SafetyInsights.org

How does attributions of ‘human error’ create hidden hazards?

Interesting article discussing a failure type in intraocular surgery using a cannula and how it’s effectively become hidden.

PS. Check out my YouTube and recent Critical Control / barrier vid: https://youtu.be/Ly1mEn_KCLQ?si=yG3yRVOM0r2eyyKI

Shout a coffee: https://buymeacoffee.com/benhutchinson

Extracts:

·     “The way we look for harm shapes what we are able to see”

·     “Medical-device safety surveillance is organized around two failure types”, being failed devices, and erring clinicians

·     But a third type exists: “harm arising at the interface between two individually compliant devices used in combination”

·     They describe a surgical procedure using a syringe to deliver fluids into the eye, but can inadvertently cause injury

·     Survey of 551 ophthalmic surgeons “found that 84% had experienced this event and 78% had witnessed resulting patient harm” and despite its prevalence “it is almost absent from the systems that are supposed to register it”

·     “Harm arising at the interface between two individually compliant medical devices falls outside vigilance taxonomies built around single-device malfunction, and is therefore systematically under- recorded”

·     Why is this failure mode invisible? First, nothing appears to have malfunctioned. The syringe’s lock and cannula thread are intact & each device met its specs. The manufacturer finds no faults in the components

·     But the issue is no the individual components “but in their combination” and “Vigilance taxonomies, organized around the malfunctioning device, have no place to put a failure that belongs to no single device”

·     The 2nd issue is a “human-factors problem rather than a regulatory one”, as when the cannula detaches “the operating team has a ready explanation: it cannot have been tightened properly. The event is understood, in the moment and in the record, as a slip by the person who assembled the syringe”

·     The human error explanation “is powerful because it is plausible, because it requires no further inquiry, and because no one in the room has an incentive to look past it”

·    “The clinician absorbs the blame; the system records, if anything, a technique error; and the contribution of the device interface to the harm is never written down”

·    “This pattern is familiar to anyone working in safety science. The reflex to locate harm in the last person to touch the system, rather than in the conditions that made the error likely”

·    This misattribution isn’t “merely a framing choice after the event; it determines whether the event is reported at all”

·    Passive reporting systems can’t “capture what the people closest to it have already filed under their own fallibility. A blame-oriented culture does not just treat clinicians unfairly. It deletes the data”

·    And because nothing is reported, no epidemiology accumulates, and because of this, the “the problem appears smaller than it is”

·        The error misattribution “that suppresses the data also harms the clinicians involved”, e.g. second victims

·        The author argues that treating these interface failures as a safety threat requires at least three changes: 1) reporting frameworks need a way to describe this type of harm in combination, 2) the “the investigative reflex must change”, e.g. “the question should be whether the system permitted, even invited, that failure, not simply whether the individual was careful enough”

·        And 3, more proactive use of trend analysis rather than waiting for passive reporting

·        “At present, interface failure is invisible because it is recorded as something else – a surgical complication, or a lapse by the person who assembled the device”

#safety #patientsafety #systemsthinking #healthcare