Intro. [Recording date: May 20, 2026.]
Russ Roberts: Today is May 20th, 2026. Before I introduce today’s guest, I want to inform listeners that we’ll be launching an EconTalk Book Club centered around The Iliad by Homer. Our first episode will feature Ido Hevroni from Shalem College in Jerusalem, who has taught The Iliad for over a decade. This will air, assuming all goes smoothly, on July 6th.
This will offer valuable context to help you dive into the book, and we plan to have at least one or two more episodes in the following weeks. We’ll be utilizing the Fagles translation, although there are numerous others available.
Now, onto our guest for today, Dr. John Mandrola. John is a cardiac electrophysiologist specializing in heart rhythm disorders. Our discussion today revolves around an article he penned for the substack Sensible Medicine, which he edits alongside former EconTalk guest Adam Cifu.
John, welcome to EconTalk.
John Mandrola: Thank you for having me. It’s a real honor to be on a podcast that has hosted Milton Friedman and Thomas Sowell, so I’m quite excited.
Russ Roberts: Well, thank you.
Russ Roberts: The title of your article is “Bravery (and Humility) Is Needed to Do Proper Medical Science.”
Let’s kick things off with some context. Imagine you’re playing basketball, tennis, or skiing, or perhaps you’re just bending down to pick something up. Suddenly, you experience excruciating knee pain, maybe even a locking sensation. You visit the doctor and discover you’ve damaged your meniscus. ChatGPT will kindly provide you with an image of this little shock absorber that protects the bones in your knee.
So, what are your options upon receiving this diagnosis?
John Mandrola: When you consult a doctor, they will likely evaluate your condition and may order an MRI or X-ray after some time. If these images reveal an anatomical abnormality, such as a meniscal tear, the doctor might recommend arthroscopic surgery.
What strikes me about these treatment plans is that a patient might improve simply due to the natural progression of the illness—what we call “natural history.” They could improve because of the surgery, or perhaps through a combination of factors, such as the placebo effect—where the expectation of recovery plays a significant role. In many medical cases, there are multiple reasons why patients experience improvement.
The study we’ll discuss is particularly interesting because it employed a proper placebo control, which is standard in drug trials but often overlooked in surgical trials.
Russ Roberts: I have a friend who’s an emergency room physician. When I ask him about a health issue and it resolves, he often shrugs it off, saying, ‘Or passage of time.’ It seems many doctors lack the perspective you’re suggesting.
But let’s delve into the study. Just to clarify, laparoscopic surgery is essentially non-invasive, correct?
John Mandrola: Well, Russ, anytime you insert a tube into a joint, I would consider it invasive. While it’s less invasive than traditional open knee surgery, it’s still a surgical procedure.
Russ Roberts: So, how prevalent is this issue? Do you have any statistics on how common these surgeries are for athletes?
John Mandrola: According to the paper’s introduction, around 700,000 arthroscopic partial meniscectomies are performed annually in the United States. So, this is a significant number.
Russ Roberts: That seems quite high. What did the study entail?
John Mandrola: It’s crucial to consider who was included in the study. The patients had chronic tears rather than acute injuries. In this study, 146 patients underwent arthroscopy—
Russ Roberts: Please explain.
John Mandrola: Arthroscopy involves inserting a tube to visualize the knee. The surgeon examines the meniscal issue and then randomizes the patients to either repair the meniscus or perform a sham operation, where they merely simulate the procedure without actually repairing anything. The patient remains unaware of whether they received genuine surgery or a sham.
Russ Roberts: Wait, how is that feasible? Wouldn’t I wake up and see the surgical team?
John Mandrola: No, you’re under light sedation. There’s also a nerve block in place, and the patients might listen to music. During sham-controlled studies, patients often cannot guess their treatment assignment, especially if the blinding is effective.
Post-surgery follow-ups are conducted by individuals who are unaware of whether the patient had actual surgery or sham surgery.
Russ Roberts: But wouldn’t I have a scar? Wouldn’t they make an incision in both cases?
John Mandrola: Correct. The elegance of this study lies in the fact that all patients undergo the arthroscopic procedure. They all have an incision made for the tube insertion; the difference is that one group receives the actual repair, while the other group has no repair performed, effectively leaving them with the same external appearance.
Russ Roberts: You mentioned in your article that there are ethical concerns here. Performing sham surgery on someone who might genuinely need it seems borderline unethical. Yet, the goal of the study is to determine whether the surgery is effective. This creates an interesting ethical dilemma: is it okay to do this for the greater good?
John Mandrola: That’s the crux of the issue, isn’t it? With 700,000 procedures performed annually, we must question the ethics of conducting surgeries that yield no better results than sham operations. Conversely, we must also weigh the ethics of operating on a mere 145 patients. There are numerous examples in cardiology where invasive surgeries have proven ineffective when subjected to placebo testing.
While I’m not an ethicist, I recognize this tension. However, without placebo-controlled studies, we cannot ascertain the efficacy of these procedures.
Russ Roberts: A somewhat ironic point to mention is that many would argue, ‘Those 700,000 procedures were done with good intentions, whereas the sham was merely a way to play with someone’s knee.’ There seems to be a discrepancy in how we view motivation. It’s fascinating how our ethical judgments are clouded by perceived intentions, ignoring that many perform surgeries for reasons beyond patient care.
John Mandrola: I understand your perspective, but one could argue that arthroscopy has a diagnostic aspect. The surgeon might discover something else while examining the knee joint.
In cardiology, we have a notable study in which patients with severe coronary disease received either a stent—an intervention—or nothing at all. Ultimately, the study revealed no difference in exercise capacity between the two groups.
They navigated the ethical concerns by allowing patients to undergo the stent procedure after a waiting period, ensuring they wouldn’t know whether they received the stent or a placebo.
Russ Roberts: That’s a clever approach.
John Mandrola: Indeed. The key takeaway is learning what works and what doesn’t.
Russ Roberts: Before we discuss the actual findings, do you have any statistics on how often actual surgeries result in complications? We’re considering not just effectiveness but also the potential harm they can cause, like infections. Is that accurate?
John Mandrola: Absolutely. This is a crucial point. Every surgery carries risks, including infections, anesthesia complications, and bleeding. If a procedure proves no more effective than a placebo, understanding these risks becomes critical.
Russ Roberts: So, what were the findings of the study?
John Mandrola: Interestingly, when examining pain score responses—subjective measures—both groups improved equally. There was no discernible difference in outcomes. The surgery group did not outperform the placebo group.
Essentially, the surgery performed 700,000 times a year yields no better results than a placebo procedure.
Russ Roberts: We need to revisit that point: why didn’t they include a no-surgery group? Surely to truly assess this, one would want that comparison. Isn’t that a significant oversight?
John Mandrola: Generally, it’s not done. I can’t pinpoint exactly why, but I suspect logistical challenges and assumptions about outcomes play a role. However, including a no-surgery group would be the most effective way to measure the true placebo effect.
Russ Roberts: So, they found no significant difference when comparing the sham surgery to the actual surgery. Ironically, the term ‘sham’ refers to the group that only received the scope, but if the surgery is ineffective, aren’t we all a bit “sham” in this context?
John Mandrola: Absolutely. A prominent group in London—the Imperial College—advocates against using the term ‘sham’ since it implies a lack of value. Instead, they prefer to call it a placebo procedure, which is more accurate.
Russ Roberts: Finally, I want to highlight a point from the study: “More patients in the surgery arm,” meaning those who underwent actual surgery, “progressed to high tibial osteotomy or total knee replacement, 12% versus 4%.” This suggests that the surgery may have weakened the knee, making a replacement more likely.
John Mandrola: Yes, those numbers are small but significant. It highlights not just the potential harm, inconvenience, and cost of surgery but also the risk of subsequent issues.
The study I referenced in Sensible Medicine was a 10-year follow-up, which is remarkable in itself. As a clinician, these studies teach us about the efficacy of procedures and the role of placebo responses, while also emphasizing the importance of the caregiver-patient relationship.
Russ Roberts: What exactly is the nocebo effect?
John Mandrola: The nocebo effect is essentially the opposite of the placebo effect. The placebo effect occurs when a patient experiences improvement because they expect it. For instance, the color or size of a pill can amplify the placebo effect, and surgery typically carries a greater expectation than a pill.
Conversely, the nocebo effect arises when patients anticipate harm, leading them to feel worse. Statins, medications used to lower cholesterol, exemplify this. In blinded trials, side effects are indistinguishable between those on statins and placebos. However, observational studies reveal that many patients report adverse effects once they know they are taking statins.
In clinical settings, we frequently encounter patients claiming statins cause them problems, often stemming from the nocebo effect.
A compelling study—the SAMSON [Self-Assessment Method for Statin Side-effects Or Nocebo] Trial—randomized 100 patients who believed they were statin-intolerant. They alternated between taking a statin, a placebo that looked identical, and no tablets at all. Remarkably, the best results occurred during the no-tablet phase, while the side effects reported were similar regardless of whether they took the statin or the placebo.
Russ Roberts: I must apologize; I’ve forgotten the source, but I recently read about a construction worker who fell off a ladder, and a long nail pierced through his boot. He was in excruciating pain until they discovered the nail hadn’t penetrated his foot at all. This is the ultimate nocebo, isn’t it?
Pain is a peculiar phenomenon, deeply intertwined with our minds. The brain possesses extraordinary powers, as evidenced by the studies you’ve discussed. The real question is how we can harness that potential.
Before we delve deeper into that, let’s address a more fundamental inquiry. We previously discussed vertebroplasty, a procedure aimed at alleviating back pain. Often, pain isn’t localized to the area where we assume it originates. For instance, my mother had a cracked vertebra. I recall that vertebroplasty trials involving placebo controls revealed no significant benefit from the actual procedure.
My mother underwent the surgery and experienced instant relief, walking out of the hospital the next day. Someone pointed out that injecting the cement might provide additional rigidity to her back, possibly offering temporary relief. However, the challenge remains: what actionable insights can we derive from these discussions? Are we suggesting that if someone has chronic pain, they should simply wait it out?
John Mandrola: The key takeaway is that we must inform patients that while pain can be distressing, it often improves over time without intervention. By conveying that surgery is no more effective than placebo, we can guide patients toward understanding that they will likely recover naturally, and we can support them through this process.
I once delivered a lecture titled “Words Can Harm and Words Can Heal.” What we glean from these sham surgeries is that patients often do improve through their natural course or through alternative measures, but they will improve. We need to partner with our patients and support them, which may sound lofty but is indeed true.
Russ Roberts: Good advice is hard to monetize. That’s one of the challenges. Meanwhile, performing surgery comes with a neat Medicare billing code, right?
John Mandrola: Every doctor will affirm that it’s significantly easier to perform surgery than it is to explain why it’s unnecessary.
Russ Roberts: Why is that?
John Mandrola: You do receive compensation for your time and the satisfaction of helping someone more directly, which is often more gratifying than wielding a knife or a catheter.
Russ Roberts: So, I walk into your office with knee pain, and you say, ‘The surgery is unnecessary. You’ll likely heal on your own.’ But patients often believe they require some form of intervention for their recovery, even if it’s just a placebo.
John Mandrola: That’s where the doctor-patient relationship is crucial. You can reassure the patient that many similar cases resolve on their own. I had a labral tear in my hip, and a surgeon told me, ‘You’ll recover naturally. Surgery won’t provide any additional benefits.’ This advice was immensely valuable to me.
Russ Roberts: We don’t offer medical advice on this program. It’s vital to take note that not all labral tears will heal independently.
John Mandrola: Exactly. My labral tear did heal on its own.
Russ Roberts: I had similar experiences with shoulder injuries. After two incidents, one of which was a rotator cuff tear, I received an injection that worked wonders. I avoided surgery for frozen shoulder and found myself fine without it. But I acknowledge that everyone’s situation is different.
John Mandrola: Absolutely. I don’t want listeners to believe that we should dismiss surgery indiscriminately. Some conditions indeed require intervention.
For example, I developed atrial fibrillation—an irregular heart rhythm. As a cardiologist, I was apprehensive about undergoing a procedure myself, so I chose to monitor it. It eventually improved, validating the notion that many conditions may not necessitate immediate intervention.
Russ Roberts: That’s fascinating.
John Mandrola: I have a patient who meticulously tracked his atrial fibrillation episodes. When we reviewed his progress, he had transitioned from numerous episodes to virtually none. Had he undergone the procedure earlier, we might have deemed it a success, even if it was his condition that improved naturally.
As clinicians, we should embrace the idea that not every ailment requires invasive action. It’s often more beneficial to adopt a conservative approach, allowing time for recovery.
Russ Roberts: I worry that my intellectual understanding of these concepts may lead to inaction. My son urges me to visit a doctor for my leg pain, but my knowledge of the placebo effect and medical reversals makes me hesitant. Is it possible that I’m masking my fear with intellectual reasoning?
John Mandrola: You and I share a similar mindset. I also have a greater fear of interventions than of the conditions themselves.
However, we must differentiate between various conditions. Some warrant careful monitoring with an expert physician. It’s entirely reasonable to adopt a wait-and-see approach for certain issues. On the other hand, there are circumstances where immediate intervention is crucial. This is where a wise physician can guide you—some conditions should be monitored, while others necessitate timely action.
Voltaire once remarked, ‘The best physician humors the patient until nature heals the patient.’ I believe this approach is underutilized but remarkably effective.
Russ Roberts: Let’s shift gears to the concept of medical reversal. We’ve previously discussed this phenomenon—where initial observational studies suggest a treatment is effective, only for randomized controlled trials to reveal otherwise. It’s alarming to consider how many procedures and medications may be ineffective or even harmful.
However, I hope there are still many treatments that genuinely work based on rigorous trials. How do we discern the reliable ones from the overhyped procedures?
John Mandrola: There are indeed numerous reversals, often stemming from overconfidence in observational studies. However, many treatments and interventions are effective, and we’re fortunate to live in a time with significant advancements in medicine.
For instance, we’ve made tremendous strides in treating congestive heart failure, enabling patients to live extended lives with chronic conditions. While we should acknowledge the reversals, we must also recognize the multitude of interventions that have proven effective.
What we learn from these reversals is the importance of humility and the necessity for rigorous evaluation before accepting treatments as standard practice.
I advocate for a culture of randomization within medical practice. For example, in Denmark, a significant portion of patients are involved in randomized trials, allowing continuous evaluation of various treatments. This culture could greatly enhance our understanding of interventions.
Russ Roberts: Your essay was titled “Bravery (and Humility).” I assume bravery refers to the courage to admit when a practice is ineffective, which is also tied to humility, correct?
John Mandrola: Yes, bravery is required to randomize patients to sham surgeries—it truly takes courage.
Russ Roberts: That’s a fair point. The recent study we discussed, the FIDELITY [Finnish Degenerative Meniscal Lesion Study] study, was a 10-year follow-up. You’d expect that findings demonstrating the ineffectiveness of meniscal repairs would lead to a decline in such surgeries. Yet, I suspect we won’t see a significant change among orthopedic surgeons.
John Mandrola: That’s a valid observation. Many issues arise when translating these highly selective studies to the clinical setting. While this study involved a limited number of patients, there remains a plethora of knee injuries and conditions that differ from those in the study.
Surgeons may argue that while this specific group showed no benefit, numerous other cases may warrant surgical intervention. The challenge lies in reconciling these selective studies with the broader spectrum of patients we encounter in clinics.
Russ Roberts: But isn’t it true that when you have a hammer, everything looks like a nail? There’s a tendency to find justifications for interventions, particularly when they’ve been performed successfully in the past. It seems challenging to resist the urge to perform surgery when it’s your livelihood at stake.
John Mandrola: Absolutely. In cardiology, we’ve known for years that stenting coronary blockages doesn’t necessarily prevent heart attacks or prolong life, yet the allure of performing an intervention remains strong. It’s not just about financial incentives; it’s also about the satisfaction derived from fixing issues.
However, the beauty of placebo-controlled trials is that they impart a sense of humility. We learn that not all interventions yield the results we anticipate, which is a valuable lesson for all medical practitioners.
Russ Roberts: I completely agree. I recall when my mother sought my advice about undergoing vertebroplasty. I was apprehensive, knowing the complexities involved and the data surrounding such procedures. The particular surgeon she consulted preferred general anesthesia, which carries additional risks.
Our tendency to err on the side of intervention is a common human trait. People often want immediate action upon discovering a potential health issue, even if it may not significantly impact their quality of life. This emotional drive can overshadow the more rational approach of monitoring the condition.
John Mandrola: I completely agree. I’ve heard you discuss the PSA test, and I share your sentiments. I prefer to avoid unnecessary testing, as it can lead to a cascade of interventions that may not be beneficial. The same applies to colonoscopies; studies suggest they don’t significantly affect longevity.
When we discover potential health issues, it’s often better for our peace of mind to refrain from invasive testing and intervention.
Russ Roberts: To wrap up our discussion, let’s revisit the concept of being a medical conservative, as you previously articulated in your 2019 essay with Adam Cifu, Vinay Prasad, and Andrew Foy. I’m curious: What exactly constitutes a ‘medical conservative’? Have you faced criticism or praise for your views over the years? And do you sense a growing recognition of our limitations as medical practitioners?
John Mandrola: First and foremost, my stance has remained unchanged. As medicine evolves, I find myself leaning even more towards a conservative approach. Secondly, I have a small podcast in cardiology called This Week in Cardiology, where I often hear from young professionals who feel inspired by our discussions. They approach me at conferences, sharing that they cannot voice these thoughts to their professors, yet they appreciate the perspectives we offer.
So, while it may seem like a lonely path, I believe we’re gradually influencing the next generation of medical practitioners.
