014) study indicated that a corroded or rusted container posed a hazard to the user.

Conclusions

This article presented a case study that illustrated a multidisciplinary approach to evaluating the causes of a fatality that resulted from using a common household product. The multidisciplinary approach enabled consideration of the causes of the incident from different perspectives. The case of the man who died as a result of failing to notice the denatured alcohol flames is similar to many others in which people who have used denatured alcohol failed to notice its invisible flames and suffered severe burns, and sometimes death, as a consequence.

Had the manufacturer of this product installed an appropriate flame arrestor at the opening of the container of denatured alcohol and made the container with noncorrodible material, the resultant burns and fatality most likely would not have occurred. Label warnings also were shown to be inadequate; however, warnings should be used to supplement the safer design of an installed flame arrestor and noncorrodible container, not as a substitute for safer design (Christensen, 1986).

The design of effective product labels and warnings − as with any other product − should be the result of design and development phases, such as product planning, hazard identification and analyses, prototype design, and evaluation (Cushman & Rosenberg, 1991; Frantz et al., 1999; Robinson, 2009). It does not appear that manufacturers of denatured alcohol have designed their product labels and warnings with such a process. We have seen no evidence of any formal design and evaluation process in documents provided by manufacturers; warning and label design appear only to be an ad hoc process based on historical precedence and outdated federal requirements.

There are a number of ways to identify hazards and evaluate risks, such as scenario analysis, fault tree analysis, and hazard assessment (Cushman & Rosenberg, 1991; Frantz et al., 1999; Robinson, 2009). Such methods should be used to discover hazards that may result from anticipated uses or misuses of a product. It can be beneficial to develop warnings and instructions early in the design process; doing so can result in the elimination of hazards that otherwise might be discovered only late in the design process (if at all) or after the product is in production and in the hands of consumers (Cushman & Rosenberg, 1991; Frantz et al., 1999; Robinson, 2009).

In the subject case, one could easily imagine the product designers deciding, early in the design process, to include a flame arrestor, which costs only a few cents, after drafting a warning about internal explosions and the potentially deadly consequences. Had the manufacturer conducted such a design and evaluation process and effectively guarded users from the product hazards, it is most likely that the man in the subject forensic case, and others like him, would not have been injured or killed by foreseeable use of this product.

Standards Identification

U.S. and International Organization for Standardization standards relevant to the associated article are as follows:

ANSI Z535.4: American National Standard for Product Safety Signs and Labels (American National Standards Institute, 2011)

Code of Federal Regulations, Title 16, Chapter II, Subchapter C Federal Hazardous Substance Act Regulations, Part 1500: Hazardous Substances and Articles; Administration and Enforcement Regulations (Consumer Product Safety Commission, 2012)

Federal Hazardous Substances Act, Pub. L. 86–613, 74 Stat. 372–81 (15 U.S.C. 1261–74) (2011)

ISO 3864-2: Graphical Symbols—Safety Colours and Safety Signs—Part 2: Design Principles for Product Safety Labels (International Organization for Standardization, 2004)

Previous research on warnings has indicated that various guidelines provided in ANSI Z535.4 (American National Standards Institute, 2011), and neglected in the Federal Hazardous Substances Act (FHSA; 2011), have been shown to improve warnings, such as providing

a signal word in high-contrast colored lettering

message panel text in high-contrast black lettering

outline format

borders around warning content

a consequence statement

symbols

concise and explicit text

related hazard, consequence, and instruction statements grouped together

(Adams & Edworthy, 1995; Braun, Sansing, Kennedy, & Silver, 1994; Braun & Silver, 1995; Chapanis, 1994; Desaulniers, 1987; Frantz, 1993; Godfrey et al., 1991; Jaynes & Boles, 1990; Kalsher, Obenauer & Weiss, 2016; Laughery & Stanush, 1989; Laughery, Vaubel, Young, Brelsford, & Rowe, 1993; Morrow, Leirer, Andrassy, Hier, & Menard, 1998; Nemire, 2016; Otsubo, 1988; Wogalter et al., 1987; Wogalter, Begley, Scancorelli, & Brelsford, 1997; Wogalter, Kalsher, & Racicot, 1992; Wogalter & Rashid, 1998; Young, 1991).

Part of the problem with the warnings on the subject can of denatured alcohol is that the labeling of this hazardous product, and others, falls under the auspices of the FHSA, which is administered by the Consumer Product Safety Commission (CPSC). A major focus of the FHSA covers labeling requirements for hazardous substances and products. The FHSA regulations define a hazardous substance as follows:

any substance or mixture of substances which is toxic, corrosive, an irritant, a strong sensitizer, flammable or combustible, or generates pressure through decomposition, heat, or other means, if such substance or mixture of substances may cause substantial personal injury or substantial illness during or as a proximate result of any customary or reasonably foreseeable handling or use, including reasonably foreseeable ingestion by children. (CPSC, 2012, 1500.3[b][4][i][A])

The FHSA labeling requirements apply only to hazardous substances that are intended, or packaged in a form suitable, for use in the household or by children. Consequently, the FHSA labeling requirements cover a wide range of household products, including charcoal briquettes, turpentine, paint strippers, methyl alcohol, and household cleaning products.

Although the FHSA labeling requirements are updated over time, they have not benefited from current warnings research and standards creation that has developed over the past several decades. The result is that a household hazardous substance can meet the FHSA labeling requirements but fail to produce a warning that sufficiently alerts, informs, and motivates a consumer to behave in a safe manner and avoid injury (Nemire, 2014; Nemire & Vigilante, 2015). The failure to employ such updated formatting may have resulted in failure by users to notice, read, understand, and follow what warnings were on the denatured alcohol container.

– Adapted from Nemire and Vigilante (2015)

Footnotes

Kenneth Nemire, president of HFE Consulting LLC, has provided forensic human factors consulting for personal injury and product liability litigation for more than 26 years, primarily in cases involving product design and liability, warnings and instructions, falls, and driver and pedestrian perception and behavior. He earned a PhD in experimental psychology from the University of California, Santa Cruz, and is a Certified Professional Ergonomist. He can be reached at knemire@HFEconsulting.com.

Glen Stevick is the principal at Berkeley Engineering and Research, Inc. (BEAR), in Berkeley, California. He received his master’s and doctorate degrees in mechanical engineering from the University of California, Berkeley (UC Berkeley), and his BS in mechanical engineering from Michigan Technological University. He worked for Chevron as an engineering mechanics specialist and founded BEAR in 1989, where he continues to provide consulting services. He has served as an instructor for the Senior Mechanical Engineering Design Course at UC Berkeley.

David Xu, a metallurgist at Berkeley Engineering and Research, Inc., in Berkeley, California, received his BS, MS, and PhD from University of California, Berkeley. His dissertation focused on the characterization of the mechanics of biomaterials. He is involved in forensic engineering, investigating failures ranging from bridges to household appliances.

References

Adams

A. S.

Edworthy

(1995). Quantifying and predicting the effects of basic text display variables on the perceived urgency of warning labels: Tradeoffs involving font size, border weight and colour. Ergonomics, 38, 2221–2237.

Aleccia

(2010). Great balls of fire! Grill injuries can ruin your cookout. NBCNews.com. Retrieved from http://www.nbcnews.com/id/38047717/ns/health-health_care/t/great-balls-fire-grill-injuries-can-ruin-your-cookout/#.VfTKmGeFND8

American National Standards Institute. (2011). ANSI Z 535.4: American national standard for product safety signs and labels. New York, NY: Author.

Bass

(1986). Products liability: Design and manufacturing defects. Colorado Springs, CO: Shepard’s/McGraw-Hill.

Braun

C. C.

Sansing

Kennedy

R. S.

Silver

N. C.

(1994). Signal word and color specifications for product warnings: An isoperformance application. In Proceedings of the Human Factors and Ergonomics Society 38th Annual Meeting (pp. 1104–1108). Santa Monica, CA: Human Factors and Ergonomics Society.

Braun

C. C.

Silver

N. C.

(1995). Interaction of signal word and colour on warning labels: Differences in perceived hazard and behavioural compliance. Ergonomics, 38, 2207–2220.

Chapanis

(1994). Hazards associated with three signal words and four colours on warning signs. Ergonomics, 37, 265–275.

Christensen

J. M.

(1986). Forensic human factors psychology. In Kurke

M. I.

Meyer

R. G.

(Eds.), Psychology in product liability and personal injury litigation (pp. 33–79). Washington, DC: Hemisphere.

Cohen

H. H.

(2009). “Spontaneous combustion” of oil-based wood finishing products: Hidden or obvious hazard? In Proceedings of the Human Factors and Ergonomics Society 50th Annual Meeting (pp. 860–864). Santa Monica, CA: Human Factors and Ergonomics Society.

10.

Consumer Product Safety Commission. (2011). Staff briefing package for ANPR on firepots and gel fuel, December 7, 2011. Bethesda, MD: Author.

11.

Consumer Product Safety Commission. (2012). Hazardous substances and articles; administration and enforcement regulations, 16 C.F.R. pt. 1500.

12.

Cushman

W. H.

Rosenberg

D. J.

(1991). Human factors in product design. New York, NY: Elsevier.

13.

Denatured Alcohol Regulations. (2005). Retrieved from http://www.legislation.gov.uk/uksi/2005/1524/pdfs/uksi_20051524_en.pdf

14.

Deppa

S. W.

Pollack-Nelson

(2014). Consumer product safety issues in forensic human factors/ergonomics. In Nemire

Cohen

H. H.

(Eds.), Guide to forensic human factors (pp. 26–37). Santa Monica, CA: Human Factors and Ergonomics Society.

15.

Desaulniers

D. R.

(1987). Layout, organization, and the effectiveness of consumer product warnings. In Proceedings of the Human Factors and Ergonomics Society 31st Annual Meeting (pp. 56–60). Santa Monica, CA: Human Factors and Ergonomics Society.

16.

Equistar. (2003). Ethyl alcohol handbook (6th ed.). Houston, TX: Author.

17.

Federal Hazardous Substances Act, Pub. L. 86–613, 74 Stat. 372–81 (15 U.S.C. 1261–74) (2011).

18.

Frantz

J. P.

(1993). Effect of location and presentation format on attention to and compliance with product warnings and instructions. Journal of Safety Research, 24, 131–154.

19.

Frantz

J. P.

Rhoades

T. P.

Lehto

M. R.

(1999). Practical considerations regarding the design and evaluation of product warnings. In Wogalter

M. S.

DeJoy

D. M.

Laughery

K. R.

(Eds.), Warnings and risk communication (pp. 291–311). Philadelphia, PA: Taylor & Francis.

20.

Godfrey

S. S.

Laughery

K. R.

Young

S. L.

Vaubel

K. P.

Brelsford

J. W.

Laughery

K. A.

Horn

(1991). The new alcohol warning labels: How noticeable are they? In Proceedings of the Human Factors and Ergonomics Society 35th Annual Meeting (pp. 446–450). Santa Monica, CA: Human Factors and Ergonomics Society.

21.

International Organization for Standardization. (2004). ISO 3864-2: Graphical symbols—Safety colours and safety signs—Part 2: Design principles for product safety labels. Geneva, Switzerland: Author.

22.

Jaynes

L. S.

Boles

D. B.

(1990). The effect of symbols on warning compliance. In Proceedings of the Human Factors and Ergonomics Society 34th Annual Meeting (pp. 984–987). Santa Monica, CA: Human Factors and Ergonomics Society.

23.

Kalsher

M. J.

Obenauer

W. G.

Weiss

C. F.

(2016). Evaluating ANSI Z535-formatted warning labels as an integrative system. In Proceedings of the Human Factors and Ergonomics Society 60th Annual Meeting (pp. 1640–1644). Santa Monica, CA: Human Factors and Ergonomics Society.

24.

Laughery

K. R.

Sr. Stanush

J. A.

(1989). Effects of warning explicitness on product perceptions. In Proceedings of the Human Factors and Ergonomics Society 33rd Annual Meeting (pp. 431–435). Santa Monica, CA: Human Factors and Ergonomics Society.

25.

Laughery

K. R.

Sr., Vaubel

K. P.

Young

S. L.

Brelsford

J. W.

Jr. Rowe

A. L.

(1993). Explicitness of consequence information in warnings. Safety Science, 16, 597–613.

26.

Martin

B. J.

(2000). The value of explicit hazard and consequence warnings for products with hidden hazards. In Proceedings of the IEA 2000/HFES 2000 Congress (pp. 4-302–4-305). Santa Monica, CA: Human Factors and Ergonomics Society.

27.

Morrow

D. G.

Leirer

V. O.

Andrassy

J. M.

Hier

C. M.

Menard

W. E.

(1998). The influence of list format and category headers on age differences in understanding medication instructions. Experimental Aging Research, 24, 231–256.

28.

Nemire

(Ed.). (2011). Special issue: Forensic human factors. Ergonomics in Design, 19(1).

29.

Nemire

(2014). The hidden hazard of an invisible flame. In Proceedings of the Human Factors and Ergonomics Society 58th Annual Meeting (pp. 1869–1873). Santa Monica, CA: Human Factors and Ergonomics Society.

30.

Nemire

(2016). Ratings of the effectiveness of warnings on consumer paint stripper products: Evaluation by university students and MTurk workers of FHSA- and ANSI Z535.4-formatted warning labels. In Proceedings of the Human Factors and Ergonomics Society 60th Annual Meeting (pp. 480–484). Santa Monica, CA: Human Factors and Ergonomics Society.

31.

Nemire

Cohen

H. H.

(Eds.). (2014). Guide to forensic human factors. Santa Monica, CA: Human Factors and Ergonomics Society.

32.

Nemire

Vigilante

W. J.

Jr. (2015). A call for updating the FHSA regulations with warnings research and to match ANSI Z535.4. In Proceedings of the Human Factors and Ergonomics Society 59th Annual Meeting (pp. 1471–1475). Santa Monica, CA: Human Factors and Ergonomics Society.

33.

Otsubo

S. M.

(1988). A behavioral study of warning labels for consumer products: Perceived danger and use of pictographs. In Proceedings of the Human Factors and Ergonomics Society 32nd Annual Meeting (pp. 536–540). Santa Monica, CA: Human Factors and Ergonomics Society.

34.

Robinson

P. A.

(2009). Writing and designing manuals and warnings (4th ed., pp. 229–309). Boca Raton, FL: CRC Press.

35.

Rogers

W. A.

Lamson

Rousseau

G. K.

(2000). Warning research: An integrative perspective. Human Factors, 42, 102–139.

36.

Sanders

M. S.

McCormick

E. J.

(1993). Human factors in engineering and design (7th ed.). New York, NY: McGraw-Hill.

37.

Stevick

Rondinone

Sagle

(2014, April). Flame arrestors for gelled alcohol containers. Presentation to Society of Forensic Engineers and Scientists, Yosemite, CA. Retrieved from https://www.researchgate.net/publication/266910002_Flame_Arrestors_for_Gelled_Alcohol_Containers

38.

Williams

K. J.

Kalsher

M. J.

Laughery

K. R.

(2006). Allocation of responsibility for injuries. In Wogalter

M. S.

(Ed.), Handbook of warnings (pp. 617–628). Mahwah, NJ: Lawrence Erlbaum.

39.

Wogalter

M. S.

Begley

P. B.

Scancorelli

L. F.

Brelsford

J. W.

(1997). Effectiveness of elevator service signs: Measurement of perceived understandability, willingness to comply, and behaviour. Applied Ergonomics, 28, 181–187.

40.

Wogalter

M. S.

Conzola

V. C.

Smith-Jackson

T. L.

(2002). Research-based guidelines for warning design and evaluation. Applied Ergonomics, 33, 219–230.

41.

Wogalter

M. S.

Conzola

V. C.

Vigilante

W. J.

Jr. (2006). Applying usability engineering principles to the design and testing of warning text. In Wogalter

M. S.

(Ed.), Handbook of warnings (pp. 487–498). Mahwah, NJ: Lawrence Erlbaum.

42.

Wogalter

M. S.

Godfrey

S. S.

Fontenelle

G. A.

Desaulniers

D. R.

Rothstein

P. R.

Laughery

K. R.

Sr. (1987). Effectiveness of warnings. Human Factors, 29, 599–612.

43.

Wogalter

M. S.

Kalsher

M. J.

Racicot

B. M.

(1992). The influence of location and pictorials on behavior compliance to warnings. In Proceedings of the Human Factors and Ergonomics Society 36th Annual Meeting (pp. 1029–1033). Santa Monica, CA: Human Factors and Ergonomics Society.

44.

Wogalter

M. S.

Rashid

(1998). A border surrounding warning sign text affects looking behavior: A field observational study. In Proceedings of the Human Factors and Ergonomics Society 42nd Annual Meeting (p. 1628). Santa Monica, CA: Human Factors and Ergonomics Society.

45.

Young

(1991). Increasing the noticeability of warnings: Effects of pictorial, color, signal icon and border. In Proceedings of the Human Factors and Ergonomics Society 35th Annual Meeting (pp. 580–584). Santa Monica, CA: Human Factors and Ergonomics Society.