Ways of knowing

Wooden blocks used to create a path for a marble

I just finished an interesting book, “Patient from Hell” (Schneider and Lane). The central thesis of the book has to do with the types of evidence that can lead to knowledge about how to treat something and how the medical profession fails to make use of some of those types of evidence. A second key point in the book is that certain types of decisions should be in the hands of patients, not just doctors. Let me try to clarify these points, since I think they are very relevant to the experience of Lyme disease. As a side note, the book itself is a little difficult to follow, this is not necessarily a recommendation of the book, just my attempt to restate their conclusions in terms that make sense to me and relate to Lyme disease.

The type of knowledge that is most commonly used in medicine is called “frequency knowledge” by Schneider and Lane. This is the knowledge that is created by trying something over and over again and seeing what happens. For example, flip a coin enough and you’ll know if it’s balanced or not. Give long term antibiotics to enough people with Lyme disease and maybe you’ll find out whether it can help them. This kind of knowledge is  relatively easy to generate in an easily controlled, repeatable situation (like flipping a coin). However, it is much harder to get solid frequency knowledge in a more complex situation where every coin you are flipping is different (or every person with Lyme disease is different). Clinical trials generate frequency knowledge, and in the case of Lyme disease, they will probably be at least partly inconclusive until many more trials are run.

A different, equally valid type of knowledge that is sometimes ignored is “process knowledge.” This kind of knowledge complements frequency knowledge, but is sometimes not considered to be truth by doctors because it is “untested”. As an example, consider the image at top right. It’s made with wooden blocks that can be put together in any number of combinations. No “studies” have been done of this particular construction, yet you can look at it and with a fair amoutn of certainty, you can predict where the red marble will end up (even though you do not have complete informatino about the structure). Not only that, you can predict with complete certainty what will happen if I remove a critical block. You are using logic to draw valid conclusions about the world that are likely to be true even though they are untested. The wonderful work being done to understand Bb’s life cycle, the ways in which it may resist antibiotics, and so on, are all examples of process knowledge. This knowledge can be used by researchers to infer potential was of treating lyme disease. It can also be used by patients and doctors to decide, now, given a lack of frequency knowledge, what the best choices are. This is one part of the  path taken by LLMDs.

Schneider and Lane argue that sound process knowledge should carry as much weight as sound frequency knowledge. Ideally, to fully understand something, you have both. Thus, the process knowledge may suggest that certain medications could help, and this could then be tested further in clinical trials. If both draw the same conclusions, you have high certainty that you should proceed. However, both don’t always exist, and in a situation with a lot of uncertainty, it’s stupid to throw out one kind of knowledge entirely as Medicine sometimes does. Worse yet, the  knowledge that comes from your history, you experiencing something over and over again is also ignored.  Frequency knowledge is often simplified down to an average, yet most of us (especially in the case of Lyme disease) are decidedly not average, and the ways in which we are not average matter and should help determine our treatment.

A final important point that Schneider and Lane make is that there is a difference between knowledge and policy. They use a very compelling example, amniocentises. This is a procedure that is used to extract amniotic fluid from the uterous of a pregnant woman so that certain tests can be run to determine whether the baby may have disabilities. The procedure carries a small risk that the baby will be spontaneously aborted. Some obstetricians and midwives used to encourage women over 35 to have this test, but not women under 35*. Why? It turns out that “the curves cross” at age 35 — the average woman over 35 has a higher risk of having a baby with a disability than she does of having an abortion due to the procedure.

This is knowledge (frequency knowledge, in fact, since the procedure has been tested with many women). However, if the doctor makes the decision for the patient without discussing the tradeoff, the doctor is making a policy decision. Some women would prefer to take the risk of an abortion so that they can know whether their child has those issues (and may even choose to abort the baby if it does). Other women would never abort and would consider taking the risk of an abortion a crime (at an extreme). The patient, not the doctor, should be able to make this decision.

The issue of knowledge versus policy decisions is very applicable to Lyme disease. One of the arguments that the IDSA camp makes against those few clinical trials of long term antibiotics that have been run is that the risks of the antibiotics are not worth the payoffs (on average, based on the frequency knowledge coming from the trials). But that tradeoff is a policy decision that should be made by patients with the help of doctors who educate them about the choice.

*I changed the sentence about counseling in response to the first comment below.

2 thoughts on “Ways of knowing

  1. Well written comment on a book (which I haven’t read) that seems to address the always difficult decision making in medicine, a science that is not exact, like math, where 2 plus 2 has always added up to 4. In medicine, there are still so many new frontiers: New diseases, new treatments, new methods, and also new failures and risks, are found it seems every day. We say, medical knowledge has a half-life of about 6 months….
    But I would like to comment specifically on one paragraph, because it is not only outdated, but also not a realistic description of how pregnant patients are counseled about what options they have to test for two possible genetic defects (Down’s Syndrome and Neural Tube defect). Until a few years ago, standard of practice was to offer amniocentesis only to women who had a relatively high risk of Down’s Syndrome – at or after age 35 – because the blood test, that only evaluates the risk of those conditions (and is offered to all pregnant women) is not valid for women after age 35. Younger women were only offered the amniocentesis if the blood test result indicated an increased risk.
    Now, standard of practice is that all pregnant women can opt to have amniocentesis, regardless of their age, in lieu of the blood test, which does not give a diagnosis, only a risk assessment.
    But all these prenatal tests are optional and ultimately are the decision of the patient. She is counseled carefully, the risks are explained, and because medical malpractice dictates caution, patients are asked to sign a consent, even if they decline the tests.
    I have no doubt that a counselor’s bias can influence a patient’s decision; ideally, counseling is supposed to be totally non-judgmental, never coercive. Many time, women ask me what I recommend: they are unsure, and ask for my opinion….

    The decision making for amniocentesis is simple compared to a patient who is suffering from Lyme: She can find out if she’s pregnant with a simple test. And prenatal care is straightforward in most cases, and the condition ends – usually with a welcome addition to the family.

    How I wish that we had a similar standard protocol for Lyme disease!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s