Wednesday, October 1, 2008
EMS Safety: Hit or Miss
There seems to be a sudden rise in EMS accidents. Sharing lessons learned may help prevent future crashes.
I’ve enjoyed producing occasional articles and stories for Rotor & Wing, our industry’s most respected helicopter-oriented periodical, during an extended span of years now.
Usually, I’ve described adventures by the U.S. Coast Guard and flying in the Caribbean or Alaska. But lately, during a helicopter operational period that seems to be increasingly identified as a conspicuously unfortunate one featuring a return to alarming accident rates in the civilian emergency medical sector of our profession, the ante has been raised a bit.
My objectives have now been shaped by those events, to become — with the very constructive encouragement of R&W — the production (or provocation) of a series of articles drawing from a cumulative assortment of experience and intended to convey what we think we might have learned about operating EMS helicopters safely up to this point in an apparently endlessly challenging history.
Having spent more than two decades investing most of my professional time in working with our own private EMS helicopter programs in rapidly growing Atlanta, it might be said that we have experienced a position more or less in the mainstream of a typical mix of challenges in the business. I hope each installment of this series of articles will address particular safety issues that have proven to be persistently problematic to EMS operations, but I don’t pretend quite enough absolute authority to attempt using a lecturer’s tone.
I unknowingly produced the first installment in February of this year, outlining a strategy for safely dealing with the ever-present, inadvertent IFR threat that has always been a constant risk factor for EMS. But I didn’t want to present only my own findings as purely conclusive. I meant to have the piece stimulate, across as wide a readership as possible, thought and discussion. We seem to have been measurably successful in our local programs here with regard to productivity and safety, as compared to industry norms, But we have learned that no one can ever know everything, and I want these presentations to be written as if from a friendly uncle, and to stimulate thought and response from all readers who might ever feel moved to share information, especially from fellow EMS operators. Responses of every ilk, from every quarter, are encouraged.
We learned long ago that accumulating knowledge and perceived expertise unavoidably positions anyone, eventually, in operational ruts. Ruts can be good, and even necessary to some extent, but by definition seeing out of ruts can be difficult. A fresh look from outside very often can come only from someone with a new perspective, so please prepare to read what we have to say always standing ready to contribute thoughts of your own, hopefully to the benefit of the collective good.
My next offering, after the inadvertent IFR story, was to have been a description of a productive way we have found to lay a philosophical groundwork on which aviation and medical experts can think of themselves as safely in bed together. That may sound provocatively patronizing. But aviation and medicine can easily be conceptualized as intrinsically at odds with each other while being compellingly attracted, and we think that finding a happy balance in this potentially "friendly fire" conflict is truly the secret to safe EMS helicopter operations. But I’ll explore this philosophical foundation in the next installment, since the appallingly tragic midair collision of two EMS helicopters in Flagstaff, Ariz. on June 29 has suddenly put helicopters on the center stage of current events once again, and timely topics should dictate changes in our presentation schedule.
There is no question that the Flagstaff midair has received wide public attention. Many medical helicopter operators have consequently been plunged into something of a low-grade worried grief, since — completely aside from the terrible loss of life incurred — we know that a black eye for anyone in helicopters is a black eye for everyone, and black eyes, rightly or wrongly, usually invoke unfortunate consequences.
We cannot know the precise cause of the accident yet, and we will not attempt to speculate here on that aspect in any context. But I would like to recount the details of a very close near-miss with which I am intimately familiar. It might seem to have quite a lot in common with the Flagstaff accident, with the additional modifier that the site of our near-collision was at a rooftop helideck above the 20th floor of a major downtown medical center. The damage and loss of life, had the collision actually occurred, would have dwarfed that of the Flagstaff accident.
To protect the innocent, let’s say that our close call happened in a typical urban-density skyline not too many years ago, and that it was generated by a series of factors setting up a true "perfect storm" potential disaster scenario. Our intent here will be to look at each of several compounding factors, and to see that corrective strategies for minimizing the effect of any one of them could have prevented the entire sequence from developing. I hope that we will also see that one particular strategy, which any program can readily utilize and standardize, ultimately saved the day.
Starting the chain of events, two company helicopters were progressing toward the same medical destination from opposite directions, during the same very late-night hour, with similar ETAs. The communications specialist monitoring the missions from the dispatch center, and providing flight following, was new on the job, and the importance of recognizing aircraft converging was probably not yet fully apparent to him.
To complicate things further, the computer-link radio system through which flight following was being accomplished was exhibiting a maddening assortment of intermittent problems, causing communication to be only marginally satisfactory. This created increased workload and frustration on board each aircraft. Additionally, aircraft No. 1 was tasked with negotiating controlled airspace until fairly late in its progress toward the destination, further compressing communications challenges. Once this aircraft was clear of controlled airspace, use was made of the standard area helicopter Unicom frequency, 123.025 VHF AM, in the form of a radio call providing location and intentions, and recognition lighting was confirmed burning brightly.
The second aircraft, meanwhile, was maneuvering for his own approach, wrestling with the same radio frustrations being experienced by aircraft No. 1. The first aircraft, upon reaching the destination from the west, overflew the rooftop helideck by 500 ft, visually surveying the landing zone, making another Unicom radio report, and setting up for a teardrop approach path to terminate in a final approach from the east. Though it was later recalled that a parcel of garbled Unicom radio traffic might have been indistinctly heard by both aircraft, both pilots were largely saturated with nursing their link radios, trying to stay current with flight following, and expediting delivery of their patients.
Aircraft No. 1 was now established on final, after having visually cleared itself through a normally descending turn. The pilot was beginning to narrow his scan to touchdown references, as the medical crew in back started their usual rhythm of intercom voice safety callouts. On close final, though, actually inside the structural boundaries of the medical center complex, and less than 100 ft above helideck touchdown, the flight nurse, on the right side of the aircraft, changed everything with the kind of unexpected verbal interruption to normal speech patterns that immediately causes the chest to tighten.
"ABORT, ABORT" came the unmistakably abnormal callout, not in panic, but in clear, distinctly exaggerated and very assertively presented enunciation.
The pilot arrested his descent with collective, widening his by-now beam-focused outside scan to his left (where building structural threats were known to be possible), expecting that perhaps a loose antenna guy wire might have been spotted or maybe a blowing garbage bag. Seeing nothing, his eyes redirected right. Nearly the entire right- side window was filled with bright airframe paint and flashing rotor blades. To his surprise and at least temporary relief, his aircraft was already set on an avoidance-profile climbout path, and he didn’t need to induce any additional high-stress maneuver heroics to avoid an actual collision. But looking down into spinning rotors as he overflew the other aircraft, he only had time to hope that aircraft No. 2 would not suddenly spot him, and make some kind of an unpredictable, panicked change to his own, very late stage approach path.
Aircraft No. 1 climbed back to recon altitude and began orbiting, trying to believe what had just happened. When aircraft No. 2 had shut down and offloaded its patient, No. 1 was finally able to land, getting through its own patient offloading duties as normally as possible. Its pilot stayed topside to visit with the pilot of No. 2. Pilot No. 1 remembers that the body language of pilot No. 2 told the whole story before words could even be exchanged. Mercifully, in many ways, pilot No. 2 had no idea what had just happened.
Needless to say, a colossal safety meeting was called even as dawn broke during the next few hours, at which the terrifying night’s events were reviewed and a long debriefing exercise began. After many hours and days of consideration and contemplation, several other follow-on meetings were spent isolating the following chronologically arranged observations and recommendations, which, interestingly, turn out to present themselves in what is probably a correctly ascending order of importance:
To begin, the long appreciated "broken habit pattern" threat should have been evident in the form of a marginal dispatch radio operator performance, made even more acute by radio technical difficulties. All parties concerned, starting with the pilots of any affected aircraft, should have seen red flags here, and should have taken whatever precautions necessary to make all essential communications as slow, deliberate and as redundant as necessary to minimize risk and maximize safety. Additionally, every scrap of intelligence comprising situational awareness should be recognized and considered by EMS pilots, allowing them to keep a good spatial/operational "big picture" of where they are and what they’re doing at all times. "Maximizing safety" through good situational-awareness habits is not always an easy thing to completely identify, but experienced operators usually know when they are doing it and, conversely, they usually know when they need to redouble their efforts by slowing down and confirming cautious proceedings, even where only communications procedures may seem substandard.
The importance of staying as visible as possible, and the liberal use of recognition lighting, simply cannot be overstated. It might be argued that saving the burn hours on a controllable searchlight might make sense in some high-traffic situations, where making one’s aircraft as visible as possible would seem to be desirable, but landing and taxi lights should light your aircraft up like Atlantic City whenever traffic conflicts could threaten, and the use of good strobes and "Chuck Light" options should always be elected when possible. High-visibility blade striping should also be considered, especially since television news helicopters frequently overfly scenes populated by EMS aircraft during daylight hours.
Substandard flight following radio communications notwithstanding, all aircraft, especially those engaged in operations within known busy airspace, and particularly in the vicinity of terminal medical destination areas, should have made it an absolute priority to exaggerate the conspicuous application of all aircraft separation radio procedures. Helicopter Unicom frequencies have been comprehensively shown as proven lifesavers many times over, and they conclusively demonstrate that good ATC can be accomplished with great effectiveness and reliability even outside controlled airspace, but only if all participants are conscientious and thorough with radio discipline, taking advantage of well-established self-separation techniques. This aspect of good radio use can be especially important to EMS aircraft in airspace known as busy only to them.
Good reconnaissance overflight habits can never be accomplished too deliberately or too carefully, even if home base recoveries, with completely "familiar" settings, are being accomplished. After all, it can never be known positively that a deck is clear in real time unless that surface is being visually checked in real time. It’s also always essential for a landing aircraft to confirm that no children’s tours of the facility are in progress, and, for that matter, that the landing zone site is not on fire, for example. Overflying the LZ with at least one pass can also be very useful on behalf of announcing to anyone who may happen to be on deck that another helicopter approach is imminent, further enhancing order and possibly even triggering a ground-initiated "wave-off" if required. It should also be noted that deliberately exaggerated approach profiles can greatly improve your apparent visibility to other aircraft. "Sneaky Pete" approaches can be great for avoiding gunfire, but they are terrible for "seeing and being seen" in the EMS environment.
The last midair collision avoidance strategy turns out to be one of the most important single techniques that can ever become part of a set of habit patterns for EMS helicopter crews. It was discovered, at least within the group I know best, as a way of confirming ground-factor security during use of emergency medical scene landing zones. We had noticed over time that visual scan patterns dramatically narrow during final approach, especially for the pilot, of course, as actual touchdown is closely approached. Interestingly, this presumably normal human tendency to increasingly focus on the landing patch as touchdown is anticipated seems to be a psychological compulsion for non-flying crewmembers also, even if seated in rear-cabin stations (as we learned one night during a scene landing as one of our pilots was luckily able to catch sight of a threatening incursion by a moving ambulance in his peripheral vision). Thereafter, we began briefing crews to develop a very deliberate discipline in always keeping their visual scan patterns wide, all the way out to 90 deg or better, even during short final approach to landing. Our intent was to maximize our chances of staying informed with regard to potentially changing circumstances on the ground at emergency landing zones. We never dreamed that one day our very lives would be literally saved by a dutifully conscientious flight nurse who would spot a conflicting helicopter where it would never be expected, so close to touchdown! Interestingly, this particular flight nurse, though a clinical master, had never enjoyed a reputation within her program that placed her in a "best aviation performer" group. But her status in our profession changed forever that night, as did our appreciation for how important detail in crew briefings can be.
These days many technologies are being developed at increasingly accelerated rates, and some within aviation promise to minimize the always present midair collision threat, which, as a function of simple math, increases as aircraft populations grow. It is inarguable that those populations are literally exploding within some EMS helicopter communities.
Traffic-alert/collision-avoidance equipment, it has already been suggested, might have prevented the Flagstaff collision. A tool that could show the presence of a conflicting aircraft would have been invaluable. But it must always remembered that tools are only tools, and they are only as good — ultimately — as the human using them.
There may have been a factor in Flagstaff that will eventually show up as something no one has ever considered, but it is statistically more likely that some of the risk elements that were part of the scenario painted above were also part of the Arizona accident. Any one of them, if addressed even partially, could have prevented disaster. It is also likely that if carefully observed details taken from lessons learned in the past are incorporated within programs as improvements to operating discipline, the chances of experiencing another Flagstaff can be minimized.