“A former pilot stated in an interview that he had received text messages on “more than one occasion” from current company pilots and medical crewmembers stating they were “scared to fly.” One nurse reported she believed the pilots were safe, but the administration and management were unsafe.”
Welcome to another Why Spotlight! This one is full of learning points and discussions on how culture influences aviation accidents.
The flight we’re covering was supposed to be routine, but it wasn’t.
About 22 minutes after departure, the helicopter encountered a snow band, the visibility deteriorated, and the pilot likely found herself inadvertently in instrument meteorological conditions (IIMC).
The helicopter began a turn, in what was probably an attempt to escape the weather. But the aircraft started descending, and never recovered.
It struck forested terrain, killing the pilot, flight nurse and flight paramedic.
The NTSB investigation found that this wasn’t simply a case of “the pilot flew into bad weather” though.
In fact, the most important lessons from this accident start long before the helicopter ever left the ground.
Let’s take a look at what we can learn from this.
💥 Accident Overview
On 29 January 2019, at 06:28 EST a Bell 407 (N191SF), departed Mount Carmel Hospital in Grove City, Ohio, on a helicopter air ambulance mission to Holzer Meigs Emergency Department.
The mission was to fly approximately 69 nautical miles southeast to Holzer Meigs Emergency Department to pick up a patient.
But there was already a warning sign.
Before Survival Flight accepted the request, two other helicopter air ambulance operators had declined it because of the weather.
The report states:
“Before contacting Survival Flight, the ERT contacted two other HAA operators, MedFlight and HealthNet Aeromedical Services. According to the ERT in a post-accident interview, MedFlight immediately turned down the flight due to weather.”
And:
“The MedFlight pilot stated in a post-accident statement that he declined the flight request because the icing probability at 1,000 ft above ground level (agl) was greater than 75 percent. He also stated that snow squalls were present on the HEMS Weather Tool, which “would reduce visibility and/or ceilings to below [VFR] minimums.”
Information the pilot was not made aware of (but should have been!).
The Survival Flight operations control specialist contacted the evening-shift pilot to assess whether the requested flight could be accepted.
Approximately 28 seconds later, the flight was approved. The report continues:
“The ERT later received a call from HealthNet refusing the flight due to weather. The vice president for HealthNet stated that the pilot declined the request due to low cloud ceilings and icing.”
The evening-shift pilot knew that the day-shift pilot was about five minutes away and that she might take the mission.
When the accident pilot arrived, the helicopter was already running and being prepared for departure.
There was no evidence that the accident pilot reviewed the weather before departure. More importantly, neither pilot completed the required preflight risk assessment.
The evening-shift pilot apparently expected the accident pilot to complete it after returning from the mission.
Weather forecasts showed snow showers and areas of IMC along the route. There were also AIRMETs for moderate turbulence and moderate icing.
Yet, the flight departed.
The helicopter climbed to just below 3,000 ft before beginning a descent. It passed through one snow band and then entered a second.
During this, the pilot likely experienced a significant reduction in visibility and inadvertently entered IMC.
The helicopter then began a descending 180° left turn, likely an attempt to escape the weather. The aircraft continued descending.
The final data showed increases in collective that may have been an attempt to climb, possibly in response to a terrain warning, perceived ground proximity, or both.
It was too late. The helicopter impacted trees.
The investigation found no evidence of an inflight mechanical failure. The aircraft was airworthy. The pilot was qualified.
So, what did the investigation team find?
Ready to learn from real aviation decisions, mistakes and challenges? Get our latest Notes from the Cockpit sent straight to your inbox, twice a month ⤵️
🔍 What Caused this Accident?
The NTSB’s probable cause did not simply say:
“Pilot entered IMC and lost altitude.”
Instead, the NTSB determined:
“Survival Flight’s inadequate management of safety, which normalized pilots’ and operations control specialists’ noncompliance with risk analysis procedures and resulted in the initiation of the flight without a comprehensive preflight weather evaluation.”
The FAA’s inadequate oversight of the operator’s risk-management programme and its failure at the time to require Part 135 operators to establish safety management systems, was also identified as a contributing factor.
This matters. Because if we focus only on the final few seconds of the flight, we miss most of the lessons we can learn here.
1️⃣ The weather wasn’t a surprise
There was enough information available to identify a weather risk.
The forecast included precipitation and marginal conditions. There were advisories for icing and turbulence.
There had already been two air ambulance operators that had rejected the mission because of the poor weather.
The NTSB concluded that the failure to obtain complete en-route weather information prevented the operational team from identifying the critical meteorological risks.
The issue was that the available information wasn’t being brought together into a meaningful risk assessment.
2️⃣ The risk assessment existed, but wasn’t really doing its job
Survival Flight had a risk assessment process, but the NTSB found that it was inadequate.
🔸It wasn’t consistently completed before every flight
🔸 It didn’t capture the fact that other air ambulance operators had already refused the flight
🔸 And the accident flight departed without the required assessment being completed.
This is an important lesson for us all: a risk assessment is only useful if it changes the decision-making process.
A form that gets completed because a procedure says it has to be completed, isn’t necessarily risk management.
That defeats the entire purpose and becomes a tickbox exercise.
The real value comes from asking:
🔸 What are the threats?
🔸 How serious are they?
🔸 What has changed?
🔸 What information are we missing?
🔸 What would make us say no?
🔸 And, most importantly: Are we actually prepared to stop the operation if the risk is too high?
Any other way is a waste of everyone’s effort, time, and SOP.
3️⃣ The safety culture allowed procedural non-compliance to become normal
This is probably one of the most uncomfortable findings in the report.
The NTSB identified a range of organisational issues, including inconsistent compliance with procedures and regulations, pressure to complete flights, gaps in operational oversight and punitive repercussions associated with safety decisions.
The Board concluded that these were indicative of a poor safety culture.
As an example:
“A former Survival Flight pilot reported that, in one instance, he was criticized by the base lead pilot for waiting on the helipad in cold weather until all engine temperature gauges were “in the green,” in accordance with the takeoff checklist, instead of immediately taking off.”
It also found that the culture likely influenced the accident pilot’s decision to conduct the flight without a proper shift-change briefing and preflight risk assessment.
This is where an accident becomes bigger than one pilot. Because when people repeatedly see procedures bypassed without consequence, the deviation itself can become normal.
Eventually: “We don’t normally do it that way” becomes “that’s just how we do it.”
And well, that’s dangerous.
4️⃣ The system had data, but wasn’t learning from it
Another interesting part of the investigation, revealed that the helicopter was equipped with flight data monitoring equipment.
However, Survival Flight didn’t have an effective Flight Data Monitoring (FDM) programme to routinely analyse the data for safety.
The NTSB found evidence of previous flights involving snow and IIMC encounters where flight deviations may have been visible in the data.
In other words, there may have been warning signs hidden inside the company’s own data.
But they weren’t being used.
A good FDM programme isn’t about catching pilots out. It’s about asking “Why is this happening?”.
If several flights show unexpected altitude changes or deviations around weather, that is an opportunity to investigate the underlying risk before something goes wrong.
Without that feedback loop, an organisation can continue repeating the same unsafe pattern without realising it.
5️⃣ The pilot’s final decision
We shouldn’t completely remove the pilot from the story. Once the helicopter entered IIMC, the pilot had to deal with an extremely demanding situation.
The final flightpath followed the company’s 180° IIMC escape manoeuvre. But the helicopter descended throughout the manoeuvre and ultimately struck terrain.
We’ve spoken about the threats of IIMC and how to mitigate them recently ⬇️
The NTSB concluded that the pilot did not maintain altitude during the escape attempt. However, the report also makes it clear that this final event happened within a much larger chain of organisational and operational failures.
And that distinction is important.
The question isn’t just: “Why didn’t the pilot maintain altitude?” (We’ve discussed this to death by now across the industry, like in our “56 seconds to live” coverage ⬇️
It’s: “Why did the pilot find herself in that situation in the first place?”
💡 What Can We Learn From This?
There are plenty of lessons here, and they go well beyond HEMS.
Don’t confuse having a procedure with managing a risk
A company can have a beautifully designed risk assessment form.
It can have weather minima, SOPs, and procedures saying exactly what should happen. But none of those things make an operation safer if people aren’t actually using them.
The real test is what happens when the operation becomes inconvenient.
This is shortcut culture, and can be hard to tackle – as we discussed here ⬇️
When the patient is waiting. When another operator has already said no. When the weather is marginal.
When you’re already running late. When somebody says: “It’ll be fine.”
That’s when safety culture shows itself.
A “no” is valuable information
Helicopter shopping is a horrible practice. It’s the practice of a flight being accepted by one company after it has been declined by other companies due to safety concerns.
To combat this, there is guidance material on how this should be handled:
“Responsibilities of communications specialists should include ascertaining, from those requesting HAA services, whether another HAA operator has previously declined to carry out a particular flight and, if so, for what reason. The response received should be conveyed to the pilot performing the risk analysis in accordance
with CFR 135.617.”
However, the investigation team states:
“A review of recorded phone conversations obtained during the investigation revealed the ERT did not report that the flight had been previously refused.”
The fact that two other operators refused the mission should have been identified as a threat.
Step 1 is to ensure that information is captured. Step 2 is to be careful about judging those crews as being “less willing” or “less capable.”
They may simply have made a different, and potentially safer, assessment of the same threat.
A mature safety system doesn’t hide those decisions. It learns from them. Another example of how culture can affect operations is shown in one of the interviews the investigation conducted:
“Pilots often felt pressure to accept flights. As an example, company management motivated bases to conduct flights by
purchasing one massage chair for any base where the pilots flew 30 flights in 1 month. According to the company’s monthly summary, the accident flight was the 26th flight for the base in January.”
A information notice from the company included a line (5) that could be interpreted as encouragement to call them after other companies refuse a flight:
Every declined flight, aborted approach, go-around, weather diversion and precautionary landing should make us ask:
“What did that crew see that we might have missed?”
Look for signs before they become accidents
A big lesson from this accident is the importance of near misses and deviations.
Survival Flight’s FDM data apparently contained evidence of previous encounters with snow and IIMC. That should have been an opportunity.
We should always ask:
“What happened today that shouldn’t have, and what can we change before something serious happens?”
Safety management is about finding those signs while there is still time to do something about them.
Safety Management is a System
The NTSB highlighted the absence of a comprehensive Safety Management System (SMS) at Survival Flight.
An SMS isn’t simply more paperwork.
Done properly, it identifies hazards, assesses risk, implements controls, checks whether those controls actually work, and continually improves the operation.
Basically:
Find the problem ➡️ manage the risk ➡️ check the mitigation ➡️ learn ➡️ improve.
And then… repeat! 🔁
This is what stops an organisation from having to wait for an accident to tell it whether its controls are effective.
💭 Conclusion
The most sobering part of this accident is how many opportunities there were to prevent it before the helicopter got anywhere near the snow.
The final mistake is often only the last hole in the Swiss cheese.
The real safety opportunity is finding the holes long before they line up.
So, let’s try and change the conversation from “why did this pilot not use the AP or other SOP to get out of IIMC”
Let’s zoom out, and learn from the actual root cause of these many many -many- IIMC accidents.
You can find the NTSB report here.
Ready to learn from real aviation decisions, mistakes and challenges? Get our latest Notes from the Cockpit sent straight to your inbox, twice a month ⤵️
3 Comments
wallacedavid1955 · August 17, 2026 at 12:26 AM
Very sad situation but a great job on your part.
Dé Jansen · August 16, 2026 at 8:03 AM
Great analysis! Well done
Janine Lythe · August 16, 2026 at 8:12 AM
Thanks so much Dé!