Articles

Epiphany

The Institute of Medicine report is a frequent ‘rebuttal’ to science based/real medicine. The argument is usually phrased something to the effect that since medicine can be dangerous, SCAM’s are legitimate. Of course, one does not follow the other. It is the equivalent of saying since you are old, bald and pudgy, I am young, have a full head of hair, and are thin. If every doctor and hospital were to vanish tomorrow like an episode of the Outer Limits, SCAM’s would be just a ineffective.

Despite the flawed logic of the comparison, I have always had an affinity for the estimates that 44,000 to 98,000 were (note the deliberate use of the past tense) killed each year in hospitals. There may be methodological flaws in the estimate but the ballpark figure is probably correct.

In 1999 there were 5000 hospitals in the US. Just one death a quarter would bring the number of deaths to 20,000, and one death a quarter is not that many deaths. Lest I sound hardhearted, everyone dies, 2.5 million a year, and often death occurs in the hospital. Against the background of the mortality of existence, a few ‘extra’ deaths would be lost in the background.

For an individual doctor, it would be indistinguishable against the background death rate in the hospital. What makes it even more difficult to track and recognize excess mortality is that each death may be due to a different breakdown in medical practice.

Amongst my many jobs is Infection Control. For twenty years I have chaired Infection Control for both the Legacy Health, a collection of 5 hospitals in the Portland-Vancouver area, as well as for Portland Adventist Medical Center. As Chair I get the joy of sitting on many other committees such as Quality Council and Pharmacy. I know all the way hospitals could kill and the endless efforts to try and improve and perfect medical practice to avoid these complications.

In 20 years of investigating outbreaks, hospital acquired infections, and deaths, I have yet to see two infection related deaths that are due to same cause. Every infection was reviewed and evaluated as a potential for improvement, and I think we practiced the best medicine we knew of at the time.

With one exception, the universal horse shit compliance with hand washing that was the norm 20 years ago. It always boggled my mind that it was difficult back in the day to get people to wash their hands. The information on efficacy was only 150 years old, after all. But otherwise we practiced state of the art medicine. With the perfect vision of hindsight, I can see that state of the art left much to desired. We didn’t have the studies to guide practice that we have now, and I anticipate that 20 years from now I will be rolling my eyes at how we practiced in 2010. I will sound just like Bones McCoy wandering throughout a 1980′s San Francisco hospital grumping about the butchers of the past and hoping I do not run into T.J. Hooker.

Unlike the hodgepodge of practices that comprise SCAM, medicine changes and mostly for the better. Change is always slow, and always painful, and more difficult to implement than one could ever anticipate, but if you read the medical literature, you have to change.

Hospital based medicine is mind bogglingly complex and difficult, and humans are limited in their ability to always function perfectly. The Institute of Medicine knew what it was doing when they entitled their report “To Err is Human.” And all too often we were not able to pinpoint a breakdown that lead to a complication or death

When I started practice back in the last century, I would have thought that hospital acquired infections were part of the price of taking care of ill and compromised patients. Sure, we can minimize infections, but wound infections, ventilator pneumonias, and line infections are going to happen. You can’t do the things we do to people and NOT get an infection.

Right?

Wrong.

What both administrations at my hospital systems have in common is a commitment to patient safety and over the last decade they have committed considerable time and money to the application of proven procedures to decrease infections and other complications of hospital care.

You cannot know best practice based on individual experience. I like to tell the residents that the three most dangerous words in medicine are “In my experience.” You need large numbers of patients and studies to guide practice. The last 15 years have seen a large number of clinical trials aimed at discovering what is the best practice to prevent everything from line related infections to deep venous thrombosis. Dozens of science based investigations whose goal was to improve patient care in the hospital, and my hospitals aggressively applied them.

The first intervention was the use of alcohol foam instead of hand washing. I have in my mind, and cannot find the reference, that if a nurse would wash her hands appropriately after every contact, he would spend 80% of their shift washing hands. Soap and water, it turned out, was not a practical solution to keeping hands clean. It is too time consuming in a busy work day, despite it’s proven efficacy in preventing infections.

Alcohol foam can be used in a fraction of the time with superior results since it is much easier to foam frequently. And once I discovered it was not to be used orally like cheese whiz, the results were even better.

The foam is now ubiquitous in the hospitals. Even when the use of the foam was 20%, the overall infection rate in the hospitals fell by half when compared to rates with hand washing. Then, over the next decade, the hand hygiene compliance rate has steady increased to around 90% and there was a corresponding steady decrease in infections. It took over a decade of consistent work and a lot of trial and error to get the rates to 90%.

At one hospital the limiting problem was no foam outside the rooms. No one would walk an extra few feet to get to the foam. But at another hospital the fire Marshall said alcohol foam in the halls was a fire hazard and we could not put alcohol in the halls. He was eventually overruled, but what are you going to do in the meantime?

And I could go on for paragraphs about the issue of finding product that minimized the number of HCW’s whose hands where turned raw by the alcohol.

Ninety percent seems to be the best we can consistently achieve with the current program for hand hygiene, and we are puzzling over how to get the rates to 100%. One approach is the “It’s ok to ask” program, where patients are encouraged to ask their provider if they washed their hands. I asked a series of patients if they would ask, and they uniformly said no, they did not want to risk angering their health care provider. I agree. It is important not to piss off the person providing your morphine. Besides, would you fly on an airline if their motto was “It’s ok to ask if the landing gear is down.”

Last year I ran a red light. It was 7 am, I was taking the kids to school and I have to make a right then an immediate left across four lanes of traffic. I make this turn everyday. I am talking to the kids and I look several times, no traffic, and make the turns.

What I did not see was that the light was red nor did I see the cop stopped on my left. I was so intent on the traffic I missed two key features in my environment.

It is the main reason, I suspect, that we cannot get hand hygiene to 100% every time, every where. The hospital has too many opportunities to focus our attention elsewhere that, for the short term, allows us to forget to foam.

Somehow, and I do not know how, I suspect we need to make foaming the default rather than optional; then our rates will get to 100%.

But foam is not the only intervention my hospitals have implemented.

Surgical check lists, best practice bundles (collections of proven interventions gathered together) to prevent ventilatory pneumonias, to prevent intravenous catheter related infections, to prevent urinary tract infections, to prevent deep venous thrombosis. Innumerable checks and balances with pharmacy to prevent medication errors.

Simple things to prevent surgical wound infections but logistically difficult to get to 100%: timing of antibiotics to within an hour of cut time, no shaving the surgical site, not letting the patent get cold post op, and tight glucose control were are associated with decreased wound infections. Next up may be no staples with orthopedic cases as a recent meta-analysis demonstrated fewer infections with sutures. That will be fun, getting surgeons to alter practice.

Over the last year my hospitals have implemented dozens of practice improvements based on the medical literature to improve outcomes and the results have been amazing. Practices that were not effective were abandoned or modified, sometimes going through multiple iterations until were discovered was worked and was practical.

As a result, at Legacy we have prevented over 200 deaths (12.5% reduction in non-risk-adjusted mortality rate, which is now 1.47% for our system that includes a regional trauma center and regional burn center as well as two NICUs, oncology program, and multiple other high-risk programs) and over 570 prevented infections (39.5% reduction in whole-house infection count) above historical data. And that is over the most recent 24 months. At Legacy it is estimated we have also saved 8 million dollars in associated costs.

A few of the hospitals have gone a year without a ventilator associated pneumonia or a catheter related infection. Every year has seen a decrease in the healthcare associated infections and other complications.

That is 100 deaths prevented a year for 5 hospitals. Multiply that for the remaining 4,995 hospitals in the US and the IOM estimates for last century seem reasonable. But not for this century and not for the decades to come.

I used to think that infections were inevitable, but no longer. There is the occasional patient who will get an infection: the badly burned, the multiple trauma. But even the trauma ICU had a marked decrease in all infections with increased infection control compliance. We had a wound infection in a 400 lib patient who literally had dirt tattooed in the palms and soles and a Hemoglobin A1c of 15 who required emergency surgery. I was not surprised that patient developed an post-op infection. We did everything correctly and still had a complication. Sometimes the barriers we have to overcome to prevent infection may be too great, but it does not stop us from trying.

But the experience of Legacy and Adventist demonstrates that aggressive adherence to proven infection control works and that the majority of health care associated infections and deaths need not happen.

I have three epiphanies in my life: my first great meal (at a restaurant called St. Estephe’s), my first great Bordeaux (oddly enough, a St. Estephe’s), and when I realized that most infections in the hospital need not happen.

This has been a real decrease in infections and death, not just playing with numbers to look better. These hospitals look at every healthcare associated infection (HAI) as an improvement opportunity and do not sweep data under the rug.

I also know personally that the numbers are real. I used to derive a significant portion of my income from hospital acquired infections. There are many reasons why my income has declined by 60% over the last decade, not the least of which being large numbers of patients that used to make up my practice (HAI’s, AIDS) have disappeared. The majority of those 570 prevented infections would have been consults. I feel like Phillip Morris making stop smoking ads.

It was not easy making these changes; it took years of committed work. People are like oil tankers and change course slowly. And some are filled with toxic waste. An interesting aspect of instituting the policies and procedures has been who fought against the changes the hardest. Docs. Not all of them, just a small subset. There is a curious subset of MD’s who feel that the data does not apply to them. They do not need to follow infection control procedures, use full barriers when placing a line, or even wash their hands. And I do not get it. I cannot figure out why some docs are so recalcitrant about doing the right thing, the proven thing. Eventually everyone complied, but some people made it more painful than it needed to be.

My hospitals made a serious commitment to providing the best care as determined by the science. It was not simple and required a surprising amount of creativity and time to apply the evidence to the real world. But the nice thing is that when you apply science to problems, you get results. Science works. Quality initiatives work. Next time you point out the deaths caused by modern medicine, leave my hospitals out of it.

It makes me wonder. There are numerous naturopathic, chiropractic, and other alternative schools and clinics involved in patient care. I am sure that they too have numerous quality improvement studies to brag about that have improved patient care and outcomes.

Think of all the practices in medicine that, eventually, have been demonstrated to be worthless, or dangerous, or flawed and that were improved or abandoned for the betterment of patient care.

So let’s start a list, shall we. The following is the top 10 list of alternative medical practices that have been modified or abandoned because of studies that demonstrated they were ineffective or dangerous and the quality initiatives that have improved patient care:

1) Disposable acupuncture needles (thanks to wales)

2)

3)

4)

5)

6)

7)

8)

9)

10)

Sorry. I found nothin’.

Perhaps it is a reflection of the perfection that is alternative medicine. Alternative medicine practices change based on evidence? No need.

Of course, I may well be wrong. As the board president of the Oregon Association of Naturopathic Physicians states, “Both MDs and NDs are trained to work from the evidence-based model of medicine, using best practices and standards of care.” I suppose my inability to find examples is due my inadequate Google and Pubmed skills to find the readily available information. I would have my 13 year old do it for me, but he is on a trip. Even if only 25% of medicine is science based, that is still 25% more than alternative practices.

Please, please, please, someone show me up. Hell, just give me some hand hygiene improvement data and let me know that, if nothing else, there is an understanding of germ theory in the alt med world. Ever since my local paper, the Oregonian, printed a picture of the local Natural Medicine School teaching acupuncture without gloves, I am not so sanguine about that understanding. It still gives me the willies to see that photograph and it looks for all the world that there are two boxes of gloves in the background, so I know they have them. It may be that all the gloves are left handed or right handed and so cannot be worn. Sometimes I pull out a glove for the right hand and it is a lefty glove and then I pull out a glove for the left hand and it is a righty, so I cannot find a pair to wear. It’s a problem.

Medicine slowly improves, too slowly sometimes. I know that 20 years ago we did not have the information to inform our practice that we do now. We did the best we could with what we knew at the time, and we do the best we can now with the information we have today. Still, despite the impressive improvements, it is a bittersweet victory. I can’t help but think what could have been, if only we had known.

Posted in: Public Health, Science and Medicine

Leave a Comment (57) ↓

57 thoughts on “Epiphany

  1. Rogue Medic says:

    Hand washing is only a part of patient contact. It is the most important part, but it is only a part.

    Patients need to be moved. When patients are moved, more than just our hands come into contact with the patient. It is not common to put a gown on before moving a patient. I have seen some ICUs that try for 100% compliance on gowning up before patient contact, even on the non-isolation patients.

    After all, it is the non-isolation patients we do gown up for. The isolation patients are already infected. We are just trying to avoid transporting their germs, not just the drug resistant germs, to non-isolation patients.

    We need gowns that work better, or drapes to put over the side of the bed to act as a barrier between the patient and the staff member’s clothing.

    It is no big deal to clean my hands after patient contact, but if my clothing touches the patient’s bed sheets, gown, or the patient, should we really act as if there is no transfer of germs – in both directions. If I wear a gown and roll a patient toward me to get linnens out from under the patient, or to put pads under the patient, I look down and find that the gown is not between my legs and the bed that I am leaning against. I am not unusually tall, so this is not something unique to me.

    Then there are the patients we are transporting to the hospital. We do not gown up for every patient. The employers would never pay for that many gowns. I end up picking some patients up and carrying them, because that is the most practical way to move the patient. People rarely have homes that comply with the Americans with Disabilities Act, because it does not apply to the home (and that is a good thing).

    I do not know which patients have which infections. I can be vaccinated for some illnesses. If I transport someone with meningitis, I be given prophylaxis. I cannot know until later which patients have which infections. Sometimes I only find out days later. I assume that most of the time, I do not find out at all.

    I love the hand cleaner. My coworkers make fun of me, because I almost never pass one without cleaning my hands. If there are a bunch in a row, I do not act like Monk and use each one equally.

    One of the big problems is that everybody seems to think that cleanliness is about protecting us from the patients. That is wrong.

    Cleanliness is about protecting the patients from us.

    We are the germ infested disease spreaders – if we do not act to prevent spreading germs to our patients.

    And that is a concept that the anti-vaccination mob will never understand, because they are not interested in doing anything to protect others from things they might do that are harmful.

    They assume that they are harmless. That attitude is extremely dangerous.

  2. BillyJoe says:

    “Sorry. I found nothin’.”

    I found an alternative to 8 :)

  3. Jojo says:

    Delicious post for a Friday. Working in the aircraft business, we have similar problems. There are so many seemingly small things that can lead to loss of life, and it can seem over whelming to try and reduce the risk to 0%. But, it’s something I’m particularly fascinated by, and it’s just as interesting to read about how it’s done in the medical community. I thoroughly enjoyed this post.

    I also ready want a margarita flavored alcohol foam dispenser too. I bet there would be a market for that.

  4. Zoe237 says:

    Brilliant!

    Now hopefully one of the other bloggers who disagrees with the 44,000-98,000 number will speak up.

    What about nurse/patient ratios? Do those make a difference? How about sending folks home too early? Overcrowdedness? Is there a correlation between SES and medical errors?

  5. DavidCT says:

    I liked the photo further down the page where the techs mixing up the herbal medicines are not even using the standards that would be expected of restaurant food handlers.

  6. WilliamLawrenceUtridge says:

    Asking in ignorance, why/how does shaving and body temperature adversely affect paitent outcomes?

  7. David Gorski says:

    1. Shaving a patient the night before surgery increases the rate of wound infection, although doing it right before starting surgery does not. These days we usually don’t shave patients before surgery anymore anyway; right after anesthesia kicks in, to cut away excess hair that can get into the wound we usually use special clippers that are much less likely to cause little skin nicks where bacteria can proliferate.

    2. Letting a patient’s body temperature drop too much during surgery (which can happen very easily due to the combination of anesthesia and open body cavities radiating the body’s heat into space) greatly increases the risk of post-surgical infections and other complications.

  8. wales says:

    Without any effort I can immediately think of at least one hygiene improvement in the alternative medicine arena, and I am by no means an expert. Disposable acupuncture needles were developed in the 70’s.

    I believe that most hypodermic needles were reusable until sometime in the 60’s.

  9. A Mark Crislip post is a lot like a good Bordeaux.

  10. wales says:

    Also, you made reference to acupuncturists not using gloves. I have not seen nurses using gloves for administering injections, are there any standards for this?

  11. Mark Crislip says:

    There is a technique of injections where gloves are not used, but the fingers are well away from the infection site.

    In the picture, and others you can find on the web, the needle entry site appears to be guided by bare fingers.

    I will have to populate the list with disposable acupuncture needles.

  12. squirrelelite says:

    @Wales,

    Reusable acupuncture needles are at least a small improvement (not that there are any good studies supporting the need to penetrate the skin at all), but if the people using them don’t wash their hands or wear gloves, there is an additional vector for cross contamination that greatly minimizes their benefit.

    Also, do you know of any good reference that tells how they were adopted and why? The real information may too obscure and confused to make a clear determination. But, I would be curious if they were started by acupuncturists themselves or the idea filtered across from the general standard medical community where it was fairly well known that needles could be a vector for cross contamination.

    And, perhaps one of the regular doctors on this blog could tell when similar changes were made in general medicine from sanitizing injection needles, surgical kits etc. to using prepackaged “disposables”. I sort of remember it happening about that time frame, but I was just a patient.

  13. wales says:

    I have friends who are acupuncturists. They tell me there is a type of needle with a clear plastic “insertion tube” covering the needle. This tube is used to guide the needle, then the tube slips off upon insertion. This prevents the fingers from contacting the insertion point.

    Squirrelelite, this link gives information about their development. They were developed by a Japanese company in 1978, and then the Japanese govt requested the company ramp up production due to AIDS. http://www.seirinamerica.com/about.html

  14. wales says:

    My acupuncturist friends tell me there is a type of needle with a clear plastic “insertion tube” covering the needle. This tube is used to guide the needle, then the tube slips off upon insertion. This prevents the fingers from contacting the insertion point. In California, disposable needles are mandatory.

    Squirrelelite, this link gives information about their development. They were developed by a Japanese company in 1978, and then the Japanese govt requested the company ramp up production due to AIDS. http://www.seirinamerica.com/about.html

  15. wales says:

    Wikipedia says the first US patent for plastic disposable syringes was issued in 1974.

  16. DevoutCatalyst says:

    “A Mark Crislip post is a lot like a good Bordeaux.”

    I was thinking Two Buck Chuck, only better, cheaper. Now go write him a glowing review…

  17. Versus says:

    A couple of years ago I was to attend a conference in San Diego and happened to notice that the American Public Health Association (APHA) was holding its convention at the same time there. I also noticed in the APHA schedule that the Chiropractic Section of the APHA (yes, there is one) was having a session during the convention. I couldn’t resist. I joined the APHA and signed up for the session. I thus spent a day with about 35 or so chiropractors in a small conference room at the far end of a convention hall that was one convention hall over from the location of the main meeting and exhibit hall.
    I learned lots of interesting stuff, such as from the chiropractor who told me that the flu did not cause death. (What?!) I also learned that chiropractors go on mission trips to third world countries without being vaccinated. But I digress.
    One DC with an MPH gave a presentation on attempts at infection control at a chiropractic school clinic. He said it was not their practice at the school to clean off the tables between patients. Also, he said that he tried to get the clinical chiropractors who taught there to clean their hands between patients with an alcohol-based cleanser. Some of the more die-hard refused because they considered this cleasner “medical” (their word). They wanted to know if they could use yogurt instead.

  18. Peter Lipson says:

    They wanted to know if they could use yogurt instead.

    Best laugh i’ve had all day. Thanks.

  19. shadowmouse says:

    “Perhaps it is a reflection of the perfection that is alternative medicine.”

    But that snake oil salesman says it treats everything! If his protocal isn’t followed to the letter it’ll be the patient’s fault he’s gonna die, not that the ‘medicine’ didn’t work!!

    That’s what it says on the Internet!! Everything on the ‘net is true!!

    Infections are caused by liver flukes! No, I’m Vitamin D deficiant! Iodine! Manuka honey! Garlic! Magnesium! No wait…MMS and DMSO! Ozone infared mega magnetic alpha water!! Don’t eat dead foods! Starve it out! Sugar is a chemical! It’s purely emotional trauma!

    Help me Obi Wan Kenobi!!!!

  20. overshoot says:

    What makes it even more difficult to track and recognize excess mortality is that each death may be due to a different breakdown in medical practice.

    I’ve been wondering whether those estimated deaths included calculated risks that turned out wrong, or otherwise didn’t work out. After all, there’s a point where the human organism is running out of margin and you’re up against balancing the odds of “the condition might kill you if untreated, but the treatment might kill you too.” Pays your money and takes your chances.

    This is a lot more on my mind lately as my mother gets older and her health gets to be more of a balancing act.

    Ever since my local paper, the Oregonian, printed a picture of the local Natural Medicine School teaching acupuncture without gloves, I am not so sanguine about that understanding.

    Damn but I love puns.

  21. overshoot says:

    Rogue Medic — one of the nice things about ski patrol is that we show up for work wearing clothes that are pretty damn good moisture barriers. Then we soak them in UV and dry air; if necessary they can be hosed down with relatively nasty disinfectants while we’re wearing them.

    For which I’ve been quite grateful a time or five.

    Dunno if Gore-Tex would work for you in the fields or not but it beats scrubs and T-shirts.

  22. Great post on a few levels!

    “What about nurse/patient ratios? ” – I recently saw a local presentation of a local analysis: the hospital-acquired infections were so low that there was not enough power to detect a diff, if one existed. This type of analysis is a challenge, since frequency of cases are small, and cases are now idiosyncratic. One idea we brainstormed was: typical staff versus substitute or rotating staff. Nurses skilled in their job may be able to incorporate all of the handwashing better than those that are less skilled, and just stepping in because someone called in sick.

    That made us also think of analyzing infections by shift: more desirable shifts have more senior, and likely better (experience), nurses. So, the data may be re-alnalyzed as a function of tenure of nursing staff.

    This gets at the theme of the challenge of infection control: implement hand-washing, and you get a lot of benefit. As you tackle more and more causes of infection, the rate drops, and the cause-of-infection becomes more varied/idiosyncratic.

    But I think some challenges, such as MRSA and C dif, are ripe for the kind of long-term attack that is desribed in this post. Lots of room for improvement.

    I saw a local presentation of a local C dif analysis. Females more likely than males. No one but me suspected some transmission from visitors to patients, or from the toilet seat (females sit down more frequently). I am not an MD, but a psychologist, so my mind wanders to other possible causal pathways – we therapists are mindful of whether a therapy client shows up alone or with a family member or friend. We ask about social networks. So, when we think abt how can infections be communicated, we think this way.

    We don’t pay attention to sneezes professionally, so we don’t really think about disease coming from sneezes.

    In that presentation discussion, someone mentioned bleaching surfaces. But we learned the hospital does not want to see white flooring and white countertops to get yellowed by bleach.

    So, sorry, your grandmother may suffer C dif because the hosp wants to keep up appearances.

    The most popular answer amongst the ppl in this presentation? We need to develop new drugs to fight C dif.

    What? That answer is better than: bleach-mop floors and counters, or make visitors wash their hands?

    It will take these types of not-obvious quality improvement efforts to make big progress on C dif and on MRSA.

    As far as SCAM nad hosp-acquired infections: the challenge has been made!!! Will anyone meet it? I won’t hold my breath.

  23. cloudskimmer says:

    Dr. Crislip,

    Whenever I see nurses and doctors outside of the hospital wearing their scrubs (so that they look cool?) I wonder about the bacteria they are taking out of the hospital, and those they take back with them. Weren’t scrubs meant to be discarded frequently to avoid this disease transfer? Didn’t medical personnel change into and out of work clothes at the hospital back in the old days?

    “My acupuncturist friends tell me there is a type of needle with a clear plastic “insertion tube” covering the needle. This tube is used to guide the needle, then the tube slips off upon insertion. This prevents the fingers from contacting the insertion point. In California, disposable needles are mandatory.”

    –but in my one unfortunate experience watching an acupuncturist, the guide was used to start the needle, tapping it to penetrate the skin, then the sleeve was discarded and the remainder of the insertion done by sliding the needle through bare and possibly contaminated fingers. It looked like a great way to insert infections subcutaneously. A nurse friend of mine told me that acupuncture is a “clean” but not sterile technique. I don’t have any acupuncturist friends.

    Since SCAM typically ignores germ theory, I’m not going to hold my breath waiting for them to clean up their act. Better to tell my friends about the risk of infections–and even death–from acupuncture.

  24. overshoot says:

    Whenever I see nurses and doctors outside of the hospital wearing their scrubs (so that they look cool?) I wonder about the bacteria they are taking out of the hospital, and those they take back with them. Weren’t scrubs meant to be discarded frequently to avoid this disease transfer? Didn’t medical personnel change into and out of work clothes at the hospital back in the old days?

    Yes, but the hospital saves a lot of money by having staff wash their own and dress on their own time rather than on the clock. The staff that I have known don’t complain about the cost of washing scrubs because they’d have to wash something anyway, scrubs are cheap, and only getting dressed once saves time.

  25. BillyJoe says:

    “How about sending folks home too early?”

    How do you define too early?

    Also I would have thought that the longer patients are in hospital the more chance they have of acquiring infections, including infections by resistant bacteria which are more prevalent in the hospital environment.

    The main drive towards early discharge, if I remember, actually came from attempts to reduce rates of DVT.

    Of course early discharge might also reflect a drive towards cost saving and, in public hospitals, reduced waiting lists.

  26. squirrelelite says:

    @Wales,

    Thanks for the link!

  27. squirrelelite says:

    @cloudskimmer,

    Maybe they were in a hurry to get to the TV studio to be on Dr Oz’s show.

  28. Great article on so many levels! I love it when I can learn something AND it can give me a chuckle too.

    “One approach is the “It’s ok to ask” program, where patients are encouraged to ask their provider if they washed their hands. I asked a series of patients if they would ask, and they uniformly said no, they did not want to risk angering their health care provider. I agree. It is important not to piss off the person providing your morphine. Besides, would you fly on an airline if their motto was “It’s ok to ask if the landing gear is down.”

    Thank you, Thank you! My son’s going in for his third of five to seven inpatient surgeries next week. I have been told or read a few times that I should ask all medical providers if they have washed their hands before they touch him.

    I’ve also read and been told that when medication is given, I should ask if the medication and dosage is correct.

    To me this feel like telling the health care provider to their face that I think they are incompetent…not something I feel comfortable doing when they are caring for my child. While I do know many people who have excellent people skills and could broach these subjects diplomatically and not ruffle feathers, I am not one of those people, particularly when I’m stressed and having your child in for surgery is a little stressful.

    I have felt like I am being an incompetent whimp. I have no idea how much I’m endangering my son by NOT asking these questions.

    Secondly, I wonder about the dangers of shared responsibility, does it make sense to set up a system that relies on a reminder from an individual that may not know their role in the system or feel competent executing it?

    Also, I love the analogy. We have two pilots in the family and I never ask them if the landing gear is down, either. I find it’s even a bit dicey to joke about their driving.

  29. manixter says:

    I’ve really integrated alcohol hand rub into my patient encounters. When I introduce myself, that is my cue that I need to “squirt and rub” first (of course, this makes social situations difficult, since I’m always looking around for my “squirt” before I shake hands). I also find that having the alcohol dispensers (p*rell or whathaveyou) being in a strategic area helps with compliance. Having them by the door to the OR makes sure I squirt and rub after inductions (after I’ve not uncommonly had my hands in someone’s mouth).
    And, of course, when in doubt, squirt and rub, since I might have forgotten…
    I’m not so sure on dry surface to dry surface contact spreading infection. If I have visible soiling (that I’ve touched a damp surface with my clothes), then I’ll change (thank god for scrubs). I haven’t read any data that supports transmission in this way (including an article on shoes and dirty OR floors). I strongly support good old fashioned soap and water to keep “things bacteria can grow on” down, with bleach for areas of body fluid soiling. Keep hospital surfaces easy to clean with minimal crevices (linoleum may be superior to wood flooring– sorry about the aesthetics!)
    Finally, if anyone has read this far, nurse to patient ratios are very important for almost any outcome you can look at– bedsores, infection, death. Some studies even show improved outcomes with nurses with more advanced degrees (don’t remember if it was BSN vs LPN or what– I’m sure I’ll be called out for my ignorance shortly).

  30. jimpurdy says:

    QUOTE:
    “You cannot know best practice based on individual experience.”

    I guess that explains why doctors don’t pay attention to individual patients. What a lousy medical system we have.

  31. BillyJoe says:

    jimpurdy,

    When you don’t have a clue, it is best not to sound so self-assured.
    In other words, spare us the arrogance of your ignorance.

    For your information:

    For a start, the “experience” referred to in that quote, as applied to your example, is that of the doctor. The doctor cannot rely on his personal experience of treating patients with the same condition to conclude which treatments work best. That requires the results of systematic reviews of clinical trials.

    Secondly, if you are referring to patient’s experiences, the patient’s experience is the start of a process only, not the end.
    Doctors start with the patient’s history. The details of the patient’s history gives them clues as to the nature of their illness. They investigate to arrive at a probable diagnosis. Then they use Science Based Medicine (not their personal experience) to arrive at a suitable treatment.

    So, thanks for failing on two levels within the space of a single sentence.

  32. BillyJoe – did a crocodile eat your dog or something? You seem unusually testy tonight (this morning.)

    Just a little point, a doctor should use SBM to arrive…

    In reality, my experience with whether individual doctor’s do this is quite varied.

    And some doctor’s really don’t seem to treat individual patients. One doctor (that I fired) basically had three treatments. You got one of the three treatments depending upon your symptoms and too bad if that treatment wasn’t appropriate for you. Health history, HA? what a waste of time. Looking at the correct patient’s records when consulting with that patient? Not required. Giving correct chances of success for a treatment in the individual patient? Why, when you can just give the whole clinic’s success rates (making sure to stack the deck with lots of young healthy patients). And this was the only clinic of this specialty that my insurance covered.

    SBM is a great goal, but it is not safe to assume that all conventional doctors are SBM doctors, EBM doctors or even good doctors.

    That said, I have had more doctors who are much better and some who really terrrific and do use SBM. It’s important to shop around.

  33. long_tail__ says:

    Another recent intervention at our hospital is the “pre-code”. If a nurse (especially a younger, less-experienced nurse) feels that something is not going well with a patient, the nurse can call the “Medical Emergency Team”. This team evaluates the patient, institutes treatments and tests, and involves the physician.

    From my point of view in the emergency department, the number of actual “codes” (cardiac or respiratory arrests) in the hospital has plummeted. (At night I may be the only physician in-house who can respond to a code.) This has been demonstrated in our Code Blue Committee’s statistics, as well.

  34. BillyJoe says:

    micheleinmichigan,

    You are perfectly correct.

    And, yes, a crocodile did eat my dog!
    How did you know?

    Well, not exactly, but something similar.
    Clyde is now with another owner having proven to be totally incompatible with our other two dogs. My brother the vet tells me that he wants to be “the only dog”. My wife assures me that her new owner is a very nice young lady. She even texted her afterwards to say that her children love her and that she took her in her trolley to the supermarket and she was perfectly behaved.
    I suppose I should be happy…but I’m not!

  35. BillyJoe – That is a bummer. I’m sorry.

  36. squirrelelite says:

    Sorry about your dog, BillyJoe.

    We had a similar situation. We had four dogs, all rescued or adopted. We had the first two for several years and after a short uneasy period, they had worked out how to get along. The third was younger and rather submissive, so he got along all right. The fourth was very nice with just people. His major fault was a tendency to lean on you leg and beat you to death with his tail!

    But, after he grew up from a puppy, he got very agressive and tried to establish dominance over the other three. After a couple bad fights and trips to the vet for repairs, we knew he couldn’t stay. Fortunately, one of my sons bought a house and took him to live there. He seems to be doing pretty well and is even getting along with their two cats.

    I don’t miss the fights, but sometimes I miss him. But, at least I get to see him when I visit my son.

  37. BillyJoe says:

    Thanks to both of you.

    Our problem was that he would suddenly and unaccountably latch onto one of the other dogs and not let go. We had to throw a blanket over the two of them and slowly and carefully separate her jaw from the other’s throat or neck.

    We did first try my sister in law, but she lives with my other sister in law (her identical left/right reversed twin!) who has a labrador. Unfortunately the same thing happened. Remarkable seeing that Clyde is a maltese/silky terrier cross.

  38. JMB says:

    “You cannot know best practice based on individual experience. I like to tell the residents that the three most dangerous words in medicine are “In my experience.” ”

    The mathematical methods of statistics and probability can be applied to the observation of clinical practice. This technically is not science in the sense that it is not trying to establish a causal relationship. It may be more correctly labeled a method for production quality used in engineering. However, if such methods are used, you can identify the most effective clinicians. If you compare the success of the decisions of the best clinicians to the best decisions based on medical science, the best clinicians will occasionally outperform SBM. In those scenarios that the best clinicians cannot perform better than SBM, there has been sufficient scientific study that all of the important factors in outcomes has been identified, and measures of the conditional probabilities are rigorous (where we would ideally be). Two scenarios in which the best clinicians can beat SBM are when there has not been sufficient scientific study, and when the scientific study is based on factors in which we know there is individual variation in powers of observation or technical skill.

    Those medical problems in which there has not been sufficient scientific study merely tell us where medical science needs to be directed to improve our outcomes. Any newly identified infectious agent, or a change in the established pattern of spread or antibiotic resistance of an infectious agent may be examples of this in infectious diseases. However, there are certain problems in which our current methods of analyzing data will prevent us from achieving any significant progress in the near future. I cannot think of a good example of this in infectious disease, it is not my area of expertise. An example from outside of infectious disease is the way clinical trials for medical screening are often based on age grouping of patients. It doesn’t make sense that women should begin screening mammography at age 40 or age 50 (as opposed to age 39, or 45, or 53), it would make more sense to calculate an age to begin screening for an individual patient based on age of menarche, genetic tests, age of first pregnancy, and age of menopause. The age groups selected for clinical trials will just result in averages of variation (and intentional randomization) in individual risk factors (length of time of exposure to estrogen, risk of exposure to ionizing radiation, and risk of genetic mutations/variants).

    The most obvious example of the second scenario in which the best clinicians may achieve better outcomes than SBM is in the field of surgery. There is variation in technical skills of the surgeon (and for that matter, in the rest of the operating team) that will have an impact on the risk versus benefit calculation in the decision to operate. The doctor considering whether to refer that patient to the surgeon, or the surgeon deciding to recommend surgery, may achieve better results than can be achieved in a decision from the scientific literature, simply based on the knowledge of an individual surgeon’s outcomes. The same also applies on the diagnostic side of the equation, anytime a human observer is relied on for the result (radiology and surgical pathology are obvious examples).

    So I would argue that although the terms “in my experience” will not lead to better science in the terms of the British empiricism of Francis Bacon, but it can lead to better patient outcomes in the terms of production quality of Edward Deming. However, to insure that it leads to better outcomes, we should apply statistical/probability methods of production quality management to monitor outcomes. The best clinicians use SBM, but they may customize decisions based on individual factors in the patient, or the healthcare team available to take care of the patient.

    On a separate note, I think this SBM site could address some of the sloppy science used in some institutions best practice guidelines committees. The editors of this site could formulate rules for SBM guidance of best practice committees. Simple guidelines like, “Read the articles cited to support the best practice guidelines, not just the abstracts”, or, “Decisions based on evidence should be cognizant of the selection criteria for the clinical series or RCT, and should not extend those decision rules beyond the population that meet those selection criteria.” Perhaps the editors here could even provide a consulting service to review practice guidelines.

  39. stewiegriffin81 says:

    JMB says:

    “It doesn’t make sense that women should begin screening mammography at age 40 or age 50 (as opposed to age 39, or 45, or 53), it would make more sense to calculate an age to begin screening for an individual patient based on age of menarche, genetic tests, age of first pregnancy, and age of menopause. The age groups selected for clinical trials will just result in averages of variation (and intentional randomization) in individual risk factors (length of time of exposure to estrogen, risk of exposure to ionizing radiation, and risk of genetic mutations/variants).”

    I disagree. There are two issues here:

    Firstly, there is the matter of practicality. To recommend general screening for entire populations, we cannot make recommendations that those populations may not be able to afford. In particular, you are asking for every woman to at some stage have a genetic test. The cost of doing genetic tests on every woman will be ridiculously expensive, and is simply not sustainable at this point (although in the future this may eventually become viable).

    Secondly, screening recommendations are individualised for increased risk where possible. Women with a higher lifetime risk of developing breast cancer have different recommendations.

  40. baldape says:

    Mark – what about using individual clip-on sanitizer dispensers.

    They could have 4 components:

    1) Just enough sanitation to be used for an entire shift, but still small enough so that, clipped next to the standard beeper, they would not interfere with any movements a caretaker would be expected to make. A trade-off may be even smaller dispensers that beep when they are almost empty, and a ready supply of fresh dispensers at the nurses’ station (see # 3 as well).

    2) A sensor that detects when it has entered a patient’s room, and beeps until the belt-wearer dispenses the alcohol into his/her palm (each doorway could have a simple transmitter to let the detector know it has entered the room… we’re talking electric-dog-fence technology here).

    3) A requirement that if a provider’s clip-on dispenser be contaminated via patient contact, it be immediately swapped out at the nurses station.

    4) Policy requiring all personnel (from the doctors to the maintenance staff) have an active clip-on at all times.

    This even helps with the “it’s okay to ask” approach you’re trying to address: I suspect much less social anxiety would be involved if patients just had to ask, “what’s that beeping noise” (not to mention the positive social pressure that comes with coworkers knowing darn well what the beeping noise is). Posters which proclaim, “Ask your provider why they are beeping” wouldn’t hurt :-).

    Crazy-talk?

  41. baldape says:

    Having an engineering background, I can’t help but stay after this idea a bit: 2 more thoughts:

    1) Each transmitter should emit a unique signal, such that the sensor can know that it should only start beeping when it encounters a new signal (and won’t beep upon exiting/re-entering the same patient’s room). The biggest death knell of any such interface is the degree to which it annoys the users (and entices them to bypass correct usage).

    2) Auditing would also be easy: administrators could simply walk down the hall with an “audit-only” transmitter, which tells a sensor to give a brief confirmation burst of beeps, allowing the administrator to quickly spot-check various departments to ensure widespread compliancy.

  42. overshoot says:

    BA, if you want to be take seriously (as distinct from just being taken):

    You have one year to file a patent application for that invention. Seriously. If you want to be magnanimous about it, by all means license the patent for peanuts or (if you can afford the costs as charity-equivalent) for nothing. Or you can assign it to some outfit who will license it broadly and cheaply, again a nonprofit being one possibility.

    Seriously. This, speaking as someone with more than 20 US patents. Either do it or make damned sure you have this documented (including site records from SBM) that it was first published here.

  43. BaldApe – “The biggest death knell of any such interface is the degree to which it annoys the users (and entices them to bypass correct usage).”

    I see you’ve already spotted the concern that I was going express. With all that beeping what would the rate of user sabatoge? Also, I’m wondering how doctors would take to it. Maybe you could find a clinic to run a small scale prototype test.

  44. BaldApe – Just because I’m a fault finder by nature, could you prevent spillage on pants? Also would the fact that the device might be under a doctor’s coat and they’d have to pull aside the coat to use it be a problem?

    But don’t mind me, I’m one of those late adopter techno skeptics. I don’t even have an ice maker in my fridge.

  45. Calli Arcale says:

    My big problem with the “it’s okay to ask” thing is that in the software industry, we have been busily engaged in demonstrating the complete worthlessness of this technique. We create software that pops up a little window saying “Are you sure?” and then asks the person to click “OK” or “Cancel”. In extreme cases, when we really really want the person to be sure they’re doing things right, we cover their entire screen with a totally not-ignorable window, every time they go to submit a certain action, and say “are you really really REALLY sure?”

    (Sidenote: my timecard application does this, and in the most egregious way, because it demands I verify my timecard is correct but does not allow me to see it without going back a screen — and after verifying, it will ask me the exact same thing again.)

    The result of this pervasive practice is that everybody who uses computers even casually is now thoroughly trained to completely and utterly ignore the request and just click “OK” so they can get on with life. Many probably don’t even remember seeing the prompt. Of course, users have to go through a period of irritation at the condescension before they get to this point.

    I expect that having the patient or the patient’s family ask would eventually come to the same result. After all, asking the question doesn’t verify anything, as the person asking is in no position to test the answer. It’s just another obstacle to doing one’s job.

    “Did you wash your hands? Y/N”

    And of course patients are uncomfortable asking; if they don’t trust the doctor to wash, how can they trust the doctor to answer honestly? I trust my doctor, so it doesn’t occur to me to ask.

    (BTW, the same phenomenon happens with click-through End User License Agreements as well, which is making them difficult to enforce and greatly reducing their value. Hardly anybody takes them seriously anymore.)

  46. JMB says:

    @stewiegriffin81

    My argument is not that we should be doing things differently now with the currently available information. My argument is that we should validate and incorporate newer methods of data analysis so that information that medical scientists provide to the healthcare practitioner will be more directly applicable to decisions. Even though a group of doctors may all be familiar with the same medical science literature, often there will be a measurable difference in the outcomes obtained by individual doctors in that group. If we apply scientific/ probability methods to study why there is a difference in outcomes, we can often find factors that were randomized by design in RCTs. The better performing doctors often have better (fuzzy) estimates of a priori or likelyhood ratio factors that were not directly measured by the experimental design. I am simply wishing (for the last 20 years) that we could obtain more complete population databases and more reliable assessment of healthcare provider results, so we would have more information that would be useful in clinical decision making.

    It does not make scientific sense that all women should start screening at a certain chronologic age. Physiologic age varies from chronologic age. It may be the only way practical, because of decisions made in designs of randomized clinical trials, and economic issues.

    I would caution that we need to separate economic issues from medical science. Economic issues can rightly guide funding of research. But a scientist should not say that there is no scientifically valid reason for a recommendation just because it is not economically feasible for a population. Most doctors are not scientists, but they should still not say that science does not support a recommendation, when it is really the economic or other practical issues that prevent a recommendation from being feasible.

    All of this is simply my opinion. I cannot make a scientific argument for it.

  47. stewiegriffin81 says:

    @ JMB

    You seem to be implying that medical scientists have not been attempting to do this. There have been decades of scientific work put into exactly what you are describing (for example, the Gail breast cancer risk assessment tool estimates the risk of an individual woman using various parameters of that individual woman).

    The problem is, the science is currently not good enough. The individualised breast cancer risk assessment tools have a low predictive probability.

    I agree with you that having reliable individualised screening recommendations would be a great thing. Unfortunately, we simply don’t know enough about human biology to be able to accurately assess this on an individual basis.

  48. BillyJoe says:

    “But a scientist should not say that there is no scientifically valid reason for a recommendation just because it is not economically feasible for a population.”

    But a scientist should also make a cost/benefit analysis to determine what is practical as opposed to what is possible.

    “Most doctors are not scientists, but they should still not say that science does not support a recommendation, when it is really the economic or other practical issues that prevent a recommendation from being feasible.”

    Ditto.

  49. JMB says:

    @stewiegriffin81
    “The problem is, the science is currently not good enough. The individualised breast cancer risk assessment tools have a low predictive probability.”

    We both would agree that the science is not yet good enough, and progress has been made. I am probably just being unrealistic about how fast progress can be made.

    In the issue of mammography screening, I would have preferred to see more progress in the modeling of the screening process, which includes risk assessment models, models of the screening tests, and models of the treatment. Breast density can be incorporated into the Gail model of risk assessment, but why isn’t it also incorporated into a model predicting the likelyhood of screening detection before detection by palpation? If we are to recommend mammogram screening to an individual woman, shouldn’t we consider that if she has dense breasts, then the mammogram is less likely to detect the tumor before palpation? And why rely on age as a predictor of breast density, when a single mammographic exam will determine the breast density? When death is a chaotic event, why not use prevention of Stage IV disease as the endpoint? Chaotic events play havoc in making predictions. What if there is a breakthrough in treatment? Will we have to wait 15 more years to determine if breast cancer screening should continue, or should we have a validated model that we can update the parameters to reflect the change in success of treatment, and use the model for the decision. Why don’t the risk assessment tools and RCTs differ for different types of breast cancer? I think that we could probably see a variation in predictions depending on tumor type and histologic grade. I suspect we would see higher predictive power for screening models for invasive ductal carcinoma, as opposed to invasive lobular carcinoma (that does not necessarily apply to risk assessment).

    @BillyJoe

    I agree that the doctor practicing medicine has practical considerations in their recommendations to patients. The scientist can also note practical concerns in recommendations to the public or political entities. I just don’t like the public (or a patient) being told a recommendation is based on the best science, when the recommendation incorporates practical concerns. Science can give us the best estimate of risk versus benefit. Science may also give us an estimate of cost per year of life saved. But science does not tell us whether a medical intervention is economically practical. Science cannot tell us whether $20,000, $50,000, or $100,000 per year of life saved is acceptable.

  50. BillyJoe says:

    Fair enough.

    Except I was going to say: What? Science can’t do a cost/benefit analysis? But I suppose we’re talking Maths here.

    Nevertheless, I think a doctor is not doing his duty – in fact it could be seem as a dereliction of duty in some cases – if he gives his patient the recommendation based on the best science and ignores the practical aspects related to cost/benefit analysis. It would raise unrealistic expectations in the patient for and grind the health system to a halt if all patients were to demand the recommendations of the best science.

  51. GinaPera says:

    Excellent treatment, Mark, of this confounding subject. It’s clear that when the dean of Stanford’s Medical School chooses handwashing as the lead story in a newsletter, as he did in the last few years, it’s a problem. Then again, my nursing friends have been complaining about this for decades.

    You wrote:

    “There is a curious subset of MD’s who feel that the data does not apply to them. They do not need to follow infection control procedures, use full barriers when placing a line, or even wash their hands. And I do not get it. I cannot figure out why some docs are so recalcitrant about doing the right thing, the proven thing. Eventually everyone complied, but some people made it more painful than it needed to be.”

    I suggest a screening among these MDs for Adult ADHD, and I’m not being sarcastic.

    Some docs with ADHD who forgo the hand-washinig might truly think the rules don’t apply to them. Others simply forget. Again and again. And rather than saying, “I can’t remember to do it,” they might save face by rationalizing that it’s not really necessary. It’s others’ hands who are germ-ridden, not theirs. Then again, “denial” can take many forms with conditions affecting the frontal-lobes. In fact, many might be quite convinced they ARE complying with the rules, it will require nothing less than videographic proof to convince them otherwise.

    On a more basic level, with ADHD it’s not a question of not knowing what to do; it’s doing what is known. You can warn all you want, teach, admonish, etc. But to follow through on all that requires

    –initiation
    –motivation
    –a tolerance for the tedious (hand-washing, even if with foam)
    –a constant sense of and respect for consequences (infections),
    –and empathy (which, even if abundant in the person with ADHD is often compromised by key neurobiological deficits in impulsivity, hyperactivity, and inattention).

    For anyone who thinks that adults with ADHD could never make it through med school, that’s dead wrong. Plenty of adults with ADHD earn advanced degrees, even if some might take longer and need more assistance. In fact, many people with ADHD do well in highly structured environments, where there is an external locus of control and a supported path to their goal (graduation). Doing the job they have prepared for, however, is an entirely different matter.

    (FYI: To those who respond with self-medicating arguments and disparagement of my character, my education, and the horse I rode in on, you’ll have to get your medication elsewhere. I take these topics seriously, but I have a life. And that life doesn’t include indulging others’ in their psychopathology. I focus my energies on helping people to understand conditions such as ADHD, which when left untreated hurt not only those people who have it but also everyone around them.)

  52. Harriet Hall says:

    GinaPera,

    You have offered a rational, plausible, testable hypothesis: that undiagnosed ADHD might contribute to physician non-compliance. If testing showed that to be the case, we would then have to ask if treating those individuals (with Ritalin, etc.) would improve compliance and would have an acceptable risk/benefit ratio. You made a good point. You could have stopped there.

    Your concluding FYI statement is bizarre, paranoid, and inappropriate.

  53. GinaPera says:

    Oh Harriet, please. Before you pontificate about what is bizarre, paranoid, and inappropriate, go back and read the inanities from a previous posting. That’s a textbook case for bizarre, paranoid, and inappropriate.

    And that truly is all I have to say on the matter. I lack the time for and interest in another round of Rashomon.

  54. wales says:

    Hey MC, I just ran across this piece that says overall hand hygiene compliance was only 34%. I believe you said in this post it was 90%, was that referring to your hospital only? If you have any citations on this subject I am interested. Thx.

    http://www.medpagetoday.com/InfectiousDisease/InfectionControl/20367

  55. Mark Crislip says:

    My institutions hover around 90% for the last two years, based on direct observation by secrete shoppers

  56. wales says:

    Kudos then, for doing it right. It’s hard to believe this is still a problem in the 21st century.

    I found it particularly interesting that there was more post-procedure than pre-procedure handwashing, and the authors’ conclusion that a form of selfishness or self-preservation was the motivation “these findings may suggest that healthcare providers are probably driven to wash their hands by their need to protect themselves more than their patients.”

    Maybe it would help for institutions to start advertising “clean hands” compliance as a method of supporting a form of institutional “herd immunity”. They could create policies and campaigns for castigating and publicly shaming those who don’t conform.

  57. wales says:

    Installing cameras (even fake ones) at all sinks might help too.

Comments are closed.