health · infrastructure
the cord still hanging by the crib
Canada tried education, voluntary industry measures, a standard and a 2009 regulation against window-blind cords that strangle young children, and deaths fell only from 1.4 to 1.0 a year; its 2021 rule removes long cords from new blinds, but it leaves the blinds already in homes, guarded by cleats and tension devices that work only if someone fits and uses them every time
Source: Government of Canada, “Corded Window Coverings Regulations: SOR/2019-97” Open ↗source: primaryLast updated: 2026-11-17
Problem statement
Young children are strangled by the cords of window blinds and shades: the pull cords hanging at the side, and the inner cords that raise the slats. Health Canada counts 39 such deaths in Canada from 1989 to November 2018, with children aged one to three most at risk. Deaths are also reported in the US, Australia and the UK. For more than 25 years Canada tried education, voluntary industry action, a national standard and, from 2009, a regulation built on that standard. Deaths fell only from 1.4 to 1.0 a year. In every Canadian death after 2009 where the product could be identified, the blind met the 2009 regulation (four such deaths are confirmed). Health Canada's diagnosis is that the rules allowed long, reachable cords, guarded by cleats, cord stops and tension devices that carers have to install and use every time. "Incident data show that they are not always properly installed or used." A stricter regulation came into force on 1 May 2021, with Health Canada stepping up enforcement from 1 May 2022: any cord a child can reach must be too short to wrap round a one-year-old's neck. But it applies to blinds sold from then on, and Health Canada counts on its own rules only for what is sold: an add-on that is not fitted to the product when it reaches the buyer, or can be taken off by hand, does not count. Health Canada's own estimate expects deaths from blinds not yet replaced to continue for years. The unsolved problem is the corded blind already hanging in a home, sometimes a rented one, where the only protection is a device that depends on an adult every day. A US evaluation for the regulator found that carers find these devices hard to fit and use and do not use them every time.
Why this matters
Health Canada's 2014 risk assessment found that the risk of dying from window-covering cords is "high for one-year-old children in Canada who live in a home where CWCs are used" (corded window coverings). Strangulation is fast and silent. Compression of the neck can cause unconsciousness within 15 seconds and death in under six minutes, and children have died with carers nearby. The impact statement cites US CPSC figures of about 11 strangulation deaths a year (1996–2012), with no significant change since. It cites the ACCC's figure of one to two child deaths a year in Australian homes, and 18 deaths in the UK from 2010 to early 2015 (RoSPA). Canada's hospital surveillance (CHIRPP) recorded 15 cord incidents from 2011 to 2016, some of which may duplicate reports Health Canada already had. Scaling from US sales data, Health Canada estimated about 11 million window-covering units sold in Canada each year, nearly 10 million of them with long accessible cords before the new rule.
What’s been tried and why it hasn’t worked
- Education and voluntary measures (1991–2009). Health Canada ran outreach and education, met industry, and in one round wrote to the industry asking it to retrofit Roman shades and roll-up blinds voluntarily. A follow-up market survey found 91% of Roman shades and 100% of roll-up products did not comply with the request. It concludes that education efforts "have not significantly helped reduce strangulation injuries and deaths".
- A national standard written into regulation (2009). The 2009 regulation incorporated the national standard. Since it came into force, Health Canada has known of 9 deaths in nine years, against 30 in the 20 years before (1.0 against 1.4 a year). Where the product could be identified, "in each case the CWC that killed the child was compliant". In one case from 2018, a 20-month-old was strangled by the cords of a blind wrapped high round the top of the blind. The maker's test reports showed it met the regulation. In a 2012 case, cords tucked high to keep them out of reach either fell or were reached by climbing.
- The consensus standards process. Health Canada took part in the US (ANSI/WCMA) and Canadian (CSA) committees in 1994, 1998 and 2010–2016. It concluded that "the consensus-based process of standards development for CWCs has not yielded a result that provides adequate protection to children". The 2018 US voluntary standard allowed long accessible cords on custom products (its later revision is described in item 5). Its inner-cord test allows a 125 cm loop under a 22.2 N pull, which a child can easily exert and which fits over a child's head.
- The 2021 regulation, and what it leaves. SOR/2019-97 (in force 1 May 2021; Health Canada spent the first year on awareness and stepped up compliance monitoring from 1 May 2022) requires any reachable cord to be no longer than 22 cm, or to form no loop over 44 cm under a 35 N pull, in every position of the blind. A cord that needs more than 35 N to pull out must retract after release. Health Canada notes that safe alternatives are common and affordable: cordless systems, wands and shrouds. A literature and patent review by a firm Health Canada contracted found at least 55 devices, methods or innovations that could make cords inaccessible. The impact statement is not consistent on how many are on sale: one passage says the 55 exist "with working functional products in the market", another that a review of one large online retailer's aftermarket products showed "several" of them do. But the rule governs products sold, not blinds already installed. With an average useful life of about 7 years, it expected risk to fall about 14% a year as households replace blinds, and it still expected 2 deaths over 20 years "due to yet-to-be-replaced CWCs". For those blinds, Health Canada's advice is to replace them with ones that have no long accessible cords, starting with children's rooms, or else to keep cords out of reach and use cleats and tension devices: the same consumer-fitted devices that the impact statement says have "repeatedly proven to be ineffective in preventing all fatalities". The impact statement keeps outreach only as "a complementary instrument" for products "already in people's homes". Blinds that break the new rule also kept being sold: after enforcement began, more than 350 products were recalled by autumn 2022 (Consumers Council of Canada), recalls continued into 2026, and some recalled blinds had been on sale for years. In Health Canada's 2023–24 compliance project, 43 of 44 products tested led to a recall, alert, stop-sale or other action, though Health Canada says that sample was targeted and not representative of the market. Recalls tell owners to stop using the product, which reaches only the owners who see the notice. No read source gives deaths after 2021.
- Devices fitted to blinds already in homes (US evidence). The Window Covering Safety Council, set up by US makers and retailers, ran a free retrofit-kit programme for consumers; its website now says the programme "has been suspended" and recommends "using only cordless products that comply with today's standards". Cord cleats, winders, clips and tension devices are sold as add-ons. In 2016 the US CPSC had Westat evaluate them: a review of products, hands-on testing by the research team, and ten focus groups and a pilot session in which 59 homeowners and renters (of 67 recruited) handled the devices. Its main findings: awareness of the hazard "does not translate to precautionary action"; only a fourth of participants had acted to keep cords from children; "difficulty using and installing safety devices" are primary reasons for not using them; and many did not recognise what the devices were for. The cleat was the best received, yet "among those participants who installed a cord cleat in their homes, most admitted that they do not use it all of the time". Older participants struggled to compress tensioner springs. Renters "may not be permitted to install" safety equipment and may not own a drill. Many devices looked flimsy and, participants feared, attractive to children. On the positive side, the research team found most aftermarket devices easy to install, with clear instructions, and Westat judged better point-of-sale information "partially helpful", but only "incremental"; it recommended cordless products, including cordless blinds fitted by managers of rental homes. The CPSC's 2022 final rule on custom window coverings found 25 deaths on continuous cord loops without a working tension device (no device, a device not fixed to the wall, or a broken one), and a cleat incident in which a four-year-old climbed a table and took the cord off the cleat. It concluded that external devices needing installation or use by the consumer are "ineffective to adequately reduce the risk", allowed only passive devices integrated with the product as sold, and noted that when homes are sold "or new renters move in, the existing window coverings, if they are functional, usually remain installed". Like the Canadian rule, it does not reach blinds already installed. The rule never took effect: the US Court of Appeals for the D.C. Circuit stayed it in January 2023 and on 12 September 2023 vacated it and sent it back to the CPSC, on the rulemaking process (notice and comment, a cost–benefit analysis that did not rely on the price of custom products, and the effective date). The industry's revised voluntary standard, ANSI/WCMA A100.1-2022, in effect for products made or imported from 1 June 2024, removes free-hanging operating cords from custom products; it too governs products, not blinds already installed. This check found no later CPSC rule, and no uptake figures for the free kits.
- Rules for blinds already installed elsewhere (Australia, UK). Australia's national installation standard (in force 2015) requires that a loose cord cannot form a loop of 220 mm or longer below 1,600 mm above the floor, using a cleat at least 1,600 mm up or a cord guide that holds a 70 N pull in any direction for 10 seconds; it covers blinds installed from 2015 onwards. Victoria closed that gap for rented homes: from 1 December 2025 every corded internal window covering in a residential rental must have an anchor installed to secure the cords and prevent loops, "with no exemptions" (Consumer Affairs Victoria's minimum-standards page adds that the rental minimum standards do not apply to an agreement that became periodic, month to month, before 29 March 2021), and it is an offence to let a new renter move into a property that does not comply. Renters may fit the anchors without the landlord's permission, and Consumer Affairs Victoria gives out a free kit of 10 plastic tension devices that are screwed into the window frame, "only suitable if you have roller blinds with looped cords and wooden window frames". No read source reports how many rentals comply or any effect on deaths. In the UK, since early 2014 it has been illegal to supply or professionally install an internal blind that does not meet the child-safety requirements of BS EN 13120:2009+A1:2014; retrofit safety devices have their own European standard, BS EN 16434:2014 (tensioning, breakaway, accumulation and non-tangling devices and inner-cord stops, "fitted ... at the time of manufacture or for retrofitting"), and a 2011 EU decision on what European standards for blinds and safety devices must require says such devices must be "resistant to the operation by young children". RoSPA (2019, 33 child deaths since 2001) found that 34% of privately rented homes surveyed had no safety mechanisms fitted on their blinds, and asked landlords to remove looped cords or fit cleats; this check found no UK rule requiring landlords to act on blinds already installed. Canada's own home-safety page (2025) still advises cleats and wall-fixed tension devices for blinds not yet replaced and says nothing about rentals.
What would unlock progress
A way to make an installed corded blind as safe as the 2021 rule requires of new ones, without depending on an adult to wind, tuck or clip the cord every time. It should be cheap enough to fit across a home or a rental building, and fittable by a tenant or a landlord's handyperson without replacing the blind. The 2021 rule already supplies a test: no reachable cord over 22 cm, and no loop over 44 cm at 35 N, in any position. The count of at least 55 patented mechanisms in Health Canada's impact statement, and its finding that cords and loops become hazardous in ways carers do not foresee (knotting, climbing, cords wrapped "out of reach"), show what a retrofit must survive. Health Canada's guidance for new products sets a further bar a retrofit should meet: a part that keeps a cord out of reach counts only if it cannot be removed by hand without a tool and does not dislodge or break under a 35 N pull. The US evaluation adds the conditions of use: carers who will not act every time, older users who cannot compress a spring, and renters who may not be allowed to drill.
How we’d know it’s being solved
The primary measure is child strangulation deaths and hospital cases from window-covering cords in Canada: Health Canada's count (1.0 a year under the 2009 rule) and CHIRPP cord incidents, with whether the product was installed before or after May 2021. For smaller tests, the measure is the 2021 rule's own: for each retrofitted blind, the longest reachable cord and the largest loop formed under a 35 N pull, in the fully open, fully closed and middle positions, after repeated use cycles and without anyone re-setting the device. A count of homes that say they own a cleat is a weak stand-in, since the impact statement and the US evaluation show cleats are owned and not used.
Where to start
Facility-free door (installed-cord audit): in student housing, a residence hall or volunteers' own rented homes (with permission), list every blind with a cord. Record its type, whether the operating and inner cords are reachable from the floor or from furniture near the window, and the reachable length and loop size. Measure the loop with a tape and a luggage scale hooked through it at 35 N, gauging the loop against a rigid template, never against a hand, wrist, neck or any other part of the body. Audit only rooms where no child is present. Also record whether a cleat or tension device is fitted and in use. Deliver the share of blinds that would fail the 2021 test, and which cord causes it. Only a tape measure and a spring scale are needed. The test is the product's own, so it reads the same in a Canadian home as in a US or European one; what differs is the mix of blinds installed, so record each blind's type and approximate age.
Make-and-test door (a retrofit that needs no daily action, bench rig only): cord cleats, cord stops, tension devices and cord winders are already sold, and so are cordless replacement blinds. A patent review for Health Canada counted at least 55 mechanisms. The ones fitted to installed blinds depend on someone using them, and the US evaluation found carers find them hard to fit and do not use them every time. Victoria's free tension-device kits (roller blinds with looped cords on wooden frames only, and sent only to Victorian mailing addresses) and the Window Covering Safety Council's suspended free kits are existing options to compare against, and the European standard for safety devices, including retrofit ones (BS EN 16434:2014, a paid standard), is the existing test regime to read first. So make a retrofit that keeps an installed corded blind within the 2021 limits on its own: for example, a shroud or auto-retracting housing that fits standard headrails. Cord shields, shrouds and winders are already sold or given away, and clip-on motors that drive a blind's existing chain are sold too (one maker's motor "clips onto your shade's beaded chain"; its page makes no child-safety claim and says nothing about inner cords), so the object must differ in all of these at once: it needs no action after fitting, it cannot be taken off by hand without a tool, it fits without drilling, and it holds the limits in every position of the blind, for the inner cords as well as the pull cord. Check what is sold first, and state in one line how it differs from the cleats, winders, shields and free kits on offer. Test it only on a bench rig: a blind mounted on a bench frame, a force gauge pulling a rigid test probe at 35 N, a loop gauge (a 44 cm circumference template) and a cycle counter, through hundreds of raise-and-lower cycles. Check that it does not dislodge or break under a 35 N pull, the bar Health Canada sets for add-on parts on new blinds, and also against the 70 N for 10 seconds that Australia's installation rule asks of a cord guide below 1,600 mm. Passing these limits on the bench is a safety test, not legal compliance: Health Canada says an add-on that is not fitted when the product reaches the buyer does not make it compliant, and the rule governs blinds sold, not blinds already installed, so the door makes no claim that a fitted blind becomes legal. Never use a person, a child or any neck form worn by a person. Then test fitting with adult volunteers reached through student housing (tenants of residence halls or shared off-campus houses, and housing maintenance staff), including some older adults, since the US evaluation found older users struggled with tensioner springs: time to fit, errors, whether it needs a drill (which renters may not be allowed to use), and whether the blind still works. The fitting barriers in the sources (no drill, no permission to drill, springs that are hard to compress) apply to renters in Canada as elsewhere, and comments to Health Canada recorded in the impact statement name "restrictive lease agreements for those who rent and are not allowed to replace CWCs that may be installed"; volunteers' lease terms are recorded, not assumed. No child at any stage, and no person puts a cord loop around any part of the body.
Design-the-test door (homes with young children): whether a retrofit stays fitted and working in homes with toddlers can only be learnt there. Design the follow-up study (recruitment through a public health service or a landlord, what to inspect at 1, 6 and 12 months) and do not run it. Any study involving homes with young children needs ethics review.
Relevant skills: mechanical and product design, test-rig building, human factors, housing and tenancy policy.
Genes
Related problems
Stuck the same way, in other fields.
field: digital & spacecheap rockets, crowded launchpadsLaunch site throughput and range operations bottleneckthe rules got in the waysource: secondaryfield: the built worldthree percent of the pipes, an outsized share of the leaksLegacy cast-iron gas pipes cause disproportionate leaks but in-place rehabilitation is unsolvedthe rules got in the waysource: primaryfield: the built worldthe bearing fails between detectorsRail wayside bearing detection and hazmat tank car crashworthinessthe rules got in the waysource: primaryfield: digital & spacerenting the nation's death fileThe United States has no national death registry — it rents one from 57 vital-records offices through a fee schedule nobody can cost, under a statute that decides who may lookthe rules got in the waysource: primarySources
- Government of Canada, "Corded Window Coverings Regulations: SOR/2019-97", Canada Gazette, Part II, Vol. 153, No. 9 (2019-05-01), Regulatory Impact Analysis Statement, accessed 2026-11-17. Link ↗
- Canada Gazette, Part I, Vol. 151, No. 24, "Corded Window Coverings Regulations" (proposed, 2017-06-17), accessed 2026-11-17. Link ↗
- US Consumer Product Safety Commission, "Safety Standard for Operating Cords on Custom Window Coverings", Federal Register, Vol. 87, No. 227, pp. 73144–73195 (2022-11-28), FR Doc. 2022-25041, accessed 2026-11-17. Link ↗
- Levi S, Benedick A, Lerner N, De Leonardis D, Huey R, "Effectiveness of Safety Devices in Reducing the Risk of Child's Access to Hazardous Cords and Loops: Final Report", Westat for the US CPSC (2016-07-12; with the Focus Group Final Report and a CPSC staff statement, October 2016), accessed 2026-11-17. Link ↗
- Consumers Council of Canada, "Health Canada: Corded Blinds Retrofits a No-No" (2023-03-10), accessed 2026-11-17. Link ↗
- Consumers Council of Canada, "Unsafe Window Blinds Still Found, Five Years After New Rules" (2026-05-05), accessed 2026-11-17. Link ↗
- Health Canada, "Notice to stakeholders for corded window coverings" (page dated 2021-05-07) (read in full through the Internet Archive copy of 2026-04-13), accessed 2026-11-17. Link ↗
- Health Canada, "Industry guidance related to the Corded Window Coverings Regulations: Safety requirements" (page dated 2025-09-17) (read in full through the Internet Archive copy of 2025-12-07), accessed 2026-11-17. Link ↗
- Health Canada, "Consumer Product Enforcement Summary Report - Compliance Verification Project 2023-2024: Corded Window Coverings Regulations" (page dated 2025-07-22) (read in full through the Internet Archive copy of 2026-01-22), accessed 2026-11-17. Link ↗
- RYSE, "Can Existing Blinds Be Motorized?" (maker's page), accessed 2026-11-17. Link ↗
- Window Covering Safety Council, "Retrofit Your Window Coverings", accessed 2026-11-17. Link ↗ Link 2 ↗
- Bureau Veritas Consumer Products Services, "U.S. CPSC Custom Window Covering Rule Vacated" (2023-09-16), accessed 2026-11-17. Link ↗
- Consumer Federation of America, "Custom Window Coverings Continue to Put Children at Risk" (2023-10-05), accessed 2026-11-17. Link ↗
- Consumer Affairs Victoria, "Are your curtain and blind cords safe?" (2025-08-22), accessed 2026-11-17. Link ↗
- Consumer Affairs Victoria, "Order a curtain and blind cord safety kit", accessed 2026-11-17. Link ↗
- Consumer Affairs Victoria, "Rental properties - minimum standards", accessed 2026-11-17. Link ↗
- ACCC Product Safety, "Blinds, curtains and window fittings mandatory standard", accessed 2026-11-17. Link ↗
- ACCC, "Corded internal window coverings: Installation guide" (November 2014), accessed 2026-11-17. Link ↗
- British Blind and Shutter Association, "2014 legal requirements for internal window blinds" (Make it Safe leaflet, 03/14), accessed 2026-11-17. Link ↗
- "BS EN 16434:2014 Internal blinds. Protection from strangulation hazards. Requirements and test methods for safety devices" (scope page; standard itself paid, not), accessed 2026-11-17. Link ↗
- "Commission Decision of 27 July 2011 on the safety requirements to be met by European standards to address certain risks posed to children by internal blinds, corded window coverings and safety devices pursuant to Directive 2001/95/EC of the European Parliament and of the Council" (2011/477/EU), annex, accessed 2026-11-17. Link ↗
- RoSPA, "Landlords should do more to eliminate blind cords dangers, says RoSPA" (2019-10-09, via WiredGov), accessed 2026-11-17. Link ↗
- Health Canada, "Window covering safety" (page dated 2025-09-17) (the site does not answer automated requests; Internet Archive copy of 2026-06-05), accessed 2026-11-17. Link ↗ Link 2 ↗
Revision notes
This brief gets checked against its sources and contains the following revisions. Future revisions get added here.
Worth knowing
- In Health Canada's 2023–24 check, 43 of 44 products tested led to a recall or other action, but Health Canada says that sample was deliberately biased and does not reflect the market as a whole.
- The finding that carers do not use add-on cord devices every time comes from US focus groups, not a controlled trial.
Full check
Sources read
- Canada Gazette Part II, impact statement and text of the Corded Window Coverings Regulations (2019): full text
- Canada Gazette Part I, the proposed regulations (2017): full text
- US CPSC, final rule on operating cords on custom window coverings (Federal Register, 2022): full text
- Westat, evaluation of cord safety devices for the US CPSC (2016), with its focus-group report and a CPSC staff statement: full text
- Consumers Council of Canada, corded blinds retrofits (2023): full text
- Consumers Council of Canada, unsafe window blinds still found (2026): full text
- Window Covering Safety Council, retrofit page: full text
- Bureau Veritas, US custom window covering rule vacated (2023): full text
- Consumer Federation of America, custom window coverings (2023): full text
- Consumer Affairs Victoria: blind-cord news page, safety-kit page and rental minimum standards: full text
- ACCC, blinds and window fittings mandatory standard page and installation guide (2014): full text
- British Blind and Shutter Association, 2014 legal requirements leaflet: full text
- BS EN 16434:2014, safety devices for internal blinds: record only (scope page; the standard is paid)
- Commission Decision 2011/477/EU, annex: full text
- RoSPA release on landlords and blind cords (2019): full text
- Health Canada: window covering safety page, industry guidance on safety requirements, 2023–2024 compliance project, and notice to stakeholders: full text (the live pages refuse direct download, so read in part at first; later read in full from archived copies)
What changed
- Stated more strongly than the source: that the blind is 'often a rented one'; no source says how many corded blinds are in rented homes, so the page now says sometimes.
- Numbers: the 15 cord incidents in Canada's hospital surveillance are now said to include some that may duplicate reports Health Canada already had; the count of at least 55 mechanisms is now said to come from a literature and patent review by a firm Health Canada contracted, and the page notes that the impact statement is not consistent on how many are on sale.
- Entry points: Victoria's free tension-device kits are now described as sent only to Victorian mailing addresses.
Left open
- the court's grounds for setting aside the US custom-blind rule [the opinion itself was not opened; taken from two reports of it]
Separate check of specific claims: A sentence that went beyond the sources about the blinds in homes was replaced with what the recall record shows, that blinds breaking the 2021 rule kept being sold after enforcement began; clip-on motors that drive a blind's chain were added to what the build starting point says is already sold; and citations were added to the source list.
Earlier full check: Seven details were corrected, including the size of the US focus groups (59 took part, of 67 recruited), the court's grounds for setting aside the US rule, the base of a UK figure on rented homes, what an EU decision requires, an exception to Victoria's rental rule, and the wording of Health Canada's advice for older blinds; a record of Canadian renters' lease limits was added.
Separate check of specific claims: A search for existing products and rules found that the US industry's free retrofit kits are suspended and that the US custom-blind rule was set aside by a court and never took effect, and added rules for installed blinds in Australia, Victoria and the UK, with matching comparisons in the build starting point.
Separate check of specific claims: Entry points: the build starting point must now differ from add-ons already sold or given away and says that passing on the bench is a safety test, not legal compliance; safety lines were added to the audit and bench tests, and the fitting test now includes older adults.
Earlier full check: A claim of similar death rates abroad was replaced with what the impact statement gives, and Health Canada's rule on add-on parts, recalls after enforcement began, its 2023–24 compliance check and a section on US evidence about add-on devices were added.