Physician, Disinfect Thyself

We spend a good bit of time in SuperFreakonomics writing about doctors’ hand hygiene: specifically, how important good hand hygiene is in order to cut down on hospital-acquired infections and yet how historically it has proven difficult to enforce. That’s because behavior change can be hard in general, but particularly so when it doesn’t necessarily benefit the people who are actually changing their behavior.

So it turns out that some of the most effective means of discouraging infections don’t rely so much on behavior change; as we write:

Among the best solutions: using disposable blood-pressure cuffs on incoming patients; infusing hospital equipment with silver ion particles to create an antimicrobial shield; and forbidding doctors to wear neckties because, as the U.K. Department of Health has noted, they “are rarely laundered,” “perform no beneficial function in patient care,” and “have been shown to be colonized by pathogens.”

If this topic interests you, then you certainly want to check out a recent Wall Street Journal article by Rebecca Smith, which offers a more in-depth examination of the role of the necktie in passing along germs. One new solution for doctors who just can’t bear to part with formal neckwear: the antimicrobial necktie. (Hat tip: Ted DiSante)

But here’s a much more interesting and broad-reaching suggestion, from a reader named Larry McColloch:

In the section on doctors washing hands, you apply economic theory and approach. It would be much better to view it as a “control systems” problem. In this regard, the doctors were operating “open loop” in that the germs on their hands are not visible. What really happened with the hand-culture experiment was that the results allowed the doctors to see the germs that were present; in control-loop theory, this is “closing the loop.”

You might want to apply the idea of feedback and control theory to many economics problems. No doctor would contaminate a patient with potentially killer germs if they knew they were contaminated regardless of economic implication. What is missing in the hospital is feedback. You are also pointing out a product for someone to make money with. If someone can make a device that easily lets doctors see germs on their hands, it will immediately change habits in the hospital.

The control systems perspective also applies to many of your other topics, such as climate change. What is lacking in both understanding and fixing climate change is some kind of feedback mechanism, although I don’t know how to get appropriate feedback.

I agree with Larry: feedback is hugely important in problem-solving and, unfortunately, a far-too-scarce commodity. That’s one reason the Internet is so powerful: if you can find a way to filter out the noise, the Internet is one of the greatest feedback loops imaginable, at least for certain kinds of problem-solving.


I've wondered why, with our hand gels and so other anti-bacterial stuff, that instead of the involved, time-consuming (and perhaps half-heartedly done) "scrubbing," our surgeons/doctors could just either submerge their arms and hands in a tub of the hand gel...OR have up-to-the-elbow latex gloves that are largely filled with the gel. So you just put your hands in the gloves (effectively "submerging" them in gel).

Wouldn't that be faster and just as effective?

Don the libertarian Democrat

Ask your doctor to wear gloves when he examines you.


Larry's observation has much truth in it, but it's not directly on target. Doctors are not stupid. They know with 100 percent certainty that there are germs on their hands when they fail to wash them enough. But when they can't see the germs with their eyes, the whole thing exists only as an abstraction in the mind that isn't powerful enough to make many doctors take action. Making the germs visible doesn't give the doctors even a shred of new information, but it might be enough to force them to make use of what info they have.


AaronS - nope!

Removing the things that the cleaning products cannot kill is far more important. Good old soap and water will do that best.

The alcohol gel is great for some bugs, but you should also wash your hands.

I agree that a feedback mechanism would be good. On the spot culturing of medical staffs hands with the results shown on hosptal website maybe?


Hand sanitizing gel (that is 60% alcohol) is largely effective, but does not kill all microbes. For example, it does not kill Clostridium difficile, a major problem in many hospitals.
Also, hand sanitizing is often performed inadequately. Medical professionals need to learn how to apply hand sanitizer appropriately and how often it can be applied before the hands need to be washed.


Prevention of the spread of germs should not be the sole responsibility of physicians but should be expected from everyone in the less than sterile healthcare environment. First, start having each patient, caregiver and visitor wash their hands before entering the premises and require all healthcare professionals to do the same throughout the day.


AaronS - It's not as effective.

Timothy K

@ AaronS There are water-less hand scrubs used in many ORs in place of the classic soap and water scrub. One example is the product Avagard, which is very similar to consumer hand sanitizer gels. I'm sure there are other products that are similar.

The issue is that there are some organisms that still need soap and water to be effectively removed from the skin, and so the first "scrub" of the day should always be soap and water.. The best example is Clostridium difficile ("C. dif") a common cause of hospital acquired intestinal infection.


How about the hand sanitizer being in a dispenser on the exam room door. Use technology to create a system which will not allow the door to be opened until the sanitizer is dispensed.

Names changed to protect innocent ;)

THE FEEDBACK LOOP WORKS!! For years, whenever I told my husband he had bad breath - he grumped that he did not have bad breath, and that I was oversensitive, and that heck yeah of course he brushed his teeth. But last Christmas, I got one of those "Breathalyzers" - but not for alcohol - but bad breath. The first Saturday, we were out, ,and I used it to check my own breath (a happy face!) and then offered it to him. He declined - saying he'd forgot to brush his teeth that morning. So I asked, - "but you did last night right??" and he hemmed & hawed and ... "uh - no." So when did you last brush your teeth? "Yesterday morning." I replied "well, at least you manned up and are being honest, and that's something. Thanks! . You need "hand-a-lyzers" Knowledge is power.


Feedback is the key. Maybe I'm watching too much CSI but having a product or a device that would instantly show how clean your hands really are would go a long way towards changing behaviors.

If that can't be done, maybe making surgeons rub their hands in petri dishes at the entrance of ORs and putting those on display for a week with the name of the physician. Delayed feedback is the next best thing... It doesn't have to be public, but if the surgeon and his colleagues can see it, you can bet they're gonna be sterile by the end of week 2.


The rubbing action kills far more bacteria than just submerging in goop would.

And while I agree that there is no reason for doctors to wear neckties, I have yet to see any evidence that getting rid of them has done any good in reducing infection rates in the UK. The hands are far and away more important.


Whoa whoa WHOA.

That last bit that Larry winded past, the one about climate me, it seems like an incredible opportunity. If there was a way to effectively calculate how the climate around us is changing because of global warming, and making it mainstream, it would be a success in helping save the world.

Sure, some would argue that the fact that global warming is occurring is already very mainstream, but just how many people now real, solid, recent statistics? Aside from "the Earth is getting hotter"?

Gadget companies would benefit by getting sponsorship from environmentalist organizations to develop modern, cheap technology, items that could be bought by many people. A gadget that notes and informs the user how he or she is contributing to global warming, plus a list with effective methods to stop, would be a success in the "green-head' market.


I'm a physician, and I wash my hands before and after each patient. Still, the evidence that handwashing reduces hospital infections in our era is only so-so (Semmelweis made a big difference in obstetrical infections with handwashing in the 1800s--no gloves then for procedures docs would always glove for now). Sure there are germs on doctor's hands, and fewer after washing. And the germs in some hospital infections are the same as those on doctor's hands. But showing that germy hands actually cause hospital infections is difficult. Studies show modest reductions in hospital infections as a result of increased handwashing, and that's enough to warrant handwashing. But an economic analysis of hospital handwashing should at least consider the possibility that the low level of evidence behind handwashing isn't enought to motivate doctors.


It is interesting that physicians have been singled out as those responsible for cross-infection through poor hand washing habits. The truth is that the physician effect is likely to be quite small compared with all the other personnel that comes into contact with the patient, especially the nurses. And if you think nurses are any better at hand-washing I suggest you spend a day observing a ward activity in any hospital.

Furthermore, try finding a sink in some of the older wards to wash your hands in. You may have to travel some distance and then find the sink full of coffee cups, discarded IV sets or some such.

Most physicians tend to visit their patients only once a day and sometimes do not even touch them which brings up a whole different set of issues...However, there is no excuse for wearing a tie in this day and age.


When my wife gave birth, it was in a nice controlled environment. Except for the computerized system they were introducing into the hospital (or at least the maternity ward).

It was on a cart, and was rolled into the room. Nurses would walk over to it and type on it, and ask other people for assistance (since they weren't fully familiar with it yet), who would type on the keyboard. I don't know where it had been before it got into our room. I do know I never saw it sanitized, or anybody cleaning their hands after using it.

I don't know if it actually represented a problem. It didn't for us, as none of us got sick.


Great article.
What I believe is that people haven't found a strong enough incentive to wash their hands. This is because even though we NEED to do it, people see it as a POSSIBILITY.

Its kind of like putting sunscreen on just because your mom says so. You really don't want to put it on because it takes up your time and whatnot but you do anyways but because your mom gave you the incentive. Now lets say you grow a little older and you found out the hard way. You go out and don't put sunscreen on. The next morning your back is super red and it burns like hot potatoes. By that time, its too late and it is a good incentive experience to start wearing sunscreen.

This however cannot be tolerated in the medical department because one of those 'learn the hard way' could actually cost someones life. Thats why, as McColloch said, you must show visually what potential danger you have in your hands. Like that, you will know washing your hands is actually doing something and that you in fact have potentially dangerous germs.



Feedback is all well and good, but unless it's automatic, that just moves the behavior change problem over one square. In other words, if you can't get docs to scrub, you won't be able to get them to use the feedback machine, either.


This is an incredibly legitimate viewpoint brought about by Larry. I believe this to be a controversial issue; however, since we cannot actually know the true implications of doctors. A doctor in his practice may be like any other first one learns about in Economics 101 whose purpose is to maximize profit. Some doctors might also believe that the end justifies the means.

What would happen if doctors were purposely utilizing ties and avoiding washing their hands in order to purposely make their patients sick thus assuring an increase in demand for the services they provide. An increase in demand while supply remains the same will inevitable raise the equilibrium price as one's health is an inelastic "good". People are willing to pay anything in order to be in pristine health.

I believe that hospitals should enforce certain regulations forcing all their doctors to comply in order to reduce the risk of being infected inside the hospital. Personally, I hate going to hospitals because every time i breathe I feel I'm breathing in a different sickness. Anything hospitals can do to change t his is greatly beneficial and might possibly raise their demand as an unintended consequence. Seeking to simply reduce infections, people that think like me and resist going to the doctor unless it is imperative due to a fear of secondary infection will be more likely to go to the doctor for less impending threats and will surely increase hospital revenue. What will it be?



As a route of transmission, I wonder how the doctors' hands compare to, collectively, the door knobs, elevator buttons, restroom controls, arm rests on the waiting room chairs, etc. I think the doctors' hands might be less of a concern because I imagine the likeliest path of disease transmission is eyes, mouth or nose ("EMN") to hand to hand to EMN, and patients and doctors are more on guard about the doctor touching the patient in the EMN than the patients are about touching themselves. On the other hand, there probably isn't nearly as much risk in the doctor feeling your knee or shoulder. If this is true, installing automatic doors, voice activated elevators, automatic water,soap and towel dispensers in the restrooms (with outward exiting doors so that you can open them with your feet) might give more bang for the buck.