Drunks In The Cockpit: A Problem?

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

On a news sensationalism scale of one to 10, stories about drunken airline pilots are about an 11. These seem to pop up in the news cycle about once a month and occasionally, just to really turbocharge them, we even get video of the pilot stumbling through security and falling down %^$faced into the cockpit as a mortified flight attendant looks on.

Diligent readers will recall that over the weekend, we hit the lottery with not just one drunken pilot story, but two, plus a bonus video of a juiced captain wobbling through security, then slurring his way through a PA announcement.

What to think of all this? As a pilot, I suspect you assume what I assume. Even though they appear on the evening news with concerning regularity, they’re such isolated incidents as to hardly merit mention, much less worry. But is something that occurs once a month isolated? A solar eclipse is a rare event; 12 times a year is not.

But I suspect you understand that these incidents are flyspecks among millions of flight operations each year. Statistically, they’re indistinguishable from the background noise. And they’re not increasing, so why worry about it? But how do you know they’re not increasing? Like me, you probably just assume it. But you could be wrong for all the contextual information news stories—including ours—give.

We have in place a reasonably tight net of random testing to snare drunks and druggies from entering the cockpit and what data is available suggests that the problem, if it even is one, is infinitesimally small and not trending in either direction. But to inform your thinking on this, it helps to know the actual numbers.

The FAA requires companies to have in place a random alcohol and drug testing program. It used to require 50 percent of all safety-related employees to be tested, but that was reduced when it was surmised that the tack was too small to hit with such a large hammer. Now it’s 25 percent.

The FAA told me this week that for 2015, 56,327 random alcohol screening tests were given, which yielded 119 people having 0.04 blood alcohol level or higher. Ten of these were pilots for a percentage of 0.017. Keep that number in mind next time you see a drunken pilot story. Since 1995, when the random testing program started, an average of 11 pilots have failed the alcohol screening. There is no trend, just a spikey graph from year to year. (High 22 in 2002; low three in 1995.)

The screening program also hunts for illicit drug use, including marijuana, opiates, cocaine, amphetamines and phencyclidine. Here, the raw numbers are a little higher for positives on random tests, but for 2015, the percentage of positives was identical to alcohol: 0.017 percent positive.

So with these low numbers, where’s the problem? There isn’t one, unless you consider 0.017 percent a valid risk. Triple it and you‘re still not up to a full point. And as far as I know, there is no recent alcohol-involved accident history related to airline flying. (GA is another story. I’ll examine it in a future blog.)

So why are we running these stories? Two reasons: One is that the flying public has a compelling reason to know if the airline pilots flying the airplanes they’re riding on are likely to be sober. With vanishingly small exceptions, they are and the stories should say this each time we run them. Ours did not (although we have in the past) and many other outlets don’t either. Next time, we will include that, probably providing a link to this blog just to put things in statistical perspective.

The second reason we run these stories is the same reason the evening news gives them 12 seconds: click bait. Readers like reading them for the same reason we used to put bad car wrecks above the fold in newspapers. Just to lend detail to this, I put on my Internal Affairs Division hat and emailed my colleague Russ Niles to ask about his decision-making on the weekend story. Here’s his reply:

“The decision to run the story was based on two factors: hard news value and the availability of the video for the guy in Indonesia. The Indonesian story came in first and I was on the fence about running it until I saw the video. It was still not a firm choice for the Monday flash until the story from Calgary came in.

These were two examples of extreme pilot behavior in a short period of time that are similar to those we occasionally run on other types of safety violations by pilots. Simply put, pilots are interested in this stuff and, yes, I believe that our audience understands that egregious safety violations, regardless of their nature, are rare among pilots. 

There is also the fact that there is pressure from you [me being editorial director] and Tim Cole [Belvoir editorial VP] to ensure that aviation-related stories that hit your CNN and ABC news alert boxes get covered in AVweb and both of these qualified under that protocol.”

As I apply ointment to the burns from my own petard, there in a nutshell you see how spectacle and competitive pressure combine to push news judgment in a direction our instincts, on further reflection, might cause us to reconsider. Had I been on the news desk over the weekend, I’d have made the same decision. And if you clicked on the Monday newsfeed and didn’t see the drunk pilots story you knew you saw over the weekend, you’d have thought we missed it.

Next time, as is my wont, I’ll make sure we include the rest of the story. Meanwhile, I can't resist leaving you with Dean Martin's and Foster Brooks' hilarious shtick on the drunk airline pilot. I dare you not to laugh.

Comments (40)

0.017 per cent. How many millions/billions of dollars wasted on this issue? Amazing! Sure glad most pt 61/91 ops don't require this nonsense. I wonder if any sightseeing (pt 91) pilots have ever had a positive test (the only pt 91 op that requires testing)? Another government solution to a nonexistent problem in aviation!

Posted by: matthew wagner | January 5, 2017 7:55 PM    Report this comment

And then there's the commercial flying public. Which will they fear more - robot pilots or the occasional drunk pilot?

Posted by: Rollin Olson | January 5, 2017 8:43 PM    Report this comment

With 28,537 domestic airline flights per day, that 0.017% figure works out to 4.85 airline flights PER DAY. I don't find that to be statistically comforting.
Apparently, we need two crewmembers to statistically prorect us from chemically-impaired pilots. Otto is looking better, Larry. Oy.

Posted by: Tom Yarsley | January 5, 2017 11:58 PM    Report this comment

I love that era of comedy. It was truly golden. Yes, it brought a big smile to my face.

Posted by: Thomas Cooke | January 6, 2017 4:03 AM    Report this comment

It would be nice to have a comparison to put the problem into perspective. F'rinstance what percentage of drivers are over the limit or on drugs or are suicidal/insane at any given time?

Posted by: Richard Montague | January 6, 2017 7:28 AM    Report this comment

>>With 28,537 domestic airline flights per day, that 0.017% figure works out to 4.85 airline flights PER DAY. I don't find that to be statistically comforting.

Actually it's twice that (almost ten flights per day) since the 0.017% value is per-pilot and there are two pilots per commercial flight. So the odds that your commercial AIRPLANE has one drunk pilot onboard is doubled, even though the risk of a bad outcome is reduced (by having the other pilot also onboard).

Just as your odds of an engine failure go up in multi-engine aircraft.

Posted by: David Bunin | January 6, 2017 7:30 AM    Report this comment

One of the proposals placed before regulators was to require commercial pilots to blow an alcohol test before every flight and submit a urine sample for quick analysis.

You can see the smart thinking behind this proposal. Such a program would cost hundreds of millions and keep drunks out of the cockpit. Never mind that there's zero connection between alcohol and U.S. airline accidents in the modern era, spending more would... what? Reduce it to less than zero?

Has sanity left the building? What's next, trigger warnings at the gate?

Posted by: Paul Bertorelli | January 6, 2017 8:17 AM    Report this comment

But Yars ... if we're gonna be fair and PC about this, Otto has to be tested, too ... maybe in a different way but ... something like an unannounced software load verification. Especially if he's the only one running the airplane. And, of course, we'll need another bureaucracy to run this ... AIR121.2b, e.g.

Posted by: Larry Stencel | January 6, 2017 8:32 AM    Report this comment

Obviously everyone has missed the simple solution. Encapsulate each pilot in a sealed environmentally controlled pod that prevents access to alcohol and other drugs for 24 hours prior to flight. The pod could be delivered to boarding gate where TSA agents (with drug and alcohol sniffing dogs) would escort the flight crew to the cockpit and lock them inside until after departure. Since the pod would be self contained, the exhaled air and other body byproducts produced by the pilots could also be scanned for evidence of use prior to pod sequestration.

This concept is merely based on a superior bureaucratic and technological solution to a non existent problem. Might be tough on the pilots on the reserve list though.

Posted by: Leo LeBoeuf | January 6, 2017 8:42 AM    Report this comment

The Dean Martin skit is good, but you might want to check out, "The Man Show's" take on drunken pilots on You Tube.

Posted by: MICHAEL BROOKER | January 6, 2017 8:52 AM    Report this comment

It could be argued that two-pilot crews are correlated, meaning that there is a greater-than-coin-toss probability that they both will be drunk!
Seriously, drunk pilots are news because they are what journalists call man-bites-dog stories... I have never heard pilot inebriation cited as a fear-of-flying concern.

Posted by: Jerry Fraser | January 6, 2017 9:44 AM    Report this comment

I seem to recall that the Air Force conducted some tests which conclusively showed improved performance on difficult (simulator) approaches after a drink or two. Not the slurring, falling down stage, obviously.

Posted by: John Wilson | January 6, 2017 11:00 AM    Report this comment

I'm never one for more government intrusion, but sometimes we need to make changes to existing procedures that are BETTER than simple laws and endless "feel-good" procedures.

Years ago, Aeroflot (Russian airlines) had a problem with drinking pilots. They instituted a cognitive screening procedure, rather than a breathalyzer. The early procedure required the crew member to replicate a series of blinking lights--much like a childs game. Quick and easy--no pass--no fly. Minimum time and intrusion.

The great thing about these tests is that it screens for more than alcohol. There are any number of reasons beyond alcohol which may impair a pilot--drugs, emotional state, personal problems, stroke, age, lack of rest. If a pilot can pass the test, let them fly. It's quick, easy, and immediate. It gets rid of "is the pilot incapacitated or not?" issues--while pilots may be drug and alcohol tested, those take too long--and don't test for any of the other issues. It gets rid of the all-too-prevalent but unreported to the public "post-mortem tests found evidence of prescription (or non-prescription) drugs--we all know that we are not to self-medicate and ground ourselves, but many of us don't. There is also the problem of pilot interaction--crewmembers are loathe to accuse another crew member of being impaired.

Rather than being MORE intrusive, cognitive screening could do away with subjective laws--medical issues and age, for example. It would go a long way toward eliminating medical uncertainty and bureaucracy. It would reassure the passengers, the airline, the pilot's union, and the FAA that the pilot is capable to make the flight that day.

THINK about it--Which is less intrusive and more effective--the existing complex set of laws subject to multiple interpretation and self-reporting, or actually TESTING prior to flight?

Posted by: jim hanson | January 6, 2017 12:02 PM    Report this comment

An interesting idea. What happens if/when 20% of stone-cold-sober pilots can't pass the test? Do we need a "personal normal" standard for each pilot, as is done for athletes' concussion protocols?

Posted by: Tom Yarsley | January 6, 2017 12:20 PM    Report this comment

Yars--The tests aren't that hard. I tried to post a link to the Australian cognitive tests, but couldn't as the site viewed it as spam. Try this--copy and paste this


See if that gets by the spam filter

Posted by: jim hanson | January 6, 2017 12:29 PM    Report this comment

Jim, what problem are you trying to solve?

Posted by: Paul Bertorelli | January 6, 2017 12:32 PM    Report this comment

If I add in the full address, the spam filter rejects it. I had to eliminate the preface hotel tango tango papa sierra colon slash slash papa delta foxtrot sierra period in order for it to post the link.

I can't even post that part separately.

Posted by: jim hanson | January 6, 2017 12:49 PM    Report this comment

Yes, you can. Just strip off the hyperlink portion and it will post. Like this:


Posted by: Paul Bertorelli | January 6, 2017 2:44 PM    Report this comment

I think you made the point in the article that it does NOT warrant coverage, and then made excuses for why it was covered.

Doesn't make much sense to me. Cover it if you feel it's newsworthy. But don't do that, and then show why you shouldn't have.

I'd have looked at it and said, "How many US airline crews were involved?" None? Well, then, what's the problem worth reporting on, and what's the point it makes? Nothing.

If we publish every foreign carrier's screwups and imply that it's a problem here, that's unfair to those of us who have to meet the higher US standards those particular carriers don't, whether they are for maintenance or for flight operations.

Posted by: Ronald Cox | January 6, 2017 5:50 PM    Report this comment

It's like playing catcher, pitcher and not letting anyone go at bat.

Posted by: Rafael Sierra | January 6, 2017 7:46 PM    Report this comment

Michael, The Man Show "drunken pilots," that was great, both episodes.
Yes, The Man Show takes the cake.

Posted by: Thomas Cooke | January 6, 2017 8:30 PM    Report this comment

Ronald, I rather thought I made both sides of the argument on the coverage decision, not that it should not be covered. What it was lacking was context: the actual data showing that intoxication in the airline pilot cohort appears to be inconsequential.

As I think I said, that's why I followed up with a blog.

I'm sure our thousands of Canadian readers will take issue with you considering them a "foreign country." The reality is we have a global audience, which is why we cover news from Canada, from Australia, from South Africa and the Pacific Rim, where we have many readers. It's the nature of the business.

Posted by: Paul Bertorelli | January 7, 2017 5:40 AM    Report this comment

As long as we're suddenly splitting hairs here ...

Ronald said, "If we publish every 'foreign carrier's' screw-ups and imply that it's a problem here, ...". He did NOT say "foreign country."

A "foreign air carrier" is clearly defined by 49 U.S.C. 40102(21) and further defined by 14 CFR 1.1 as, "a person, not a citizen of the United States, undertaking by any means, directly or indirectly, to provide foreign air transportation." (The above definitions lifted directly from a DOT and FAA websites). I'd say an operation where a Slovakian Captain flying a Canadian airplane or an Indonesian Captain flying an Indonesian airplane certainly qualifies under these definitions.

I read your blog and wondered the same thing Ronald did but chose not to say anything about it. I 'got' the intent of your commentary, however, and moved on. Likewise, I did not take Ronald's comment as derogatory toward anyone ... foreign or domestic. He was just questioning your reasoning for writing this blog in the manner you did. Maybe he's a "domestic" Captain and took exception to being lumped into a the same storyline, on purpose or by accident? Since you have a penchant for empirical data," maybe you ought to provide us a further explanation of how many drunk "foreign" vs "domestic" pilots there are in any manner you find consistent with supporting your premise?

In your response, you re-explained your rationale but then found it necessary to retort, "I'm sure our thousands of Canadian readers will take issue with you considering them a "foreign country." " SO ... did any of those "foreigners" send you a message complaining about this? Where's the "data?" Since you chose to 'chide' Ronald in the public forum, I'd say you owe him an apology in the public forum. Neither of you are guilty of anything ... save -- maybe -- for being overly sensitive. You know what they say about "glass houses."

Has "PC" now invaded Avweb?

Posted by: Larry Stencel | January 7, 2017 9:33 AM    Report this comment

I shall arrange a poll and get back to ASAP. Your lips to God's ears.

Posted by: Paul Bertorelli | January 7, 2017 10:25 AM    Report this comment

Since you chose to 'chide' Ronald in the public forum, I'd say you owe him an apology in the public forum. "

This statement defines political correctness. Mr. Cox avoided in-depth comprehension of the blog's intent and therefore lashed out - it's the new/old power struggle of self-recognition, dontcha know. And, in today's world, it comes also from the very top - how can intelligence agencies know more than I do? Tsk, tsk.

I didn't like the Man Show skits much, but love the roasting type humor of old. Guess I'm old school, or just old.

Posted by: Dave Miller | January 7, 2017 10:51 AM    Report this comment

Those are the figures from milk drinking and corn eating USA. Certainly high enough to cause concern -- four flights a day.
The Russians know they have a serious problem and have (had) their flashing light tests -- but as the drunk snow plough driver on the runway which killed the boss of Total (and crew and others) shows, Russia can be a very crazy place.
Some of the wilder parts of Africa and South America will, I am sure, also give frightening figures.
Having said that a lot of it is cultural. Air France pilots used to always have wine with their meals because in France wine is not seen as alcoholic the way beer or spirits are. Even today pregnant French women are told not to drink and to only have a glass or two of wine at mealtimes. When Air France finally said no wine -- only around 20 years ago, they had to boost pay to compensate for the hardship. And it was after that date that they had a nasty patch of accidents.

Posted by: John Patson | January 7, 2017 11:18 AM    Report this comment

Guys, no one is chiding or lashing out. It's just a discussion, no?

You, too, Larry. I know you're trying to be as big a trouble maker as I am, but I'm a trained professional with many years of experience. I have to work today, but you can go outside. Sounds appealing.

Posted by: Paul Bertorelli | January 7, 2017 11:37 AM    Report this comment

Just make sure that you reveal to all of us ... foreigners and natives alike ... the epistemology of your methods. We don't want any of that there 'voodoo science' to invade this space, too.

Posted by: Larry Stencel | January 7, 2017 11:47 AM    Report this comment

I carefully and purposely chose the somewhat passive verb 'chiding' as opposed to a stronger form to ensure that no one took my comment as an overt attack aimed at starting a flame war That said, I took your retort to Ronald as something that had to be challenged. More than one person wondered what the end game here was so the communication was less than optimum, irrespective of your 'pro' status. In discussions, sometimes a statement has to be responded to ... that's all. Until then, I was mostly silent except for keeping OTTO inflated. :-)

Yes ... you ARE correct. Retirement is HIGHLY UNDERRATED.
ME ... a trouble maker? I see myself as an irreverent and wry humorist serving a distinct purpose.

Now can we get back to beer?? Or was it wine? :-)

One last comment ... I read your blogs because I carry things away from them. I'm sure your other readers do too? This time it's ... remember the eight hour rule ++ if you're gonna pull back on the yoke. Sage advice.

BTW: I see now where an airliner crashed into a drone ... also 'overseas.' Can we revisit your drone blogs / positions now?

Posted by: Larry Stencel | January 7, 2017 12:33 PM    Report this comment

I'm still working on "epistemology" and you guys already kissed.

Posted by: Rafael Sierra | January 7, 2017 1:03 PM    Report this comment

When/where drones are outlawed, only outlaws will have drones.

Posted by: Tom Yarsley | January 7, 2017 9:02 PM    Report this comment

I agree with Larry. And to YARS, drones can cause annoyance, trouble, damage or injury. Therefore they can be a nuisance. .

Posted by: Rafael Sierra | January 8, 2017 9:25 AM    Report this comment


Yes, drones can cause all of those. And they can be a nuisance to some. That does not justify outlawing them, any more than drunk driving justifies outlawing the consumption of alcohol (we tried that), or that the Fort Lauderdale Airport shooting justifies repealing the Second Amendment.

I'll bet that Sully thinks that birds can be annoying and can cause damage, too. And they can and do. Light aircraft also can cause annoyance, trouble, damage, or injury. And yes - there are plenty of people who think that we should outlaw them, too. Keep that in mind when you look down your nose at "lesser" things that fly.

Compliance with the law is a good way to attenuate many kinds of annoyances.

Simplest and most-effective solution? Outlaw the manufacture, sale, and outdoors operation of non-geo-fenced drones - and jail violators for ten years.

Posted by: Tom Yarsley | January 8, 2017 11:27 AM    Report this comment

"BTW: I see now where an airliner crashed into a drone ... also 'overseas.' Can we revisit your drone blogs / positions now?"

*Possibly* crashed into a drone. The last time there was a "this is definitely it, a drone crashed into an airliner" report, it turned out not to be a drone. I will wait and see on this one too before jumping to any conclusions.

"Simplest and most-effective solution? Outlaw the manufacture, sale, and outdoors operation of non-geo-fenced drones - and jail violators for ten years."

Though that would unfairly punish those people who know how to responsibly operate a drone. I just found out that there's an ultralight airport within 5 miles of my home, which I never realized after years of looking at sectionals of the area. But I did notice it after I bought a drone (hey, it's easy to get a commercial sUAS certificate, so I figured I might as well buy a drone too to practice flying it) and loaded the "b4u fly" app. Since I know the rules, there's no reason why I can't fly a drone around my home and not bother any possible ultralights, but a geo-fenced drone would prevent me from doing so.

But if we're talking such extremes, why not also outlaw the manufacture, sale, and operation of non-geo-fenced "little airplanes" and jail violators for ten years, so that they don't operate around noise-sensitive areas?

Posted by: Gary Baluha | January 9, 2017 7:56 AM    Report this comment

"But if we're talking such extremes, why not also outlaw the manufacture, sale, and operation of non-geo-fenced "little airplanes" and jail violators for ten years, so that they don't operate around noise-sensitive areas?"

Expect to see it in our lifetimes.

Posted by: Tom Yarsley | January 9, 2017 8:08 AM    Report this comment

Outlawing non geo-fenced drones would be no more effective than gun laws that seek to separate bad actors from firearms. All these laws do is appease the fraidy cats while punishing the innocent. Just like many gun laws. The laws themselves are often too over broad and of questionable constitutionality and enforcing them is utterly impossible. There are too many ways around these laws and this is technological tsunami that regulators and legislatures will never keep up with it.

Short of doing nothing, the best we can hope for, in my view, is reasonable regulation coupled with training and education. Sanctions against egregiously reckless operation give regulation teeth, but the rest of it is absorbed by the fact that all new technologies involve risk/gain ratio. The gains here are considerable, the risks still relatively unknown.

I would remind you that three people were killed in a Texas midair last week. Presumably the pilots were trained and competent and we do have regulations against running into each other. We seem to accept that risk. Same should apply to drones.

Posted by: Paul Bertorelli | January 9, 2017 9:14 AM    Report this comment

Paul, one significant difference with drones is that the population that comprises the "we" who accept certain risks is fundamentally different from the traditional pilot population.

Gun laws rely on people to obey them. Most do; some don't. Geo-fencing relies upon technology to enforce regulatory compliance. Very few drone operators would attempt to defeat geo-fencing technology, if for no other reason than that they simply don't have the skills to do it. With apologies to Cyndi Lauper, these girls just wanna have fun.

If a law / rule / procedure is important enough, the best and often ONLY way to ensure compliance is through automation - REMOVE the ability not to comply. I say this with some experience as well as with some bias, since that's how I make the bulk of my living.

Posted by: Tom Yarsley | January 9, 2017 9:39 AM    Report this comment

"Very few drone operators would attempt to defeat geo-fencing technology, if for no other reason than that they simply don't have the skills to do it."

Few is a relative, Tom. Thousands of users hack into drone software to change parameters, defeat altitude and geo limitations and they circulate this on the technical forums. These are the guys you wish to stop and they are the very ones you can't, because they'll go around you. Every time. There's a video on YouTube with a piece of foil taped over the GPS antenna on an Inspire. An astonishing number of people build their own drones.

I know you're a automation fundamentalist, but it has its limits and we're seeing them every day. As for the at-risk community, I'm thinking of the general public, not pilots. They accept the sharing-the-airspace-with-little-airplanes risk. I actually think I'm being guilty of over generalizing in assuming pilots don't accept the drone airspace share risk. I think they generally do.

Posted by: Paul Bertorelli | January 9, 2017 10:01 AM    Report this comment

By the way, related to this, last night 60 Minutes had a report on emerging autonomous war fighting capability. An Air Force general discussed the worry that programming a machine to kill humans crosses an ethical line.

He's right. Once we do that, we're in an entirely new era.

Posted by: Paul Bertorelli | January 9, 2017 10:07 AM    Report this comment

Few IS relative, Paul. I have to put the hackers in the category of "don't worry about things over which you have no control." But I (wrongly?) think that the vast majority of DJI-class drones couldn't be bothered to hack them. THOSE are they guys I'm interested in stopping, because those are the guys for whom technology-based constraints will be effective. Geo-fencing is a great way to prevent the perfect from becoming an enemy of the very very good.

Land mines and water mines are machines that are programmed to kill people. At least hunter-killer robots are more discriminating than "booby-traps" are.

I LOVE that phrase "automation fundamentalist." I hadn't heard it before, but I'm going to share it with a colleague this afternoon. It may inform some work we're collaborating on.

Posted by: Tom Yarsley | January 9, 2017 11:05 AM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?


Enter your information below to begin your FREE registration