Seeking medical cert help!!

[PeterNSteinmetz]

You keep talking and have still yet to support anything you’ve said.

Isn't that a bit of an exaggeration? I have noted exactly what the meaning of the sport pilot data is and why I think it is somewhat meaningful, particularly as compared to the only argument in support of the efficacy of FAA medical regulation, which is an ergo propter hoc fallacy and consists of practically no data at all.

And evidently my offer to provide some pointers was ignored. Would you like me to provide you some pointers to this literature? And if so, will you have a look at some of it?

Or would you prefer that I find one example of regulatory failure to discuss? Clearly the whole literature can not be summarized in a quick post here.

Here is a perhaps more practical and important question -- what sort of evidence would persuade you that FAA medical regulation is not helpful or necessary? Is there any that could in fact be practically produced which would persuade you of that?

 

[PeterNSteinmetz]

Google is your friend. Further, it found that out of 54,295,899 commercial flights during the period from 1993 through 1998, there were two accidents attributed to a medical incapacitation of the pilot. Two.

Does 2 in 54,295,899 strike you as “extremely, extremely rare”?

So here we are getting somewhere, thanks for finding that. So the 2 number refers to accidents caused by incapacitation, which is different from the original assertion which dealt with medical incapacities being "extremely, extremely rare". The rate of those, plus the closely related impairments is evidently 0.04549 per 100,000 hours.

That is also quite rare. But the functional question is whether it is significantly less than the rate in sport pilots? That is probably much harder to estimate given the difficulties of estimating the denominator (I have tried but not found good numbers on that. If you Google-fu can turn some data up, I am happy to try and see if a statistical comparison can be made.) If one had those two sets of numbers, one could try and draw a better conclusion about whether the regulations were CORRELATED with a lower rate of incapacitations or accidents.

Then to use this as evidence that this low rate implies the FAA aeromedical regulation have a positive effect, one would have to decide how of that low rate is due to the regulations, versus say the airline fitness for duty requirements. These are hard things to estimate.

A better way to study this would be to compare the rates of accidents before and after the introduction of regulations. Unfortunately, I don't think this was done historically.

 

[PeterNSteinmetz]

Why do you keep asking if I want to see the mass of evidence you have to refute the safety influence of FAA regulations instead of just posting it?

Please DO NOT misquote me (straw man fallacy). What I said was there existed

the large body of literature having to do with the unintended side effects of regulation, the effects of regulatory capture, and the typical behaviors of regulatory agencies

If you would like a pointer to start into that literature, from a pro-freedom point of view, please see the already mentioned book by Mises on Bureaucracy. It is a free download at https://mises.org/library/bureaucracy. Even the first few chapters will lay out parts of the basic argument.

In terms of the effects of FAA aeromedical regulation on safety, really closer to the OP here, I have summarized the direct data as follows in a post above as

One can discuss the power of that dataset to detect a change of a certain magnitude. However, it is incorrect to say “no conclusion can be drawn”. It is valid to say that those data are not terribly strong, I would agree there.

To be objective, they must be weighed appropriately against the other evidence. I find the evidence in favor of the hypothesis that FAA aeromedical regulation improves the safety of flight to be quite weak. It consists almost solely of the fact that we've had those regulations and our commercial safety record is good.

I believe you have read that post.

I believe the large body of literature on the ineffectiveness of coercive regulation bears on this question INDIRECTLY and suggests that we should skeptical of claims that FAA aeromedical regulation is effective without good evidence.

But again, the more practical question, what sort of evidence would persuade you that FAA medical regulation is not helpful or necessary? Is there any that could in fact be practically produced which would persuade you of that?

If it is not possible to name some, then this is effectively a non-falsifiable belief.

 

[PeterNSteinmetz]

To no one in particular: I've noticed that people always wants the burden of proof to be on the other side.

That is true of course, but the proper resolution is that the burden of proof is on he who asserts existence. To adopt any other standard is to make it trivial to prove absurdities, for example, that there are non-corporeal dragons in my garage. Carl Sagan writes amusingly about this.

 

[PeterNSteinmetz]

Just wondering: What is the difference between a coercive regulation and a non-coercive regulation? Can you give an example of each?

Good question, thanks. We have had a long related discussion before.

I would define a coercive regulation as any regulation to which you have not voluntarily agreed where the ultimate enforcement for non-compliance involves the use of physical force or threat thereof. Please note I said ultimate, so this would include any regulation in which if you don't comply, and you subsequently also refuse to comply with the consequences, force will be used against you.

Example of a coercive regulation - you must obtain a FAA medical certificate before flying your own aircraft. There is a threat of ultimate use of force if you don't comply and you did not voluntarily agree to this. As you have noted, the threat is pretty gentle and is often resolved without any violence, but that is only because people comply. Almost every regulation currently by the government falls into this category.

A non-coercive regulation would be any regulation which you have freely agreed to or is not ultimately backed by the use of physical force or the threat thereof.

Example of a non-coercive regulation - you must keep your lawn appropriately watered in your neighborhood per your association agreement where you voluntarily purchased a house. In this case, you voluntarily agreed to purchase the house and comply by the terms of the association agreement. These are often backed by contract law, so there is an ultimate threat of the use of violence for failure to comply.

Aviation has of course been heavily regulated from the very beginning, so I can imagine a lot of scenarios where things could be argued to be somewhat murky in terms of these distinctions. But I think a fair number of cases are clear.

I have speculated with my wife that it might be possible to create non-coercive regulations created by the government. The government could have regulations where the only punishment is being placed on a list of regulation violators, but no physical violence would ever possibly be used, there is no threat. Unclear how much deterrent that would actually create, though maybe some, especially in this day and age of readily accessible data on the web. Mostly just a thought experiment though.

 

[PeterNSteinmetz]

I'm not buying this so-called distinction. I was under the impression that ALL agreements were backed by contract law. Furthermore, just like you voluntarily agree to purchase a house, you voluntarily decide to become a pilot. However, under that logic, ALL regulation of pilots would be "non-coercive."

I think one of the distinctions here is that you can buy a house somewhere else or not buy one at all and still live within the general geographic area where you reside. Also most neighborhoods actually don't have such association agreements. That is more voluntary than the regulations enforced on anyone who happens to want to fly in the USA.

So if you don't want to think of that as a useful categorical distinction, perhaps we could agree that one is more coercive than the other on a continuum?

I think that coercion, in a finer analysis, really is best regarded as being on a scale and as having different types. What I think is really potentially bad is the use of physical violence or threat thereof. Much more dangerous than say being "coerced" by someone just verbally berating you or offering you a lot of money to do something you don't want to. Physical violence can irrevocably injure or kill you. A verbal berating or not taking money doesn't do that.

I agree that non-coercive regulations are considerably rarer than coercive ones, unfortunately, in our present state of affairs.

To try and argue that everyone who happens to live in a particular geographic area has somehow "agreed" to follow some set of rules is essentially to argue for Rousseau's social contract theory. There is a long history of debate about that. https://en.wikipedia.org/wiki/Social_contract.

When we are talking about governmental laws and regulations in this regard the major problems with that theory are that this supposed "contract" was never available for us to observe and agree to when we could have made a meaningful choice about it. There is a big difference with a housing association agreement -- the contract was available before you decided to purchase the house and you could read it and hire an attorney and then decide whether or not it was for you.

Personally I much prefer a natural rights based approach. You only have a right to use physical force or threat thereof in retaliation or to prevent violence from being used against you. You can pool together with others in your neighborhood that right and hire police, courts, etc. You can also choose to enter into a contract, complete with covenants of non-fraud, faithful execution, etc which will potentially be enforced using physical force, but that is up to you, and voluntary. But if someone hasn't entered into a voluntary contract with you and hasn't used physical force against you, or presented a clear and present danger of doing so, you have no right to use physical force against them or threaten to do so.

Applying such theory to the question of aeromedical regulation. One might try and make an argument that simply flying a heavy airplane over someone's house or person is sufficiently dangerous that it constitutes a threat to possibly inflict physical force and harm on someone. I don't know the numbers off hand, but it seems that threat is deminimus and fairly small by comparison to the possible threats we tolerate in everyday life. And even if such a threat were regarded as not deminimus and worth worrying about, then one would have to demonstrate that the threat is sufficiently higher in the case of those who don't have an FAA medical certificate compared to those which do in order to justify the use of a coercive regulation. I could be convinced by the data, but despite my repeated queries here, haven't seen that.

I am not convinced on either count -- either that the threat is more than deminimus or that the difference is significantly different. Therefore I argue that such imposition of a regulation, backed by the potential use of physical force, is not justified from a natural rights perspective.

Maybe one could argue that certain dangerous behaviors in an airplane, such as diving at high speeds towards someone's house, are sufficiently dangerous that one is justified in having a law or regulation to forbid it. But that is a different question from whether having an aeromedical certificate indicates a sufficiently reduced risk that it should be reflected in law.

 

[PeterNSteinmetz]

The results of the review showed that sport pilot accidents were caused by medical incapacitation 3.7% of the time, while private pilots with Class 3 medicals had accidents that cited medical incapacitation 2.5% of the time. Medical incapacitation represents either a probable cause or contributing factor in aviation accidents less than 5% of the time. The author concluded that although there is a slightly higher rate where the NTSB cited incapacitation among pilots not receiving regular aeromedical evaluations, the difference was not statistically significant.

(emphasis added)

I think this is the sort of weaker evidence that argues that medical certification does not improve the safety of flight. As noted by others, it is not conclusive by any stretch.

Clearly something causes commercial flights in the US to be orders of magnitude safer than private flights by pilots holding either a medical certificate or not. The fact that there is such a large difference between commercial and non-commercial strongly suggests that there is some other factor than having a medical certificate which matters. I believe many have suggested it is the stricter standards for training which have made the largest difference over the last several decades.

 

[PeterNSteinmetz]

@Palmpilot It occurs to me, given our long prior discussions of this subject, that I should perhaps ask a question similar to the one I have asked others here. I really do think it helps to clarify whether something is a falsifiable belief or non-falsifiable.

What evidence would persuade you that the existing aeromedical regulations do not effectively improve the safety of flight? At least evidence which could in principle exist and would be persuasive.

Because if there is some, then one can often tease these issues apart further by asking how close the present evidence is to such. OTOH, if there is none which would be persuasive, then the belief that aeromedical regulations effectively improve the safety of flight is just that, a belief, like a belief in a religion. It is best then to just agree to disagree I think.

 

[PeterNSteinmetz]

If you're saying that the house-buying example is not applicable to FAA regulations, I agree with you, but you're the one who brought it up.



So far, I haven't seen even ONE example of a non-coercive regulation. That's why I'm not buying the distinction. Regulations have the force of law, and they can be enforced with real penalties, ultimately backed up by the threat of fines or imprisonment. ALL regulations are coercive, and that means that this whole discussion of "coercive" regulations is a red herring, because it distracts from the real issues, which are whether the particular regulations being considered accomplish a legitimate public purpose, and whether they do more good than harm.

Hmm. Somehow it seems like we are not understanding each other here.

Would you agree there is a distinction in the level of coerciveness between 1) choosing to buy a house in a neighborhood and executing a home association agreement and 2) finding oneself in a country where there are regulations requiring you to obtain a medical certificate simply because you want to fly an airplane?

BTW, I think nearly all my discussion regarding aeromedical regulation in this thread used the term “coercive regulation”. I don’t think I ever raised a distinction in this thread. So in a sense I agree that isn’t the main point. You could go back and substitute “regulation” for “coercive regulation” everywhere in those arguments, I believe. I was actually a little puzzled why that question was raised.

 

[PeterNSteinmetz]

There is also the fact the commercial levels of medical certification are stricter than those for third class and below. I do believe, however, that the level of recurrent training and the procedures required in commercial operations are responsible for the lion's share of the airlines' exemplary safety record.

True, but the differences in the strictness of the aeromedical standards are pretty minor, aren’t they? I guess the main ones would just be the timeframes of validity.

It strikes me it would be hard to ascribe much of the large difference in safety between commercial flight and private flight to the aeromedical standards differences when the apparent difference between a 3rd class and no certificate in terms of safety are so small and not significant.

 

[PeterNSteinmetz]

It would be just as valid to ask what evidence would persuade me that the existing aeromedical regulations do effectively improve the safety of flight, because I don't have a firm belief one way or the other. I just feel that the discussion would be more useful if red herrings were left out of it.

I think the question of what is an appropriate level of evidence does enter into the political domain. Inevitably one has to ask, when are we justified enforcing a regulation — how strong should the evidence be before we are willing to do that?

I imagine we would agree that the government should not enforce a regulation preventing people from eating Doritos in flight just because someone at the FAA hates Doritos and prefers pretzels as a snack.

On the other side of the continuum of proof, I imagine we would also agree that people and by extension the government are justified in enforcing a regulation forbidding people from dive-bombing near other people’s houses dropping a kilogram of high explosive.

The latter is fairly obviously a clear and present danger to others and the former is just silly. The evidence for these seems rather clear and does not give rise to a scientific empirical question.

The question of the effectiveness of existing aeromedical regulations is a more subtle one. Thus the question of the standard of evidence to be applied in debating such regulations will become more critical.

I don’t mean it to be a distraction or red herring. But I think it is important to understand which differences in evaluation are due to different standards of evidence, reflecting a different view of the moral justification for regulations, and which are due to differences in interpreting the limited data. The former is essentially a philosophical difference; the latter is essentially empirical.

This is why I ask about the standards of evidence required for enforcing such regulations. Thus casting my query in the other light, how much evidence do you think should be required to enforce an aeromedical regulation? Is a good hunch sufficient? Some suggestive data? A preponderance of evidence they will reduce clear and present danger? Clear and convincing data that it will reduce a clear and present danger? Proof beyond a reasonable doubt?

The answer to that and it’s justification I think is inherently a philosophical or moral question. I honestly can’t see a way to avoid it. Certainly one can simply agree to a certain standard without debating it, but I do think in such subtle questions as this one has to have some agreed upon standard for discussions of the empirical questions to be truly meaningful.

I think it is particularly important in a case such as this, where I suspect you and I have a similar view of the data itself and its limitations. Our remaining differences may simply reflect different standards of evidence to be applied when trying to justify enforcement of a regulation.

 

[PeterNSteinmetz]

I agree that there are unnecessary regulations, but figuring out which ones is the hard part!

Ok, but what standard of evidence should be applied to determine whether enforcement of the regulation is justified ? Or necessary?

I think it is clear that will need to be decided on before any such discussion of necessity is meaningful and any progress can be made on deciding which is which.

 

[PeterNSteinmetz]

Avoiding being subject to a regulation does not make that regulation non-coercive. It just means that it doesn't apply to you.

...
In essence, my argument is that there is no such thing as a non-coercive regulation, and therefore the repeated characterization of a particular body of regulations as "coercive" seems like a red herring.

Ok, if it is in your view a red herring, then do you really want to continue the discussion on that point presently? I don’t think it matters a bit to the meaning of the actual arguments with respect to the efficacy of aeromedical regulations.

I suspect there is some point where we are not understanding each other about this, but it is not important to the main subject here. If you like, perhaps we can discuss further in a separate thread at a later time.

 

[PeterNSteinmetz]

Either that or you have done absolutely zero research on the matter (which is, of course, entirely possible).
...
It's actually quite a fascinating history. And it's astounding to me that someone who claims this is all pointless and irrelevant apparently doesn't know anything about it.

Firstly, please don't start making assumptions about what I do and do not know. Particularly apparently without having checked my CV, which is available online. These really are a form of personal attack because they are raising characteristics of the speaker. I do believe you when you say that you are astounded based on your current knowledge, so I will let the attack aspect of it go this time. I will attempt to explain some of the reasons that what you have cited there does not address my arguments in this thread in a following post.

I agree these are interesting questions and there is a lot of fascinating history. I'm glad you have enjoyed looking into some of it.

 

[PeterNSteinmetz]

I just took a half hour and with two google searches learned about the history of aeromedical research and certification in the United States dating to 1912. We have had aeromedical research centers conducting research on human and medical factors in aviation since 1926. Much of the early research centered on military aviation. The idea that this is all just voodoo and has no impact on safety is just preposterous. The earliest research was done on hypoxia and he effects of altitude, which resulted in regulations on oxygen use.

I don't believe I ever tried to argue that "this is all just voodoo" -- so the straw man fallacy there. Indeed, the regulations regarding the use of oxygen appear to be based on good science in the aeromedical and other fields of high altitude research.

My argument has been that there is no good data or research to show that the regulations of aeromedical certification of pilots have an overall positive effect on the safety of the traveling public. The only data which has been adduced so far in favor that it does is the study which @Stan Cooper kindly posted a reference to. It does show a small effect in absolute terms which failed to reach statistical significance. (Statistical significance is determined by the size of the samples, the deviation amongst the observations, and the magnitude of the effects, not by someone just considering it to be because of the size of the effect.) There are various possible problems with that study -- that it just looked at the rate of medical certification in fatal accidents rather than estimating actual overall rates; that it did not apparently account for age differences, which almost certainly impact the rate of medical incapacitation, etc. Those weaken further, beyond the lack of statistical significance, the ability to use that to conclude that medical certification, per se, improves the safety of the traveling public.

So the lack of statistical significance in the one small and somewhat poor study is one (not entirely technically correct) argument that there is no positive effect on traveler's safety of such regulations. The others are more general as I have noted at length in posts above which I believe you read previously.

The question thus comes down to how much evidence should there be in order to justify the use of government regulation? -- to make people do things they might otherwise not want to do through the threat or use of physical force? In cases where the evidence is mixed, the answer to that will determine how one views policy. That is why I asked @Palmpilot what his standard was. He has not responded. What is the standard of evidence you think should apply? Is a hunch good enough? Some weak data that means it may possibly be true? The preponderance of evidence? Clear and convincing evidence? Proof beyond a reasonable doubt?

Depending on the answer to that question about the standard of evidence, in the presence of weak data, reasonable people will disagree. The more interesting discussion from a policy perspective is then which standard of evidence to use.

 

[PeterNSteinmetz]

I'm no statistician, but I'd consider a 50% higher rate of incidence to be statistically significant. Furthermore, I'd love to see the per 100,000-hours flown rate for both. I'll eat my sneakers if it doesn't show an extraordinarily higher per rate for sport pilots, given the above.

Agreed the rates would be a much better piece of evidence, though can be hard to estimate because the denominator is difficult to obtain.

Given the existing data, I personally wouldn't offer to eat my sneakers, they probably taste pretty bad.

What could be done here is to go back and try to expand the dataset to get the ages of the pilots in those incidents to do an age adjustment and see if the difference continues to exist under those circumstances. If you are seriously interested in that, I am willing to help, but will ask that you do a fair amount of the work in the data gathering aspects.

 

[PeterNSteinmetz]

The reason I didn't respond is that I don't have an opinion on that, and formulating one would would require more time than I am willing to devote to the subject.

Fair enough and an honest response. It just means that it is hard to debate the policy implications of the available weak evidence. My invitation stands. I am willing to try and help improve the evidence if people want to pitch on the data gathering side.

 

[PeterNSteinmetz]

Why would anyone care enough to lie about their opinions on a message board?

Don't know either. I suspect there are a lot of people that rather than admitting they don't have the time presently to think carefully about something and come to a reasoned opinion, would just spout off whatever. I imagine we could both find some examples of that.

 

[PeterNSteinmetz]

That would only represent an 'admission' in a person's mind if they considered spending time on a particular subject to be some kind of moral obligation. It isn't.

I guess some might interpret it that way, which might account for some people's reluctance. Though certainly that was not how I intended the meaning. Here's a definition - "ad·mis·sion noun 1. a statement acknowledging the truth of something." So doesn't always mean something like admitting guilt or a moral failure. Clearly I agree that there is no obligation of any kind for someone to spend time on a subject. OTOH, I guess I would argue that just spouting off nonsense is not a good thing to do, though a fault easily dealt with by ignoring such people.

 

[PeterNSteinmetz]

Upon what do you base your allegation that this is based on "weak data" and "hunches"?

Please see post #89, which lays that out in detail, including how one might interpret a result like this which failed to reach statistical signficance. If you want to discuss that study from a scientific perspective, please lay out a case how it supports the proposition that FAA aeromedical certification improves the safety of flight, or adduce other actual data to support that proposition.

And as previously noted, saying "I think the effect size is large so it must be significant" is not a valid statistical or scientific argument. Saying "there are a whole bunch of studies about oxygen requirements" or some other unrelated aeromedical issue also is not a valid argument for the proposition above.

The way this whole discussion started was that I asked "Is there evidence that FAA medical certification improves the safety of flight, perhaps for commercial pilots?". The only actual data which has been adduced here is the study which failed to failed to reach significance. I would say that in answer to my question that is either no data (the more technically precise interpretation of such a failure to obtain a statistically significant result) or rather weak data if one wants to simply try and estimate the means. Is there some other data that can be adduced here -- we haven't seen it posted despite much discussion and a fair amount of searching by several of the participants.

Please also note that I never said that dataset was just a hunch. Rather I asked a question about the interpretation of such data in terms of policy. Near the end of post #89 I asked a general question about what you thought about policy and what sort of evidence is needed to justify a government regulation. Without an answer to that general question, it really is not possible to meaningfully interpret this evidence in terms of its policy implications. So if you don't wish to answer that, that is fine, but I don't see any point in our having a further discussion about the policy implications.