Snake Oil

I was reading an article last week about PSA being an ineffective marker for testing for prostate cancer. This morning I was thinking about chronic pain affecting simple daily habits and fibromyalgia. And at the breakfast table I blurted out, Medical testing is the new Snake Oil.

Medical testing can undoubtably be useful. But even in a well-designed test and with reliable results, what you end up with is often like a photograph of a hand over a cookie jar: what’s really happening in that picture? Is the hand about to steal cookies? Has it just filled the jar? We’re not sure. We have some information, but it’s important not to mistake that for the whole picture.

The first stumbling block is that not all tests are well-designed: many test for factors that are not actually indicative of what you’re looking for, or factors that are not binary (for example, thyroid tests start with TSH levels. If those levels are high, you know that the thyroid is being asked to create more hormone. But the cause is not binary – there are many factors involved. And in fact, the result is not binary either: if the levels are low, this is not a reliable indicator that there isn’t a thyroid problem. Again, there are many factors involved.)

And well-designed or not, lab results are frequently unreliable. Paul Bergner references an experiment done wherein six people submitted blood samples. The blood samples were then divided and sent to laboratories. Each person’s sample was submitted to six different laboratories, and duplicate samples were sent to each laboratory with different identifiers (as if they were different patients). When the results came back, more than 50% of the results were drastically different for the same samples across laboratories. Even more interesting, the samples that were duplicates labeled as unique samples frequently showed different results within the same laboratory!
The reason for this isn’t nefarious, it’s simple: lab testers are human. They have a bad day, they drink too much caffeine, they get tired, just like the rest of us. Human errors happen, even in laboratories. Additionally, equipment has error margins, or a test sample might have been left out longer in one round than in a previous round, or a thousand other variables.

Even in a well-crafted home, not every line is straight and level. Why do we expect laboratory tests to be different?

And yet, when something feels wrong in our bodies, off we go to get a test. And based on that test, we make decisions – or decisions are made for us.
Let’s return to our prostate example: If PSA levels are not a good indicator of whether or not a person actually has prostate cancer, and whether, if he does, that cancer is in any way threatening, then why are we using it? What happens when someone’s PSA is elevated? (we’ll assume for this example that the test was not false positive, which happens with as many as 50-75% of test results) A series of treatment protocols are started, all aimed at “saving lives”, all of which are costly and associated with very high risk rates of urinary incontinence, erectile dysfunction, and other complications.

According to the Chief Medical Officer of the American Cancer Society, of 48 men treated for prostate cancer based on PSA screening, only one required intervention.

PSA is not the only test that fits this mold. The same example can be made with mammography, and many other tests. Other examples can be made where tests let us down – they fail to find the problem they’re looking for. Celiac, gluten sensitivity, and thyroid problems fall into this category.

People put a great deal of trust in medical testing. But when I look at the way that we’re using the tool, all I can think of is Snake Oil: A good-for-what-ails-you mystery miracle cure that promises more than it could ever deliver; all yours for only a little faith and a little more money. This is not to say scientists and doctors are the same as snake oil salesmen–they’re not trying to con anyone, they’re doing what they’ve come to believe will help bring certainty and ease suffering. But the tests themselves are, all too often, little better than snake oil when it comes to identifying the causes and cures of disease.

We all need a great deal more education in the area of health care. We need to take the time to understand how our bodies work (how they actually work, not how the media tells us they work). We need to have a good understanding of the foundational factors that influence health: sufficient sleep, optimal diet, frequent movement, significant time out of doors. When we understand our bodies, we don’t need to run for a test every time there’s an ache or a pain. We don’t expect a test to tell us something shocking we haven’t noticed before. In this scenario, we can use well-designed tests to help us make informed decisions, but this requires that tests are optimally designed and calibrated, not just employed because they were the least expensive option, or the most commonly performed, or what was available.

What can we do about it? First and foremost, get education. When testing is necessary, be particularly picky about which tests are performed. Don’t settle for “most common” – make sure that it’s “best available”. Recognize that tests are often flawed, and have the test done more than once to confirm a positive OR a negative. Ensure that you have an artful, experienced expert to interpret your tests: lab test interpretation can be tricky, and it takes creative, flexible thinking, in addition to skill, to make sure nothing is overlooked. And finally, get lots of opinions, across disciplines: practitioners frequently see what they’re familiar with, so get the opinions of practitioners in different fields to make sure you have a whole picture.


  1. Heather@Mommypotamus on 11 December, 2011 at 10:00 pm

    Great article, Katja! So interesting about the variance in lab results. I’m curious, are there any alternative methods of diagnostic testing you think are more beneficial? Though not truly diagnostic I often visit my herbalist when I’m trying to unravel a mystery with my body. She can usually tell me quite a bit based on how my body responds via muscle testing. And now she has this sensor thingy that I have to put my hand on for five minutes. Not really sure what it is . . .

  2. katja on 12 December, 2011 at 4:59 pm

    I know a lot of folks who use muscle testing, but I don’t – I find it too unreliable.
    When we eat things that we are allergic to, or eat things that are otherwise problematic for us, we get an endorphin response. This is part of the reason that people often crave the most the thing they’re allergic to – because they’re getting a “boost” when they eat it in the form of an endorphin response.

    The problem is that, when you’re muscle testing, it’s impossible to tell whether your body is responding positively to a thing because it is causing a positive vital response, or because there is an endorphin response. The body responds with energy in both situations, how can you tell which is which? There is in fact no real way, since muscle testing is only looking for that vital response, there’s no real way to differentiate between types of vital response. So I stay away from it – I’ve seen too many people come in who were sure they could safely have this or that because they muscle tested, only to find out that when they finally did do an elimination and rechallenge, what had been muscle tested was wrong.

    There are a lot of these types of testing modalities, and they’re not bad, they’re just only useful as far as they’re useful. Some I like are pulse testing – after you’ve eaten something, does your pulse race? It’s a nice way of getting some information, but it’s also fallible: sometimes the negative response you have doesn’t come till later, or comes in some other way. I like glucose testing, if you have access to a glucometer and are willing to bleed a little – if what you’ve eaten is spiking your insulin, that’s a problem. Again there, you won’t catch every problem, but you’ll find a lot of them. I even use some conventional testing some times, though again, you have to be picky: pick a good lab, and recognize that it’s only worth what it’s worth.

    All these things can be useful. Usually, I start by looking at the body, at the person, at the energy levels. Usually, if you’re muscle testing something, you know you might have an issue with it – you’re testing a finite list of “suspects”. Usually these suspects are the ones you “would die without” – after all, isn’t that the definition of addiction? I’d rather see folks try elimination with those suspects. I might follow up with pulse testing or a glucometer or whatever, to help a person who needs extra validation, but generally once we’ve explained things in terms of addiction (and explained the differences between what humans ate for the first eons vs. the last few hundred years), people are pretty ready to dive in with elimination.

    It’s not the easy way, but then again, the “easy way” often ends up to get us into more trouble! (I am remembering any number of “great ideas” i’ve had over the years that were supposed to be “easier”…)

  3. Heather@Mommypotamus on 14 December, 2011 at 12:32 am

    Very interesting! I’ve done pulse testing and muscle testing but had never considered the generic nature of the vital response. Do you think that herbs and such that a person had never taken before would yield a more reliable response since the body would not have had the opportunity to make an endorphin connection?

    And oh, I have no idea what you’re talking about. ALL of my “shortcuts” and “easier ways” have worked through the years. Hee!

  4. katja on 19 December, 2011 at 1:49 pm

    Hm. I think that, if muscle testing is reliable, then an herb that a person has never taken before would register as it would according to that person: if it’s caffeine, and the person is type A, it will probably register as “zoooom – great!” and if it was chamomile for the same person, you’re just as likely to get a negative response, because that person doesn’t want to relax. They think relaxation is bad bad BADEVILDEATH! and anything that suggests that to the body is going to get the muscle-testing equivalent of “pbbbbbt!”

    So you could use muscle testing to do that, but it doesn’t excuse you from having to actually know what’s right for the person 🙂

    A lot of students I see unfortunately often want to use muscle testing as a way so that they don’t have to learn all the infinite intricacies of the herbs, or of the person they’re working with – or as a way to circumvent their own lack of confidence. Unfortunately, you have to study, you have to observe, and you have to find your spine – whether you muscle test or not! 🙂

Leave a Comment

You must be logged in to post a comment.


Join our newsletter for more herby goodness!

Get our newsletter delivered right to your inbox. You'll be first to hear about free mini-courses, podcast episodes, and other goodies about holistic herbalism.