Why Suspension Trauma Makes Rescue a Timed Event

Why this matters

An arrested fall is not a survived fall. It is a hazard that has changed shape. The worker is alive, the equipment did its job, and a second clock has started that nothing in the harness slows down. This card is about that clock: what starts it, what it depends on, and why the only number that matters is the one your own crew produced with a stopwatch rather than the one written in the plan. The rescue procedure itself belongs to a sibling article; what this one owns is the timing, and the reason a fall-arrest system with no measured rescue time is an incomplete system rather than a complete one with a gap.

What the harness does to a body that stops moving

Standing up, your leg muscles work as a pump. Every time you shift weight they squeeze the deep veins and push blood back toward the heart. Hanging motionless in a harness removes that pump entirely and adds a second problem on top of it: the leg straps bear on the upper thigh, and under the worker's full weight they compress the soft tissue and the veins running through it.

Blood pools in the legs. Venous return falls, so the volume arriving at the heart falls, so cardiac output falls, so pressure at the brain falls. The worker feels light-headed, then nauseated, then greys out. The condition has a name in the literature, orthostatic intolerance, and OSHA published a Safety and Health Information Bulletin on it (SHIB 03-24-2004, updated in 2011) specifically because fall-arrest programs were treating the arrest as the end of the event.

The cruel part is the feedback loop. A conscious worker who can push against the structure with their legs, or stand in a set of suspension relief steps, restores some of the pump and buys time. A worker who has been knocked out by the fall, or by striking the structure during it, cannot do any of that, and their onset is faster than a conscious worker's for the same harness and the same suspension.

OSHA's bulletin puts symptom onset in minutes rather than tens of minutes, and it deliberately does not publish one safe interval, because onset depends on the person, the posture, the harness fit, the temperature and whether the worker is conscious. Treat that absence as information: there is no number you are allowed to plan up to, only a measured rescue time you are trying to drive down.

What this card deliberately leaves out, and why

Three things, on purpose.

No treatment protocol. What is done to the worker after they are off suspension is a medical decision. The older field advice to hold a rescued worker upright for a set period is contested, and OSHA's own updated bulletin reflects that debate rather than settling it. Your field action is to call emergency medical services early, tell them explicitly that the worker was suspended in a harness and for how long, and follow their direction. Do not carry a positioning rule from a training deck into a real event.

No rescue procedure. Sequence, authority and the specific ways of getting a hanging worker to a surface are covered in the related SOP. Splitting them is not tidiness: the clock is a planning fact you use before the job, and the procedure is what you execute during it.

No harness inspection content. The library's fall-protection overview owns donning, fit and inspection. Fit matters here only in one direction worth naming: a leg strap riding high and tight on the thigh loads soft tissue harder than one bearing lower and wider, so a badly fitted harness shortens the clock in the same event.

What shortens the clock and what lengthens it

Read this as a list of what to change, not a scoring system. None of these produce a number you can add to a plan.

• Unconsciousness shortens it, sharply. No muscle pump, no relief-step deployment, no ability to shift. Any fall where the worker can strike structure on the way down should be planned as if the worker will arrive unconscious.
• A vertical, legs-dangling posture is the worst case. A worker who can get a foot on structure, or who lands seated in a positioning device, is not in the same condition.
• Suspension relief steps lengthen it, conditionally. These are the small stirrup straps stowed on the harness that a worker deploys and stands in to take load off the thighs and restore the leg pump. They require a conscious worker with a free hand. They are worth having on every harness and they are worth zero minutes in your written plan, because the plan has to survive the unconscious case.
• Cold lengthens nothing and shortens some things. Exposure runs in parallel with the suspension problem and adds its own timeline in winter work.
• Pre-existing cardiovascular conditions matter and you do not know your crew's. This is a reason to shorten the plan, not a reason to ask.

The only number that counts is the one you measured

Rescue plans fail on the wrong quantity. A plan that says "call the fire department, average response seven minutes" has recorded an arrival figure and used it as an off-suspension figure. Those are different quantities by a wide margin, and the difference is where people die.

Three properties of that seven-minute figure need stating before it enters any arithmetic. Its basis is dispatch-to-curb, not call-to-worker. Its character is a central value with a spread around it, not an upper bound, so half of all responses are slower than it by construction. And it covers one segment of a chain that also contains noticing the fall, placing the call, getting responders through a gate and up to a roof, and then the rescue itself. An average of one segment cannot be used as a bound on the whole chain.

The quantity your plan needs is time from arrest to off suspension, measured end to end, on your own site, by your own people.

Worked example: a shop puts a stopwatch on its own plan

A mechanical contractor runs rooftop unit change-outs on commercial low-slope roofs. Two techs, a personal fall arrest system anchored to a permanent roof anchor, and a written plan naming the local fire department as the rescue resource. They ran a drill with a weighted manikin.

Each qualifier from the section above gets its own line, so the corrections are visible rather than assumed.

Segment                                          Clock    Basis / character
Arrest (clock starts here, not at discovery)     0:00     event
Partner notices (he was told a drill was on)     2:10     optimistic, drill artifact
Voice contact attempted, no response             2:40     measured
911 placed from the roof                         3:20     measured
Foreman reaches the roof                         4:00     measured
Self-rescue ruled out (worker unresponsive)      5:30     decision point
Relief steps credit none requires consciousness
First engine at the curb (3:20 + 7:00 avg)      10:20     average, not a bound
Engine crew on the roof (gate, stairs, hatch)   13:00     measured on this site
Technical rescue unit, worker off suspension    26:00     measured on this site

Read the last two lines against the plan the shop had written. The plan's assumed figure was 7 minutes. The measured time to off suspension was 26 minutes. The assumed figure covered 7 of 26 minutes, about 27% of the actual exposure, and it was the only number in the document.

Two of those lines deserve their correction stated out loud rather than implied. The 2:10 notice time is optimistic because the partner knew a drill was running; on a real call he is inside the unit with a panel off and the clock runs longer, so the shop recorded it as a floor and not as a value. And the 10:20 arrival is built on an average, so the shop wrote it into the record as "10:20 or later, half the time later" rather than as a plus-or-minus figure. An average used as a bound is the single most common defect in a rescue plan.

What they changed. They pre-rigged a lowering system that lives in the same bag as the anchorage kit, so it is on the roof whenever the anchorage is in use, and they trained two people on it. Re-drill, same site, same manikin:

Arrest                                           0:00
Partner notices                                  2:10   (unchanged, same optimism)
911 placed                                       3:20   (unchanged, still placed early)
Foreman on the roof with the rescue kit          4:00
Worker lowered to the deck, off suspension       7:40

Time to off suspension fell from 26:00 to 7:40, which is under a third of the original. The 911 call did not move and should not: emergency medical services are still needed for a worker who has been suspended, and calling them at 3:20 while rescuing at 7:40 is the correct sequence, not a redundancy.

What they refused to claim. The 7:40 came from a manikin in known conditions with two trained people already on site. A real event with an unresponsive adult adds handling time. The shop recorded 7:40 as a floor, wrote "assume the real figure is larger and re-drill quarterly" beside it, and did not treat the improvement as headroom to accept a longer notice time. That refusal is the whole discipline of this card: every optimistic term in a rescue chain compounds in the same direction, and the direction is against the worker.

How to verify you got this right

Do not audit the plan document. Audit the chain.

Pick a live job with an anchorage in use and ask the crew one question: from where you are standing right now, who gets this person off suspension, with what equipment, and where is it. If the answer names a phone number rather than a person and a piece of hardware that is currently within reach, the plan has an arrival time in it and not a rescue time.

Then time it once, with a weighted manikin at the real anchorage on a real roof, not in a training bay. Record the segments separately, because a single end-to-end number hides which segment is worth fixing. In the example above, the segment worth fixing was not the fire department, which performed as expected; it was the 13 minutes between the engine crew reaching the roof at 13:00 and the technical rescue unit completing the lower at 26:00.

Finally, write the measured figure into the plan in place of the assumed one, with its basis in the same line. "Time from arrest to off suspension, measured 7:40 with a manikin, treat as a floor" is a usable sentence. "Fire department, seven minutes" is not, and it will read as adequate to everyone who signs it.

References

• 29 CFR 1926.502(d), personal fall arrest systems for construction, which requires the employer to provide for prompt rescue of a fallen employee or to assure that employees are able to rescue themselves; the general-industry counterpart is 29 CFR 1910.140.
• OSHA Safety and Health Information Bulletin SHIB 03-24-2004, updated 2011, Suspension Trauma / Orthostatic Intolerance.
• ANSI/ASSP Z359.4, safety requirements for assisted-rescue and self-rescue systems, in the edition your employer's fall protection program or your contract adopts; it binds through that adoption, not on its own.
• See related: universal-getting-a-suspended-worker-down for the rescue procedure and decision authority, and universal-what-fall-restraint-is-and-why-it-is-the-better-answer for the system that removes this clock entirely.
• See related: universal-fall-protection-systems for harness fit, inspection and system components.