Skip to content
problem genome

One of 592 unsolved problems in the Problem Genome, a collection written up for student design and engineering teams.

health · materials

the sleeve over the back burner

In Japan clothing fires kill about 100 people a year, mostly older adults, and in Tokyo homes the stove starts about 80% of clothing fires that hurt someone; sleeve-sensing stoves now exist, but for stoves already in homes the official advice is a flame-resistant arm cover, which few older people in a Tokyo-area survey knew of or owned, and a warning video shown to students raised perceived risk but not intentions

Source: Consumer Affairs Agency (消費者庁) Open ↗source: primaryLast updated: 2026-11-17

Problem statement

In Japan, a fire in the clothes a person is wearing (着衣着火, "clothing ignition") kills about 100 people a year, and more than 80% of them are 65 or older. Nationally, more of these deaths happen at bonfires and other open flames than while cooking: of 101 clothing-ignition deaths in 2022, 13 happened while cooking, and of 97 in 2023, 15 did. But in Tokyo homes the kitchen stove is where most clothing fires that hurt someone start. There, about 80% of clothing fires that caused a death or injury in 2020–2024 started at the stove (186 people, 8 of them killed), and people in their 80s were the largest group. A sleeve catches as the cook reaches over a lit gas burner for a pot or a bottle of seasoning at the back. Warnings come round every autumn and winter, and the Consumer Affairs Agency and the Tokyo Fire Department give the same advice: don't lean over the flame, keep the flame under the pan, avoid wide sleeves, and wear flame-resistant (防炎) aprons and arm covers. In the national product-safety agency's (NITE's) accident records, serious accidents from pans left on the flame nearly ended after stove safety devices were made mandatory in 2008, which NITE credits to the devices, but clothing-ignition accidents went on. Gas stoves that sense a hand or sleeve over the burner and turn the flame down are now sold. For the stoves already in homes, though, NITE's answer is the flame-resistant arm cover or apron. In a 2013 survey of 3,000 people aged 60 and over in and around Tokyo, 83.8% did not know such products existed and only 15 people owned any. In a small controlled test with students, a warning video raised perceived danger slightly but not the intention to be careful. The unsolved problem is a way to protect the sleeve at the stoves people already have, one that older cooks will actually use every day, in place of one more warning.

Why this matters

The Consumer Affairs Agency's count of national fire statistics gives 572 clothing-ignition deaths over 2015–2020, of which 493 were aged 65 or older. The yearly figures (87, 81, 92, 111, 106, 95) show no downward trend, and the Fire and Disaster Management Agency's tables give 101 in 2022 and 97 in 2023. Of those two years' deaths, 13 and 15 happened while cooking (炊事中), 13 of the 13 and 12 of the 15 in people aged 66 and over; bonfires (26 and 27) and other handling of fire account for more. Counting only fires in homes, the agency's final figures are 41 clothing-ignition deaths in 2022 and 38 in 2023. Injuries are where the stove leads. The Tokyo Fire Department counts 186 people killed or hurt by stove clothing fires in Tokyo homes in 2020–2024 (8 deaths, 178 injuries), with women more than three times as many as men. Of 86 clothing-ignition accidents reported to the Consumer Affairs Agency's hospital network between December 2010 and October 2021, 50 (58%) needed admission to hospital and 4 people died, and the stove was the most common source indoors (about 60%). The national product-safety agency (NITE) found that, since 2007, of 33 gas-stove deaths and serious injuries in people aged 60 and over whose cause could be judged, 13 (39%) were clothing ignition. A study of 130 burn patients aged 60 and over at one hospital found that flame burns were among the strongest predictors of a severe burn, and that in patients aged 75 and over, burns on the back were the strongest. Citing an earlier study of sleeve fires at gas stoves, its authors explain this by flames spreading from the arm to the back and chest before they can be put out.

What’s been tried and why it hasn’t worked

  1. Stove safety devices and sensing stoves. Since 2008 Japanese law has required every burner on a household gas stove to have an oil-overheating guard and a flame-failure device. NITE's accident data show that accidents from pans left on the flame fell after 2008 and now almost never happen, while clothing-ignition accidents, though few, "continue to occur" (our translation). NITE's 2026 report says no regulation directly prevents clothing ignition. Under a government award scheme for products that reduce misuse accidents, begun in 2025, two stoves have been recognised for cutting this risk: Paloma's アバンセ ("Avance" is our spelling), whose top-plate sensors detect a hand or sleeve and turn the flame to minimum, and Rinnai's SAFULL+, which limits the flame so it does not spill out from under a 20 cm pan. NITE's modelled risk assessment (not an observed outcome) finds such sensors give a large risk reduction. For gas stoves already in use, it points to flame-resistant aprons and arm covers.
  2. Repeated public warnings. The 2021 Consumer Affairs Agency notice gives a checklist (keep things off the back of the stove, take off scarves and stoles, avoid flared sleeves and hems, keep the flame under the pan). It warns that fluffy, napped or worn cotton and rayon fabric can "flash" across its surface in about a second. It quotes the Tokyo Fire Department's finding that about 80% of clothing ignitions in Tokyo homes in 2018–2020 started at the stove and about 10% at candles. The 2014 Tokyo survey report itself says (our translation) that "every year, as autumn and winter arrive, we see warnings to older people about clothing ignition." Deaths stayed near 100 a year from 2015 to 2020, and were 101 in 2022 and 97 in 2023 (see above).
  3. Flame-resistant products, recommended but rarely owned. The agency's notice recommends aprons and arm covers with flame-resistant (防炎) performance, which "do not easily spread fire" even if lit, "especially" for older people (our translation). In the 2014 Tokyo survey (fieldwork in November and December 2013, an internet panel aged 60 and over, n = 3,000), only 16.2% (486) knew of flame-resistant products such as pyjamas or arm covers, and only 3.1% of those (15 people) owned any. Of those 15, 6 owned pyjamas, 4 curtains, 2 arm covers and 1 an apron. Asked whether they would buy clothing that is hard to set alight, 52.2% said yes: 3.9% "even if a little expensive", 12.7% if a little more expensive but well designed, and 35.6% only "if it is not more expensive" (our translation). The other 47.8% said they would not go out of their way to buy it. The report ends by calling for repeated warnings and for flame-resistant products "that are easy to buy and easy to take into daily life" (our translation) to be developed and spread.
  4. A warning video, tested against slides alone. Suzuki and Kobayashi made a 2-minute-16-second video of cotton and polyester fabric burning on a chicken wing, to show how fabric melts and sticks to flesh. They showed it to one lecture class of university students, many of them hoping to work in childcare and teaching, after a slide explaining clothing ignition. The other class saw the slide only (303 valid answers; the two classes were the groups, not randomised individuals). Perceived danger was higher with the video (9.56 against 9.31 on a 0–10 scale, p < .05, d = 0.226). Intention to be careful did not differ (9.42 in both groups, p = .98). The authors note that a ceiling effect may explain this, and conclude that the video had an effect on awareness but none was found on whether people would act. They also note that clothing ignition is not only a children's problem: the harm is greater the older the age group.
  5. What the injury pattern says about why advice fails. In the Tokyo survey, 261 respondents reported a near miss or injury from clothing catching fire. The kitchen stove was the source in 233 cases, and in 192 of those the sleeve or arm caught. Women reported three times as many stove cases as men (175 against 58), and women aged 80 and over still reported 34 incidents from all sources. Respondents' own explanations include reaching across a lit front burner to a pot at the back, moving pots without turning the flame off, and (our translation) "doing several things at once has become impossible as I've got older." The burn study's authors write that older people's burns come from objects they use every day, perhaps with a sense that "I'm still fine," and call for "not only warnings about daily life but changes to the living environment" that allow for sensory decline and dementia (our translation). Advice asks the cook to remember, every time, a distance from the flame that they cannot easily judge while reaching.

What would unlock progress

Something that works at the moment the sleeve crosses the flame without relying on the cook to remember, on the stoves people already have. Sensing stoves address this for new purchases. The open part is the installed stove: a garment or cover older cooks actually choose to wear (the survey shows price and design decide this for most of the half who would buy), a way of gathering the sleeve that does not depend on a separate purchase, a retrofit that does at an existing stove some of what the sensing stoves do, or a change to what sits behind the burners so the reach never happens. Each needs testing with older cooks for whether they keep using it, not just whether they like it, since the one controlled test we found showed that a warning video changed what people thought but not what they planned to do.

How we’d know it’s being solved

The primary measure is the number of people killed or hurt by clothing fires at the stove: the Fire and Disaster Management Agency's yearly count of clothing-ignition deaths while cooking (13 in 2022, 15 in 2023) and the Tokyo Fire Department's count of stove clothing-fire casualties in homes (186 over 2020–2024), with the age split. Small trials can't move those numbers. For them, the measure is behaviour, not stated intention (the warning-video test shows the gap): the share of days a protective item is actually worn at the stove over several weeks, and the number of sleeve-over-flame events in observed cooking. A survey repeating the Tokyo questions would show whether knowing about flame-resistant products and owning them have moved from 16.2% knowing and 3.1% of those owning (2013).

Where to start

Facility-free door (reach map): with adult volunteers cooking a set menu at an unlit stove in a student-housing or teaching kitchen, film and count how often and how far a sleeve passes over each burner zone when reaching for back pots, lids and seasonings, and how much the layout (what is stored behind the stove) changes the count. Deliver a sleeve-zone map and a ranked list of layout changes. No flame and no special facility are needed.

Make-and-test door (older cooks): flame-resistant aprons and arm covers are already sold, and new sensing stoves already turn the flame down when a sleeve comes near. So do not make a new flame-resistant fabric, a new stove, or a copy of the covers on sale. Make something for the stove a cook already owns that answers what the survey found decides buying, price and design, plus the practical cost of putting on one more thing: for example, a cuff gather that works on the clothes the cook already has on, so no separate garment is needed. State in one line how it differs from the covers on sale and from ordinary sleeve clips and arm bands. Wear-test it with older adults reached through a senior centre or a community kitchen, at an unlit stove only, over several cooking sessions in one week. Measure whether they put it on unprompted (a simple wear log), whether it gets in the way of cooking, and whether it keeps the sleeve inside the zone the reach map marks as safe. Prototypes are never worn at a lit stove, in the sessions or at home; how the materials behave near a flame is a question for the lab door, tested on fabric alone. The burn study links risk to dementia: if people with dementia are included, the deliverable is the test design, and it needs ethics review.

Lab door (access line): comparing how everyday winter garments (worn fleece, napped sweaters, cotton and rayon blouses) and labelled flame-resistant covers behave against a gas-burner flame needs a fume hood or fire-test space and a supervisor trained in fire testing. A team without that access should design the test protocol and hand it to a textile or fire-safety lab. No flame test is ever run on a person or on clothing someone is wearing.

Relevant skills: product and textile design, human factors, observational study design, survey design.

Genes

Related problems

Stuck the same way, in other fields.

field: society & learningthe senior gym few older australians useAustralian councils install park exercise equipment built for older people, but in two Melbourne parks under 3% of older visitors were seen using it, a randomised trial of QR codes and guided sessions did not raise use among people over 60, and the council programme that did raise it runs on weekly volunteer sessionsbuilt without the people who'd live with itsource: primaryfield: the built worldgetting the fire alarm to the pillowIn Japan in 2006–2009, about 12,000 light-alarm units were shipped against 67.7 million home fire alarms, and only about 1 alarm in 10 could connect to a light or shaker; a free 2011–2013 scheme for deaf people on public assistance reached 3,613 homes, and in three Tokyo wards today's equipment grants mostly cover only severe certified hearing loss, while sleep tests abroad found shakers and a low-pitched sound wake most hard-of-hearing sleepersbuilt without the people who'd live with itsource: primaryfield: food & the living worldthe cupful that moves the musselsInvasive mussels ride between lakes in the cupful of water left in a wakeboat's ballast tank, and the existing decontamination takes up to an hour per boatbuilt without the people who'd live with itsource: secondaryfield: food & the living worldthe warning system nobody maintainsFlood early warning systems often lack the ongoing funding that keeps them runningbuilt without the people who'd live with itsource: primary

Sources

  • Consumer Affairs Agency (消費者庁), "着衣着火に御用心!毎年約100人の方が亡くなっています!-火に近づき過ぎない!火力の調節、適切な服装で事故予防-" News Release 令和3年11月17日 (Japan, 2021-11-17) (notice page), accessed 2026-11-17. Link ↗ Link 2 ↗
  • 東京都生活文化局消費生活部「ヒヤリ・ハット調査『シニア世代における衣服・履物の危険』報告書」平成26年9月 (Tokyo Metropolitan Government, Bureau of Citizens and Cultural Affairs, Consumer Affairs Division, September 2014), accessed 2026-11-17. Link ↗
  • 鈴木 公啓, 小林 久美「着衣着火啓発の取り組み(2) ―繊維の燃焼映像資料の作成の試みおよびその効果検証―」未来の保育と教育 ― 東京未来大学保育・教職センター紀要 ― 6: 57-64, 2019, doi:10.24603/tfumhk.6.0_57, accessed 2026-11-17. Link ↗
  • 総務省消防庁 防災情報室「令和5年(1~12月)における火災の状況(確定値)」消防の動き and「令和4年(1~12月)における火災の状況(確定値)」, accessed 2026-11-17. Link ↗ Link 2 ↗
  • 菊池 奈々子, 山田 律子「前期高齢者と後期高齢者における熱傷の特徴と「熱傷の重症度」への影響要因」老年看護学 22(1): 51-60, 2017, doi:10.20696/jagn.22.1_51, accessed 2026-11-17. Link ↗
  • 独立行政法人製品評価技術基盤機構(NITE)「ガスこんろにおける着衣着火の事故防止対策報告書」令和8年8月 (report page; annex 別紙3), accessed 2026-11-17. Link ↗ Link 2 ↗ Link 3 ↗
  • 東京消防庁「STOP!着衣着火」(STOP!住宅火災シリーズ①)令和8年1月発行, accessed 2026-11-17. Link ↗
  • 総務省消防庁『消防白書』資料1-1-22「死亡に至った経過と年齢別の死者発生状況」令和5年版(令和4年中) (data) and 令和6年版(令和5年中), accessed 2026-11-17. Link ↗ Link 2 ↗ Link 3 ↗
Revision notes

This brief gets checked against its sources and contains the following revisions. Future revisions get added here.

Worth knowing

  • Nationally, more clothing-fire deaths happen at bonfires and other open flames than while cooking; the stove leads on injuries in Tokyo homes.
  • The one controlled test of a warning found was with university students in two lecture classes, not with older adults.

Full check

Sources read

  • Consumer Affairs Agency, clothing-ignition notice (2021), with its notice page and data appendix: full text (read in Japanese)
  • Tokyo Metropolitan Government, survey of hazards from clothing and footwear for older people (2014): full text (read in Japanese; the clothing-ignition section and summary read, the rest searched)
  • Suzuki and Kobayashi, testing a fabric-burning warning video (Tokyo Future University bulletin, 2019): full text (read in Japanese)
  • Kobayashi and Suzuki, companion workshop report (2019): full text (read in Japanese; not used)
  • Kikuchi and Yamada, burns in older adults (gerontological nursing journal, 2017): full text (read in Japanese)
  • Fire and Disaster Management Agency, final fire bulletins for 2022 and 2023: full text (read in Japanese)
  • Fire and Disaster Management Agency, white-paper tables of fire deaths by cause and age (2022, 2023): full text
  • NITE, report on clothing ignition at gas stoves (2026), with its annex: full text (read in Japanese)
  • Tokyo Fire Department, clothing-ignition leaflet (2026): full text (read in Japanese)

What changed

  • Stated more strongly than the source: that the stove is where most clothing fires that hurt someone start, with no place given; only the Tokyo Fire Department's figures show this, for Tokyo homes, so the page now says so; that three named bodies repeat the same advice each autumn; the page now says warnings come round every autumn and winter, and that the agency and the fire department give the same advice; that clothing-ignition accidents fell after 2008; the product-safety agency says they are few but continue; a warning video as 'the only controlled test'; it is now the one controlled test we found.
  • Numbers: the 34 incidents were reported by women aged 80 and over, not women in their 80s; the 3.1% who owned flame-resistant products is now given as a share of the 16.2% who knew of them, not of all respondents.
  • Entry points: the build starting point now also asks how the design differs from ordinary sleeve clips and arm bands.

Left open

  • that about 80% of clothing fires in Tokyo homes in 2018–2020 started at the stove [cited through the agency's notice; the fire department's original was not read]

Separate check of specific claims: The Paloma stove's name is now given in the product-safety agency's Japanese script, with 'Avance' marked as our spelling, and a related page was named.

Earlier full check: Nine details were corrected, including that the product-safety agency credits the 2008 safety devices for the fall in left-pan accidents rather than this being stated as fact, 'more than three times' as many women as men among those hurt, the hospital-network cases called accidents rather than injuries, translated quotations marked as ours, and an unsourced closing sentence removed.

Separate check of specific claims: Entry points: the build starting point now says not to copy covers already sold and leads with a cuff gather for the clothes a cook already wears; prototypes are never worn at a lit stove, in sessions or at home; and an ethics line was added for people with dementia.

Earlier full check: National fire tables showed that the stove leads on clothing-fire injuries in Tokyo homes but not on deaths nationally, so the long title and main text were rewritten to say this (display title unchanged); sleeve-sensing stoves now on sale were added, and the page now focuses on stoves already in homes.