Sky Anxiety

     Over the last 12 months, the fatal EMS helicopter crashes have come in rapid succession. First were Whittier, AK, and Cherokee, AL, last December, respectively killing four and three. Then South Padre Island, TX, in February and La Crosse, WI, in May, three each. Huntsville, TX, in June, four more. Two collided in Flagstaff, AZ, that same month, claiming seven, then another fell in Burney, IN, in August, taking another three. Finally, as this story neared completion in late September, a Maryland State Police medevac crashed in suburban Washington, DC, killing four.

     That's eight accidents and, pilots and patients and crew, a total of 31 dead in the last year. It could have been worse—another half-dozen events saw everyone emerge alive.

     It's easy, if natural, to ask what equipment or practices might help reduce the spiraling yardsticks of air-medical crashes and fatalities. It's harder to look at the proliferating use of helicopter EMS (HEMS) resources and try to decipher the optimum balance of benefit, risk and safety, and the most likely path to achieving it.

     But as all the metrics climb—aircraft, missions, flight hours, crashes and deaths—it becomes increasingly worth doing.

     "At this year's National EMS Memorial in Roanoke, half the people inducted were air medical people," notes Texas emergency physician Bryan Bledsoe, DO, FACEP, perhaps the most prominent critic of the accelerating use of HEMS resources (not all were killed in 2007). "The flight paramedics and flight nurses, they're the victims in this. And those are smart, dedicated people who just want to help people."

     Certainly, EMS can ill-afford to lose those.

     As recently as 2002, researchers with the University of Chicago Aeromedical Network calculated the number of medical helicopters in the U.S. at around 400. Today, that's doubled.

     Some reasons for that growth are clear. The population's greying. Rural hospitals and EDs have closed. Top trauma resources and specialty care are consolidated. The importance of timely interventions in situations like stroke and cardiac arrest is ever more widely reflected in protocols and practice. Even in urban areas, traffic congestion and overstretched ground resources may sometimes make flight a faster option. Concomitantly, total EMS flight hours have risen steadily in the U.S., from fewer than 200,000 in 1998 to more than 400,000 in 2006.

     In terms of relative safety, at least until this year, things have been improving. Reviewing a decade's worth of accidents from 1992–2001, UCAN found an average of 3.8 accidents per 100,000 hours flown. Through June, the rate for 2008 was…3.8 accidents per 100,000 hours flown. However, that rate represents a major increase from 2007, and a reversal of four straight years of decline.

     So what's happening this year?

     "What's causing the accidents is probably very simply poor decision-making and a lack of situational awareness while flying at night in marginal weather conditions and people hitting something they didn't see," says Ed MacDonald, lead pilot for PHI Air Medical in Santa Fe, NM, and chair of the National EMS Pilots Association's Safety Committee. "The question we always ask is, why were they out there? Why did they keep going? Why did they keep pushing?"

     The obvious answer is that there was a patient in need. But the longer answer also involves things like indications and procedures for air-medical use, financial realities, restrictions on oversight and the overall design of communities' emergency response networks.

     "Right now, we don't have a thoughtfully designed system," says Tom Judge, executive director of LifeFlight of Maine and past president of the Association of Air Medical Services (AAMS). "It's not unlike many other things in medicine. It's not well-integrated into EMS. There needs to be more medical oversight and oversight of use, and making sure people are making the right decisions about risk, benefit and cost. It's a lot of money—it's 10 times the cost to be put in a helicopter as in a ground ambulance. So we really want to make sure that when we do that, there's a benefit attached to that cost and that risk. Right now that's not well done."

     There's evidence to support that contention. In a 2003 article for this magazine, Bledsoe cited numerous studies finding high rates of overtriage, and flown patients who didn't benefit from flying, whose injuries turned out to be minor or who were discharged directly from the ED (see that article and its references at www.emsresponder.com). More recently, in a 2006 Journal of Trauma article reviewing 22 previous helicopter trauma transport studies, his team found that 69% of flown patients had nonlife-threatening injuries.

     "You always want to err on the side of caution, but we're finding rates that are ridiculously high," Bledsoe says. "In our study, two out of three, by all objective criteria, had minor injuries. These people get an $8,000 helicopter ride, then they're not even admitted to the hospital. Is our prehospital assessment so poor that we can't even begin to judge who might even need hospital admission?"

     Casting a wide net is important, of course; you wouldn't want to miss any patient air resources might help. Triage isn't always a precise art, and sometimes critical-care ground resources aren't available. So it's natural that some patients are flown who might have been fine by ground. The questions are, what level of overtriage is acceptable, and what kinds of decision-making processes are leading to people being flown?

     "You always try to get it right, but given the circumstances, you don't always know," says Ed Eroe, CEO of Airlift Northwest in Seattle and also a past president of AAMS. "A fall from 20 feet is usually a criteria [to be flown]. Well, sometimes you come upon a patient who fell from 20 feet, and they're perfectly fine. That's why it has to be as protocol-driven as possible."

     This challenge isn't an unfamiliar one in medicine: making a resource quickly and reliably available to those who can truly benefit from it, while not utilizing it needlessly for those who can't. But with the recent growth of air medical services outstripping the ability of care systems to systematically integrate them and the ability of researchers to examine and evaluate all this increased use, that's not happening. Instead, perhaps predictably, these new big-gun resources are being enthusiastically used.

     "What I've started hearing from our clinical team members in recent years is, 'Why are we even flying this patient?'" says MacDonald, who also cochairs AAMS' Safety Committee. "We seem to be flying more and more patients who might be better handled by ground or even fixed wing. My concern is that I see a lot of programs not doing what we did in the early days: We'd go back to a hospital or EMS provider and say, 'You know, guys, that may not have been an appropriate patient to fly.' There was some sort of follow-up and education process for sending hospitals and prehospital personnel. Too many programs now are afraid they'll never be called again if they constructively criticize their customers."

     That's the concept of post-flight review: Evaluating a transport afterward to determine if it was truly warranted. This is how dispatch guidelines can be refined, the evidence body for HEMS benefit grown, and risk reduced. It's an important component of smart air-medical systems.

     "Most programs don't want to be called just to be called; you always try to do utilization review back," says Eroe, who also consults for Fitch & Associates. "As part of accreditation, you need to have a utilization and review report where you look at the transports. And that might mean going back to the referring agency and saying 'Hey, this probably could have gone by ground.'"

     Beyond those sorts of back-end controls, there are also guidelines available on the front end to help systems determine when air resources should be utilized. The National Association of EMS Physicians offers a set, endorsed by the Air Medical Physicians Association. Others exist. Controlling activation is an obvious way to control crash risk. "Wherever I've been, we've actively tried to get those out to people," says Eroe. But those occasional stubborn high overtriage and discharge rates make you wonder about what they encompass and if and how they're used.

     "When you see very high overtriage numbers—and there's places where it's 40%, 50%, 60%—you have to wonder if a system is really using guidelines," says Judge. "How are these decisions being made, and who is looking at this in retrospect? Is there effective medical oversight? Because we need to look at this stuff in retrospect: Was this the right thing to do for this patient?"

     That always should be the bottom line. But sometimes, other pressures can be hard to filter out.

     Offering air resources isn't cheap. Aircraft cost millions to obtain and maintain and use. Public and hospital-based programs may have such services financially supported. But much of the recent growth has been among private, for-profit community-based services, which typically must sustain themselves.

     The benefit of these services is that they can bring air-medical availability to rural and underserved locations. The caveat: To survive, they have to do a certain amount of business.

     "The incentives are to fly people; otherwise you don't get paid," says Judge. "You're not paid for the readiness costs of having the service available, so you have to get as many people as you can in the helicopter."

     With this reality comes the challenge of separating business pressures from flight decisions. Overt disregard for crew safety is obviously rare, but go-or-not decisions are seldom black and white. How bad is that weather cell? How rough is that terrain? Darkness and poor visibility are obvious dangers, but what happens if you hold back? Could it harm the patient? What if your service fails, leaving your community without? Decide quickly, someone's waiting.

     Plus, if you turn down a request, there's often another service that will take it. "Helicopter shopping" has been so problematic that the the main accreditation body for air-medical providers, CAMTS (the Commission on Accreditation of Medical Transport Systems), offers a video discussing its hazards.

     In short, competitive pressures can exist—that's the reality. Most people handle them responsibly. Occasionally, some don't.

     "At NEMSPA, we've received a couple of complaints in the last several years from pilots saying that somebody—a program director, fellow crew member, business manager—intimated that they need to have more flights to survive, or to do better," says MacDonald. "I'd call those managers unenlightened at best. I don't think there's a lot of them. But they could potentially cause a pilot or a crew to take a marginal flight in weather conditions that they otherwise would have had the sense not to take."

     Other competitively based behaviors also appear to work against safety. The Fort Worth Star-Telegram last year reported that after a local air service hired as its medical director a physician who also served that function for several local ground services, those services regularly called for that company's helicopters at scenes, even when others were closer. Bledsoe cites companies in his state wooing fire and law enforcement providers with minimal medical training who can call them independently of EMS. "They bypass the paramedics and go to the Texas Highway Patrol or local VFD, give 'em a landing zone course and a ride in a helicopter, and then they're beholden," Bledsoe says. "No EMS crew is going to come in and tell a Texas Ranger to turn a helicopter around. It just borders on unethical."

     That's hopefully uncommon, and individual instances of legal, if medically questionable, behavior shouldn't discredit a whole industry. No one wants to harm patients or providers. But in a for-profit world, survival requires money, money requires transports, and getting transports requires a business to approach its business like a business.

     "It comes back to the issue of the marketplace, no matter what we say as administrators," says Judge. "There's not an operator in this country who says, 'We're going to do this slipshod, we're going to cut corners.' But the bottom line is, in many places there are lots of aircraft, and that becomes incredible pressure. If you have lots of aircraft trying to do the same thing in the same place, and none of them are being reimbursed adequately, you build yourself into a risk. It comes back to the design of incentives and disincentives."

     The legislative backdrop to these issues makes addressing them tricky. While medicine is regulated at the state level, aviation is a federal jurisdiction per the 1978 Airline Deregulation Act. Civilian operators operate under parts 91 and 135 of the Federal Aviation Regulations. Part 91, which provides general flight rules, governs flights when no patients are on board; Part 135, applicable during transports, sets additional requirements for on-demand operations. The ability of states to regulate HEMS operations is therefore restricted; they are federally pre-empted in all areas of price, route and service.

     This dichotomy can hamstring states wishing to exercise greater control. Texas and Tennessee recently failed in efforts to tighten their rules. Overall, degrees of oversight vary widely; five states don't even license their air ambulances. Nine states and counties in two others, though, currently require their air-medical operators to be accredited, signifying they've met certain standards of patient care and safety. A court decision from Colorado this year upheld a requirement of CAMTS accreditation for operators there.

     But while some CAMTS standards reflect the FARs fairly closely, others may pose conflicts.

     "There are a lot of grey areas," says CAMTS executive director Eileen Frazer. "We get into areas that aren't covered in the regulations, but could be judged as superceding the regulations because they're not there, or are partially there, but not to the detail we have."

     Potentially clearing the playing field is an anticipated bill from Sen. Elizabeth Dole (R-NC) that would amend the ADA to grant states greater regulatory latitude for air-med operations. Other members of Congress are also looking at crash issues and touting equipment and procedural upgrades, fielding bills requiring things like terrain awareness warning systems (TAWS), standardized risk evaluation and flight dispatch procedures, and flight data and cockpit voice recorders.

     The FAA, which earlier this year pledged its full attention to the matter, has at its disposal various recommendations from the National Transportation Safety Board (also planning hearings). In a 2006 report finding that 29 of 55 crashes it studied could have been prevented, the NTSB identified four primary safety issues: less-stringent requirements for flights without patients on board (33 of the 55 occurred under Part 91 conditions); a lack of EMS flight risk evaluation programs (most operators studied had none); a lack of consistent, comprehensive dispatch procedures (which could have prevented 11 of the 55 crashes); and no requirements for safety technology such as TAWS and night-vision goggles (which could have prevented 17 and 13 respectively).

     In response, the FAA issued a number of notices and advisories, including encouraging risk-assessment procedures and use of technology like TAWS, NVGs and radar altimeters (see its Regulatory and Guidance Library at https://rgl.faa.gov), but made nothing mandatory. Now many are wondering if that should change.

     "There has to be some teeth to it," says Bledsoe. "Not all services are equal. Just because there's a paramedic and a nurse in the helicopter doesn't mean the helicopter's well equipped or the personnel are well trained. It can be a real nebulous area."

     What this all points to, at its core, is a need for greater integration. For air-medical resources, in whatever numbers they exist, to optimally enhance the emergency care provided to a community and minimize the risks of flight to EMS providers, they must be part of an organized system of care in which they're known quantities, called in defined ways and for defined reasons, for patients whom they can most clearly benefit. As personnel must be controlled and accounted for on a scene, so assets must be controlled and accounted for in a system. The elements of care work best in concert.

     Improving integration was the subject of a 2007 report produced by AAMS, the National Association of EMS Physicians, and the National Association of EMS Officials' Air Medical Task Force. Cowritten by Judge and published in Prehospital Emergency Care, it laid out principles for state regulation of the medical aspects of air-medical services and attempted to clarify what areas states might look to cover. It assessed things like criteria for use, coordination with 9-1-1 systems, who's authorized to request, coordination with scene units and hospital staff, destination decisions and more.

     Bringing these resources under a centralized control, with clear, standardized procedures for use, could blunt some of the perils that accompany having multiple competing providers in a community. The theme has been echoed in documents like the Rural and Frontier EMS Agenda for the Future and the IOM's Crossroads report, and by experts like the Foundation for Air Medical Research and Evaluation.

     "It has to be handled as part of the EMS system, and the closest helicopter that's capable should be dispatched," says Bledsoe. "First we need to assess the capabilities and limitations of these aircraft, and where they are geographically. Then, just like you would with an ambulance, you call the closest helicopter. And we have to revise the criteria for using them—they need to be set by the surgeons.

     "Basically, we all need to sit down, look at the literature, and really determine what the benefits and risks are, and balance them. And then we need 100% auditing of every flight."

     Flying patients, Judge notes, is a therapy in and of itself. As with any medical therapy, it must be constantly evaluated for effectiveness.

     "We have to look at it as we would anything in the hospital: Does this patient really need this?" he says. "We're not doing that, and that's where we see these continued questions about use. And frankly, the questions should be asked. I think there's very good science to say there's a group of patients who need access to medical helicopters. But they aren't free, and all of us have to be responsible about how we target therapies. We have to be able to discriminate. And I think we can; the tools are there to do it. But it's not being done universally, and that's a big problem."

     Other measures that could improve the safety of the air-medical environment:

• AMRM—The FAA laid out guidelines for training in Air Medical Resource Management in 2005. It's a system to teach effective team-building and communication skills among members of flight crews. One of its key concepts is that any member of the operation can express thoughts or doubts, seek more information and appropriately question decisions. In some cases, one team member who feels unsafe can stop an entire operation.
• H-TAWS—Once limited to fixed-wing craft, TAWS has progressed to where it's more feasible now for helicopters (H-TAWS). "There were false positives on the early ones because you flew so close to the ground," says Eroe. "The technology has improved now, and I think you'll see people putting this in place."
• Night-vision goggles—These are recommended but not required. Not only are they expensive, but the FAA must individually approve each aircraft using them. It's short of inspectors to do so. "They just don't have the people to get it put in place," says Judge. "There's no question they're a huge safety enhancement."
• Better localized weather reporting—EMS helicopters typically travel locally and at lower altitudes than fixed-wing craft. That means they're more susceptible to small, localized cells of weather. "Beyond the airports, we don't have much in the way of AWOS (Automated Weather Observing Systems) or weather-reporting stations," says Eroe. "The biggest improvement I think we could make is that we don't have good weather reporting everywhere we go. It would improve safety dramatically if we had access to that."

     In some areas, services that turn down a flight contact other services nearby and tell them why. AirMed International has created a Web hub for this purpose at www.weatherturndown.com.

     ADAMS Atlas and Database of Air Medical Services—www.adamsairmed.org
A comprehensive mapping and directory of air providers by state.

     Air Medical Safety Advisory Council—www.amsac.org
An operator-driven forum for sharing safety information and initiatives.

     FAA Regulatory and Guidance Library—https://rgl.faa.gov
A compendium of federal regulatory, certification and safety information documents.

     FlightWeb—www.flightweb.com
News, information and resources for air-medical professionals.

     HEMS Low-Altitude Flight Tool—www.weather.aero/hems
A resource for calculating ceiling and visibility for medical flights.

     International Helicopter Safety Team—www.ihst.org
An international coalition of organizations working for increased safety.

     Vision Zero—https://visionzero.aams.org
A safety program aimed at reducing the air-med accident rate to zero.

     HEMS accidents and deaths this decade: