The nocebo effect

I have become quite infatuated with hypnosis lately, having done an introductory course. This led to renewed interest into the psychology of everyday medical communication and thus I arrived at the topic of the nocebo effect. There is a lot of literature about the subject, but not very many actually delve into the details.

So here is my short description of the theory.

  • Patients in hospitals (and in particular the ER, OR and ICU) are scared and confused, which leads to high suggestibility.
  • Highly suggestible people can take in every word you utter, so you have to be careful what you say.
  • Most of our daily smalltalky utterances influence negatively.
  • Highly relevant communication such as informed consent, speaking with relatives, is particularly prone to such effects. So we should precisely filter what we tell patients, e.g. about medication side effects to spare them the nocebo side effects. In how far this is ethical, is not too clear.

This might all seem pretty obvious, but the consequences are immense.

  • Our soothing remarks like “just a wee sting here”, “some cold” and so forth mainly serve to soothe ourselves, the will hurt the patient literally. So the best thing is to dump them.
  • If you really want to soothe a patient for a painful procedure, use techniques of everyday hypnosis, like focusing attention on something nice, using narratives and engage the mind in doing things actively instead of hoping for passive tolerance.
  • Our empathic questions about nausea and headache in situations where patients could but should not have these symptoms can really provoke them. It is way better to use a positive formulation, such as “Is there anything we can do to make you feel even better?” [Of course, if you need to know about symptoms for your differential diagnosis, then you still have to ask for them, but remember that in a properly taken history, the patient is supposed to volunteer all relevant information and the review of pertinent symptoms asked in quick succession after the opening phase of the interview should basically all be answered in the negative.]
  • Telling patients about side effects of drugs and procedures should be done carefully, perhaps asking the patient if he or she would be open to forgo a detailed listing if that helps them. Also always use the positive numbers instead of the negative ones and be sure to use the proper modality (90% do fine with LP, a few might get headache). [But beware, people who tend to want to make a choice each and everywhere might react differently to this use of modality, so you have to fine-tune your statement, adapting to personality styles.]

Obviously, I am not an expert on this subject, but my talk should help to interest you in the field. Here are some introductory references.


Information management

I always was a fierce defendant of evidence-based medicine. After all, what harm is there in finding, using and being able to reject the available evidence for a particular question. Although we never learned it in medical school, I did some postgrad courses on medical literature appraisal, ploughed through the User’s Guide and tried to understand the mechanics of metaanalysis. I also follow most of the relevant Neurology journals. Still, I am unable to answer any given bedside clinical question in the way EBM would have it. It just takes too long and I am too dumb. Properly reading a study takes hours and then I am not able to find all weaknesses that internet blogs find in it.

So either I am unsuited for  EBM or it does not work. Also, it just explains how to answer questions, not how to raise them.

With these fears in the background, I stumbled upon this paper and this talk, explaining a basic approach to what I would call informed medicine. I would add to the information in the article some core principles of work with medical literature:

  1. Try to stay current with the big journals (because they matter), but beware of their pharma influence.
  2. Try to stay current with some more independent and small journals (such as in PMC and the British journals).
  3. Follow informed discussions on the web (such as CCML, emcrit, …) in places where you reliably found good doctors.
  4. Don’t be an early adopter, unless a study has very small NNTs and has been replicated.
  5. Mistrust studies with thousands of participants, unless their (so-called scientific) results have been replicated in clinical practice.
  6. If facing a patient with a problem, read up in good textbooks written by people you trust and who are too old to be influenced by every small whiff of medical wind.
  7. Then confront this with something like uptodate.
  8. Finally read some reviews.

As a final remark, I offer the following strategy for finding a treatment for your patient: do a search on RCT with limits to 1970-1990 on your subject – you will find many rightfully abandoned approaches, but also some gems to be recovered.

A manifest checklist

We had a few near misses and annoyed team members the last weeks, so we did some probing, asked around and tried to identify what went wrong. It turns out – as so often – that team communication had deteriorated over the last months, so that the various members did not understand eachother anymore. This led to exchanges about our communication culture and to the project of a team timeout with a stroke team “preflight checklist” that we go through for each patient in the morning:

  • What is the disease of the patient?
  • Is he stable?
  • What is the plan for today?
  • Are all drugs ordered and ready?
  • Have all labs been drawn?
  • Is there any impediment to mobilization or oral nutrition?

We learned a lot from projects like the Provonost study and the surgical safety checklist. As far as I understand the author of “The checklist manifesto” – a Brigham and Women’s surgeon named Atul Gawande – checklists form the condensation point for interdisciplinary communication, allowing all team members to contribute to and understand the common goals and prioritize their work.

Of course, we will see how it works out but I have been fascinated by Gawande’s exposition and expect great results…


In general, medical students suffer a lot through bad education. Neurology adds to this the very special Neurophobia.

In my quite personal opinion, a lot could be improved if we actually knew what students should do and learn in our hospital. Is it history taking and (neurological) examination? Bedside manners? Basic Neurology? Advanced Neurology? Fascination for the field? The dreadful everyday life of a doctor?

Further confusion comes from the fact that we happily deploy medical students for some tasks (phlebotomy in some hospitals, mini mentals, walking speed measurements, writing discharge letters) that we really don’t like to do ourselves.

The “Junge Neurologen” is an initiative of not so young Neurologists trying to improve education and career options for Neurologists-to-be in Germany. Valuable!

Of course, we cannot simply revolutionize the way our medical education and system works, but there are some messages I’d like to convey:

  • Students are there to learn basic stuff, core skills (history, examination, localization, differential) and bedside manners.
  • They should treat patients (themselves, that is).
  • They should love their time on our ward – what we don’t like, they won’t. What we didn’t like then, won’t be that great today.
  • They should be and feel responsible.
  • They should respect the basic virtues of punctuality, politeness and preparedness – if they don’t, teach them to.
  • You should fetch them from where they are: how far are they? what do they know? what do they want to learn?
  • Don’t show them how bad medicine can be, show them how great it is.
  • Avoid Neurophobia by concentrating on common diseases, on basic skills, normal examination techniques, standard examinations and keep from rare three-eponym-diseases in your differential, weird examination techniques no one can recognize.

References: the only international journal that regularly issues articles on medical education in Neurology is Practical Neurology – here are some pearls:

Induction, deduction and how experts think in medicine

The age old description of how science is (supposed to be) working to generate theories, refute them, induce statements from data and use the Popper/Pearson model to work on hypotheses does not apply to the non-scientific field of medicine. Since the seventies people have wondered how good doctors think and how to bring young ones to become good doctors. We discuss the various approaches to patients and their pathophysiology, to clinical data and thereby involve definitions of pattern matching, heuristics, induction, abduction and deduction as well as Bayes statistics to develop an understanding why working in medicine becomes more complicated and less efficient during the first years – what is known as the U-shaped learning curve of becoming a medical professional.

This is all but theoretical, but of course, you have to read the classic texts if you want to understand the details. Here is a short review on the main part of subject of today’s talk.

Psychosocial history

With 5-7 new patients a day it becomes hard to remember all their stories, if you only recap their biological data. Furthermore, those who only remember the “medically pertinent” mumbo-jumbo tend to deliver worse care than those that capture the psychological and social background of their patients.
We read the article “Complicated lives” in the NEJM this year together in today’s session and concluded that learning, knowing and remembering more about our patients will make us better doctors. In handovers it really makes much sense to add some critical psychosocial details – it leads to better recall.

Shift work

We are going to establish a 3-shift-system on our Stroke Unit coming month. So I wondered what literature there is about the best way to implement it. Rumour and a few recommendations have it that a forward shifting system (morning, late, night) is better than the converse, yet there is scant data to substantiate it. Even less obvious is how many night shifts in a row are best for a) the hospital b) the doctor.

If you can come up with some evidence, pls let me know.

Otherwise we can just list a few myths that are supposed to help to cope:

How to stay awake during long nights:

  • Keep your room brightly lit and not too warm.
  • Use coffee not for staying awake but to get into the mood.
  • If you start to nod off, try a power sprint (100 m), get fresh air, drink and eat light.
  • If you feel a dangerous dip, try to make room for a power nap.
How to get to sleep after a night shift:
  • Make time for decompression after work before going to sleep
  • Find something easy to read in bed
  • No coffee 6h before going to sleep
  • Eat light 1h before going to sleep
  • Use helpers such as ear plugs, night covers, thick curtains etc.
  • Use a standard routine (same book, same muzak etc.)
  • Don’t force your sleep: Nothing more dangerous than calculating your lack of sleep while trying to sleep. If it doesn’t happen, so be it.
  • Avoid sleeping pills and alcohol.
How to resynchronize:
  • Use everything you would do for jet lag resynchronization.
  • L-melatonin doesn’t help, but the retarded form might. Just for 1-2 days – if you have big resync issues.
  • Get into the normal rhythm as quickly as possible.
How to make the best of it:
  • Do morning business (shopping, bills, buerocratic stuff)
  • Sports (skiing on tuesdays!!!)
  • Show up at your spouse’s work with flowers (if applicable)
  • Show up at your children’s daycare/school etc.
  • Don’t expect to be able to do more than before on your free days – you need them to recover!