Curmudgeon's Corner: SIO-ASCO and the Difference Between Type I and Type II Errors.
In June 2018, ASCO officially adopted guidelines for the integrative care of breast cancer patients that were initially created by a committee of naturopathic physicians for the Society of Integrative Oncology (SIO). Some consider this adoption to be good news, a mile marker in integrative cancer care. For we perpetually underdog naturopaths even to be noticed by ASCO is exciting, something like being offered a seat at the grownup's table. To have ASCO endorse ND work, well this sounds fantastic, at least when you're reading the headlines.
Reactions have varied depending where on the medical spectrum the viewer seems to practice. Some have heralded this as a landmark achievement for integrative medicine: Fred Hutchinson's website describes "... ASCO's endorsement represents a milestone in the integrative oncology field," (1)
And for others this adoption is solely an excuse to try out new derogatory terms: "The advance of quackademic medicine in oncology continues apace." (2)
For most of us as we read the actual guidelines, we can't help but be left feeling puzzled, surprised by how much good information just seems to have been left out.
All this sadly reminds me of the late Steve Austin. He explained all of this decades ago, and this only serves to remind me of how much I miss him.
The best way to make sense out of ASCO's adoption of SIO's Guidelines for Integrative Care of Breast Cancer is to go back to when Steve Austin, ND, explained the difference between NDs and MDs was what each profession was afraid of. I think I read this in his and his wife Cathy Hitchcock's 1994 book Breast Cancer. (3)
This is far enough back that I may need to find a real copy of the book and type the text by hand. Before I do that, let me back up and provide some more details so you understand what has me thinking about all this.
ASCO, the association of the very straight MDs who specialize in cancer treatment, made a shocking step recently by accepting a series of guidelines from SIO, that is the Society of Integrative Oncology, the relatively open-minded group that investigates and sometimes incorporates alternative therapies into their care of cancer patients. There is normally a wide gulf between the two groups. SIO lets naturopathic doctors become members.
SIO was founded in 2003 as a professional organization for those practicing integrative oncology, so still a relatively young group. Claiming to be integrative, they were obligated to start accepting naturopathic doctors as members. Eager beavers that we are, naturopathic doctors quickly rose to take leadership positions in the group. Heather Greenlee, ND, served as president in 2014, and Suzanna Zick, ND, in 2015. In fact, it was Heather Greenlee, ND, PhD, who co-chaired the SIO guideline task force that developed the set of guidelines for integrative adjunctive care for during and after breast cancer treatment. These guidelines were published in 2017. In June 2018, ASCO pretty much adopted them, with a few added caveats.
This is a massive achievement from the perspective of naturopathic medicine building bridges and working across the spectrum of medical thought. There has been a great deal of back patting and congratulations. Indeed, this congruency of thinking is unprecedented in the realm of naturopathic and conventional medicine. I'll buy Dr. Greenlee a drink next time I see her, for sure.
There is a slight problem with the recommendations, at least from the perspective of anyone who actually practices naturopathic oncology and sees patients; the guidelines are underwhelming to say the least. They are so conservative and so dismissive of so many interventions that naturopathic medicine takes for granted that Heather and her ND colleagues who were part of this task force are getting equal parts abuse along with their deserved congratulations.
According to the guidelines the only acceptable therapies aside from the standard medical interventions are yoga, meditation, and acupuncture. Diet, exercise, or supplements do not have adequate data in the view of the straight oncology world to be suggested.
The SIO guidelines were based on an analysis of peer-reviewed randomized controlled trials published between 1990 and 2013. More than half of the patients in any study considered had to have breast cancer. The integrative therapy being studied had to be used along with standard therapy and it had to have a noticeable effect on symptoms or side effects of treatment. The bar was set high.
It is hard to believe these few recommendations are all that were found supported by adequate enough evidence to satisfy the members of the ASCO task force that took up this challenge. Medical oncology has standards for the level of evidence that are incomprehensible to the average ND or even for those who focuses on naturopathic oncology in practice, even for those who claim that their practices are evidence based.
In fact, various colleagues of my acquaintance are in such disbelief they are questioning the wisdom of these eminent colleagues on the other side of the spectrum.
Thus, I come back to Steve Austin, ND. He believed that the difference between MDs and NDs all boiled down to different types of error and our respective needs to avoid them.
In statistics there are two fundamental types of error, labeled simply Type I and Type II errors. In statistical hypothesis testing, a type I error is the incorrect rejection of a true null hypothesis (a "false positive"), while a type II error is the failure to reject a false null hypothesis (a "false negative"). A null hypothesis is the statement being tested, usually that there is no difference between two populations.
Let's try an example. There's an ongoing debate on whether taking oral curcumin will reduce radiation dermatitis. Back in 2013, a paper written by Julie Ryan and colleagues told us that they had given 6 grams oral curcumin per day to 30 breast cancer patients during their course of radiotherapy. (4) The null hypothesis for their study was that there would be no difference in frequency or intensity in radiation dermatitis between the placebo and the curcumin groups of women.
In this example, a Type I error would be incorrectly concluding that the null hypothesis was false and that frequency of dermatitis differed between the groups, that is the curcumin did something when, if truth be known, it didn't. In simple words, it is a false positive, believing curcumin is useful when in fact it isn't.
A type II error, "is the failure to reject a false null hypothesis (a "false negative")" so in this example it would be thinking there was no difference between the two groups when in reality there is. Again, in simple language a false negative.
Ryan et al reported that "... fewer curcumin-treated patients had moist desquamation (28.6% vs. 87.5%; P = 0.002)." and that is where the matter has stood until this year when Ryan and her team reported the results of a second study.
In a second much larger study that included 686 patients, the statistical analysis did not show a significant difference in skin damage from radiation so the null hypothesis, the curcumin did not help, stands. A type II error would be if we failed to find a reduction in dermatitis in the experimental group when in fact curcumin did help. (5)
In the late Steve Austin's words,
Many medical doctors tend to make the philosophical mistake of unconsciously assuming that if something is not proven, it doesn't work. Consciously we can all understand the fallacy of this position. It's tantamount to suggesting that before the link between vitamin C deficiency and scurvy was proven, people who ate foods rich in vitamin C were no better off than sailors on hardtack....As a result of the 'If it's not proven, it can't work' philosophy numerous effective treatments and preventive agents have been and are currently being ignored by many conventional medical doctors. Scientists refer to the mistake made when a useful intervention is considered useless because of lack of absolute proof as a 'type II error.' Conventional medical doctors make many type II errors, but not without good reason. If you deal with dangerous substances, you must be sure that these therapies do something useful before prescribing them to patients. The implementation of a therapy may be postponed by a few years while it's being proven. Although some precious time is lost in the process, the alternative--employing dangerous treatments, some of which turn out to be useless--could be a disaster.... But this kind of thinking doesn't make the same sense when talking about vitamin D. It's not chemotherapy. If a substance is inexpensive, has been proven safe, and reduces the risk of cancer in animals, should you really wait ten years for proof to develop before taking it, especially when you may not have ten years to wait? (3)
The difference between naturopathic medicine of the sort practiced by naturopathic oncologists and the standard of care world views practiced by ASCO's membership is what sort of errors we are willing to risk because of the difference in therapies we choose to use. We are fairly certain that there is little risk in taking curcumin for pretty much any patient. We do not mind taking it by accident, that is thinking it will help when in fact it does us little good. We are willing to risk a false positive. If in the end it turns out to provide no benefit, then as they say "no harm done."
If instead of curcumin we were contemplating starting the patient on a three-month course of doxorubicin, a particularly unpleasant form of chemotherapy, we would want to be absolutely certain it will provide benefit to the patient.
So, this is the difference between us and them. We are both interested in improving patient outcome. It's just a difference in the tools we are accustomed to using. For naturopathic doctors, the starting assumption is that the tools are safe and won't hurt. For the medical oncologist, the starting assumption is that the tools are dangerous and should only be used with the utmost of caution. For the naturopathic oncologist, the equation is the opposite: "It won't hurt, and it might help."
Perhaps this explanation will help some understand the SIO-ASCO breast cancer guidelines. They attempt to bridge two vastly different world views, and translating between these different paradigms does not come easy.
(1.) The full ASCO integrative breast cancer guidelines can be read here: https://www.fredhutch.org/en/news/center-news/2018/06/asco-endorses-integrative-oncology-guidelines-for-breast-cancer-patients.html
(3.) Hitchcock C, Austin S. Breast Cancer: What You Should Know (But May...). Prima Publishing. 1994:208-209.
(4.) Ryan JL, et al. Curcumin for radiation dermatitis: a randomized, double-blind, placebo-controlled clinical trial of thirty breast cancer patients. Radiat Res. 2013 Jul;180(l):34-43.
(5.) Wolf JR, et al. Oral curcumin for radiation dermatitis: a URCC NCORP study of 686 breast cancer patients. Support Care Cancer. 2018 May;26(5):1543-1552.
by Jacob Schor, ND, FABNO
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Society of Integrative Oncology-American Society of Clinical Oncology|
|Date:||Jan 1, 2019|
|Previous Article:||Genetic Expression and Lifestyle Choices.|
|Next Article:||Healing with Homeopathy: Finding the Remedy from the Miasm.|