Keywords: Hot flashes, magnesium; placebos, ethics

Magnesium and Hot Flashes: The Ethics of Placebos

By Jacob Schor, ND, FABNO

The American Medical Association has struggled to come to grips with the use of placebos for years. In 2006, the AMA’s Council on Ethical and Judicial Affairs (CEJA) decided that it is unethical for a physician to use a placebo in the guise of therapy without the “knowledge and cooperation of the patient.”¹ This opinion remains in place; a 2008 publication by the Council in the Journal of Clinical Ethics declared the following:

Ultimately, the deceptive use of placebos is not ethically acceptable because it may harm patients to a greater degree than it helps them. This is particularly true in cases when placebos are utilized to serve the convenience of the physician rather than to promote the welfare of the patient. Perhaps the most pernicious use of placebos is for mollifying a patient who is demanding, displays a difficult personality, or has a complex problem that has become frustrating to the physician. Placebos should never be used in this way because it is fundamentally inconsistent with physicians’ professional obligations to promote patients’ welfare and respect the autonomy of patients.²

Some people live in a world where ethical decisions are black and white choices. Some of us find we live in a more complex world, not just many shades of grey, but full rainbows of color, and ethical decisions are less clear. Prescribing a placebo is one of such area.

As an example of a placebo that isn’t clear cut, consider the changing status of magnesium for hot flashes. For the last half-decade, we’ve suggested it to breast cancer patients. And it seems that taking magnesium often helped.

The idea to use magnesium oxide originated about six years ago when two pilot studies reported that doing so reduced the number and intensity of hot flashes experienced by women taking either tamoxifen or aromatase inhibitors.

The first of these studies was presented in 2010, at the American Society of Clinical Oncology (ASCO) meeting in Chicago. Herrada reported on a pilot study of 22 women who received 400 mg of magnesium oxide three times a day for a month. All the women were receiving adjuvant treatment for breast cancer. “Ten (45%) pts achieved a complete resolution of hot flashes. Ten (45%) pts experienced at least a 50% reduction in the number of hot flashes per day. In two (10%) patients, no changes in the number of hot flashes were noted.”³

Results of a second pilot study were reported in 2011. A group of 25 breast cancer patients, also receiving some form of adjuvant treatment, were given 400 mg of magnesium oxide per day for four weeks, escalating to 800 mg if needed. Hot flash scores were significantly reduced. “Of 25 patients, 14 (56%) had a >50% reduction in hot flash score, and 19 (76%) had a >25% reduction. Fatigue, sweating, and distress were all significantly reduced.”⁴

Together these two studies, albeit small, were enough to suggest potential benefit. Thus, for the last five years we have recommended that women in similar situations try taking magnesium; it seemed helpful in about the same proportion of cases as hinted by these studies.

Then last year, in 2015, something confusing happened. A large placebo-controlled randomized trial was published that called these early results into question. The lead author of this new and far more comprehensive trial is Haeseong Park, the lead investigator from the 2011 pilot study.

Park’s new study enrolled 289 women between December 2011 and March 2013. All the women were postmenopausal with a history of breast cancer and had bothersome hot flashes. They were randomized into treatment groups who were given either magnesium oxide 800 mg or 1,200 mg daily or placebo. Treatment lasted a total of eight weeks.

Mean hot flash scores, mean hot flash frequencies, and associated changes during the treatment period were similar for all groups, including those receiving placebo. Taking magnesium increased incidence of diarrhea, not a surprise as this is a well-known side effect. There was no statistically significant difference in other toxicities or quality-of-life measures observed, in other words all the women did equally well.⁵ The study concluded that magnesium didn’t do anything. The reality is more complicated. Hot flashes dropped steadily during the study. But they dropped equally in the placebo group and the magnesium-taking groups. Magnesium oxide did not offer any more benefit than placebo did. Hot flash intensity and frequency changed by a bit more than 30% during the eight-weeks of the trial for both magnesium and placebo groups. See:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4442087/figure/F2/

It is not that magnesium wasn’t associated with improvement; it was. The problem is that the placebo worked just as well. These data suggest that the benefit of taking magnesium oxide for hot flashes is a placebo effect. Both of the earlier pilot studies were open label without placebo arms.

When we simply had data from the 2010 and 2011 trials, we could honestly tell patients that the evidence suggests that magnesium is helpful for hot flashes. What should we tell them now? Should we tell patients that magnesium works no better than a placebo, but that it still may help them? What about patients who have taken magnesium for years and think it helps? Do we tell them to stop taking it as its benefit is only as a placebo? Or to keep taking it because for them a placebo appears to work?

This brings to mind Ted Kaptchuk’s somewhat mind boggling study from 2010 on placebo effect in irritable bowel syndrome (IBS) that suggested it doesn’t matter what patients know. In Kaptchuk’s study, 80 women with IBS were given either “placebo pills” or no treatment. Despite the fact that those receiving placebo knew they were placebo, they still nevertheless did significantly better than the no-treatment control group.⁶

Some will point out that magnesium oxide is typically not our first choice for magnesium supplementation as it is poorly absorbed and tends to have a laxative effect. As the first two studies demonstrated benefit, this might not matter. The magnesium oxide seemed to work well enough. The fact that magnesium oxide is a laxative may enhance its placebo effect, and we should think twice before switching to “better forms” of magnesium. Benefit may have nothing to do with absorption.

Hot flashes appear to be quite responsive to placebo intervention.⁷ Placebo benefit has been reported in several prior randomized studies. In Boekhout’s 2006 review, about 25% of 1174 patients who received placebo or the intervention reported hot flashes reduced by at least 50%, and 15% had greater than 75% reduction.⁸ In Sloan’s 2001 review of seven randomized trials, the 375 patients who received placebo had an average 25% decrease in hot flash frequency and intensity.⁹ Hot flashes may just be particularly sensitive to placebo effect.

The steady decrease in hot flashes seen in the 2015 Park study may not be the result of placebo or magnesium at all; it may be the natural course of the symptoms. Hot flashes may simply improve over time, or the improvement may be a limitation of using diaries to record intensity and frequency. Women recorded their symptoms, and one might imagine that mere habituation and tedium led to a decrease in symptom recording.

We should also note that this same research group has a history of investigating several hot flash treatments and proving them ineffective, including vitamin E in 1998,¹⁰ soy in 2000,¹¹ black cohosh in 2006,¹² acupuncture in 2007,¹³ flax seed lignins in 2012,¹⁴ and paced breathing in 2013.¹⁵ Like magnesium oxide, black cohosh had shown benefit in an earlier pilot study.¹⁶

There is, in all likelihood, a “…psychological component of hot flashes. In fact, a previous small study demonstrated that distressing or problematic hot flashes were predicted by depression, anxiety, and low self-image, but not frequency of hot flashes. Therefore, women with equal numbers, and perhaps severity, of hot flashes may have very different emotional responses to their hot flashes; some desiring treatment, considering the symptoms bothersome, and others, not.”¹⁷

The more women perceive that they ‘have control over their symptoms’ the less bothersome hot flashes seem to be.¹⁸ Thus, taking magnesium or anything else that offers a sense of control over those ‘bothersome hot flashes’, even if the sense of control is an illusion, may in a way prove useful to take.

We should also consider that women with a history of breast cancer whose hot flashes are treated using the commonly prescribed drugs, venlafaxine, gabapentin or clonidine, are at high risk for adverse events. A November 2016 Cochrane review of 12 studies with a total of 1467 participants, reports that 81% of those women in the treatment group and only 19 % in the control group had adverse reactions.¹⁹ With this in mind, trying magnesium, even if it is only a placebo, might be a safer first option. Yet, ethically, it would now be considered a placebo and a violation of the patient’s autonomy. Perhaps, the ethically correct thing to do is to tell the patient magnesium is a placebo and suggest they take it anyway? Their symptoms might drop by 30%. I have yet to suggest this in this way to a patient but am looking forward to the conversation that follows.

Some conservative elements in the medical community think that any and all treatments prescribed by complementary and alternative medicine (CAM) practitioners are efficacious only because they elicit a placebo response. This opinion was harshly echoed in an online posting by Valerie Jones: “As far as I can tell, no CAM treatment has been proven effective beyond placebo…. Therefore, the effects of these treatments cannot be explained by inherent mechanisms of action, but rather the mind’s perception of their value. In essence, the majority of CAM treatments are likely to be placebo therapies, with different levels of associated ritual.”²⁰

While, to us, this seems to be a ludicrous generalization, it is a perception worth noting, as this is how many medical practitioners view our therapies. Being aware of this should make us more mindful of the evidence that does recommend what we do and more conscious of when the claims of benefit are well founded and when they are weak.

When confronted with lack of solid evidence, I have a catch phrase that helps me keep it all in perspective: “prunes, parachutes and poppies.” Prunes, because until quite recently there were no randomized controlled trials that prove they had laxative action yet they worked fine.²¹ Parachutes because we will never have a placebo controlled randomized trial to prove they are helpful, yet we are fairly certain they work. And poppies are worth remembering as an example of an herbal medicine whose effect is clearly not placebo.

In the case of hot flashes, it appears that placebos have relatively strong effect; that has left us with many remedies that work some but not all of the time.