- Email: [email protected]
Clothing-related burns in New South Wales, Australia: Impact of legislation on a continuing problem
burns 41 (2015) 58–64
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Clothing-related burns in New South Wales, Australia: Impact of legislation on a continuing problem Lara A. Harvey a,*, Siobhan Connolley b, John G. Harvey c a Falls and Injury Prevention Group, Neuroscience Research Australia, University of New South Wales, Sydney, NSW 2052, Australia b Statewide Burn Injury Service, NSW Agency for Clinical Innovation, NSW, Australia c Burn Unit, Children’s Hospital at Westmead, Sydney, NSW, Australia
article info
abstract
Article history:
To combat the risk of nightwear burns a mandatory standard regulating the design,
Received 10 June 2014
flammability and labelling requirements of children’s nightwear was introduced in
Received in revised form
Australia in 1987. This population-based study examined the trends, characteristics and
16 September 2014
causes of clothing-related burns to inform a review of the current standard, and to facilitate
Accepted 9 October 2014
the development of targeted prevention strategies. Clothing-related burns for 1998–2013 were identified from hospitalisation data for all hospitals in NSW and detailed information regarding circumstance of injury from a burn
Keywords:
data registry. To investigate percentage annual change (PAC) in trends negative binomial
Prevention
regression analysis was performed.
Legislation
There were 541 hospitalisations for clothing-related burns, 18% were nightwear-related
Regulations
and 82% were for other clothing. All clothing burns decreased by an estimated 4% per year
Clothing-burns
(95% CI
Injuries
(PAC
6.2 to
2.1). Nightwear-related burns decreased by a significantly higher rate
7.4%; 95% CI
12.5 to
2.1) than other clothing (PAC
2.5%; 95%CI
4.7 to
0.1).
Exposure to open heat source (campfire/bonfire) was the most common cause, followed by cooking. Of factors known to be associated with clothing burns, accelerant use was reported in 27% of cases, cigarettes 17%, loose skirt or dress 8%, and angle grinders in 6% of cases. Hospitalisations for clothing burns are relatively uncommon in NSW and rates, particularly of nightwear burns, have decreased over the last 15 years. Strategies for continued reduction of these injuries include increasing the scope of the current clothing standard or developing new standards to include all children’s clothing and adult nightwear, and increasing community awareness of the risk associated with open heat sources, accelerant use and loose clothing. # 2014 Elsevier Ltd and ISBI. All rights reserved.
* Corresponding author. Tel.: +61 2 9385 1852; fax: +61 2 9385 6040. E-mail address: [email protected] (L.A. Harvey). http://dx.doi.org/10.1016/j.burns.2014.10.013 0305-4179/# 2014 Elsevier Ltd and ISBI. All rights reserved.
burns 41 (2015) 58–64
1.
Introduction
Despite major advances in the prevention of clothing-related burns seen in many countries over the last 60 years they continue to constitute a significant burn hazard worldwide, particularly in low to middle income countries where loose and/or flammable clothing catching fire remains a major cause of flame burns [1–3]. Although now comparatively uncommon in high income countries, clothing-related burns remain potentially devastating injuries. In the United States (US) in 2009, it was estimated that 4300 clothing-related burns are treated in hospital emergency departments each year and 120 people die [4]. The decrease in morbidity and mortality associated with clothing burns experienced in high income countries has been attributed in part to a change in fashion towards tight fitting clothing and the introduction of synthetic fabrics that do not burn readily, combined with the introduction of regulations restricting the flammability of clothing [5]. The first of these fabric regulations, the Flammable Fabric Act (FFA) was introduced in the US in 1953 in recognition of the high burn danger posed by brushed rayon sweaters and cowboy chaps [6]. Much of the early regulatory efforts thereafter focussed on children’s nightwear, as it was recognised at the time that the majority of severe injuries were attributable to loose-fitting easily ignited night attire [5]. Regulations introduced in the United Kingdom (UK) the following year addressed the flammability of fabrics specifically in children’s night dresses. In 1967, the UK regulations increased in scope to include adult nightdresses [7] and in the same year the US Flammable Fabric Act was amended to include all articles of wearing apparel and interior furnishings [6]. In 1973 additional standards were introduced in the US to regulate the flammability of young children’s sleepwear (sizes 0–6X) and increased to include larger sizes (sizes 7–14) in 1975 [8]. In Australia, the clothing standard AS 1249 ‘Safe design rules for children’s night clothes’, was first introduced in 1972, updated to ‘Children’s night clothes having reduced fire hazard’ in 1976 [9] and mandated under national legislation in 1978 [10]. The standard has undergone several subsequent amendments and currently includes some daywear or underwear items that may commonly be used as nightwear such as boxer shorts and infants all-in-ones [11]. Several studies have shown that the incidence of nightwear-related burns in children reduced significantly compared to other clothing burns following the introduction of nightwear standards/regulations in the US, UK and New Zealand [12–14]. In particular a study undertaken at the Shriner’s Burns Institute in Boston demonstrated that the percentage of nightwear related burns dropped from 32% to 4% of all burn admissions after the nightwear regulations were enforced [13]. Since these early reports, few studies have explored the long-term impact of the introduction of clothing regulations and standards. One study investigating impact of UK regulations found little to no sustained impact [15], and there has been little published on trends in other nonnightwear clothing burns. In Australia, the clothing standard AS/NZS 1249:2003 underwent public comment and review in 2013. Our aim
59
was to examine the trends, characteristics and causes of clothing burns at a population level in NSW, to inform the clothing standards review, and to facilitate the development of targeted prevention strategies.
2.
Methods
2.1.
Data sources
2.1.1.
Admitted patients data collection (APDC)
The APDC is a routinely collected census of all private and public hospitalisations across NSW. The APDC records ‘episodes of care’ in hospital which end with the discharge, transfer or death of the patient. All Hospitalisations due to burns and scalds were identified by ICD-10-AM external cause codes ‘X00-X19’ and of these, clothing-related burns were identified using ICD-10-AM external cause code ‘X05’ Exposure to ignition or melting of nightwear and ‘X06’ Exposure to ignition and melting of other clothing and apparel for the 15 years financial period 1998–1999 to 2012–2013. Records where the mode of admission was a transfer from another hospital or change in service category were excluded to minimise multiple counting of cases for the same injury. Cases admitted to a hospital in NSW but usually residing another state were also excluded as hospitalisation rates were calculated against the NSW resident population. All NSW residents admitted to interstate hospitals are included in the APDC and were included in the analysis. Numbers and rates for 2011–2012 and 2012–2013 include an imputed estimate of these interstate hospitalisations (n = 184).
2.1.2. Agency for clinical innovation NSW statewide burn injury service (SBIS) burns data registry The SBIS burns data registry contains admission records and case details for all patients admitted to the three designated Burns Units in NSW (two adult, one pediatric) and captures both hospitalisations and outpatient clinic visits. Detailed demographic, etiological and clinical data items are collected including a free text description of the circumstances resulting in the burn injury. Cases are classified according to mechanism of injury. Clothing-related burns were identified by mechanism of injury ‘Flame-clothing and bedding’ for the 6 years period for which full data coverage is available, 2007–2008 to 2012–2013. The free text descriptions of the circumstances of injury were reviewed and admissions for house fires and bedding-related burns were excluded.
2.2.
Statistical analysis
Analysis was performed using SAS Enterprise Guide 5.1 [16]. Direct age and sex standardised rates per 100,000 population were calculated along with the percentage annual change in the hospitalisation rate. Ninety-five percent confidence intervals were calculated assuming a Poisson distribution [17]. Agespecific population estimates for NSW at 31 December for each of the years studied were used to calculate rates. These estimates correspond to the mid-point of each financial year of hospitalisation data, and were interpolated from the Australian Bureau of Statistics population estimates at 30
60
burns 41 (2015) 58–64
June each year [18]. The Australian estimated resident population for 2001 was used as the standard population. To correct for over dispersion, negative binomial regression was used to examine the statistical significance of changes in trends over time [19]. To examine differences between nightwear and other clothing burns chi-squared tests of independence were used. Hospital costs were calculated using the Australian-Refined-Diagnostic-Related Group (AR-DRG) codes recorded in the NSW APDC for each episode of care. The AR-DRG is allocated after hospital discharge and is intended to provide an indication of the resources required by the hospital to treat the individual, including all inhospital medical and allied health treatment, nursing, diagnostics and pharmaceuticals. Average AR-DRG version 5.2 costs were applied to each episode of care in accordance with the NSW Cost of Care Standards 2009/10 [20].
3.2.
Demographic characteristics
3.
Results
Hospitalisations by age group, sex and clothing type are shown in Table 2. A third of all clothing related burns were adults and a further quarter were children aged 5–14 years. Nightwear burns were more common among females (59%) whilst burns related to other clothing and apparel were more common amongst males (68%) ( p < 0.0001). There were no nightwear clothing burns in infants during the study period. The highest rate of hospital admissions was for non-nightwear clothing in boys aged 5–14 years. The majority of people hospitalised for clothing-related burns were Australian born (87%), and this did not differ between nightwear and other clothing and apparel ( p = 0.3813). Admissions for clothing-related burns occurred in every month of the year, with the peak in the winter months of June–August. There was a significant difference in pattern with nightwear peaking in June and other clothing and apparel related burns peaking later in August ( p = 0.0022).
3.1.
Temporal trends
3.3.
There were 541 hospitalisations across all age groups for clothing-related burns for the 15 years period of the study. The annual number of hospitalisations ranged from 59 in 1998–1999 to 20 in 2011–2012 (Fig. 1). Of these 18% (99) were as a result of nightwear and 82% (442) were other clothing and apparel. Age and sex standardised hospitalisation rates by year for all burns and scalds, clothing-related burns, nightwear burns and other clothing and apparel burns are shown in Table 1. All clothing-related burns have decreased over the study period by an estimated 4.1% per year (95%CI 6.2 to 2.1, p < 0.0001) from 0.91 per 100,000 in 1998–1999 to 0.38 per 100,000 in 2012–2013. Nightwear-related burns decreased by an estimated 7.4% per year (95%CI 12.5 to 2.1, p < 0.007) whilst other clothing and apparel decreased by 2.5% (95%CI 4.7 to 0.1). Hospitalisations for all burns and scalds combined decreased by 1.0% per year (95%CI 1.5 to 0.5, p < 0.0001) over the same period.
Burn and hospitalisation characteristics
The most common location of burn was the trunk (36%) followed by hip and lower limb (29%). There was no difference in location between nightwear and other clothing and apparel burns ( p = 0.2122). Of the 411 hospitalisations where the depth of burn was recorded the majority were for full thickness burns (67%), partial thickness (31%) and erythema (3%) (Table 3). There was no difference in proportion of depth of burn between nightwear and other clothing and apparel ( p = 0.1407). The majority of hospitalisations were for short duration (less than 2 days) with 7% being for greater than 4 weeks. Length of stay for nightwear ranged from 1 to 80 days and clothing and other apparel for 119 days. There was no difference in LOS between nightwear and other clothing ( p = 0.0998). There were 24 in-hospital deaths (4%) for all clothingrelated burns, with a significantly higher proportion of deaths
60
X05 Exposure to ignion or melng of nightwear 50
Number of hospalisaons
X06 Exposure to ignion or melng of other clothing and apparel 40
30
20
10
0 1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
2008-09
2009-10
2010-11
2011-12
2012-13
Fig. 1 – Number of hospitalisations for clothing-related burns, by clothing type, NSW, Australia, 1998–1999 to 2012–2013.
61
burns 41 (2015) 58–64
Table 1 – Age and sex standardised hospitalisation rates per 100,000 for burns, by mechanism and year, NSW, Australia, 1998–1999 to 2012–2013. Year
1998–1999 1999–2000 2000–2001 2001–2002 2002–2003 2003–2004 2004–2005 2005–2006 2006–2007 2007–2008 2008–2009 2009–2010 2010–2011 2011–2012 2012–2013
All burns
All clothing burns
Nightwear burns
Other clothing and apparel burns
38.4 36.9 37.1 35.1 30.7 31.3 27.1 30.0 30.1 32.1 28.3 29.0 28.9 30.1 28.9
0.91 0.56 0.80 0.76 0.49 0.49 0.50 0.43 0.59 0.48 0.40 0.52 0.43 0.28 0.38
0.34 0.09 0.06 0.23 0.14 0.10 0.08 0.11 0.07 0.10 0.03 0.01 0.03 0.05 0.03
0.58 0.46 0.74 0.53 0.35 0.39 0.42 0.33 0.52 0.38 0.37 0.51 0.40 0.23 0.36
reported in 64 cases (27%), cigarette 41 (17%), alcohol use/ intoxication 12 (5%) and loose flowing clothing (skirt or dress) 19 (8%).
4.
This state-wide population based study examined the incidence, characteristics and circumstances leading to clothingrelated burns over a 15 years period in NSW, providing the most comprehensive profile available to date. It shows that hospitalisation for clothing-related burns is relatively rare, comprising only 2% of all burns hospital admissions for NSW over the study period. Although not common, clothing-related burns were severe injuries with over half being full thickness and a quarter requiring hospitalisation for a period of greater than a week. The cost of these injuries was just under half a million dollars annually for acute hospital care alone, and this is likely to be a substantial underestimate of the true cost of providing burn care in hospital, as the NSW peer group average cost estimates (AR-DRG groupings) have recently been shown to significantly underestimate actual hospitalisation costs for both adult and pediatric trauma patients [21,22]. Hospitalisation rates for all clothing-related burns have decreased over the 15 years study period by 4% per year in this study. In contrast, rates for all burns and scalds combined have decreased by just 1% over the same period making it unlikely that the observed decrease for clothing-related burns is solely a reflection of changes in hospital admission practice for burns. The relatively stable hospitalisation rate for all burns and scalds in NSW documented in our study is consistent with recently reported national trends [23]. In Australia, the current iteration of clothing standard AS/ NZS 1249:2003 [11] specifies the design, flammability performance and labelling requirements for children’s nightwear garments for sizes 00-14. The standard stipulates that all children’s nightwear must display an appropriate hazard classification, thus the approach is not one of banning the sale of dangerous nightwear, but to inform consumers of the potential hazard and permit them to make their own judgement as to whether they consider the fire risk is acceptable. Our study provides evidence of the sustained impact of these standards, with no hospital admissions for nightwear-related burns in infants and low hospitalisation rates in toddlers and children over the study period. Whilst it is
resulting from nightwear (9%) compared to other clothing and apparel (3%) ( p = 0.0128). The annual average acute hospitalisation cost for all clothing-related burns was AUS $446,529 of which nightwear contributed $74,139 and other clothing and apparel contributed $372,390.
3.4.
Discussion
Circumstances leading to injury
There were 237 patients requiring hospitalisation identified from the SBIS burns registry, which was 34% more than identified from the APDC for the same period. There was no difference in sex distribution between hospitalised patients identified in the APDC and the burns registry ( p = 0.3638), however registry data identified a higher proportion of adult clothing-related burns but less for children ( p < 0.0001). Nightwear-related burns identified in the free text constituted only 8% of all clothing burns in the registry data, which was lower than that identified in APDC (18%) ( p = 0.0004). The circumstances surrounding the burns are summarised in Table 4. Exposure to an open heat source such as a campfire or bonfire was the most common cause of clothing burn, followed by cooking, children playing with matches or cigarette lighters and use of angle grinders. Of factors known to be associated with clothing burns, accelerant use was
Table 2 – Age and sex distribution of hospitalisations for clothing-related burns, by clothing type, NSW 1998–1999 to 2012– 2013. Age group
Night wear Male n
Infant (<1 year) Toddler (1–4 years) Child (5–14 years) Youth (15–24 years) Adult (25–64 years) Elderly (65+ years)
0 8 19 1 10 3
Total
41
Female
Rate (95% CI)
n
0.29 0.28 0.01 0.04 0.05
0 13 12 3 16 14
(0.13–0.58) (0.17–0.43) (0.00–0.08) (0.02–0.07) (0.01–0.16)
Other clothing and apparel
58
Rate (95% CI) 0.50 0.18 0.04 0.06 0.18
(0.27–0.86) (0.09–0.32) (0.01–0.13) (0.03–0.09) (0.10–0.19)
Male n 2 21 96 47 101 32 299
Female
Rate (95% CI)
n
Rate (95% CI)
0.29 0.77 1.39 0.66 0.37 0.52
0 23 41 14 49 16
0.71 0.62 0.21 0.18 0.21
(0.03-1.04) (0.48–1.17) (1.13–1.70) (0.49–0.88) (0.30–0.45) (0.36–0.74)
143
(0.45–1.06) (0.45–0.85) (0.11–0.35) (0.13–0.24) (0.12–0.34)
62
burns 41 (2015) 58–64
Table 3 – Depth of burn and length of stay for hospitalisations for clothing-related burns, by clothing type, NSW 1998–1999 to 2012–2013. Characteristic
Depth of burna Erythema Partial thickness Full thickness LOSb <2 days 2–7 days 8–28 days >28 days
Nightwear n (%)
Other clothing n (%)
Total n (%)
2 (2.8) 22 (30.6) 48 (66.7)
11 (3.2) 145 (42.8) 183 (54.0)
13 (3.2) 167 (40.6) 231 (56.2)
70 17 24 11
261 (49.0) 121 (22.7) 117 (21.9) 34 (6.4)
331 (50.5) 138 (21.1) 141 (21.5) 45 (6.9)
(57.4) (13.9) (19.7) (9.0)
a
Includes 411 cases (130 other/unspecified excluded). Includes records for new admissions, transfers and type changes (n = 655).
b
possible that the low hospitalisation rates may relate to an increased awareness of the danger of clothing burns in the general population, the sustained decrease in nightwear burns (7.4% per year) was significantly higher than that seen for other clothing and apparel (2.5% per year). Further, only 18% of the clothing-related burn hospitalisations, (on average two cases a year over the last 5 years) were for nightwear burns, which is in contradiction to early epidemiological studies which reported up to 60% of pediatric clothing-related burns admissions were due to nightwear [13,24]. Traditionally two vulnerable groups have been identified as being disproportionately at risk of nightwear related burns – young children who do not realise these the dangers and indulge in fire play, and elders in whom the reaction time is slowed [7,15,24]. Our study highlighted that whilst there were
Table 4 – Circumstances surrounding clothing-related burns, NSW 2007–2008 to 2012–2013. Circumstance Campfire/bonfire/open fire Cooking Child playing with matches/ cigarette lighter Candle Angle grinding Welding Not adequately described/other Smoking-related- when awake Smoking in bed/sleep with cigarette BBQ Heater Self-inflicted Fixing car Game/prank Other work-related Assault Sparkler/fireworks Kerosene lamp/gas bottle explosion Total Source: SBIS Burns Data Registry.
Male
Female
Persons
51 (32.7) 6 (3.9) 16 (10.3)
14 (17.5) 19 (23.8) 6 (7.5)
65 (27.5) 25 (10.6) 22 (9.3)
5 15 14 9 6 3
(3.2) (9.6) (9.0) (5.8) (3.9) (1.9)
12 0 0 2 5 6
(15.0) (0) (0) (2.5) (6.3) (7.5)
17 15 14 11 11 9
(7.2) (6.4) (5.9) (4.7) (4.7) (3.8)
3 1 4 6 6 6 2 1 2
(1.9) (0.6) (2.6) (3.9) (3.9) (3.9) (1.3) (0.6) (1.3)
5 6 3 0 0 0 1 1 0
(6.3) (6.3) (7.5) (3.8) (0) (0) (1.3) (1.3) (0)
8 7 7 6 6 6 3 2 2
(3.4) (3.0) (3.0) (2.5) (2.5) (2.5) (1.3) (0.8) (0.8)
156
80
236 (100)
no hospitalisations for infants, nightwear burns in toddlers and children under 15 years still occur. The equal numbers of males and females in this age group may reflect a trend towards wearing large or adult sized T-shirts which are not considered night attire, and not currently covered by the Australian standards. A similar phenomenon has been reported in the UK [25]. In older people, nightwear-related burns were more common in females perhaps reflecting that the general trend away from loose nightdresses has not been taken up by the elderly. Again this pattern is consistent with studies from the UK [7,15] and highlights the potential benefit of either increasing the scope of the Australian nightwear standard to include all age groups or to develop an additional standard to address adult nightwear. In our data nightwear burns were severe injuries and although the proportion with full thickness burns or length of time in hospital did not differ significantly from other clothing-related burns they resulted in a higher proportion of in-hospital deaths (9%), signalling a high priority for prevention. Currently in Australia there are no safety requirements for children’s daywear clothing, or any adult clothing. In our study other clothing-related burns were more common in males, with highest hospitalisation rate in boys aged 5–14 years, perhaps associated with fire play and risk taking behaviour. It is well recognised that clothing burns involve three key factors: the presence of a heat source, the behaviour of the wearer and the flammability of the fabric [6]. The presence of a heat source remains a large and potentially modifiable issue in our data. The peak of admissions occurred in winter months and evidence from the burn registry highlight that open fires and cooking were the most common causes of burn. Behaviour was also a major factor, with just under a third of all admissions reporting use of an accelerant such as petrol. Even garments complying with the requirements of the standard will give little or no protection against burns where flammable liquids are involved. Regular education on the dangers of accelerants and heat sources may be beneficial. Smokers’ materials such as cigarettes, lighters and matches have been previously implicated as a major risk factor for clothing fire [15,25]. In our study 17% of the hospitalisations involved smokers materials which is less than previous reports of 29% from the UK [15] perhaps reflecting that smoking rates in NSW have decreased from 23.5% to 16.2% over the last 10 years period [26]. If this decreased trend in smoking continues, coupled with the introduction of a mandatory safety standard for child resistant cigarette lighters in 1997 [27] and reduced fire cigarettes in 2010 [28] the number of clothing-related burns caused by smokers material may continue to decrease. Candles have been shown to pose a fire risk, contributing up to 3% of clothing burns in the UK [15] and 1.2% in our study. The design and fit of garments is a critical element in the hazard. Loose fitting garments with long flowing design and billowing sleeves are hazardous anywhere near open flames. Case reports have highlighted the risk of saris, long loose scarfs and grass skirts [29] and gypsy-style skirts [30]. Eight percent of clothing burns in our study involved the ignition of dress or skirt. Targeted education into the risk posed by loose flowing clothing is clearly needed.
burns 41 (2015) 58–64
Finally our study identified a high proportion of clothing burns resulting from the use of angle grinders. Angle grinders are known to be one of the most common power tools to cause injuries in the workplace in Australia [31], however in our data 40% of the angle grinder burns were sustained during do-ityourself (DIY) activities at home. Injuries related to DIY activities pose a particular problem for prevention as is extremely difficult to control safety education, and impossible to regulate or enforce the use of protective equipment. Home safety surveys report that around 20% of Australian households have an angle grinder [32] and only 5% of DIYers who sustained an angle grinder injury were wearing personal protective equipment at the time of injury [33]. Educational material highlighting the potential risks and appropriate safety measures should be developed and provided at the point of sale or hire of DIY equipment. The population coverage and relatively long time frame are strengths of this study, however there are several limitations that must be considered. To identify clothing-related burns in the routinely collected administrative hospitalisation database we were reliant on the accuracy of external cause coding. Whilst much effort has been expended in Australia to ensure the accuracy of diagnosis codes a recent clinical audit which evaluated the accuracy of coding for external causes in Australian hospitals found 29% disagreement for the coding block X00-X10 ‘smoke, fire and flames’ [34]. In our study the incidence of clothing-related burns identified from the APDC would appear to be under numerated by around 30% when compared to the incidence recorded in the SBIS Burn Registry, which in itself is provides an underestimation as it only includes hospitalisations for the three designated burns units in NSW. Further research linking APDC hospitalisation data to the SBIS Burns registry data is warranted to determine the impact of coding inconsistencies on the identification of burns. In addition, less severe burns treated in emergency departments or outpatient burns clinics are not included in the APDC leading to further underestimation of the total burden of clothing-related burn injury in NSW.
5.
Conclusions
Hospitalisations for clothing burns are relatively uncommon in NSW and rates, particularly of nightwear burns, have decreased over the last 15 years. Important strategies for continued reduction of these injuries include increasing the scope of the clothing standards or developing new standards to include all children’s clothing and adult nightwear, and increasing community awareness of the risk associated with open heat sources, accelerant use and loose clothing. Improved coding of clothing burns in hospital data, including the circumstances of injury could provide a more comprehensive information base from which to plan and monitor prevention strategies.
Conflict of interest There were no conflicts of interest.
63
Acknowledgements Lara Harvey was supported by the NSW Ministry for Health. The authors thank the Centre for Population Health at the NSW Ministry of Health for providing access to the hospitalisation data via Secure Analytics for Population Health Research and Intelligence (SAPHaRI) and the Statewide Burn Injury Service at the NSW Agency for Clinical Innovation for providing access to the registry data analysed in this study.
references
[1] Golshan A, Patel C, Hyder AA. A systematic review of the epidemiology of unintentional burn injuries in South Asia. J Public Health 2013;35:384–96. [2] Ansari-Moghaddam A, Baghbanian A, Dogoonchi M, Chooban B, Mostaghim-Roudi M, Torkfar G. Epidemiology of burn injuries in south-eastern Iran: a retrospective study. J Pak Med Assoc 2013;63:1476–81. [3] Kalayi G, Muhammad I. Clothing burns in Zaria. Burns 1994;20:356–9. [4] Damant G. Clothing-related burn casualties: an overlooked problem?. GBH International; 2009. [5] Liao CC, Rossignol AM. Landmarks in burn prevention. Burns 2000;26:422–34. [6] Oglesbay FB. The flammable fabrics problem. 1969. Inj Prev 1998;4:317–20. [7] Warne CA. The continuing problem of serious burns involving the ignition of clothing particularly nightwear. Fire Mater 1979;3:195–201. [8] US Consumer Products Safety Commission. Children’s sleep-wear standards: sizes 0-6X: 16 CFR 1615 (FF3-71); sizes 7-14: 16 CFR 1616 (FF5-74). [9] Gordon PG, Pressley TA. The fire hazard of children’s nightwear: the Australian experience in developing clothing fire hazard standards. Burns 1978;5:13–8. [10] Commonwealth of Australia. Trade practices (Consumer Product Safety Standards) (Children’s Nightwear and Paper patterns for Childrens’s Nightwear) regulations 2007. [11] Standards Australia Children’s nightwear and limited daywear having reduced fire hazard AS/NZS 1249; 2003. [12] Laing RM, Bryant V. Prevention of burn injuries to children involving nightwear. N Z Med J 1991;104:363–5. [13] McLoughlin E, Clarke N, Stahl K, Crawford JD. One pediatric burn unit’s experience with sleepwear related injuries. 1977. Inj Prev 1998;4:313–6. [14] Knudson MS, Bolieu SL, Larson DL. Children’s sleepwear flammability standards: have they worked. Burns 1980;6:255–60. [15] Horrocks AR, Nazare´ S, Kandola B. The particular flammability hazards of nightwear. Fire Saf J 2004;39:259–76. [16] SAS Institute. SAS enterprise guide 5.1. North Carolina: Cary; 2013. [17] Dobson A, Kuulasmaa K, Eberle E, Schere J. Confidence intervals for weighted sums of Poisson parameters. Stat Med 1991;10:457–762. [18] Centre for Epidemiology and Evidence. Statisitcal application for population health research and intellegence (SaPHaRI); 2014. [19] Berry J, Harrison J. A guide to statistical methods for injury surveillance. Adelaide: Australian Institute of Health and Welfare; 2006. [20] NSW Ministry of Health. NSW costs of care standards 2009/ 10. Sydney: NSW Ministry of Health; 2011.
64
burns 41 (2015) 58–64
[21] Mitchell RJ, Curtis K, Holland AJ, Balogh ZJ, Evans J, Wilson KL. Acute costs and predictors of higher treatment costs for major paediatric trauma in New South Wales, Australia. J Paediatr Child Health 2013;49:557–63. [22] Curtis K, Lam M, Mitchell R, Dickson C, McDonnell K. Major trauma: the unseen financial burden to trauma centres, a descriptive multicentre analysis. Aust Health Rev 2014;38:30–7. [23] Pointer S. Trends in hospitalised injury, Australia: 1999-00 to 2010-11. Injury research and statistics series 86 INJCAT 162, Canberra: AIHW; 2013. [24] Stanwick RS. Clothing burns in Canadian children. Can Med Assoc J 1985;132:1143. [25] Wilson DI, Bailie FB. Night attire burns in young girls – the return of an old adversary. Burns 1999;25:269–71. [26] Health Statistics NSW. Current smoking in adults by sex, NSW 2002 to 2012 online report. Sydney: NSW Ministry of Health; 2014. [27] Commonwealth of Australia. Trade practices (consumer product safety standard) (disposable cigarette lighters) regulations 1997.
[28] Commonwealth of Australia. Trade practices (consumer product safety standard) (Reduced fire risk cigarettesamendment regulations 2009) (No 1). [29] Barss P, Wallace K. Grass-skirt burns in Papua New Guinea. Lancet 1983;321:733–4. [30] Leong SC, Emecheta IE, James MI. The flaming gypsy skirt injury. Injury 2007;38:122–4. [31] Hockey R, Miles E. Workplace injury Brisbane. Queensland Injury Surveillance Unit; 1999. [32] Australian Bureau of Statistics. Safety in the home Cat: 4387.2. Australian Bureau of Statistics; 1998. [33] Ashby K. Injuries associated with do-it-yourself maintance activities. Hazard.Melbourne: Monash University Accident Research Centre; 1999. [34] McKenzie K, Enraght-Moony E, Waller G, Walker S, Harrison J, McClure R. Causes of injuries resulting in hospitalisation in Australia: assessing coder agreement on external causes. Injury Prev 2009;15:188–96.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Clothing-related burns in New South Wales, Australia: Impact of legislation on a continuing problem Lara A. Harvey a,*, Siobhan Connolley b, John G. Harvey c a Falls and Injury Prevention Group, Neuroscience Research Australia, University of New South Wales, Sydney, NSW 2052, Australia b Statewide Burn Injury Service, NSW Agency for Clinical Innovation, NSW, Australia c Burn Unit, Children’s Hospital at Westmead, Sydney, NSW, Australia
article info
abstract
Article history:
To combat the risk of nightwear burns a mandatory standard regulating the design,
Received 10 June 2014
flammability and labelling requirements of children’s nightwear was introduced in
Received in revised form
Australia in 1987. This population-based study examined the trends, characteristics and
16 September 2014
causes of clothing-related burns to inform a review of the current standard, and to facilitate
Accepted 9 October 2014
the development of targeted prevention strategies. Clothing-related burns for 1998–2013 were identified from hospitalisation data for all hospitals in NSW and detailed information regarding circumstance of injury from a burn
Keywords:
data registry. To investigate percentage annual change (PAC) in trends negative binomial
Prevention
regression analysis was performed.
Legislation
There were 541 hospitalisations for clothing-related burns, 18% were nightwear-related
Regulations
and 82% were for other clothing. All clothing burns decreased by an estimated 4% per year
Clothing-burns
(95% CI
Injuries
(PAC
6.2 to
2.1). Nightwear-related burns decreased by a significantly higher rate
7.4%; 95% CI
12.5 to
2.1) than other clothing (PAC
2.5%; 95%CI
4.7 to
0.1).
Exposure to open heat source (campfire/bonfire) was the most common cause, followed by cooking. Of factors known to be associated with clothing burns, accelerant use was reported in 27% of cases, cigarettes 17%, loose skirt or dress 8%, and angle grinders in 6% of cases. Hospitalisations for clothing burns are relatively uncommon in NSW and rates, particularly of nightwear burns, have decreased over the last 15 years. Strategies for continued reduction of these injuries include increasing the scope of the current clothing standard or developing new standards to include all children’s clothing and adult nightwear, and increasing community awareness of the risk associated with open heat sources, accelerant use and loose clothing. # 2014 Elsevier Ltd and ISBI. All rights reserved.
* Corresponding author. Tel.: +61 2 9385 1852; fax: +61 2 9385 6040. E-mail address: [email protected] (L.A. Harvey). http://dx.doi.org/10.1016/j.burns.2014.10.013 0305-4179/# 2014 Elsevier Ltd and ISBI. All rights reserved.
burns 41 (2015) 58–64
1.
Introduction
Despite major advances in the prevention of clothing-related burns seen in many countries over the last 60 years they continue to constitute a significant burn hazard worldwide, particularly in low to middle income countries where loose and/or flammable clothing catching fire remains a major cause of flame burns [1–3]. Although now comparatively uncommon in high income countries, clothing-related burns remain potentially devastating injuries. In the United States (US) in 2009, it was estimated that 4300 clothing-related burns are treated in hospital emergency departments each year and 120 people die [4]. The decrease in morbidity and mortality associated with clothing burns experienced in high income countries has been attributed in part to a change in fashion towards tight fitting clothing and the introduction of synthetic fabrics that do not burn readily, combined with the introduction of regulations restricting the flammability of clothing [5]. The first of these fabric regulations, the Flammable Fabric Act (FFA) was introduced in the US in 1953 in recognition of the high burn danger posed by brushed rayon sweaters and cowboy chaps [6]. Much of the early regulatory efforts thereafter focussed on children’s nightwear, as it was recognised at the time that the majority of severe injuries were attributable to loose-fitting easily ignited night attire [5]. Regulations introduced in the United Kingdom (UK) the following year addressed the flammability of fabrics specifically in children’s night dresses. In 1967, the UK regulations increased in scope to include adult nightdresses [7] and in the same year the US Flammable Fabric Act was amended to include all articles of wearing apparel and interior furnishings [6]. In 1973 additional standards were introduced in the US to regulate the flammability of young children’s sleepwear (sizes 0–6X) and increased to include larger sizes (sizes 7–14) in 1975 [8]. In Australia, the clothing standard AS 1249 ‘Safe design rules for children’s night clothes’, was first introduced in 1972, updated to ‘Children’s night clothes having reduced fire hazard’ in 1976 [9] and mandated under national legislation in 1978 [10]. The standard has undergone several subsequent amendments and currently includes some daywear or underwear items that may commonly be used as nightwear such as boxer shorts and infants all-in-ones [11]. Several studies have shown that the incidence of nightwear-related burns in children reduced significantly compared to other clothing burns following the introduction of nightwear standards/regulations in the US, UK and New Zealand [12–14]. In particular a study undertaken at the Shriner’s Burns Institute in Boston demonstrated that the percentage of nightwear related burns dropped from 32% to 4% of all burn admissions after the nightwear regulations were enforced [13]. Since these early reports, few studies have explored the long-term impact of the introduction of clothing regulations and standards. One study investigating impact of UK regulations found little to no sustained impact [15], and there has been little published on trends in other nonnightwear clothing burns. In Australia, the clothing standard AS/NZS 1249:2003 underwent public comment and review in 2013. Our aim
59
was to examine the trends, characteristics and causes of clothing burns at a population level in NSW, to inform the clothing standards review, and to facilitate the development of targeted prevention strategies.
2.
Methods
2.1.
Data sources
2.1.1.
Admitted patients data collection (APDC)
The APDC is a routinely collected census of all private and public hospitalisations across NSW. The APDC records ‘episodes of care’ in hospital which end with the discharge, transfer or death of the patient. All Hospitalisations due to burns and scalds were identified by ICD-10-AM external cause codes ‘X00-X19’ and of these, clothing-related burns were identified using ICD-10-AM external cause code ‘X05’ Exposure to ignition or melting of nightwear and ‘X06’ Exposure to ignition and melting of other clothing and apparel for the 15 years financial period 1998–1999 to 2012–2013. Records where the mode of admission was a transfer from another hospital or change in service category were excluded to minimise multiple counting of cases for the same injury. Cases admitted to a hospital in NSW but usually residing another state were also excluded as hospitalisation rates were calculated against the NSW resident population. All NSW residents admitted to interstate hospitals are included in the APDC and were included in the analysis. Numbers and rates for 2011–2012 and 2012–2013 include an imputed estimate of these interstate hospitalisations (n = 184).
2.1.2. Agency for clinical innovation NSW statewide burn injury service (SBIS) burns data registry The SBIS burns data registry contains admission records and case details for all patients admitted to the three designated Burns Units in NSW (two adult, one pediatric) and captures both hospitalisations and outpatient clinic visits. Detailed demographic, etiological and clinical data items are collected including a free text description of the circumstances resulting in the burn injury. Cases are classified according to mechanism of injury. Clothing-related burns were identified by mechanism of injury ‘Flame-clothing and bedding’ for the 6 years period for which full data coverage is available, 2007–2008 to 2012–2013. The free text descriptions of the circumstances of injury were reviewed and admissions for house fires and bedding-related burns were excluded.
2.2.
Statistical analysis
Analysis was performed using SAS Enterprise Guide 5.1 [16]. Direct age and sex standardised rates per 100,000 population were calculated along with the percentage annual change in the hospitalisation rate. Ninety-five percent confidence intervals were calculated assuming a Poisson distribution [17]. Agespecific population estimates for NSW at 31 December for each of the years studied were used to calculate rates. These estimates correspond to the mid-point of each financial year of hospitalisation data, and were interpolated from the Australian Bureau of Statistics population estimates at 30
60
burns 41 (2015) 58–64
June each year [18]. The Australian estimated resident population for 2001 was used as the standard population. To correct for over dispersion, negative binomial regression was used to examine the statistical significance of changes in trends over time [19]. To examine differences between nightwear and other clothing burns chi-squared tests of independence were used. Hospital costs were calculated using the Australian-Refined-Diagnostic-Related Group (AR-DRG) codes recorded in the NSW APDC for each episode of care. The AR-DRG is allocated after hospital discharge and is intended to provide an indication of the resources required by the hospital to treat the individual, including all inhospital medical and allied health treatment, nursing, diagnostics and pharmaceuticals. Average AR-DRG version 5.2 costs were applied to each episode of care in accordance with the NSW Cost of Care Standards 2009/10 [20].
3.2.
Demographic characteristics
3.
Results
Hospitalisations by age group, sex and clothing type are shown in Table 2. A third of all clothing related burns were adults and a further quarter were children aged 5–14 years. Nightwear burns were more common among females (59%) whilst burns related to other clothing and apparel were more common amongst males (68%) ( p < 0.0001). There were no nightwear clothing burns in infants during the study period. The highest rate of hospital admissions was for non-nightwear clothing in boys aged 5–14 years. The majority of people hospitalised for clothing-related burns were Australian born (87%), and this did not differ between nightwear and other clothing and apparel ( p = 0.3813). Admissions for clothing-related burns occurred in every month of the year, with the peak in the winter months of June–August. There was a significant difference in pattern with nightwear peaking in June and other clothing and apparel related burns peaking later in August ( p = 0.0022).
3.1.
Temporal trends
3.3.
There were 541 hospitalisations across all age groups for clothing-related burns for the 15 years period of the study. The annual number of hospitalisations ranged from 59 in 1998–1999 to 20 in 2011–2012 (Fig. 1). Of these 18% (99) were as a result of nightwear and 82% (442) were other clothing and apparel. Age and sex standardised hospitalisation rates by year for all burns and scalds, clothing-related burns, nightwear burns and other clothing and apparel burns are shown in Table 1. All clothing-related burns have decreased over the study period by an estimated 4.1% per year (95%CI 6.2 to 2.1, p < 0.0001) from 0.91 per 100,000 in 1998–1999 to 0.38 per 100,000 in 2012–2013. Nightwear-related burns decreased by an estimated 7.4% per year (95%CI 12.5 to 2.1, p < 0.007) whilst other clothing and apparel decreased by 2.5% (95%CI 4.7 to 0.1). Hospitalisations for all burns and scalds combined decreased by 1.0% per year (95%CI 1.5 to 0.5, p < 0.0001) over the same period.
Burn and hospitalisation characteristics
The most common location of burn was the trunk (36%) followed by hip and lower limb (29%). There was no difference in location between nightwear and other clothing and apparel burns ( p = 0.2122). Of the 411 hospitalisations where the depth of burn was recorded the majority were for full thickness burns (67%), partial thickness (31%) and erythema (3%) (Table 3). There was no difference in proportion of depth of burn between nightwear and other clothing and apparel ( p = 0.1407). The majority of hospitalisations were for short duration (less than 2 days) with 7% being for greater than 4 weeks. Length of stay for nightwear ranged from 1 to 80 days and clothing and other apparel for 119 days. There was no difference in LOS between nightwear and other clothing ( p = 0.0998). There were 24 in-hospital deaths (4%) for all clothingrelated burns, with a significantly higher proportion of deaths
60
X05 Exposure to ignion or melng of nightwear 50
Number of hospalisaons
X06 Exposure to ignion or melng of other clothing and apparel 40
30
20
10
0 1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
2008-09
2009-10
2010-11
2011-12
2012-13
Fig. 1 – Number of hospitalisations for clothing-related burns, by clothing type, NSW, Australia, 1998–1999 to 2012–2013.
61
burns 41 (2015) 58–64
Table 1 – Age and sex standardised hospitalisation rates per 100,000 for burns, by mechanism and year, NSW, Australia, 1998–1999 to 2012–2013. Year
1998–1999 1999–2000 2000–2001 2001–2002 2002–2003 2003–2004 2004–2005 2005–2006 2006–2007 2007–2008 2008–2009 2009–2010 2010–2011 2011–2012 2012–2013
All burns
All clothing burns
Nightwear burns
Other clothing and apparel burns
38.4 36.9 37.1 35.1 30.7 31.3 27.1 30.0 30.1 32.1 28.3 29.0 28.9 30.1 28.9
0.91 0.56 0.80 0.76 0.49 0.49 0.50 0.43 0.59 0.48 0.40 0.52 0.43 0.28 0.38
0.34 0.09 0.06 0.23 0.14 0.10 0.08 0.11 0.07 0.10 0.03 0.01 0.03 0.05 0.03
0.58 0.46 0.74 0.53 0.35 0.39 0.42 0.33 0.52 0.38 0.37 0.51 0.40 0.23 0.36
reported in 64 cases (27%), cigarette 41 (17%), alcohol use/ intoxication 12 (5%) and loose flowing clothing (skirt or dress) 19 (8%).
4.
This state-wide population based study examined the incidence, characteristics and circumstances leading to clothingrelated burns over a 15 years period in NSW, providing the most comprehensive profile available to date. It shows that hospitalisation for clothing-related burns is relatively rare, comprising only 2% of all burns hospital admissions for NSW over the study period. Although not common, clothing-related burns were severe injuries with over half being full thickness and a quarter requiring hospitalisation for a period of greater than a week. The cost of these injuries was just under half a million dollars annually for acute hospital care alone, and this is likely to be a substantial underestimate of the true cost of providing burn care in hospital, as the NSW peer group average cost estimates (AR-DRG groupings) have recently been shown to significantly underestimate actual hospitalisation costs for both adult and pediatric trauma patients [21,22]. Hospitalisation rates for all clothing-related burns have decreased over the 15 years study period by 4% per year in this study. In contrast, rates for all burns and scalds combined have decreased by just 1% over the same period making it unlikely that the observed decrease for clothing-related burns is solely a reflection of changes in hospital admission practice for burns. The relatively stable hospitalisation rate for all burns and scalds in NSW documented in our study is consistent with recently reported national trends [23]. In Australia, the current iteration of clothing standard AS/ NZS 1249:2003 [11] specifies the design, flammability performance and labelling requirements for children’s nightwear garments for sizes 00-14. The standard stipulates that all children’s nightwear must display an appropriate hazard classification, thus the approach is not one of banning the sale of dangerous nightwear, but to inform consumers of the potential hazard and permit them to make their own judgement as to whether they consider the fire risk is acceptable. Our study provides evidence of the sustained impact of these standards, with no hospital admissions for nightwear-related burns in infants and low hospitalisation rates in toddlers and children over the study period. Whilst it is
resulting from nightwear (9%) compared to other clothing and apparel (3%) ( p = 0.0128). The annual average acute hospitalisation cost for all clothing-related burns was AUS $446,529 of which nightwear contributed $74,139 and other clothing and apparel contributed $372,390.
3.4.
Discussion
Circumstances leading to injury
There were 237 patients requiring hospitalisation identified from the SBIS burns registry, which was 34% more than identified from the APDC for the same period. There was no difference in sex distribution between hospitalised patients identified in the APDC and the burns registry ( p = 0.3638), however registry data identified a higher proportion of adult clothing-related burns but less for children ( p < 0.0001). Nightwear-related burns identified in the free text constituted only 8% of all clothing burns in the registry data, which was lower than that identified in APDC (18%) ( p = 0.0004). The circumstances surrounding the burns are summarised in Table 4. Exposure to an open heat source such as a campfire or bonfire was the most common cause of clothing burn, followed by cooking, children playing with matches or cigarette lighters and use of angle grinders. Of factors known to be associated with clothing burns, accelerant use was
Table 2 – Age and sex distribution of hospitalisations for clothing-related burns, by clothing type, NSW 1998–1999 to 2012– 2013. Age group
Night wear Male n
Infant (<1 year) Toddler (1–4 years) Child (5–14 years) Youth (15–24 years) Adult (25–64 years) Elderly (65+ years)
0 8 19 1 10 3
Total
41
Female
Rate (95% CI)
n
0.29 0.28 0.01 0.04 0.05
0 13 12 3 16 14
(0.13–0.58) (0.17–0.43) (0.00–0.08) (0.02–0.07) (0.01–0.16)
Other clothing and apparel
58
Rate (95% CI) 0.50 0.18 0.04 0.06 0.18
(0.27–0.86) (0.09–0.32) (0.01–0.13) (0.03–0.09) (0.10–0.19)
Male n 2 21 96 47 101 32 299
Female
Rate (95% CI)
n
Rate (95% CI)
0.29 0.77 1.39 0.66 0.37 0.52
0 23 41 14 49 16
0.71 0.62 0.21 0.18 0.21
(0.03-1.04) (0.48–1.17) (1.13–1.70) (0.49–0.88) (0.30–0.45) (0.36–0.74)
143
(0.45–1.06) (0.45–0.85) (0.11–0.35) (0.13–0.24) (0.12–0.34)
62
burns 41 (2015) 58–64
Table 3 – Depth of burn and length of stay for hospitalisations for clothing-related burns, by clothing type, NSW 1998–1999 to 2012–2013. Characteristic
Depth of burna Erythema Partial thickness Full thickness LOSb <2 days 2–7 days 8–28 days >28 days
Nightwear n (%)
Other clothing n (%)
Total n (%)
2 (2.8) 22 (30.6) 48 (66.7)
11 (3.2) 145 (42.8) 183 (54.0)
13 (3.2) 167 (40.6) 231 (56.2)
70 17 24 11
261 (49.0) 121 (22.7) 117 (21.9) 34 (6.4)
331 (50.5) 138 (21.1) 141 (21.5) 45 (6.9)
(57.4) (13.9) (19.7) (9.0)
a
Includes 411 cases (130 other/unspecified excluded). Includes records for new admissions, transfers and type changes (n = 655).
b
possible that the low hospitalisation rates may relate to an increased awareness of the danger of clothing burns in the general population, the sustained decrease in nightwear burns (7.4% per year) was significantly higher than that seen for other clothing and apparel (2.5% per year). Further, only 18% of the clothing-related burn hospitalisations, (on average two cases a year over the last 5 years) were for nightwear burns, which is in contradiction to early epidemiological studies which reported up to 60% of pediatric clothing-related burns admissions were due to nightwear [13,24]. Traditionally two vulnerable groups have been identified as being disproportionately at risk of nightwear related burns – young children who do not realise these the dangers and indulge in fire play, and elders in whom the reaction time is slowed [7,15,24]. Our study highlighted that whilst there were
Table 4 – Circumstances surrounding clothing-related burns, NSW 2007–2008 to 2012–2013. Circumstance Campfire/bonfire/open fire Cooking Child playing with matches/ cigarette lighter Candle Angle grinding Welding Not adequately described/other Smoking-related- when awake Smoking in bed/sleep with cigarette BBQ Heater Self-inflicted Fixing car Game/prank Other work-related Assault Sparkler/fireworks Kerosene lamp/gas bottle explosion Total Source: SBIS Burns Data Registry.
Male
Female
Persons
51 (32.7) 6 (3.9) 16 (10.3)
14 (17.5) 19 (23.8) 6 (7.5)
65 (27.5) 25 (10.6) 22 (9.3)
5 15 14 9 6 3
(3.2) (9.6) (9.0) (5.8) (3.9) (1.9)
12 0 0 2 5 6
(15.0) (0) (0) (2.5) (6.3) (7.5)
17 15 14 11 11 9
(7.2) (6.4) (5.9) (4.7) (4.7) (3.8)
3 1 4 6 6 6 2 1 2
(1.9) (0.6) (2.6) (3.9) (3.9) (3.9) (1.3) (0.6) (1.3)
5 6 3 0 0 0 1 1 0
(6.3) (6.3) (7.5) (3.8) (0) (0) (1.3) (1.3) (0)
8 7 7 6 6 6 3 2 2
(3.4) (3.0) (3.0) (2.5) (2.5) (2.5) (1.3) (0.8) (0.8)
156
80
236 (100)
no hospitalisations for infants, nightwear burns in toddlers and children under 15 years still occur. The equal numbers of males and females in this age group may reflect a trend towards wearing large or adult sized T-shirts which are not considered night attire, and not currently covered by the Australian standards. A similar phenomenon has been reported in the UK [25]. In older people, nightwear-related burns were more common in females perhaps reflecting that the general trend away from loose nightdresses has not been taken up by the elderly. Again this pattern is consistent with studies from the UK [7,15] and highlights the potential benefit of either increasing the scope of the Australian nightwear standard to include all age groups or to develop an additional standard to address adult nightwear. In our data nightwear burns were severe injuries and although the proportion with full thickness burns or length of time in hospital did not differ significantly from other clothing-related burns they resulted in a higher proportion of in-hospital deaths (9%), signalling a high priority for prevention. Currently in Australia there are no safety requirements for children’s daywear clothing, or any adult clothing. In our study other clothing-related burns were more common in males, with highest hospitalisation rate in boys aged 5–14 years, perhaps associated with fire play and risk taking behaviour. It is well recognised that clothing burns involve three key factors: the presence of a heat source, the behaviour of the wearer and the flammability of the fabric [6]. The presence of a heat source remains a large and potentially modifiable issue in our data. The peak of admissions occurred in winter months and evidence from the burn registry highlight that open fires and cooking were the most common causes of burn. Behaviour was also a major factor, with just under a third of all admissions reporting use of an accelerant such as petrol. Even garments complying with the requirements of the standard will give little or no protection against burns where flammable liquids are involved. Regular education on the dangers of accelerants and heat sources may be beneficial. Smokers’ materials such as cigarettes, lighters and matches have been previously implicated as a major risk factor for clothing fire [15,25]. In our study 17% of the hospitalisations involved smokers materials which is less than previous reports of 29% from the UK [15] perhaps reflecting that smoking rates in NSW have decreased from 23.5% to 16.2% over the last 10 years period [26]. If this decreased trend in smoking continues, coupled with the introduction of a mandatory safety standard for child resistant cigarette lighters in 1997 [27] and reduced fire cigarettes in 2010 [28] the number of clothing-related burns caused by smokers material may continue to decrease. Candles have been shown to pose a fire risk, contributing up to 3% of clothing burns in the UK [15] and 1.2% in our study. The design and fit of garments is a critical element in the hazard. Loose fitting garments with long flowing design and billowing sleeves are hazardous anywhere near open flames. Case reports have highlighted the risk of saris, long loose scarfs and grass skirts [29] and gypsy-style skirts [30]. Eight percent of clothing burns in our study involved the ignition of dress or skirt. Targeted education into the risk posed by loose flowing clothing is clearly needed.
burns 41 (2015) 58–64
Finally our study identified a high proportion of clothing burns resulting from the use of angle grinders. Angle grinders are known to be one of the most common power tools to cause injuries in the workplace in Australia [31], however in our data 40% of the angle grinder burns were sustained during do-ityourself (DIY) activities at home. Injuries related to DIY activities pose a particular problem for prevention as is extremely difficult to control safety education, and impossible to regulate or enforce the use of protective equipment. Home safety surveys report that around 20% of Australian households have an angle grinder [32] and only 5% of DIYers who sustained an angle grinder injury were wearing personal protective equipment at the time of injury [33]. Educational material highlighting the potential risks and appropriate safety measures should be developed and provided at the point of sale or hire of DIY equipment. The population coverage and relatively long time frame are strengths of this study, however there are several limitations that must be considered. To identify clothing-related burns in the routinely collected administrative hospitalisation database we were reliant on the accuracy of external cause coding. Whilst much effort has been expended in Australia to ensure the accuracy of diagnosis codes a recent clinical audit which evaluated the accuracy of coding for external causes in Australian hospitals found 29% disagreement for the coding block X00-X10 ‘smoke, fire and flames’ [34]. In our study the incidence of clothing-related burns identified from the APDC would appear to be under numerated by around 30% when compared to the incidence recorded in the SBIS Burn Registry, which in itself is provides an underestimation as it only includes hospitalisations for the three designated burns units in NSW. Further research linking APDC hospitalisation data to the SBIS Burns registry data is warranted to determine the impact of coding inconsistencies on the identification of burns. In addition, less severe burns treated in emergency departments or outpatient burns clinics are not included in the APDC leading to further underestimation of the total burden of clothing-related burn injury in NSW.
5.
Conclusions
Hospitalisations for clothing burns are relatively uncommon in NSW and rates, particularly of nightwear burns, have decreased over the last 15 years. Important strategies for continued reduction of these injuries include increasing the scope of the clothing standards or developing new standards to include all children’s clothing and adult nightwear, and increasing community awareness of the risk associated with open heat sources, accelerant use and loose clothing. Improved coding of clothing burns in hospital data, including the circumstances of injury could provide a more comprehensive information base from which to plan and monitor prevention strategies.
Conflict of interest There were no conflicts of interest.
63
Acknowledgements Lara Harvey was supported by the NSW Ministry for Health. The authors thank the Centre for Population Health at the NSW Ministry of Health for providing access to the hospitalisation data via Secure Analytics for Population Health Research and Intelligence (SAPHaRI) and the Statewide Burn Injury Service at the NSW Agency for Clinical Innovation for providing access to the registry data analysed in this study.
references
[1] Golshan A, Patel C, Hyder AA. A systematic review of the epidemiology of unintentional burn injuries in South Asia. J Public Health 2013;35:384–96. [2] Ansari-Moghaddam A, Baghbanian A, Dogoonchi M, Chooban B, Mostaghim-Roudi M, Torkfar G. Epidemiology of burn injuries in south-eastern Iran: a retrospective study. J Pak Med Assoc 2013;63:1476–81. [3] Kalayi G, Muhammad I. Clothing burns in Zaria. Burns 1994;20:356–9. [4] Damant G. Clothing-related burn casualties: an overlooked problem?. GBH International; 2009. [5] Liao CC, Rossignol AM. Landmarks in burn prevention. Burns 2000;26:422–34. [6] Oglesbay FB. The flammable fabrics problem. 1969. Inj Prev 1998;4:317–20. [7] Warne CA. The continuing problem of serious burns involving the ignition of clothing particularly nightwear. Fire Mater 1979;3:195–201. [8] US Consumer Products Safety Commission. Children’s sleep-wear standards: sizes 0-6X: 16 CFR 1615 (FF3-71); sizes 7-14: 16 CFR 1616 (FF5-74). [9] Gordon PG, Pressley TA. The fire hazard of children’s nightwear: the Australian experience in developing clothing fire hazard standards. Burns 1978;5:13–8. [10] Commonwealth of Australia. Trade practices (Consumer Product Safety Standards) (Children’s Nightwear and Paper patterns for Childrens’s Nightwear) regulations 2007. [11] Standards Australia Children’s nightwear and limited daywear having reduced fire hazard AS/NZS 1249; 2003. [12] Laing RM, Bryant V. Prevention of burn injuries to children involving nightwear. N Z Med J 1991;104:363–5. [13] McLoughlin E, Clarke N, Stahl K, Crawford JD. One pediatric burn unit’s experience with sleepwear related injuries. 1977. Inj Prev 1998;4:313–6. [14] Knudson MS, Bolieu SL, Larson DL. Children’s sleepwear flammability standards: have they worked. Burns 1980;6:255–60. [15] Horrocks AR, Nazare´ S, Kandola B. The particular flammability hazards of nightwear. Fire Saf J 2004;39:259–76. [16] SAS Institute. SAS enterprise guide 5.1. North Carolina: Cary; 2013. [17] Dobson A, Kuulasmaa K, Eberle E, Schere J. Confidence intervals for weighted sums of Poisson parameters. Stat Med 1991;10:457–762. [18] Centre for Epidemiology and Evidence. Statisitcal application for population health research and intellegence (SaPHaRI); 2014. [19] Berry J, Harrison J. A guide to statistical methods for injury surveillance. Adelaide: Australian Institute of Health and Welfare; 2006. [20] NSW Ministry of Health. NSW costs of care standards 2009/ 10. Sydney: NSW Ministry of Health; 2011.
64
burns 41 (2015) 58–64
[21] Mitchell RJ, Curtis K, Holland AJ, Balogh ZJ, Evans J, Wilson KL. Acute costs and predictors of higher treatment costs for major paediatric trauma in New South Wales, Australia. J Paediatr Child Health 2013;49:557–63. [22] Curtis K, Lam M, Mitchell R, Dickson C, McDonnell K. Major trauma: the unseen financial burden to trauma centres, a descriptive multicentre analysis. Aust Health Rev 2014;38:30–7. [23] Pointer S. Trends in hospitalised injury, Australia: 1999-00 to 2010-11. Injury research and statistics series 86 INJCAT 162, Canberra: AIHW; 2013. [24] Stanwick RS. Clothing burns in Canadian children. Can Med Assoc J 1985;132:1143. [25] Wilson DI, Bailie FB. Night attire burns in young girls – the return of an old adversary. Burns 1999;25:269–71. [26] Health Statistics NSW. Current smoking in adults by sex, NSW 2002 to 2012 online report. Sydney: NSW Ministry of Health; 2014. [27] Commonwealth of Australia. Trade practices (consumer product safety standard) (disposable cigarette lighters) regulations 1997.
[28] Commonwealth of Australia. Trade practices (consumer product safety standard) (Reduced fire risk cigarettesamendment regulations 2009) (No 1). [29] Barss P, Wallace K. Grass-skirt burns in Papua New Guinea. Lancet 1983;321:733–4. [30] Leong SC, Emecheta IE, James MI. The flaming gypsy skirt injury. Injury 2007;38:122–4. [31] Hockey R, Miles E. Workplace injury Brisbane. Queensland Injury Surveillance Unit; 1999. [32] Australian Bureau of Statistics. Safety in the home Cat: 4387.2. Australian Bureau of Statistics; 1998. [33] Ashby K. Injuries associated with do-it-yourself maintance activities. Hazard.Melbourne: Monash University Accident Research Centre; 1999. [34] McKenzie K, Enraght-Moony E, Waller G, Walker S, Harrison J, McClure R. Causes of injuries resulting in hospitalisation in Australia: assessing coder agreement on external causes. Injury Prev 2009;15:188–96.