12
Fire Technology, 46, 629–649, 2010 2009 Springer Science+Business Media, LLC. Manufactured in The United States DOI: 10.1007/s10694-009-0113-9
Clothing-Related Burn Casualties: An Overlooked Problem? James F. Hoebel*, U.S. Consumer Product Safety Commission, 13506 Star Flower Court, Chantilly, VA 20151, USA Gordon H. Damant, California Bureau of Home Furnishings and Thermal Insulation, 3550 Watt Avenue, Suite 5, Sacramento, CA 95821, USA Steven M. Spivak, Fire Protection Engineering, University of Maryland and Stantex Consulting, 6301 Beachway Drive, Falls Church, VA 22044-1510, USA Geoffrey N. Berlin, 5330 Chelsen Wood Drive, Johns Creek, GA 30097, USA Received: 23 January 2009/Accepted: 29 September 2009
Abstract. Between 1997 and 2006, more than 4,300 serious burn injuries per year in the United States were associated with clothing. Ages 5–14 had the highest average annual burn injury rate, and ages 25–64 had the lowest rate. There were 120 deaths per year in the United States associated with clothing burns between 1999 and 2004. The death rate for those over 65 was six times the national average. The General Wearing Apparel Standard has regulated the flammability of clothing worn in the United States since 1953. Nearly all of the clothing-related injuries and deaths have occurred in fires involving apparel that complied with this Standard. Despite the size of this problem, there is no organized national activity under way to begin to address these casualties. Experience with the Children’s Sleepwear Flammability Standards, issued in the 1970s, suggests that safer garments can be manufactured that would prevent many clothing burn injuries and deaths. A more stringent up-to-date flammability standard, production of safer garments, use of warning labels, and educated consumers are needed. Keywords: Burns, Injuries, Deaths, Clothing, Standards, Flammability
1. Introduction Every year around the world, thousands of deaths and injuries occur from the ignition and burning of textile-related materials and products. It is estimated that about 50% of fire deaths in the United States result from products covered by the US Flammable Fabrics Act being the first item to ignite [1, 2]. These items include upholstered furniture, mattresses and bedding, draperies, curtains, floor coverings, wall coverings, and clothing. All are implicated in the loss of life and life-changing injuries that can result from these textile-related fires. * Correspondence should be addressed to: James F. Hoebel, E-mail:
[email protected] The authors all serve in a pro-bono capacity as science advisors to the National Association of State Fire Marshals in the United States.
630
Fire Technology 2010
These deaths and injuries occur as a result of ignition followed by flame propagation or smoldering of the textile materials used in the design of these products, as well as the consequences of human behavior. The greatest tragedy is that these fire-related deaths and injuries in many instances are either avoidable altogether or could be greatly mitigated. Materials of improved flammability performance are readily available from manufacturers and material suppliers. In addition, product manufacturers rarely warn their customers about the potential ignitability and concomitant fire dangers of everyday clothing and consumer products. Most textile products sold in the United States are controlled by the provisions of the Flammable Fabrics Act (FFA) [2]. The FFA was first enacted by the US Congress in 1953. Originally, the FFA itself included an existing commercial standard for the flammability of general wearing apparel. This regulation is codified in the Code of Federal Regulations as the Standard for the Flammability of Clothing Textiles, 16 CFR 1610 [3], and is still commonly known as the General Wearing Apparel Standard. The 16 CFR 1610 Standard was intended to remove from the market only highly flammable children’s clothing and sheer negligee-type fabrics used for ladies’ garments. By the late 1960s, however, Congress decided that provisions of the FFA should be expanded beyond the limited scope of the General Wearing Apparel Standard. In 1967, Congress extended the scope of the FFA to include other textile-related products such as home furnishings. This 1967 amendment to the Flammable Fabrics Act changed some administrative responsibilities but did not change the minimum requirements of the 1953 standard. The US Department of Commerce was given the responsibility of developing and issuing new flammability standards. In 1973, the newly created US Consumer Product Safety Commission (CPSC) assumed these responsibilities for developing and issuing new standards under the FFA and for enforcing existing standards. In 1968, the US Department of Commerce issued a ‘‘Notice of Finding That Flammability Standard or Other Regulation May Be Needed and Institution of Proceedings’’ [4], which eventually concluded that the existing General Wearing Apparel Flammability Standard was inadequate to address the specific problems associated with children’s sleepwear. Based on a more realistic test method, two additional standards were issued to regulate the flammability of children’s sleepwear sizes 0–6X and 7–14 [5]. While neither the General Wearing Apparel Standard nor the Children’s Sleepwear Standards require the testing of actual garments (only fabric specimens), the vertical strip test method used in the Children’s Sleepwear Standards provides a more realistic flame source and fabric orientation, thus making it much more representative of typical accident scenarios. After the standards were promulgated, clothing-related burns decreased, due in great measure to a dramatic drop in the frequency of incidents associated with the ignition of children’s sleepwear. Since the advent of the two Children’s Sleepwear Standards, most of the national activity in the United States to further reduce the fire losses associated with textile-related consumer products has targeted home furnishings and products that ignite those furnishings. The CPSC has been pursuing national standards and other activities involving mattresses and bedding, upholstered furniture,
Clothing-Related Burn Casualties
631
cigarette lighters, cigarettes, heating and cooking equipment, and candles. Significant progress has been made; however, the problems associated with the flammability of clothing (apparel) continue to be disregarded. And to the contrary, the only major clothing-related regulatory action in the past 20 years has been to reduce the scope of protection covered by the Children’s Sleepwear Standards by allowing firms to market relatively safe, snug-fitting garments that have not been tested to meet the requirements of the standards [6]. The purpose of this paper is to examine recent clothing-related thermal burn data to establish the scope of the problem, to ascertain trends in the frequency or rate of injuries and deaths, and to determine the need for renewed national interest and expedited action to address the issue and protect the public.
2. Clothing Burn Data Sources The most widely consulted fire data source used by analysts to estimate fire losses associated with individual consumer products combines two inputs. These are the National Fire Incident Reporting System (NFIRS), administered by the US Fire Administration, and the National Fire Protection Association’s (NFPA) annual survey of fire losses. This combination fire data source provides reasonably accurate national estimates of fire losses associated with such products as upholstered furniture, cigarettes, cigarette lighters, and mattresses. This source, however, is based on data collected from residential fires attended by the fire services. As such, it is not especially appropriate for developing national fire loss estimates for clothing-related burns. As evidenced by CPSC reports, fire-related burn injuries attributed to clothing usually do not involve the fire services [7, 8]. Rather, they involve direct ignition of the garment and subsequent burn injuries requiring medical treatment without producing a fire that requires a response by the fire services. For this study, we have focused on the national extent of thermal burn injuries and deaths. The most complete sources for these data are thermal burn injury data collected by CPSC through the National Electronic Injury Surveillance System (NEISS) and death data collected by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC), Department of Health and Human Services. According to CPSC, NEISS is a national ‘‘probability sample’’ [9] of approximately 100 hospital emergency departments in the United States and its territories. Data from the NEISS sample are weighted based on the sample design to produce national estimates of the number of consumer product-related injuries treated in hospital emergency rooms. The sample is adjusted periodically to permit the tracking of trends across time. These data were provided by the CPSC’s National Injury Information Clearinghouse [10]. Data are presented for the tenyear period 1997 through 2006. The most recent data available at the time of our analysis were from emergency room cases collected in 2006. The death data were downloaded from the CDC’s WONDER website [11]. The death data collected by NCHS are actual counts of deaths registered with the
632
Fire Technology 2010
system, not national estimates as in the NEISS information. NCHS states that, ‘‘It is believed that more than 99% of the births and deaths occurring in this country are registered’’ [12]. The number of deaths cited in this paper are the sum of those filed under the Underlying Cause of Death codes X05 (Exposure to Ignition or Melting of Nightwear) and X06 (Exposure to Ignition or Melting of Other Clothing and Apparel). We do not know how many deaths were not registered that should have been filed under X05 or X06. Death data before 1999 were not considered because the Underlying Cause of Death codes were changed after 1998, making comparisons before and after this change questionable. Conversion factors have been applied by some analysts to adjust for this change, but there appears to be a ‘‘disconnect’’ in these data despite the adjustment.1 We examined death data for the six-year period 1999 through 2004. Again, 2004 data were the most recent available. We used annual estimates of the US population by age groups, obtained from the web site of the US Census Bureau’s Population Division, to derive injury and death rates as casualties per million of population in the United States [13].
3. Results and Analysis: Burn Injuries Table 1 displays the national estimates of the number of thermal burn injuries treated in hospital emergency rooms each year over the ten-year period 1997– 2006. The numbers of injuries range from 3,728 to 5,008 per year over the ten year period. It is believed that the differences in values are the result of natural fluctuations in thermal burn injury data and are not the result of any external influences. Regression analysis did not yield statistically significant results. Therefore, a Mann–Kendall statistic value2 was calculated to determine the presence of any trend that was not necessarily apparent, though the magnitude of trending cannot be derived from this statistic. No substantive increase or decrease in the number of injuries was observed over the time period as confirmed by a Mann–Kendall statistic of zero. The NEISS estimates of clothing thermal burn injuries treated in hospital emergency rooms were aggregated by apparel categories, age groupings, and year. Five discrete apparel categories that are standard codes within the NEISS system were examined: Nightwear, Daywear, Outerwear, ‘‘Other,’’ and ‘‘Not specified.’’ The categories of Clothing Accessories, Costumes and Masks, and Footwear were judged to be outside the scope of our effort (this is consistent with the scope used by CPSC staff to analyze clothing thermal burn injuries). It is believed that the categories examined in our present study represent virtually all clothing generally worn in the US. Groupings by age were those that are frequently used by CPSC to analyze age effects: 0–4, 5–14, 15–24, 25–64, and 65+ years. 1 These death data show a substantial decline in deaths between 1998 and 1999. We believe that this decline, if it is real, would be reflected in a similar decline in the NEISS injury data between these years. However, no such decline was observed (see Table 1). 2 The Mann–Kendall equations are presented in Annex D.
Clothing-Related Burn Casualties
633
Table 1
National Estimates of Clothing Thermal Burn Injuries Treated in Emergency Roomsa, 1997–2006 Year
Injuries
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
4,051 4,315 4,941 4,133 4,343 4,157 4,023 5,008 4,508 3,728
Source: National Electronic Injury Surveillance System, National Injury Information Clearinghouse, U.S. Consumer Product Safety Commission a Product Codes include Nightwear, Daywear, Outerwear, Other, and Not specified. Product Codes for Clothing Accessories, Costumes or Masks, and Footwear are not included
The detailed thermal burn injury estimates organized by apparel category and age group for individual years are found in Annex A. These data illustrate considerable year-to-year variability that is considered to be due to natural fluctuations in the data. No statistically significant trends are again observed. However, a downward trend is suggested by negative Mann–Kendall statistic values for all of the ages except the 15–24 age group. For this age group, the Mann–Kendall statistic value of 16 (s = 0.36) indicates an upward trend. No reason is obvious for this difference in behavior. In the two most populated clothing types—Daywear and Outerwear—both types showed similar positive Mann–Kendall statistic values. Table 2 shows the average thermal burn injury estimates for each apparel category and age group over the ten-year period examined. Table 2
National Estimates of Clothing Thermal Burn Injuries Treated in Emergency Rooms by Age and Clothing Typea, Average Annual Estimate, 1997–2006 Age: Total Nightwear Daywear Outerwear Other Not specified
All
0–4
5–14
15–24
25–64
65+
4,321 357b 3,205 249 48 462
272 8 190 17 2 54
942 32 801 34 1 74
850 13 697 51 18 70
1,644 136 1,175 115 25 193
606 160 343 32 2 70
Source: Annex A a Product Codes include Nightwear, Daywear, Outerwear, Other, and Not specified. Product Codes for Clothing Accessories, Costumes or Masks, and Footwear are not included b See Footnote b to Annex A
634
Fire Technology 2010
The frequency of total burn injuries averages more than 4,300 victims over this period. Daywear accounts for over 74% of all thermal burn injuries, with 53% of Daywear injuries associated with individuals under 25 years old. This is an unfavorable and disproportionate representation, since the age group constituted only 35% of the population. In contrast, the population 25–64 years old experienced a more favorable 37% of Daywear injuries while constituting 52% of the population. Thermal burn injury rates were derived by dividing the injury frequency data in Annex A by the US population of the age groupings in each year. Annex B is a table of the population data by age grouping for each of the ten years that was used to compute the injury rates. Table 3 lists total injury rate for clothing burn injuries for each year between 1997 and 2006. Annex C, the source of data in Table 3, presents the national estimates of clothing thermal burn emergency room injury rates per million of population by individual year, clothing type, and age grouping. Like the frequency data in Annex A, the injury rates for different clothing types and age groups show considerable year-to-year variation, but no statistically significant trends were identified. These rates range from 12.45 to 18.12 injuries per million of population. There appears to be no appreciable increase or decrease in the rates over the ten-year period. While all values are within three standard deviations of the average, which is confirmed by a Chi-Square statistic value of 1.508 (p value of 0.003 for 9 degrees of freedom), it is believed that the differences in values are the result of natural fluctuations in thermal burn injury data and are not the result of external influences. A downward trend is suggested by the Mann–Kendall statistic value of -14 (s = -0.31) and a regression analysis, but it is not statistically significant. Because of the year-to-year variation, it is difficult to examine effects on a yearly basis, but considering the ten-year averages is useful. Annex C was used to calculate the ten-year average injury rates, thereby minimizing the year-to-year variation. These results are presented in Table 4. The average injury rate over all ten years, all product categories, and all ages is 15.21 injuries per million. This table shows that ages 5–14 had the highest injury Table 3
National Estimates of Clothing Thermal Burn Emergency Room Injury Rates per Million Population, 1997–2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source: Annex C
15.13 15.97 18.12 14.64 15.23 14.43 13.83 17.06 15.20 12.45
Clothing-Related Burn Casualties
635
Table 4
National Estimates of Clothing Thermal Burn Emergency Room Injury Rate per Million Population, by Clothing Typea and Age, Average Annual Rate, 1997–2006 Age: Total Nightwear Daywear Outerwear Other Not specified
All
0–4
5–14
15–24
25–64
65+
15.21 1.26 11.27 0.87 0.17 1.64
13.97 0.41 9.78 0.88 0.10 2.80
23.43 0.80 19.93 0.84 0.02 1.85
21.11 0.34 17.28 1.26 0.45 1.78
11.05 0.92 7.88 0.67 0.18 1.31
17.03 4.49 9.63 0.87 0.05 1.99
Source: Annex C a Product Codes include Nightwear, Daywear, Outerwear, Other, and Not specified. Product Codes for Clothing Accessories, Costumes or Masks, and Footwear are not included
average rate for both Nightwear and Daywear. For Daywear, this age group also has nearly double the injury rate compared to the other age groups except when compared to the 15–24 age group, which was similar. And for Nightwear, this age group also has nearly double the injury rate when compared to the younger age groups but significantly less than the 65+ age group. An analysis of Variance confirmed that the variation among age groups (p value of 0.56) was significantly more important than the variance among clothing types (p value of 3.6E-05). Perhaps the most surprising observation was that the injury rates by age did not show the expected: substantially higher rates for the very young and the elderly. Like the average injury data in Table 2, the highest clothing burn injury category was Daywear. The injury rate for Daywear was generally an order of magnitude greater than the other product categories. Every Daywear cell in Table 4 was higher than any other single cell.
4. Results and Analysis: Burn Deaths Burn death data for the six-year period 1999–2004 are presented in Tables 5 through 8. Table 5 contains frequency of death data for each of the six years and five age groupings. The NCHS data were not aggregated by apparel categories like the NEISS injury data. Table 5 illustrates that consistently over 100 people died in the United States in each of the years examined. The number of deaths ranged between 107 and 130. The Mann–Kendall statistic value of -1 (s = -0.02) suggests a slight downward trend, despite the ever-increasing population. Table 6 summarizes the average number of deaths by age grouping over the six years. On average, 120 people died from clothing thermal burns per year between 1999 and 2004. Note that the distribution of deaths by age is dramatically different
636
Fire Technology 2010
Table 5
Deaths Associated with Clothing Thermal Burns, by Year and Age Group, 1999–2004 Age:
All
0–4
5–14
15–24
25–64
65+
1999 2000 2001 2002 2003 2004 6 yr total
118 125 130 116 107 126 722
0 0 1 0 1 0 2
2 1 0 2 1 0 6
0 4 0 0 0 0 4
29 24 25 28 30 32 168
87 96 104 86 75 94 542
Source: U.S. Centers for Disease Control and Prevention, National Center for Health Statistics, Compressed Mortality Files for Underlying Cause of Death X05 [Exposure to Ignition or Melting of Nightwear] and X06 [Exposure to Ignition or Melting of Other Clothing and Apparel], 1999–2004 [Underlying Cause of Death Codes were changed after 1998, making comparison of data before and after this change questionable], CDC WONDER On-Line Database)
Table 6
Deaths Associated with Clothing Thermal Burns by Age, Annual Mean Occurrences, 1999 Through 2004 Age:
All
0–4
5–14
15–24
25–64
65+
Total
120
0
1
1
28
90
Source: Table 5
Table 7
Death Rate per Million Associated with Clothing Thermal Burns by Year and Age Group, 1999–2004 Age:
All
0–4
5–14
15–24
25–64
65+
1999 2000 2001 2002 2003 2004
0.43 0.44 0.46 0.40 0.37 0.43
0 0 0.05 0 0.05 0
0.05 0.02 0 0.05 0.02 0
0 0.10 0 0 0 0
0.20 0.16 0.17 0.19 0.20 0.21
2.52 2.74 2.94 2.42 2.09 2.59
Source: Table 5 and Annex B
from the injury age distribution shown in Table 4. Here, fully 75% of the victims were 65 or older. Death rates per million were computed using population estimates in Annex B. These rates are presented in Table 7.
Clothing-Related Burn Casualties
637
Table 8
Death Rate per Million Associated with Clothing Thermal Burns by Age, Annual Mean, 1999–2004 Age:
All
0–4
5–14
15–24
25–64
65+
Annual Average
0.42
0.02
0.02
0
0.19
2.55
Source: Table 7
The death rates for all ages ranged from 0.37 to 0.46. The Mann–Kendall statistic has a value of -5 (s = -0.11), suggesting a downward trend slightly larger than the non-normalized data in Table 5 because of an ever-increasing population. The annual average death rate by age over the period is shown in Table 8. The overall death rate was 0.42 deaths per million of population. The death rate for those over 65 is even more accentuated. This number, 2.55 deaths per million, was six times the national average. This phenomenon was not seen in the case of injury rate. It is undoubtedly influenced by impaired mobility and the reduced immunological resistance of seniors to the effects of severe burns.
5. CPSC Report on Clothing Textiles At Tab B of the CPSC Briefing Package ‘‘Draft Final Rule Updating the Flammability Standard for Clothing Textiles 16 CFR 1610’’ is a CPSC report entitled ‘‘General Wearing Apparel Fires—Fatalities and Emergency Department Treated Injuries,’’ December 27, 2007 [14]. Like this paper, the CPSC report examined death and injury data related to consumer clothing utilizing the same sources. While there were some differences in the data processing, similar results were observed. CPSC looked at injury data from 1996 to 2005. CPSC injury and injury rate estimates were generally about 10% lower than ours (between 4% and 16%) for the same years we examined using the same sources. Both studies agree that there were no notable trends. CPSC also noted the higher injury rate for Daywear, and the higher rates for ages 5–14 and 15–24. The CPSC report noted that about 25% of the emergency room cases involving clothing were subsequently admitted to the hospital or another medical facility. By contrast, CPSC reports only about 5% of all consumer-product related injury cases are hospitalized. This dramatic difference testifies to the high severity of clothing burn injuries and the associated high costs of hospital and surgical care. These facts have been known for many years. CPSC examined NCHS mortality data back to 1980 and reported a general decline in deaths since then. However, their graph appears to show a decline in the 1980s followed by a leveling-off in the 1990s until 1999, when a major decline occurred, just when the Underlying Cause of Death codes changed. No other reason for this major decline in one year has been offered. The magnitude of this decline was much greater than the adjustment used to correct the more recent NCHS data. This sharp decline was the ‘‘disconnect’’ noted in our analysis, and the reason for not analyzing death data for the years prior to 1999 (injury data
638
Fire Technology 2010
were not affected by this change). The CPSC death data analysis shows no change since 1999, the same lack of a trend that we observed. CPSC observations about the number of deaths, the death rates, and trends since 1999 are similar to ours.
6. Discussion The General Wearing Apparel Standard (16 CFR 1610) requires igniting a specimen, measuring flame spread, and rating the product. It is a minimum standard that is more than 55 years old, and it has been essentially unchanged over this period. It was mandated by the U.S. Congress in 1953 to remove from the marketplace only those few textiles and garments (such as long rayon pile cowboy chaps and rayon ‘‘torch’’ sweaters) that were the most highly flammable. CPSC enforcement experience demonstrates that the standard continues to do this job. Occasionally, CPSC finds such garments and takes action to remove them [15]. But these actions are seldom taken. The fact remains that virtually all clothing textiles fully meet the requirements of this standard. The vast majority of textile products subject to the test specified in this standard readily pass and are designated by the standard as ‘‘Class 1 normal flammability.’’ This result unfortunately implies that they are ‘‘safe’’ to use in clothing. The concept of ‘‘normal flammability’’ in the context of this test is misleading as a measure of safety. ‘‘Normal flammability’’ is merely a term defined by the General Wearing Apparel Standard. Ordinary newsprint paper passes this test [16]. Therefore, consumers should be aware that a ‘‘does not ignite’’ result under these test conditions does not mean that a fabric will not ignite under real-world conditions. Although some shortcomings of the General Wearing Apparel Standard were addressed by the development and promulgation of the Children’s Sleepwear Standards (initiated in the late 1960s), the flammability of most apparel continued to be regulated by the General Wearing Apparel Standard. In the mid-1970s, CPSC established a project to consider an improved standard, and funded the National Institute of Standards and Technology (NIST) (then the National Bureau of Standards) to develop an improved test method. A new test method was developed that measured both ignition propensity and heat transfer rate. It was dubbed the ‘‘mushroom’’ test because of the shape of the apparatus, which simulated part of the human torso. In addition, the flammability of the fabric was related to garment size, coverage, and design criteria as is most appropriate. These were submitted to CPSC for consideration [17]. However, the project was discontinued by CPSC at this point and never revived. The Standard for the Flammability of Clothing Textiles, 16 CFR 1610, is less stringent than the Flammability Standards for Children’s Sleepwear, 16 CFR 1615 and 1616. The test method used in 1610 to measure flammability is dramatically different from the test method used in 1615 and 1616, although both methods test fabric specimens, not actual clothing. The General Wearing Apparel Standard’s test method in 1610 applies a small needle-like flame for 1 s to the surface of a fabric strip supported at a 45 angle. In contrast, the Children’s Sleepwear Standards’ test method (1615 or 1616) applies a larger flame for 3 s to the bottom edge of a fabric strip suspended vertically. Therefore, the Children’s Sleepwear
Clothing-Related Burn Casualties
639
Standards’ test method presents a more aggressive ignition condition for each of these four test parameters: flame intensity, flame exposure time, flame impingement location, and fabric orientation. Furthermore, the General Wearing Apparel Standard (1610) measures only the rate of flame spread up the fabric test specimen, and permits a fairly rapid flame spread. This is not surprising since the purpose of the standard was to remove from the market only those garments that were highly flammable. The Children’s Sleepwear Standards, on the other hand, require the fabric specimen, once ignited, to extinguish before burning seven inches on average, a much tougher hurdle to clear. As a result, many fabrics that easily pass the General Wearing Apparel Standard fail the Children’s Sleepwear Standards. And, as far as we know, any fabric that passes the Children’s Sleepwear Standards will also pass the General Wearing Apparel Standard. The purpose of the Children’s Sleepwear Standards is to protect young children through the age of 12 years from the unreasonable risk of thermal burns caused by burning sleepwear. The Children’s Sleepwear Standards brought about certain changes to the materials and construction of children’s sleepwear in order to meet the requirements of the standards. These changes resulted in safer sleepwear and contributed to a reduction in burn injuries and deaths since the 1970s [18]. This experience suggests that similar changes to other apparel would prevent many more burn injuries and deaths. It is well known that different fibers and fabrics burn in distinct ways. Cotton, cotton-polyester or rayon-polyester blends, rayon, acetate, acrylic, and linen (or flax) can ignite easily and burn rapidly in many fabric and garment configurations. This is especially true in light- to medium-weight fabrics and loose-fitting garments. Nylon, polyester, olefins, wool, silk, and flame-resistant fibers will burn relatively slowly and usually self-extinguish upon removal of the flame [19]. Garments constructed of the first grouping will likely fail the requirements of the Children’s Sleepwear Standards, while garments constructed of the second grouping will usually pass. Our experience shows that almost all of the 4,300 persons who are seriously injured and the 120 who die each year do so in fires involving apparel that complies with the General Wearing Apparel Standard. Both the frequency and severity of such clothing-related burn casualties represent a major problem that is not going away. There has been no reduction in injuries or deaths over the ten year period examined. To our knowledge, there is no organized national activity underway to begin addressing the products that are associated with these casualties: no standards under development and no voluntary industry movement toward safer fabrics. Public education efforts primarily have concentrated on dealing with clothing ignitions after they occur rather than burn prevention. A notable example is the successful Stop, Drop and Roll campaign. The only CPSC clothing-related activities over the ten-year period we examined were directed toward (a) reducing the scope of coverage of the Children’s Sleepwear Standards [7], (b) making minor changes to the General Wearing Apparel Standard ‘‘to better reflect current consumer practices, modernized testing equipment and clarifying several aspects of the Standard’’ [14], and (c)
640
Fire Technology 2010
educating manufacturers and laboratories on the test procedures of the General Wearing Apparel Standard.3 None of these efforts was intended nor would be expected to produce a measurable reduction in clothing related burn injuries and deaths. These activities were designed to alleviate industry problems in complying with the standards involved. There is a lengthy legislative and scientific documentation raising questions about the continued reliance on 16 CFR 1610 and its small-flame, 45 test method as the prevailing federal regulation in the United States and Canada [16, 20–28]. Elaboration of these historical and legislative developments also is provided in several of the references [29–33]. There have been many calls for a more modernized approach to the testing and fire hazard assessment of general wearing apparel and clothing textiles.
7. Conclusions Every day in this country, people are subjected to burns and sometimes die in clothing-related fires. Many of these losses are preventable; yet, the Federal standard regulating the flammability of these products remains unchanged. Regulators could build on the proven approach of the Children’s Sleepwear Standards, and develop a modernized clothing flammability standard that adequately protects the public. The textile and apparel industries could voluntarily move toward the production of safer garments by redesigning their products and by offering safer alternatives—enhancing consumer choice. This could be supplemented by the adoption of and conformance with a voluntary consensus flammability standard covering all apparel. Greater use of effective, visible, demonstrable clothing fire warning labels would also be helpful. It appears that most consumers do not recognize or appreciate that much of the clothing they purchase is so vulnerable to easy ignition and burning. Partnerships could be created to develop and distribute consumer fire safety information to inform and educate consumers on safer garment constructions, including fiber choice, fit, and design. Consumers could incorporate such information into their purchase decisions. Of the Consumer Product Safety Commission’s two results-oriented Hazard Reduction Strategic Goals, one is to ‘‘reduce the rate of deaths from fire-related causes by 20 percent from 1998 to 2013’’ [34]. Priority is to be given to staff activities directed towards these strategic goals. We described above the three clothing-related activities undertaken by the CPSC over the ten-year period examined. It may be difficult to quarrel with these activities, which have reduced the industry burdens related to the General Apparel Standard while not increasing the risk of injury. However, it would be productive if CPSC could allocate some of its resources to address the existing clothing burn injury and death problem described herein and further contribute to the CPSC strategic goal of reducing fire deaths. 3
Several times a year, ASTM International offers technical and professional training courses on Regulatory Compliance for the Flammability of Wearing Apparel and Children’s Sleepwear, in which CPSC representatives participate.
Clothing-Related Burn Casualties
641
The CPSC uses several criteria to help select the products/hazards to address in the future [35]. Apparel ranks relatively high on the six appropriate criteria:4 a) frequency and severity of injuries: 4,300 injuries and 120 deaths annually is in a similar range with other subjects of important CPSC projects, such as upholstered furniture (870 injuries and 540 deaths) (see Reference [8], Table 2b, page 10, and Table 2c, page 11), candles (1,370 injuries, 160 deaths) (see Reference [8], Table 2b, page 10, and Table 2c, page 11), ranges (1,980 injuries, 160 deaths) (see Reference [8], Table 1b, page 6, and Table 1c, page 7), cigarette lighters (1,100 injuries, 160 deaths pre-standard) [36], and mattresses (open flame) (1680 injuries, 330 deaths pre-standard) [37]; b) amenability of injury reduction through CPSC action: the reduction in injuries and deaths resulting from the Children’s Sleepwear Standards strongly suggests that similar action would reduce the toll from apparel burns; c) costs and benefits from CPSC action: the children’s sleepwear experience provides a preliminary indication that similar action for apparel would be costbeneficial.5 Of course, CPSC is required by the Flammable Fabrics Act and the Consumer Product Safety Act to assess this effect before taking final action; d) unforeseen nature of the risk: while many consumers may be aware of the existence of the risk of thermal burn injury associated with apparel, we believe that most consumers do not appreciate the extent nor severity of this risk, believing that the General Wearing Apparel Standard provides adequate protection; e) vulnerability of the population at risk: the data herein, particularly the death data, show that the elderly are at a significant higher risk than the rest of the population; and f) probability of exposure to hazard: the exposure of the public to the apparel thermal burn hazard is substantially greater than the exposure to any of the other products that have been the subjects of CPSC actions. Clothing-related casualties will undoubtedly continue unless changes are made to the way that clothing is manufactured and labeled or unless consumer behavior can be modified. The effects of the Children’s Sleepwear Standards demonstrate that the frequency of deaths and injuries will decline if changes are made to the fabrics used in apparel. Faced with this challenge, the textile, apparel, and retailing industries will surely respond, as they did in the case of children’s sleepwear, to achieve ‘‘reasonable, technologically practicable, and appropriate [3]’’ solutions to meet new fire safety mandatory or voluntary regulations. We believe that progress in reducing deaths and burn injuries from clothing accidents can be achieved by the textile industry’s developing safer fabrics, by apparel manufacturers’ voluntarily changing the fabrics they procure, by regulators’ applying their existing authority, and by the public’s becoming more informed, leading to better purchasing decisions. 4
The other two criteria are ‘‘chronic illness and future injuries’’ and ‘‘additional criteria.’’ Other factors (such as changes in raw material prices unrelated to standardization) can exert effects that exceed the effects of regulation. Standardization itself does not necessarily mean an increase in cost. 5
642
Fire Technology 2010
Acknowledgement The authors wish to acknowledge the efforts of Karen Suhr of the National Association of State Fire Marshals (www.firemarshals.org), who assisted in the editing and formatting of this article for submittal.
Appendices Annex A
National Estimates of Clothing Thermal Burn Injuries Treated in Emergency Rooms, Breakdown by Year, Clothing Typea and Age, 1997–2006 Age: 1997 Total Nightwear Daywear Outerwear Other Not specified 1998 Total Nightwear Daywear Outerwear Other Not specified 1999 Total Nightwear Daywear Outerwear Other Not specified 2000 Total Nightwear Daywear Outerwear Other Not specified 2001 Total Nightwear Daywear Outerwear Other Not specified
All
0–4
5–14
15–24
25–64
65+
4,051 422b 2,694 278 75 582
249 – 183 24 – 42
898 – 757 75 – 66
384 – 253 23 – 108
1,716 85 1,034 157 75 365
725 262 463 – – –
4,315 144 3,300 56 19 796
445 – 370 – 19 56
1,242 – 1,066 19 – 157
384 – 347 18 – 19
1,582 105 1,139 18 – 320
663 39 380 – – 244
4,941 547 3,693 188 101 412
259 5 196 18 – 40
1,427 101 1,245 18 – 63
1,157 67 920 67 – 103
1,553 105 1,126 86 83 153
548 269 207 – 18 54
4,133 369 2,776 375 129 484
341 6 161 42 – 132
905 17 655 146 – 87
679 – 519 6 35 119
1,959 259 1,355 164 94 87
253 87 89 17 – 60
4,343 375 3,090 436 – 442
272 68 158 – – 46
878 114 572 67 – 125
1,008 7 874 76 – 51
1,742 84 1,316 208 – 134
442 102 170 85 – 85
Clothing-Related Burn Casualties
643
Annex A
continued Age: 2002 Total Nightwear Daywear Outerwear Other Not specified 2003 Total Nightwear Daywear Outerwear Other Not specified 2004 Total Nightwear Daywear Outerwear Other Not specified 2005 Total Nightwear Daywear Outerwear Other Not specified 2006 Total Nightwear Daywear Outerwear Other Not specified
All
0–4
5–14
15–24
25–64
65+
4,157 434 2,947 158 75 543
319 – 171 6 – 142
690 15 642 – 6 27
673 – 466 69 69 69
1,752 298 1,096 83 – 275
727 121 575 – – 31
4,023 286 3,263 195 78 201
241 – 235 – – 6
812 6 728 – – 78
831 – 662 74 78 17
1,461 134 1,106 121 – 100
678 146 532 – – –
5,008 523 3,764 156 – 565
109 – 109 – – –
1,055 70 979 – – 6
1,590 59 1,344 – – 187
1,489 155 1,058 86 – 190
765 239 275 70 – 181
4,508 273 3,552 349 – 334
286 – 202 84 – –
761 – 671 15 – 75
891 – 778 98 – 15
1,836 93 1,463 68 – 212
734 180 440 83 – 31
3,728 197 2,973 298 – 260
198 – 119 – – 70
755 – 696 – – 59
899 – 803 80 – 16
1,354 46 1,058 158 – 92
520 151 294 60 – 15
Source: National Electronic Injury Surveillance System, National Injury Information Clearinghouse, U.S. Consumer Product Safety Commission a Product Codes include Nightwear, Daywear, Outerwear, Other, and Not specified. Product Codes for Clothing Accessories, Costumes or Masks, and Footwear are not included b The data printouts provided by CPSC presented the number of injuries for each clothing type by year and the percentage of this number that occurred to each age grouping. For 1997, only 82.3% of the national estimate for nightwear injuries (422) was allocated by age. Apparently 75 injuries were unallocated. This discrepancy does not affect any of the observations in this document. In all other printouts, 100% of the national estimates were allocated
644
Fire Technology 2010
Annex B
US Population, 1997–2006, by Age Group, in Millionsa Age: 1997* 1998* 1999* 2000** 2001** 2002** 2003** 2004** 2005** 2006**
All
0–4
5–14
15–24
25–64
65+
267.784 270.248 272.691 282.217 285.226 288.126 290.796 293.638 296.507 299.398
19.099 18.989 18.942 19.188 19.354 19.544 19.783 20.069 20.315 20.418
38.851 39.171 39.495 41.099 41.140 41.111 40.986 40.766 40.436 40.337
36.634 37.220 37.774 39.397 40.140 40.734 41.206 41.474 42.116 42.435
139.014 140.482 141.940 147.454 149.260 151.142 152.864 154.798 156.853 158.948
34.185 34.385 34.540 35.078 35.333 35.594 35.958 36.309 36.787 37.260
Source: Population Division, U.S. Census Bureau, *Resident Population Estimates of the U.S. by Age and Sex, April 1, 1990 to July 1, 1999 or **Annual Estimates of the Population by Sex and Five-Year Age Groups for the U.S., April 1, 2000 to July 1, 2006 [‘‘Vintage 2006’’] a The data represent the population as of July 1 of each year. Population estimates are updated annually by the Census Bureau. However, the effect of these updates on the calculated burn injury or death rates in the tables does not change the analyses and conclusions in this document
Annex C
National Estimate of Clothing Thermal Burn Emergency Room Injury Rate per Million Population, by Year, Clothing Typea and Age, 1997–2006 Age: 1997 Total Nightwear Daywear Outerwear Other Not specified 1998 Total Nightwear Daywear Outerwear Other Not specified 1999 Total Nightwear Daywear Outerwear Other Not specified
All
0–4
5–14
15–24
25–64
65+
15.13 1.58 10.06 1.04 0.28 2.17
13.04 – 9.58 1.26 – 2.20
23.11 – 19.48 1.93 – 1.70
10.48 – 6.91 0.63 – 2.95
12.34 0.61 7.44 1.13 0.54 2.63
21.21 7.66 13.54 – – –
15.97 0.53 12.21 0.21 0.07 2.95
23.43 – 19.48 – 1.00 2.95
31.71 – 27.21 0.49 – 4.01
10.32 – 9.32 0.48 – 0.51
11.26 0.75 8.11 0.13 – 2.28
19.28 1.13 11.05 – – 7.10
18.12 2.01 13.54 0.69 0.37 1.51
13.67 0.26 10.35 0.95 – 2.11
36.13 2.56 31.52 0.46 – 1.60
30.63 1.77 24.36 1.77 – 2.73
10.94 0.74 7.93 0.61 0.58 1.08
15.87 7.79 5.99 – 0.52 1.56
Clothing-Related Burn Casualties
645
Annex C
continued Age: 2000 Total Nightwear Daywear Outerwear Other Not specified 2001 Total Nightwear Daywear Outerwear Other Not specified 2002 Total Nightwear Daywear Outerwear Other Not specified 2003 Total Nightwear Daywear Outerwear Other Not specified 2004 Total Nightwear Daywear Outerwear Other Not specified 2005 Total Nightwear Daywear Outerwear Other Not specified
All
0–4
5–14
15–24
25–64
65+
14.64 1.31 9.84 1.33 0.46 1.71
17.77 0.31 8.39 2.19 – 6.88
22.02 0.41 15.94 3.55 – 2.12
17.23 – 13.17 0.15 0.89 3.02
13.29 1.76 9.19 1.11 0.64 0.59
7.21 2.48 2.54 0.48 – 1.71
15.23 1.31 10.83 1.53 – 1.55
14.05 3.51 8.16 – – 2.38
21.34 2.77 13.99 1.63 – 3.04
25.11 0.17 21.77 1.89 – 1.27
11.67 0.56 8.82 1.39 – 0.90
12.51 2.89 4.81 2.41 – 2.41
14.43 1.51 10.23 0.55 0.26 1.89
16.32 – 8.75 0.31 – 7.27
16.78 0.36 15.62 – 0.15 0.66
16.52 – 11.46 1.69 1.69 1.69
11.59 1.97 7.25 0.55 – 1.82
20.42 3.40 16.15 – – 0.87
13.83 0.98 11.22 0.67 0.27 0.69
12.18 – 11.88 – – 0.30
19.81 0.15 17.76 – – 1.90
20.17 – 16.06 1.80 1.89 0.41
9.56 0.88 7.24 0.79 – 0.65
18.86 4.06 14.80 – – –
17.06 1.78 12.82 0.53 – 1.92
5.43 – 5.43 – – –
25.88 1.72 24.02 – – 0.15
38.34 1.43 32.41 – – 4.51
9.62 1.00 6.83 0.56 – 1.23
21.07 6.58 7.57 1.93 – 4.98
15.20 0.92 11.98 1.18 – 1.13
14.08 – 9.94 4.13 – –
18.82 – 16.59 0.37 – 1.85
21.16 – 18.47 2.33 – 0.36
11.71 0.59 9.33 0.43 – 1.35
19.95 4.89 11.96 2.26 – 0.84
646
Fire Technology 2010
Annex C
continued Age:
All
0–4
5–14
15–24
25–64
65+
2006 Total Nightwear Daywear Outerwear Other Not specified
12.45 0.66 9.93 1.00 – 0.87
9.70 – 5.83 – – 3.87
18.72 – 17.25 – – 1.46
21.19 – 18.92 1.89 – 0.38
8.52 0.30 6.66 0.99 – 0.58
13.96 4.05 7.89 1.61 – 0.40
Source: Annex A and Annex B a Product Codes include Nightwear, Daywear, Outerwear, Other, and Not specified. Product Codes for Clothing Accessories, Costumes or Masks, and Footwear are not included
Annex D: The Mann–Kendall Statistic Let: x1,…,xn be a sequence of measurements over time. To test the null hypothesis, H0: x1,…,xn come from a population where the random variables are independent and identically distributed, H1: x1,…,xn follow a monotonic (e.g., increasing or decreasing) trend over time. The Mann–Kendall test statistic is calculated as S ¼ where 8 <1 sgnðxj xk Þ ¼ 0 : 1
if if if
Pn1 Pn k¼1
j¼kþ1
sgn xj xk
xj xk > 0 xj xk ¼ 0 xj xk < 0
S is asymptotically normally distributed. The mean and variance of S are given by EðSÞ ¼ 0
VarðSÞ ¼
o 8n P < nðn1Þð2nþ5Þ pj¼1 tj ðtj 1Þð2tj þ5Þ : fnðn1Þð2nþ5Þg 18
18
if ties no ties
where p is the number of tied groups in the data set and is tj the number of data points in the jth tied group.
Clothing-Related Burn Casualties
647
References 1. Damant GH (2008) Burning issues—the role of textile related products in United States fires. In: Flame Retardants 2008. Queen Elizabeth Conference Center, London, England, February 12–13, 2008 2. Flammable Fabrics Act, 15 U.S.C. §§ 1191–1204 3. Standard for the Flammability of Clothing Textiles (also known as General Wearing Apparel Standard), 16 CFR Part 1610 4. Wearing apparel: notice of finding that flammability standard or other regulation may be needed and institution of proceedings. In: Federal Register, vol 33, no 207. Department of Commerce, Office of the Secretary, October 23, 1968, p 15662 5. Standard for the flammability of children’s sleepwear: sizes 0 through 6X, 16 CFR Part 1615 and standard for the flammability of children’s sleepwear: sizes 7 through 14, 16 CFR Part 1616 6. Final rule: standard for the flammability of children’s sleepwear: sizes 0 through 6X (16 CFR Part 1615) to exclude garments sized for infants nine months of age or younger and tight-fitting sleepwear garments; and standard for the flammability of children’s sleepwear: sizes 7 through 14 (16 CFR Part 1616), to exclude tight-fitting sleepwear garments. In: Federal Register, vol 61, no 175. U.S. Consumer Product Safety Commission, September 9, 1996, pp 47633–47649 7. Miller D (2005) Estimates of fire injuries treated in hospital emergency departments, July 2002–June 2003. Table 4. Estimated number of fire injuries treated in emergency departments by selected products first ignited and fire department attendance. U.S. Consumer Product Safety Commission, January 2005, p 11 8. Chowdhury R, Greene M, Miller D (2007) 2002–2004 residential fire loss estimates. Table 2c. Estimated residential structure fire injuries, selected products, 2002–2004. U.S. Consumer Product Safety Commission, July 2007, p 11 9. National Electronic Injury Surveillance System (NEISS), On-line, Explanation of NEISS estimates obtained through the CPSC Web-site. U.S. Consumer Product Safety Commission. http://www.cpsc.gov/Neiss/webestimates.html, September 1999 10. Product Summary Tables of NEISS Data of Thermal Burn Injuries for Product Codes 1644, 1645, 1646, 1658, and 1677. The data columns present national estimates, sample counts, age percents (0–4, 5–14, 15–24, 25–64, and 65+) and disposition percents, for calendar years 1994–2006. Office of Information Services, National Injury Information Clearinghouse, U.S. Consumer Product Safety Commission, Washington, DC 20207, provided by response to several letter requests 11. U.S. Centers for Disease Control and Prevention’s Wide-ranging ONline Data for Epidemiologic Research (WONDER) system provides access to a wide array of public health information. http://wonder.cdc.gov/ 12. National Center for Health Statistics. http://www.cdc.gov/nchs/datawh/nchsdefs/regarea.htm 13. National Population Estimates Program, U.S. Census Bureau, Population Division. http://www.census.gov/popest/national/asrh/ 14. Draft final rule updating the Flammability Standard for Clothing Textiles 16 CFR 1610, briefing package, U.S. Consumer Product Safety Commission, January 11, 2008 15. For example, ‘‘BCBG Max Azria Fined $75,000 for Selling Dangerously Flammable Sweaters,’’ News from CPSC, U.S. Consumer Product Safety Commission, October 11, 2000; ‘‘CPSC, Coville Inc. Announce Recall of Garments Made From a Flammable Fleece Fabric,’’ News from CPSC, U.S. Consumer Product Safety Commission, February 28, 1995; and ‘‘CPSC, Guess? Inc. Announce Recall of Reverse Fleece Women’s Jackets,’’ News from CPSC, U.S. Consumer Product Safety Commission, December 8, 1997
648
Fire Technology 2010
16. Bliss D (2002) Comments on Standard for Flammability of Clothing Textiles; Advance Notice of Proposed Rulemaking, Federal Register, September 12, 2002. Submission to the U.S. Consumer Product Safety Commission on ANPR to Amend the Standard for the Flammability of Clothing Textiles, 16 CFR Part 1610, November 18, 2002. http://www.cpsc.gov/library/foia/foia03/pubcom/clothing.PDF 17. Braun E, Cobble VB, Peacock RD, Krasny JF, Stratton AK, Helzer S (1976) Back-up report for the proposed standard for the flammability of general wearing apparel: final report. National Bureau of Standards, Gaithersburg, MD; U.S. Consumer Product Safety Commission, Washington, DC, NBSIR 76-1072, June 1976 18. Tinsworth DK (1985) Hazard data related to the flammability of wearing apparel. U.S. Consumer Product Safety Commission, April 1985 19. ‘‘What Smart Shoppers Know About Nightwear Safety,’’ Brochure, American Association of Retired Persons with support from U.S. Consumer Product Safety Commission, National Bureau of Standards, and National Fire Protection Association and input from representatives of Federal agencies, the fiber, textile, and apparel industries, retailers, and consumers. April 1988 20. ‘‘Investigation of Fabrics Involved in Wearing Apparel Fires,’’ American Academy of Pediatrics, Committee On Accident Prevention, Samuel E. Southard, Chair, In: Pediatrics, vol 34, November 1964, pp. 728–733 21. ‘‘Flammable Fabrics,’’ Third and Fourth Annual Reports by the U.S. Department of Health, Education, and Welfare, FY 1971, published in 1972, and FY 1972, published in 1973, respectively 22. Hearings Before the Consumer Subcommittee of the Committee on Commerce, U.S. Senate, First Session on S.1003 to Amend the Flammable Fabrics Act: To Increase the Protection Afforded Consumers Against Injurious Flammable Fabrics. U.S. Government Printing Office, 1967, pp 83–93, 158–161, etc 23. LBJ Signs Flammability Act: surgeon assails it as weak. In: Woman’s Wear Daily, December 15, 1967, p 32 24. Spivak SM (2002) Clothing ANPR—Comments on 16 CFR Part 1610, U.S. Consumer Product Safety Commission. Submission to the U.S. Consumer Product Safety Commission on ANPR to Amend the Standard for the Flammability of Clothing Textiles, 16 CFR Part 1610, November 12, 2002. http://www.cpsc.gov/library/foia/foia03/pub com/clothing.PDF 25. Hoebel JF (2002) Clothing ANPR. Submission to the U.S. Consumer Product Safety Commission on ANPR to Amend the Standard for the Flammability of Clothing Textiles, 16 CFR Part 1610, November 12, 2002. http://www.cpsc.gov/library/foia/foia03/ pubcom/clothing.pdf 26. Krasny JF (1982) Flammability evaluation methods for textiles. In: Lewin M et al (eds) Flame-retardant polymeric materials, vol 3. Plenum Publishing Corporation, pp 155– 188 27. Needles HL (1980) The use of textile flammability standards by expert witnesses in product liability. J Consumer Product Flammability 7:213–217 28. Needles HL (1991) Current issues in clothing flammability: testing and regulation. In: Proceedings of the international conference on fire safety, vol 16 29. Mullins HC, Panagakos A (1971) Flammable Fabrics Act Protection: Fire Resistants v. Industry Resistance. In: The George Washington Law Review, vol 39, no 3, March 1971, pp 608–632 30. Miller JJ (1973) Dressed to kill: the Flammable Fabrics Act of 1953—twenty years in retrospect. In: Cumberland-Samford Law Review, pp 358–377
Clothing-Related Burn Casualties
649
31. Armstrong DD (1971) The Human Torch—an exegesis about the clothes we wear. In: South Carolina Law Review, vol 23, pp 787–803 32. Campbell DC, Vargo JF (1976) The Flammable Fabrics Act and Strict Liability in Tort. In: Indiana Law Review, vol 9, no 2, January 1976, pp 395–417 33. Weinstein SS (1979) Flammable fabrics. In: Case & Comment, November–December 1979, pp 3–12 34. Strategic Plan, U.S. Consumer Product Safety Commission, September 2003 35. Policy on Establishing Priorities for Commission Action, 16 CFR Part 1009, Section 1009.8 36. Safety Standard for Cigarette Lighters, 16 CFR Part 1210, Section 1210.5 Findings 37. Standard for the flammability (open flame) of mattress sets, 16 CFR Part 1633, supplemental information, Section D, risk of injury. In: Federal Register, vol 17, no 50, March 15, 2006, p 13471