*Foolproof: Why Safety Can Be Dangerous and How Danger Makes Us Safe*

That is the new and excellent book by Greg Ip, no fluff here substance all around. From the book’s home page:

How the very things we create to protect ourselves, like money market funds or anti-lock brakes, end up being the biggest threats to our safety and wellbeing.

Here is one excerpt:

The experiment found that people with no impairment to the brain’s emotional center were much more conservative.  After losing money on one coin toss, only 40 percent of them agreed to invest on the next — but 85 percent of the brain-damaged patients did.  By the end of the game, the brain-damaged patients had earned an average of $25.70 while the healthy players averaged $22.60.

And another:

By Spellberg’s reckoning, the odds of an adverse reaction to an antibiotic, such as an allergic reaction, are about 1 in 10, whereas the odds that someone will suffer because antibiotics were wrongly withheld are about 1 in 10,000.  Nonetheless, most physicians do not want to run the risk of letting a patient suffer when an antibiotic could help…His research in Nepal produced the depressing finding that antibiotic resistance was highest in communities with the most doctors.

Spellberg thinks trying to persuade doctors not to prescribe antibiotics is a doomed strategy.  Better, he says, to develop tests that rapidly identify what bug a patient has and thus whether an antibiotic is needed.

Strongly recommended, devoured my copy in a single sitting right away, due out this coming Tuesday.  By the way here is the FT review by Andrew Hill.


1 in 10? Sounds high to me. Do they count every little thing as an "allergic reaction"?

I'm sure the risk of life threatening reactions is much lower than 1 in 10. Besides, doctors have a good idea about which antibiotics are more likely to induce serious reactions. So the risk is highly heterogeneous.


"1 in 10? Sounds high to me"

It should. See "Self-reported antibiotic allergy incidence and prevalence: age and sex effects" (PubMed)

"RESULTS: Antibiotics account for a majority of drug allergy entries. Antibiotic classes with higher historical use have higher allergy prevalence. Female patients use more antibiotics than males, and have higher allergy prevalence rates for all classes of antibiotics. There is a steady increase in antibiotic allergy prevalence with aging for both sexes. Females have higher allergy incidence rates for all classes of antibiotics. Antibiotic allergy incidence in female patients is highest for sulfas, 3.4%, compared with 1%-1.5% for all other classes of antibiotics. Antibiotic allergy incidence in males also is highest for sulfas, 2.2%, compared with 1.1% for penicillins and 0.5%-0.6% for all other classes of antibiotics."

Note that self-reported rates are almost certainly overstatements.


"whereas the odds that someone will suffer because antibiotics were wrongly withheld are about 1 in 10,000"

Only in a country where mass fatalities from infectious disease are a thing of the past, would anyone dare to write such a thing. I would be the first to argue that the decline in deaths from infectious disease is not solely (or even primarily) a consequence of antibiotics. Chlorinated water, HVAC, and the Haber–Bosch process are probably more significant.

However, only 1 in 10,000 would suffer without antibiotics?

Do we have any doctors in the house who can comment on that claim?

I think this is referring to errors where antibiotics are needed but not given--not all antibiotic prescribing.

I'm no doctor, but I have to guess that the risk associated with having antibiotics withheld depends on the case. A 15-year-old with an annoying sinus infection and a 70-year-old with pneumonia might both get antibiotic prescriptions, but only one is at much risk of dying without them.

Doctors aren't naive. No doctor prescribes an otherwise health 15-year old an antibiotic as easily as to a 70 year old with co-morbidities.

@Peter - Your skepticism is well placed on both counts. You already demolished the "1 in 10" bad outcome part. The other statistic is so vague that it's hard to refute.

Since the majority of antibiotics are given to placate entitled patients who have viral infections and who don't need them, then yes, in the vast majority of cases in the United States antibiotics are given needlessly. Is that number 9,999 our of 10,000? I can't imagine that to be true.

There are two general situations where antibiotics are appropriately given, and one where it's questionable. One, when there is a known bacterial infection. When I took doxycycline for Lyme Disease after noting a classic target lesion, I could be sure I was at high risk (though not 100%) of having a worse outcome if I didn't take it. Some patients in this category have a 100% chance of a worse outcome without antibiotics (i.e., they have septicemia), and some merely have a 'significant' chance (e.g., TB). In either case, you treat.

Two, a patient with a known infection risk who gets prophylactic treatment, as an individual with a heart valve about to get dental work. The odds of a bad outcome without antibiotics is very small, but if it happens it can be catastrophic, and so the risk/benefit swings to antibiotic treatment, even though the odds of, say, infective endocarditis from a dental procedure is unlikely even with a heart valve. It's the same reason we wear a seat belt, even though the risk of an MVA is small, and there is a tiny but real chance of being trapped in a burning car.

Three, a patient with some vague symptoms that may or may not benefit from antibiotics, and the doctor is just guessing. In these cases the odds of a bad outcome range from very small, to infinitesimal. Unfortunately, we live in a world where most patients are eager to take the risk of a bad reaction, and most doctors want happy patients who "got something" from their long wait in the waiting room.


All good points. You wrote.

"Since the majority of antibiotics are given to placate entitled patients who have viral infections and who don’t need them, then yes, in the vast majority of cases in the United States antibiotics are given needlessly."

I agree (and doctors tell me the same thing). However, the original statement was

"whereas the odds that someone will suffer because antibiotics were wrongly withheld are about 1 in 10,000"

Wrongly withheld means that they would have potentially done some good. That rules out all of the viral infection cases. Only 1 in 10,000 would suffer when antibiotics where actually needed?

I think the 1 in 10,000 sounds rather low, but it's worth noting that "antibiotics were wrongly withheld" is not the same as "antibiotics were actually needed". There are some times cases where the possibility of bacterial infection is unclear and eventually infection is ruled out. In these cases, before a bacterial infection was ruled out, antibiotics could be wrongly withheld, but ultimately not needed.

Here's one study:

" In a 20-day observational prospective study from an Italian University hospital, among 171 inpatients undergoing antibiotic treatment, 7 (4%) patients experienced an ADR, of whom two (28%) may have been allergic reactions"

That sounds more like 1 in 100.

One patient had a skin rash. Hardly life threatening.

This is asinine.

Let me give you brain damage if brain damage is so great!
If you get an infection, DONT take antibiotics!

What does "wrongly withheld" even mean? How often does this happen? Give me a break. Antibiotics are the reason idiots like you keep posting drivel, when you should have died of sepsis long ago with your head stuck so far up your rear.

Seriously. Mr. Ip isn't lining up to have his brain damaged, is he? Does he forego driving cars with anti-lock brakes and seat belts? What a stupid, stupid premise.

That Greg Ip could even think such stupid thoughts is proof that General Jack D. Ripper was right: "war is too important to be left to the generals." and medicine is too important to be left to anyone who is NOT A DOCTOR!

People who read your 1 in 10 statistic on allergic reactions, the vast majority of which are benign symptoms like urticaria, might actually die because they think they are at a greater risk of harm from taking antibiotics. Posting stupid shit like this is incredibly irresponsible...look at anti-vaxxers...we do not need anti-antibacterialers.

There is a reason medical school is 4 years long, just to learn the MINIMUM, ALMOST NOTHING of how medicine works.

You shouldn't be getting antibiotics unless there is a serious reason to believe that your long term health is in danger.


The 1 in 10 statistic sounds quite bogus & needless scaremongering.

I don't think you understand how this works. Anti-biotic usage is a net-gain for the individual using, but can be a net loss for society due to increasing resistance to antibiotics. This means that a responsible public figure would encourage less use of antibiotics, due to the fact that private decision makers will not internalize the cost to others of increased resistance. Proof that economics should be left to the economists, not to doctors.

Doctor's are the only people who care about antibiotic resistance, actually.

Economists care about patient satisfaction metrics, so if a patient is unhappy you didn't give him azithromycin for his common cold, you get penalized.

Economists should not be practicing medicine. They have no idea how much they do not know.

Doctor’s are the only people who care about antibiotic resistance, actually.

So you're an apostrophe-abusing lunatic arguing with figments of your imagination on this site who care about antibiotic resistance?

If so, that's the argument he should stick to. Not needless scare mongering about allergies.

My thesis advisor in economics did a lot of work on antiboitic resistance. Among other things, he demonstrated quite clearly that producers have an exceedingly strong incentive for partially effective measures to maximize profits, but which are ineffective at eradicating the source of the problem.

Antibiotic producers clearly are not interested in doctors prescribing less, but think doctors will come around as time progresses.

But never mind doctors. You can get this stuff over the counter in most of the world.

Why only producers? Even consumers have high incentives in getting antibiotics when the smallest risk exists to hell with the resistance developed and the population level problems in the long run.

Doctors too are more rewarded / penalized by their impact on patients rather than how much they do to prevent antibiotic resistance. Why risk that one guy in 1,00,000 with no obvious risk factors where because you didn't prescribe an antibiotic his cold turned into pneumonia?

'Allergic reaction' as in 'gluten sensitive'?

Maybe [most of the women -and they're mostly women - who order 'gluten free' in cafes, aren't].

But, real allergic reactions? One in ten?

Not credible.

Yes, the majority of cases of "gluten sensitivity" are actually misdiagnosed chronic Lyme disease.

"an adverse reaction to an antibiotic, such as an allergic reaction, are about 1 in 10"

I think 1 in 10 here is for 'any' adverse reaction, not just allergic ones. Now that I'm more informed, I certainly wouldn't take antibiotics for anything other than an acute or life threatening infection. When I was in my early 20s I went to a clinic asking about a lump near my parotid gland. The person who I thought was a doctor and turned out to be a nurse practitioner prescribed antibiotics without doing a biopsy or any other tests and shortly thereafter I developed lactose intolerance.

I'd recommend a class on Causality.

As a full time trader, I can relate entirely with the comment on brain damage. Fear is such an incapacitating emotion which prevents us from following what we truly know is the the best course of action. To be successful in the long run, we need to develop intelligent strategies, and then follow them like blind morons.

I've done more damage to myself by overriding my systems and rules - especially exit rules - than I can to admit.

Couldn't pull the trigger on that mental stop-loss in the great gruyere trade-off of '09?

Fighting antibiotic resistance by massively reducing the use of antibiotics is probably one of the lowest hanging fruit in global public health. It is easily implemented by outlawing over the counter access everywhere and drumming it in to doctors that it is only to be used in the worst cases. Also, a highly repetitive and assertive lecture is needed to ensure that patients finish their course of antibiotics even if they feel better.

Not low hanging at all. It needs fighting that inherent human tendency to maximize personal benefit. And thinking with short horizons instead of long.

OTC access is not an issue in the West. If you want to make OTC access disappear in the developing world by mere diktat you have no clue of ground reality. Ain't happening.

I acknowledge the problem. All I'm saying is there's nothing low hanging about this fruit.

He means cutting antibiotic use by 80% by prohibiting added antibiotics to livestock feed, and then banning antiseptic chemicals in common cleaners.

Those use of antibiotics have no purpose other than profit yet cause significant public harms.

No, the mulp, he did not mention livestock, nor did he hint that he was referring to them. Stop posting until you can be right about something for once.

Not talking about livestock.

I've bought over the counter a couple times in developing countries. Acknowledged that it might not be that low, but I think it's hanging low enough.

Let's see, if safety equipment is making life so dangerous, why aren't more Americans dying in car wrecks? -

'The number of deaths – and deaths relative to the total population – have declined over the last two decades. From 1979 to 2005, the number of deaths per year decreased 14.97% while the number of deaths per capita decreased by 35.46%. In 2010, there were an estimated 5,419,000 crashes (30,296 fatal crashes), killing 32,999 and injuring 2,239,000.[2] The 32,479 traffic fatalities in 2011 were the lowest in 62 years (1949). Records indicate that there has been a total of 3,613,732 motor vehicle deaths in the United States from 1899 to 2013.' https://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_in_U.S._by_year

Wait, this is a GMU econ professor run web site - of course it has little interest in empirical reality based on actual data, and no time to type 'american auto fatality rate' into google, and then read the first sentences of the top result.

This time you're right. These statistics get even more interesting when you realize the first Corvette had 150 hp compared to a 2015 Camry with 180 hp. When you consider an entry level pick-up truck today has more than 300 HP, it's remarkable that people don't die more.

Engineers should have done something with cars and roads to reduce fatalities since the humans behind the wheel are as capable as drivers from 1979.

The idea that "horsepower kills" is Naderite tripe. There is no evidence for it, and there never has been any.

Cars like Camaros have higher death rates because of who drives them, not because of the horsepower.

This is actually the kind of stuff that Ip is talking about.

Insurance people know something, they will charge more to insure a car with more horsepower no matter who drives.

For motorcycles there are interesting statistics. There are two populations: sport and supersport bike riders. Average age of fatally injured rider: 27 and 30 (year 2000)/34 (year 2005).Rider death rate per 10K registered motorcycles: 10.7 and 22.5 respectively. Supersport bike riders have twice the fatality rate of sport bike riders. Unless there's better data, it seems the only difference between sport and supersport bikes is extra horsepower and bike stability that lets you ride faster before falling.


I like fun cars with powerful engines but the numbers are there. I don't know where's the equilibrium point, perhaps 100 bhp for a 1500 kg empty car, or perhaps 300 bhp. But it's sure that a 1,000 bhp road car is not a good choice. Car races in tracks should be more popular.

One guess here is that the risk that people are worried about when driving is "I'll wreck my car" rather than "I'll die wrecking my car." That would make risk balancing behavior not really apply so much to things like airbags and improved EMT services that bring down the death rate from car wrecks. You'd still expect some effect from ABS and 4WD systems. (And anecdotally, an awful lot of people really do seem to think a 4WD car means they can drive fast on icy roads. Turns out, they can get going on icy roads pretty well, but they can't *stop* too well.)

You're still assuming the same people are riding sport and supersport bikes, they're not. Also how insurance is priced varies by provider, some price by engine displacement which is not very well correlated with horsepower and some price by motorcycle type, I've never heard of an insurance company pricing by horsepower. Before you try to argue the link between HP and displacement I'll point out a HD Fatboy has a 1690cc v-twin with 78hp and a Suzuki GSX R1000 has 999cc of displacement and produces 182hp.

@ZZZ - exactly.

A lot of this is simply self-selection - thrill seekers are gonna get the Gixxer6s of the world - while Gold Wing riders tend to be older and faaaaar more conservative - Harley riders somewhere in the middle.

Same with cars - you don't see much overlap between those who choose Camrys and those who wish to drive Hellcats.

Insurance people know something, they will charge more to insure a car with more horsepower no matter who drives.

I drive a car with, IIRC, 390 HP (A Mercedes S550). I pay $600 a year for insurance. When I changed from a 200 hp sedan to it, my cost only went up a bit, accounting for the higher value of the car.

No, they're not charging a detectable premium for HP, but they're charging a premium on high HP cars that young males will tend to buy

Agree with cheesetrader. I would definitely bet on self selection rather than HP itself.

@p_a: I wonder how many auto-induced fatalities (from pollution) will result from the typical German lying and cheating over at VW

"Behind these policies is the superstitious belief that vigorous physical contact and make-believe violence will beget immediate and future real physical harms — magical thinking that fundamentally misunderstands how children play and learn. Prohibiting rough-and-tumble play doesn’t make recess safer or kids less apt to hurt others. To the contrary, the bans deprive children of the very experiences they need to master peaceful social interactions.
Contrary to what squeamish authorities seem to think, it’s the kids who don’t engage in rough-and-tumble play who actually tend to be more violent later on in life. So, says Carlson, forbidding playful physical contact “stokes the fire as opposed to diminishing it.”


This is fun until a bullied kid takes a stand in the only was he has learned: vigorous physical contact. It happened to me in junior high school. The only time I decided to fight the bully back and hit him, he cried with teachers and I ended up being expelled from there. In the long term, it was a great lesson: adults are hypocrites =)

Thats not a bad lesson to learn. Better lesson, people who run public schools are stupid.

Some just what a trouble free day.

Oh Tyler; if you hadn't cancelled your order of Antifragile, the material would be mostly old hat.

Agreed -- Antifragile rehash.

Bunkum is rare on MR but this is a steaming pile of bunkum. It is a farrago of made-up statistics, illogical claptrap and ideological blindness.


See if you can guess when seat belts became compulsory.

I notice a theme in several posts: how important it is for doctors to know the correct odds and apply them in treatment.

However concerning the risks of bacterial resistance, I think overuse in animal hsbandry is muh more important and could be addressed with a pigou tax.

Many of you are poking at claims Ip never makes...

They are poking holes in IP's books as you have portrayed him. If they are hitting strawmen, the fault lies with you.

The Long-Term Effect of ABS in Passenger Cars and LTVs by NTHSA

"ABS has close to a zero net effect on fatal crash involvements. Fatal run-off-road crashes of passenger cars increased by a statistically significant 9 percent (90% confidence bounds: 3% to 15% increase), offset by a significant 13-percent reduction in fatal collisions with pedestrians (confidence bounds: 5% to 20%) and a significant 12-percent reduction in collisions with other vehicles on wet roads (confidence bounds: 3% to 20%). ABS is quite effective in nonfatal crashes, reducing the overall crash-involvement rate by 6 percent in passenger cars (confidence bounds: 4% to 8%) and by 8 percent in LTVs (confidence bounds: 3% to 11%)."


So, the driver has the same fatality rate with ABS or not, but externalities such as fatal injuries to pedestrians or collisions with other cars are reduced. It seems the best way to describe ABS is "externality minimizing device"......such a failure.

Or TC has accurately portrayed the book and it's thesis is easily refuted.

Tyler, you don't actually expect us to read the book, do you?!

Perhaps in the coin-flipping test, the higher average for brain-damaged folks was achieved due to the long-tail effect? The bulk were (slightly) worse than the conservative folks but a few outliers pulled up the average? Sort of the St. Petersburg paradox put into practice.

Or not.

At any rate I'll be ordering this book!

Isn't this just another defense of the rational actor model promoted by libertarians. Drivers become more reckless when disc brakes and guardrails are installed because they are rational actors not because they aren't. Here's a quote from a 2012 book by Daniel Kahneman on the differing world views of those who adhere to the rational actor school (as opposed to the behaviorists) that I found this morning at DeLong's blog: "I once heard Gary Becker [argue] that we should consider the possibility of explaining the so-called obesity epidemic by people’s belief that a cure for diabetes will soon become available…" See, bad drivers and obese people are acting rationally: if people believed they wouldn't be protected from their own bad behavior (as for example by disc brakes and guardrails or a cure for diabetes), they wouldn't drive recklessly or be obese.

Seems crazy to say that drivers are worse without mention plunging death rates. If anything, drivers are acting semirationally by being fearless.

Those bad drivers on the 405 expect to walk away from any mistake.

There is no Amazon link and the book is from Hachette... Has this dispute rekindled?

I found the same thing, but if you go to the Amazon site and search for it, it's there.

The links aren't working for me, and I don't have an FT sub, so I will just give a reaction based on the above.

This looks like a way to recast Thaler and Damasio into a "new" and "surprising" freakonomics.

As with the decline of the Freakonomics franchise it might be looking for surprise more than wisdom.

"By Spellberg’s reckoning, the odds of an adverse reaction to an antibiotic, such as an allergic reaction, are about 1 in 10, whereas the odds that someone will suffer because antibiotics were wrongly withheld are about 1 in 10,000. Nonetheless, most physicians do not want to run the risk of letting a patient suffer when an antibiotic could help "

The usually excellent Ip is being sloppy. As a physician, that 1 in 10 stat is silly; we generally only care about anaphylactic allergic reactions. Ip does not mention the odds of anaphylactic allergic reactions (hint, they're very different) vs the generally benign "allergic" reactions that get bundled into the statistics. Five minute headaches after a dose of levofloxacin are not a big deal. If I fail to prescribe an antibiotic and my patient dies, or suffers permanent organ damage...that is a big deal.

And of course communities with more doctors have more antibiotic resistance. Why on earth would it be otherwise? They also have less people dying of easily treatable infections.

Val DeLa,

It would appear that you are a practicing physician (true?). Does

"whereas the odds that someone will suffer because antibiotics were wrongly withheld are about 1 in 10,000"

Sound correct to you?

"It would appear that you are a practicing physician (true?). Does “whereas the odds that someone will suffer because antibiotics were wrongly withheld are about 1 in 10,000” Sound correct to you?"

I'm not a physician, but I can search the Internet, and cipher nearly as well as Jethro Bodine.

Here's a webpage (http://www.livescience.com/23100-sore-throat-antibiotics-guidelines.html) that says:

"About 15 million people in the United States see doctors for sore throats every year, and up to 70 percent receive antibiotics, but a much smaller percentage actually have strep throat: approximately 20 to 30 percent of children, and 5 to 15 percent of adults."

So let's say approximately 10 million (70 percent of 15 million) receive antibiotics for sore throat. And let's say 70 percent of those who receive them are adults, and only 10 percent have strep. So that's about 700,000 adults with strep who get antibiotics. And the other 30 percent of them are children, and 25% of them have strep. So that's about 750,000 children with strep who gets antibiotics. So the total is 1.45 million people with strep who get antibiotics. Let's further make the very conservative assumption that only 20 percent of them really needed the antibiotics (even though they had strep). So that's 290,000 people who had strep and really *needed* the antibiotics. If the one-in-10,000 is true, and if *every single other instance of antibiotic usage was unnecessary* then there would be 2.9 billion administrations of unnecessary antibiotics every year.

Seems unlikely.

P.S. Then there are the ~100,000 MRSA infections every year in the U.S. http://www.mrsasurvivors.org/statistics. All people who receive antibiotics for MRSA probably need them. So applying the 1-in-10,000, that would be another approximately 1 billion unnecessary antibiotic administrations.

Why would billions of unnecessary administrations annually seem unlikely? Some people ask for an antibiotic every time they get a runny nose. And then stop taking it two days later when the runny nose goes away ... a major cause of resistance, since failing to kill off the entire bacterial invasion before transmitting it to someone else increases the odds of being the host of a new strain of bacterial resistance.

You really think it plausible that each man, woman, and child in the country averages 13 unnecessary antibiotic prescriptions a year?

Thanks for the point. For some reason I skipped the part about it being just about the US and was thinking about the global picture.

But I think he's mostly playing numbers games with incomparable data/stats.

I think mostly missing from the picture is that yes, antibiotics are rarely actually necessary, but this does not account for people being able to get back to work sooner. If people have even just 1-2 fewer sick days per year due to taking antibiotics (in the short run, ignoring risk of antibiotic resistance) then the numbers start to tilt quite heavily in favour of taking more antibiotics than strictly needed from a life savings perspective.

I believe dangerous is only when we are not assure of our ability and activity, I am working with full confidence in Forex market and that is with OctaFX broker which is a really first class company with having all top benefits which includes low spread of 0.2 pips, high leverage up to 1.500 plus their fairy tale platform cTrader, it completes the whole package for me and allows me to be successful without having to face much trouble at all.

FwvByPwVHKgWzVXiOc 7730

Comments for this post are closed