On ‘Blind Spots’

Dr. Makary's new book is taking the country by storm. He is being interviewed everywhere. I tried to order a hard copy of his book from Amazon, but it is sold out and on backorder. So I ordered it on Kindle and have started reading it. I confess that the chapter on the WHI (on hormone replacement therapy, HRT, in post-menopausal women) is the first chapter I read.

I find it interesting that if one is able to hang around long enough, some mistakes eventually get corrected.

Let me explain.

In Nov 2015, I incurred the ire of Russ Roberts, the conservative or libertarian economist (at least I think he is conservative; he may disagree). He had an episode of EconTalk with a guest, Robert Aronowitz, a physician at the University of Pennsylvania, who had just written a book, "Risky Medicine" regarding medical screening and attempts at risk reduction through screening, which is a fraught endeavor. The episode was very good, and I listened with great interest. The difficulty in making decisions on screening for various diseases (mammograms, PSAs, etc.) was explored in considerable detail.

Toward the end of the podcast, Roberts asked his guest if any study in medicine gave us a completely solid answer to what should be done in a given medical decision process. To my great surprise and chagrin, Dr. Aronowitz said, yes, there is. He then cited the Women's Health Initiative study that he said definitively showed that we should avoid the use of HRT, except for as short a time as possible to treat severe menopausal symptoms. Roberts noted his response and did not challenge it in any way. That was some 13 years after the WHI had been stopped and the study published. I read the study when it was published. It had so many flaws it was hard to know where to begin. Not only was it not THE definitive study to guide medical recommendations on Menopause, but it was also one of the worst studies ever done. In my opinion, the study methods constituted medical malpractice.

I placed a comment to that effect, detailing the study's vast shortcomings, stating that I felt the guest, and Dr. Roberts by extension, were guilty of promulgating false and misleading information.

To wit: The study participants' average age was 62, a decade or more post-menopausal. It was not applicable to women a decade younger early in menopause. The study was stopped before statistical significance was reached regarding breast cancer, but the study was promulgated as if it had proved that estrogens cause breast cancer. (Never proven before or since). The study did not report the weights of the participants. Obesity is associated with a higher risk of breast cancer. No information could be gleaned from the study regarding the relation between obesity, HRT, and breast cancer. Premarin was used, which had been pretty much supplanted by estradiol when the study was undertaken. Topical estrogens were not considered, despite the common use of transdermal or vaginal estrogens at the time the study was done. Could there be a difference between the route of delivery of the estrogen? In reading the study, it appeared that women who were smokers were not required to quit smoking before being placed on HRT. This is medical malpractice, as HRT and smoking increase the risk of vascular events. Further, no one was starting women in their mid-60s on HRT.

Why was this older cohort selected for this study? Perhaps because the question of menopausal symptoms was no longer a consideration in whether to start HRT and would not be a factor in a blinded prospective study. But that decision compounded the whole problem with the study.

I received a rather harsh private reprimand from Russ Roberts (justified? Probably.) for my supposed lack of civility, with a threat to ban me from the site if such behavior continued. No mention of any response to the issues I had raised. The guest posted a note that cardiovascular events were shown to increase with HRT but elided the issue of starting HRT early post-menopausally vs. a decade later. That makes all the difference. But it was true that cardiovascular problems increased when HRT was started a decade after menopause, and that is a solid conclusion from the study.

I was practicing in the 1990s when the Study was underway. It was not a minor or trivial study. It was the largest study ever done on women's health. It had been started with great fanfare. Bernadette Healy was the head of the NIH, the first woman (a cardiologist) to achieve that distinction. There was an intense discussion about the fact that women's health issues had been neglected in Medicine, that research had not been oriented toward women's health. This was the study that was going to end that discrepancy, that unfairness, that "blind spot." (As an aside, the current issue of Endocrinology Today features Monica M. Latonia, PhD, and Jane E Reusch, MD. Their work on women's health issues and reproductive medicine, and their attendance at Joe and Jill Biden's White House Initiative on Women's Health Research meeting at the University of Colorado, was ostensibly part of an initiative to "fundamentally change how our country approaches and funds research on women's health"--as if that had not been thought of or done before).

There was at the time of the WHI study a distinctly feminist trope to the study. Women were in charge and were going to right the wrongs; that women, just like men, were able to control their own fate and the resources of society to their advantage.

There was also a considerable consensus in the medical literature and among medical academics that it was obvious to them that estrogens caused breast cancer, and that HRT caused breast cancer. They were sure that adequate studies to prove the causality had not been done. This study was going to correct that deficiency.

That bias persists, although neither the WHI study nor any other study or set of studies has demonstrated it. Some studies show the opposite (as Dr. Makary notes in his book). It was an academic article of faith.

I stopped listening to EconTalk podcasts then.

Out of interest, I checked to see if Dr. Roberts, now President of Shalem College in Jerusalem and a Fellow of the Hoover Institution, had a podcast with Dr. Makary, and sure enough, he did. So I listened and heard Dr. Makary thoroughly deconstruct the WHI study and note how wrong the conclusions drawn from the study were, and how harmful those conclusions have been to women's health in the U.S. A very good presentation.

To counter that just a bit: There have been a lot of practitioners in the US who understood from the outset that the study was bunkum and had no application to HRT in menopausal women, but just to women a decade or more past menopause. Indeed, there is no time constraint on how long HRT can be used if started early in menopause, as long as untoward events do not supervene. I would say that the majority of Gyn physicians in my experience have ignored the study, even if not from the outset.

To make it personal, my wife required a full hysterectomy at age 50 and has been on estrogen transdermally ever since (e.g., for 28 years). Our daughter, an OB-GYN physician trained at USC fully agrees with that treatment. Yearly mammograms have been OK (perhaps because she is on estrogen). She had severe osteoporosis at age 50, having grown up in utter poverty in the northeast of Brazil, with marginal nutrition as a child. A 2-year course of Forteo (teriparatide) corrected her osteoporosis, and estrogen treatment has maintained her bone density stable for the subsequent 25 years. (In fact, her bone density has gradually improved over that time.)

I was hoping that Dr. Roberts might mention his prior podcast with Dr. Aronowitz in 2015. He made no mention of it... But it is in the archive of EconTalk podcasts.

Dr. Roberts got it right now, somewhat belatedly, and has corrected a long-ago mistake.