Industries / Industrial equipment and services / Industrial services / Rope access
Rescue
Rescue is getting a hurt or stuck technician off the ropes quickly, usually by the team on site, because a person hanging still in a harness can faint within minutes.
5
Milestones achieved, from 2002 to 2024.
0
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2002
Faints found only in training
A review for Britain's Health and Safety Executive (HSE) by Paul Seddon, "Harness suspension: review and evaluation of existing information", noted that a rescue plan was an essential part of fall protection arrangements. A request posted on IRATA's website for six months produced no reports of fainting at work; the only cases Seddon found happened in rescue training, when people were deliberately hung motionless.
Credit
Paul Seddon, HSE contract research report 451/2002
2009
Lay the casualty flat
Rescuers had long been told to keep a person rescued from a harness sitting up, not lying down. A review by the Health and Safety Laboratory and the University of Birmingham traced the advice to an unpublished paper at a 1972 conference of mountain rescue doctors in Innsbruck, repeated by later authors without evidence. It found no study testing the sitting posture and recommended standard first aid: lay an unconscious casualty flat and rescue anyone faint or unconscious as soon as safely possible. It cites a 1998 study in which hanging still and upright brought most healthy people close to fainting within an hour, and a fifth within 10 minutes.
LessonAdvice passed from manual to manual for 37 years had never been tested, and checking the original evidence reversed it.
Credit
A. Adisesh and colleagues (HSL) and C. Lee and K. Porter (University of Birmingham), HSE research report RR708
2019
Fainting timed in volunteers
At Eurac Research in Bolzano, Italy, 20 healthy volunteers hung in a sit harness for up to 60 minutes, with and without climbing first. Twelve of the 40 tests (30%) were stopped because the volunteer was close to fainting, after 44.7 minutes on average and 13.4 minutes at the earliest. Blood pooled in the legs, but a reflex through the vagus nerve, which slows the heart and drops blood pressure, brought on the collapse, so its timing could not be predicted, and everyone recovered within 5 minutes lying flat.
Credit
Simon Rauch and colleagues, Institute of Mountain Emergency Medicine, Eurac Research; European Journal of Applied Physiology
2023
A rescue without spare ropes
On 31 May 2023 a technician inspecting a seawater discharge pipe on Hess's Stampede platform in the Gulf of Mexico fell 90 ft (27 m), stopped 8 ft (2.4 m) above the water and hurt his leg. His supervisor had threaded the descender at the anchor wrongly and left three entry-level (Level 1) technicians unsupervised. The rescue plan named the supervisor as the only person to haul a casualty up, and no third set of ropes was on site for rescue. Another rope access team nearby rigged extra ropes and brought him back aboard. BSEE, the US offshore safety regulator, concluded that without that team "prompt rescue would not have taken place".
LessonA rescue plan needs its own ropes and enough people to use them, rigged before work starts.
Credit
BSEE investigation report, approved 8 August 2024
2024
Raised legs buy time
In 2023 the International Commission for Mountain Emergency Medicine (ICAR MedCom) reviewed 23 articles and recommended rescuing anyone hanging in a harness as soon as possible, even without symptoms, moving the legs while waiting, and laying the casualty flat once down. A University of Birmingham study published in March 2024 hung 24 people in climbing harnesses at an indoor wall for up to 45 minutes: those with their legs raised lasted 43.05 ± 6.7 minutes on average, those with legs dangling 33.35 ± 9.02 minutes.
Credit
Simon Rauch and colleagues for ICAR MedCom; Sophie Willihnganz, Zubair Ahmed and Justine Lee, Wilderness and Environmental Medicine
Open
01
Start the suspension registry
A public register of every reported case of a worker or climber left hanging in a harness, built from regulators' investigation reports and published case reports. Each case records how long the person hung, who rescued them and how, what symptoms they had and how it ended, with a link to its source.
The briefCredit · unclaimed02
Check rescue plans before work
A tool that reads a written rescue plan and method statement for a rope access job and answers a fixed list of about 40 questions drawn from investigations and rule texts: is a spare set of ropes rigged for rescue, how many trained people can haul or lower, is the rescue kit set up before work starts, is anyone working alone, are there two separately anchored ropes, is the area below closed off, is edge protection planned. Each answer points to the line of the plan that supports it, and each gap is flagged for a supervisor to fix.
The briefCredit · unclaimed
Open 01 · the brief
Start the suspension registry
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public register of every reported case of a worker or climber left hanging in a harness, built from regulators' investigation reports and published case reports. Each case records how long the person hung, who rescued them and how, what symptoms they had and how it ended, with a link to its source.
Why it is still open
ICAR MedCom wrote in 2023 that "an international registry on suspension syndrome is warranted" and that some case reports could not be analysed because their data were incomplete. Seddon's six-month appeal on IRATA's website in 2002 produced no reports. Web, GitHub and ClinicalTrials.gov searches in November 2026 found no such registry.
Built with what exists
public reports from regulators such as BSEE and WorkSafeBC and safety flashes from IMCA, the International Marine Contractors Association; case reports found through PubMed Central; a model that answers fixed-answer questions about each report, such as Google's Gemini Decision 1 (Note: "A thousand questions, one pass"); a shared spreadsheet or small database; two people to check the coding.
First step
Enter the case reports listed in the ICAR MedCom review and the Stampede case from BSEE's report.
Done looks like
At least 50 cases from at least five countries, each coded for time hanging, time to rescue and outcome; two people agree on the coding of at least 20; the register published with a form for adding new cases.
Open 02 · the brief
Check rescue plans before work
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A tool that reads a written rescue plan and method statement for a rope access job and answers a fixed list of about 40 questions drawn from investigations and rule texts: is a spare set of ropes rigged for rescue, how many trained people can haul or lower, is the rescue kit set up before work starts, is anyone working alone, are there two separately anchored ropes, is the area below closed off, is edge protection planned. Each answer points to the line of the plan that supports it, and each gap is flagged for a supervisor to fix.
Why it is still open
BSEE found that the Stampede plan named one person to haul and had no third set of ropes, and that the operator's first report wrongly said rescue gear had been set up. California's rope access rule requires a written code of safe practices that covers rescue, and ICAR MedCom recommends that rope work never be done alone. Software found in this search confirms that a current rescue plan is on file before a work permit is issued, as iFactory's permit software does, or drafts plans from templates; none was found that checks what a finished plan says and publishes how often it is right.
Built with what exists
a model that answers many fixed yes-or-no questions about one document in a single pass, such as Google's Gemini Decision 1 (Note: "A thousand questions, one pass"); questions drawn from the BSEE findings, the ICAR MedCom recommendations and the free rule texts (California section 3270.1, the 2001 EU directive, Britain's Work at Height Regulations, British Columbia's Part 34); sample plans from rope access firms and training providers, used with permission and with names removed.
First step
Turn the BSEE report into ten checkable questions and write five sample plans, each missing one item.
Done looks like
Sixty plans, 30 of them seeded with one known gap each; the tool flags at least 27 of the 30 gaps and wrongly flags no more than 3 of the 30 sound plans, scored against the review of two IRATA or SPRAT Level 3 technicians.