← Rena Malik, MD Podcast
Rena Malik, MD Podcast
Why Full Body MRI Scans Are a Dangerous Waste of Money for Most People
01:41:37
In this episode, Dr. Rena Malik, MD and Dr. Matthew Davenport discuss the realities of whole body MRI scans, highlighting the risks of overdiagnosis, unnecessary procedures, and psychological harm from incidental findings. They provide listeners with expert advice on choosing evidence-based imaging tests and offer practical guidance for prioritizing long-term health and well-being.
Loading summary
Transcript162 lines
- [00:01]
A
This episode is brought to you by State Farm. Listening to this podcast. Smart move. Being financially savvy. Smart move. Another smart move. Having State Farm help you create a competitive price when you choose to bundle home and auto bundling. Just another way to save with a personal price plan like a good neighbor. State Farm is there. Prices are based on rating plans that vary by state. Coverage options are selected by the customer. Availability, amount of discounts and savings and eligibility vary by state.
- [00:30]
B
Limu Emu and Doug Here we have the Limu Emu in its natural habitat helping people customize their car insurance and save hundreds with Liberty Mutual. Fascinating. It's accompanied by his natural ally, Doug. Uh, Limu is that guy with the binoculars watching us. Cut the camera. They see us. Only pay for what you need@libertymutual.com Liberty Liberty Liberty Liberty Savings vary unwritten by Liberty Mutual Insurance company and affiliates Excludes Massachusetts. This whole body mri. My idea is based on the idea that I'm going to find something and I'm going to save somebody's life early. That's what it's based on. And gosh, if it did that, I can tell you the medical societies would be endorsing it full throated.
- [01:14]
A
They'd say yes, be covered by insurance.
- [01:16]
B
Be covered by insurance. They showed essentially by doing prostate cancer screening, it was confirmed to result in a 13% decrease in prostate cancer related mortality. Which is great. That's amazing. And if I had $2,500 throw around, I'd literally hire a personal trainer and a dietitian and that is going to prevent so much of my disease in my life. I mean, they say something like 70 to 80% of cardiovascular disease is potentially preventable if you just like do the necessary steps.
- [01:43]
A
Imagine getting an imaging test, hoping for peace of mind, only to find unexpected something that plunges you into months of terrifying uncertainty, biopsies and even unnecessary surgery, all for something that wasn't even that serious. This isn't just a hypothetical scenario. It's the reality for too many people. Welcome back to The Rena Malik, MD podcast, your trusted guide for leveling up your health, sexual health and relationships with evidence based tools. I'm your host, Dr. Rena Malik, urologist and pelvic surgeon. And today we're exploring the complex and often emotional landscape of full body MRIs and incidental findings. Joining me is Dr. Matthew Davenport, Service chief and vice Chair of Radiology at Michigan Medicine. He is an internationally recognized authority who has shaped national imaging policy through his leadership on the FDA Medical Imaging Drugs Advisory Committee and as former chair of the American College of Radiology's Committee on Drugs and Contrast media, with over 250 publications to his name, he's at the forefront of patient safety and imaging quality. Together, we'll share real world consequences of imaging tests, including full body mri, bust common myths, and discuss how to navigate these decisions with clarity and confidence. We want to make sure that you're taking care of your health the smart way. I'm so excited to have you here. I think there's so much misconception about what radiologists do, what is safe in terms of imaging, and we actually had a lot of questions from our audience, so I'm really excited to dive in. So tell me. I think the biggest question that I, that I personally have, and I think, I'm sure a lot of people do, is right now we're seeing a lot of celebrities, influencers, talking about full body MRI scans. So what is the value in getting a full body MRI scan? Should we be doing it? And what are some of the biggest misconceptions?
- [03:47]
B
Yeah, absolutely. So whole body MRI is essentially going inside an MRI machine and getting pictures taken of you from head to toe. And the goal of doing that is to try and find things that maybe you weren't aware of inside your body. Maybe an aneurysm or a cancer or other things that you might find inside yourself you weren't aware existed. That's the concept behind it. Now, the reason why it's a little bit of a problem is because whole body mri, when performed in an average risk patient population, so patients who don't have any particular risk of cancer, they're just an average adult living their life. What ends up happening is you end up detecting an overwhelmingly large amount of things that wouldn't otherwise help you to detect them. And this word that gets used a lot is over diagnosis or finding things that wouldn't otherwise hurt you. But now you're stuck with it. And once you've identified it, it oftentimes will increase rather than decrease uncertainty. And that uncertainty that's created leads to a lot of downstream problems for you as a consumer, because someone will then take invasive steps to try to figure out what that uncertain thing is inside your body. There are certain patient populations that do benefit from whole body mri. Those would be rare patients who have certain genetic syndromes that substantially increase their risk of cancer. The one that traditionally gets thrown around is Li Fraumeni syndrome. There are others. For example, patients who have Von Hippel Lindau syndrome will often get imaging of a lot of the different parts of their body on a regular basis. These sort of high risk cancer populations could benefit from doing MRI because the pretest odds, the pretest probability of important disease is high enough that when you find something, the thing you find is very likely to be really important. If your pretest odds of finding important disease are low, which is true in the general population, then the average thing you find ends up creating uncertainty and actually weirdly harm to you from the efforts to investigate what that thing is. So right now this is being explored primarily by for profit companies that are interested in selling a product to a consumer base. And it's not really recommended by medical society. And we'll get into this later on in our discussion today. But the average patient is more likely to be hurt than helped actually by this technique, most likely.
- [06:21]
A
And is there data behind that? Like, do we know that if the average person gets an mri, what percentage are going to end up having sort of invasive testing that may actually harm them rather than do them good?
- [06:32]
B
Yeah, great question. So if we take an imaging test of any part of your body, let's say just the abdomen and pelvis, for example, not even the whole body, but just the abdomen and pelvis. With MRI, about 20 to 40% of people are going to find something inside yourself. And the thing that you find will provoke some uncertainty on the downstream end. And it's not 100% of those that end up getting worked up, but many of them do a lot of time that can result in additional imaging tests like MRIs or CT scans, or in some cases PET scans or other imaging tests, or invasive tests like endoscopies or biopsies, or in some cases, surgeries to manage the thing that we found inside the body. The largest study that I'm familiar with that's investigated, how likely are you to detect a cancer inside your body? Getting a whole body MRI? In an average risk population, it's about 2%. And 2% is actually a pretty big number. I would say 2% is a pretty big number. But the twist in the data there is that cancer is not monolithic. When we hear the word cancer, we think that all cancer is bad and all cancer needs to be detected early, all cancer needs to be treated. And that's actually not true. There's a lot of cancers which are indolent, which means they don't really grow that much, or they kind of stay where they are and don't mind their own business. Basically that detecting it actually increases your harm because of efforts to try and treat that specific disease. So there Are no studies that show that doing whole body MRI in a general risk population increases the length of your life or increases the quality of your life. And for screening studies that are done on big populations, those are kind of the outcomes we really need to see. If someone's doing a screening test of the whole body, we want to be able to see that in a randomized trial you see that patients who undergo screening end up having lower risk of mortality or have a lower risk of developing advanced stage tumors. And that data does not exist for whole body mri. And there's some specific examples we can go into later in which well intentioned efforts to screen end up causing substantial harm to individuals and to populations, which I can share with you later.
- [08:43]
A
I would love that. Before we do that, let's just take it one step back. What is an mri? How is it different from a CT and a PET scan?
- [08:49]
B
Absolutely. MRI is an acronym. It stands for magnetic resonance imaging. And it uses a very strong magnet that when you get put inside the MRI machine, it kind of gives yourself a little magnetic moment and then through doing some sophisticated things can measure different properties of tissues inside your body. The nice thing about MRI is it does not use ionizing radiation, which is like with CT scans, ionizing radiation is associated with a small increased risk of carcinogenesis. MRI does not really have that. The whole body MRI techniques that are being used in these for profit companies typically do not use contrast material. We may get into this a little bit later, which is something that gets injected into your vein to help you see things a little bit more clearly. They tend not to use that. So MRI is a way to see very sophisticated things inside your body at a pretty detailed level. It's really good for evaluating soft tissues. Maybe if you're a football fan and you watch someone gets a knee injury or something, they'll commonly get a knee MRI to figure out what's going on with the ligaments. It's also used for solid organs like the liver or the kidneys or the brain. So it's a way of evaluating anatomy and physiology of tissue. And a PET scan, a PET scan is, that's another acronym that stands for positron emission tomography. And what that usually does that. When people usually. When they say PET scan, what they usually mean is measuring the metabolism of sugar. And it's commonly used in patients to try and identify things in which sugar consumption is higher, like for example, tumors or inflammation or infection. PET scans do use ionizing radiation.
- [10:29]
A
And so the whole body MRIs are not using any contrast material, would they be any different, like in terms of harm or benefit if they were using contrast?
- [10:40]
B
Yeah, good question. For patient populations that have an indication for whole body mri, these cancer syndromes I mentioned to you before, they actually do use contrast material and they use it because it allows you to detect and characterize things that are inside your body. In that case, the benefits of giving contrast outweigh the harms because that pre test probability of disease is so high in an average or general risk population. These companies are electing not to use contrast material because it raises a lot of considerations for them as a company, and I don't want to speak on behalf of them, but it involves putting an IV catheter in. It involves being administered a contrast material that might have direct potential harm to you. It's rare for that to occur, but you're sort of trading off diagnostic accuracy for harms from the contrast.
- [11:31]
A
You mentioned substantial harm and you've seen some instances where this could lead to substantial harm. Because I'm sitting here as a, say, an individual and I'm thinking, well, I'm okay, I want to know what's there. I don't care what the potential harm is because I can make the decision if I want to do something invasive or not. Once I know that information is there and if I think it's something that has a low risk, my doctor tells me, yes, it's there, but it's low risk, I'm not going to then go do something about it. So I'm not worried about the harm. But then what if they find something that's really concerning and then I'm willing to take that risk.
- [12:02]
B
Right? Yeah. And that, by the way, that logic is very intuitive. And if I was a layperson, I probably think the exact same way. Frankly, most healthcare providers think that way too. I know a lot of healthcare providers, they have that same logic. Here's the challenge is that once you identify something, it's very hard to unsee that thing. I'll give you a specific example. In patients who are over the age of 50, it's very common for you to have a thyroid nodule, which means your thyroid gland, which is in your neck. It's very common for you to have a nodule that grows in your thyroid gland. It's also very common for thyroid nodules to have small cancers in them. This logic, well, if you're an older patient who's maybe over the age of 50, thyroid nodules are common. Cancer in thyroid nodules is common. Therefore we should go look for it in Fact, this is exactly what was done in the United States and also more formally in South Korea. What we saw when this happened, this well intentioned effort to go find common cancer, you ended up seeing a dramatic increase in the incidence of thyroid cancer in both countries. In the US it was by about 3 or 4x and in South Korea it was like by 9 to 11x. So a dramatic increase in the detection of cancer. Now we would think that's great, we've identified early stage cancer, now we can treat it. And in both of those examples, that's exactly what happened, is there was a well intentioned, thoughtful effort. Well, we found the cancer, let's go treat it. And lots of people had their thyroid glands taken out. Tens of thousands of patients, lots of people had injury to the nerves in their neck which caused problems. They lost these little glands called the hypoparathyroid glands that are in their neck and caused some hormonal complications. And many of these patients ended up on lifelong hormone supplementation. And only when you looked back in retrospect and said, let's see how we did, let's see if we've helped the population, we detected all these thousands of additional cancers. How did we do? And what you saw is that the mortality from thyroid cancer did not change. So you end up with this tidal wave of detection of cancers. And everyone in the process thinks we're doing a good thing. The patient rationally thinks, I've detected an early cancer and they probably told all their friends and relatives, my life was saved. The doctor, well intentioned, says, we've detected a cancer early, we've saved this person's life. The payers, whether it's a government agency or a third party payer, well, this is a reasonable thing to do. Cancer should be treated. All the incentives and all the ideas are along the lines of we're doing a good thing. And only in retrospect you say, actually we didn't help. What we ended up doing actually was we caused a ton of unnecessary surgeries. This was described in the New England Journal of Medicine articles, a series of them in the mid 2010s, and they described it as a quote, epidemic of thyroid cancer screening. And the steps that were taken are, please stop, please stop looking. Because once you look and you find it's very hard to unwind that we see this in a lot of other situations. In urology, for example, in small renal masses and urology, for example, in low risk prostate cancer, there's interest in pursuing things like active surveillance and sort of holding back and not doing it, but gosh, it's hard to do. And the average patient, when they're told you may have a cancer, they're going to be pretty motivated to do something about that. And it comes back to what kind of cancer are we detecting? That's the crux of this. What kind of cancer? So if I have a very low pretest probability of aggressive disease, which is the average risk population, if I have a low pretest probability disease, the kinds of cancers I'm overwhelmingly going to find are those that are indolent. So you can think of cancers and kind of, this is very crude and I apologize, but kind of three basic buckets. You've got your indolent cancers that are very unlikely to actually hurt you, but they are common and dwell in your body. You have some that are aggressive which could go from development to metastatic disease really fast, like maybe in six months. And you have those which are kind of in the middle where they have a stepwise progression of going from something indolent to something a little bit more sinister to something that's bad. If you think about those three kinds of cancers, the ones that you want to to target for screening are the ones that are in the middle that this stepwise progression, and we see this in patients who have rather randomized trials which recommend screening for things like colorectal cancer or prostate cancer or breast cancer with certain kinds of cancers are being targeted and kind of an intermediate growth phase. The really aggressive cancers, the ones that kind of develop and go from zero to horrible real fast, they're not really well suited to screening because odds that you randomly happen to image the person within maybe a one to two month window that you have before you can even do something, pour the cats out of the bag, it's just statistically unlikely. And the indolent cancers, if you end up detecting a lot of those, you end up driving a lot of treatments, a lot of biopsies and all the complications from surgeries and stuff that creates an illusion of benefit, but it's not actually helping. In fact, this over diagnosis problem is prevalent actually in prostate cancer. And it's one of the reasons why it's generates so much kind of feelings and complication around what's the right kind of prostate cancer screening to do. A recent study just came out a couple weeks ago, a very famous study was conducted over a 20 year period, enrolled like 160,000 men randomized to get PSA screening or not. Now obviously in all these trials, a little bit of cross contamination where people kind of choose to do something they're not supposed to do, according to the trial. But they showed essentially by doing prostate cancer screening, it was confirmed to result in a 13% decrease in prostate cancer related mortality. Which is great. That's amazing. That's amazing. And that says to me we have randomized trial evidence saying this is the right thing to do for this patient population that's of a particular risk. And what they also found in that study, and what they also highlighted is that patients who opted into screening had a 30% increased detection of cancer. And a lot of that cancer that was being detected was over diagnosis. Some of it was appropriate, and you're getting some treatment benefit of that survival benefit, which is that mortality benefit, which is exactly what you want. And some of what's being detected, a lot of what's being detected is over diagnosis and being smart with the screening strategy and trying to angle it in such a way that it targets the higher risk ones and minimizes low risk ones. It's a combination of doing two things. One is being smart, like you said, and two is targeting the right population. The population has to have sufficient risk to justify doing it. Otherwise the harm starts to outweigh the benefits and you end up hurting patient populations. It's the same reason we don't do, for example, colonoscopy based screening in 20 year olds. Because the harms outweigh the benefits.
- [18:39]
A
Yeah, yeah, it is. It is scary. So think about it. You find a nodule in your thyroid, you have a surgery, you injure your vocal, your nerves in your neck, and you have chronic pain for the rest of your life for something that you didn't need. Right, that's. I mean, that's the one that we can take from your example. But many of these things can happen. You can have very unfortunate circumstances, even from biopsies, even from endoscopies, these sort of tests that you do to diagnose an incidental finding on an mri.
- [19:07]
B
That's right. And I would say this is really common. And we end up dwelling a lot in anecdotes with this topic because we lack randomized trial data around it. So one side says there was an anecdote where someone found a finding and my life was saved. They decided to give an anecdote of, you know. Yes, but here's all the terrible harms that could happen. I can give you one example of this. This is a very dramatic example and I have permission to share this from the person involved. It actually was published in a New England Journal of Medicine evidence article. So essentially the person goes in because they have some shoulder pain, and their plan is to get their shoulder injected because their shoulder hurts. They get an MRI to evaluate the shoulder. And on the shoulder mri, they found a lung nodule. They said, oh, no, found a lung nodule. What are we gonna do about that? So this is a classic example of an incidental finding. I mentioned before. About 20 to 40% of all these advanced imaging studies have insulin findings. This is a classic example of one. So this individual is maybe in her late 40s, early 50s, has no particular risk of any lung disease. They're not a smoker, but nonetheless, they found a nodule. They said, well, we're not sure what it is. You have a low pretest probability of disease. But we have now generated uncertainty. We found something, so why don't we do a chest CT and figure out if there's other nodules? So they did, and they found actually lots of nodules all over the chest. There were a whole bunch of them. And the person was obviously terrified, and we didn't know what it was. And they thought, well, maybe it could be metastatic disease from some unknown cancer, maybe it could be some bad infection. We're not sure what it is. Why don't we do a PET scan? We talked about that earlier. And so she's gotten the radiation from the ct. Now she's going to get the radiation from the PET ct. And it turns out that none of the lung nodules showed any significant uptake. But what they did find on the PET scan was they found something in the pelvis, and there was a little bit of uptake around the. We call it the adnexa, kind of like where the ovary is. And the PET scan said, you know, the lung nodules are not showing any uptake, but maybe this problem is down in the ovary. Why don't we go take a look at that? So they ended up with a pelvis mri. So counting the studies here, we've got a CT scan, we have a PET scan, we have an MRI of the pelvis. MRI of the pelvis says the ovary is actually normal. We think maybe that was just some artifact from the ureter coming by, but there's some findings in the uterus, and there's a bunch of masses in the uterus. They're probably all fibroids, which are benign masses, but one of them looks a little funny. And what if those lung nodules are actually something funny from this uterine mass? And what was recommended was to remove the uterus. This person now has Surgery to take out the uterus for this lung nodules, which they were all benign masses. And I'm going to cut to the chase, but after this hysterectomy was done, the person developed a wound infection, an allergic reaction to some of the antibiotics that they were taking. They've had, I think, three colonoscopies because they still know what the heck the lung nodules are. And they were being offered a biopsy for the lung nodules. And the person said, stop, please stop. I'm being tortured. And this is what happens as you get on. They call this. There's a famous description of this called a cascade of care, and it's well described by primary care doctors where one thing begets another thing begets another thing and you end up in this vortex and you can't get yourself out again. And this is the risk, unfortunately, of doing sensitive imaging tests in low risk patient populations, because you're going to overwhelmingly detect these indolent or unimportant findings. And by the way, that the conclusion of the story is these lung nodules have no meaning in her life and she's completely fine.
- [22:35]
A
And was her uterine mass something.
- [22:37]
B
They were benign, they were fibroids, benign. So it was all just a rabbit chase. It was all pointless. But once you get on the train, you can't get off the train. The person's living their life. They're a totally functional, happy person. But for about a year of their life, they were terrified they were going to die of metastatic disease. And they had an unnecessary surgery and complications from surgery. And that is the problem when we do this, if we have time. I like to just describe one thing. We have screening tests. Screening tests are affected by common biases. You can think about whole body MRI as a screen essentially of your body. And some of these biases of screening, which are very well known, are length bias, lead time bias over diagnosis bias. So lead time bias, what that basically means is let's say that you have cancer that's detected and you live for three more years. Your parent survival was three years with that cancer. Now let's say we detected that cancer actually two years earlier by some imaging test. Well, now I've got two years from me knowing about it and then the symptoms develop and I got three. So two plus three is five in that patient. Even if I do nothing, even if I don't treat the disease, even if I've given them no therapy, their apparent survival is longer. And this is a common bias of screening, a more insidious bias which I think is very common with whole body MRI and other things. Like it is something called length bias and it goes like this. Let's imagine you have two types of bad things. One is really indolent and unlikely to hurt you over a 20 year time frame and one is something that's very aggressive and is going to go from like just starting to going to kill you within one year. And let's say they're equally prevalent in the population. And I randomly sample the population, just randomly choose, you know, I'm going to image you on this year. This year. The thing that I'm going to overwhelmingly detect are the things that are going to be in your body for 20 years as opposed to things can be in your body for six months or 12 months because I'm randomly sampling the population. This length bias helps explain why a lot of the things we find when we're over detecting end up being those indolent or low risk disease because we have flipped or inverted the ratio of important disease.
- [24:45]
A
Yeah, yeah. It's very complex. And I think the thing is we try to make it so simple, right. Like get this test, you'll find something, you might find something bad and then you can do something about it.
- [24:55]
B
Yes.
- [24:55]
A
And I think as physicians we have to take the time to explain these things. Right. And the hard part is you talk to somebody who, who these are obviously cash pay things. Right. They're $2,500 or so to get a whole body MRI and somebody who's got the disposable Inc. Like it's no big deal. I want to get this test.
- [25:15]
B
That's right.
- [25:16]
A
But if you had $2,500 to spend on improving your health, what would you recommend someone do rather than get a whole body mri.
- [25:22]
B
Yeah, that's a great question. So much of the disease that's in our population right now, a huge chunk of it, probably more than 50% of it, is modified by a preventable risk factor. So If I had $2,500 and I said I really want to maximize my health, I'd hire a personal trainer. I do my aerobic and my strength training exercises. I'd hire a dietitian to make sure my diet was good. If I was smoking, I'd invest in something to help me stop smoking. I'd make sure I engaged with evidence based, appropriate screening practices, which could be things like making sure your lipids are checked, making sure your sugar is good, making sure your blood pressure is okay. If I'm a former smoker, I'm going to make sure That I get my aorta looked at when I'm in my late 50s, 60s to make sure I don't have an aneurysm. I'm going to engage in my appropriate screening regimens. Like for example, if a patient's over the age of 45, I'm going to get my colonoscopy, I'm going to have a conversation with my doctor about whether it's an appropriate thing for me to do to engage in prostate cancer screening. If I'm a former smoker, I'm going to do lung cancer screening. I'm going to do the things that are evidence based. And If I had $2,500 to throw around, I'd literally hire a personal trainer and a dietitian. And that is going to prevent so much of my disease in my life. I mean they say something like 70 to 80% of cardiovascular disease is potentially preventable if you just like do the necessary steps. This whole body MRI idea is based on the idea that I'm going to find something and I'm going to save somebody's life early. That's what it's based on. And gosh, if it did that, I can tell you the medical societies would be endorsing it full throated.
- [26:53]
A
They'd say yes, be covered by insurance.
- [26:54]
B
Be covered by insurance. One of the weird things about this, a little bit of a side tangent. But the business model is good because it's not covered.
- [27:04]
A
Yeah, of course the business model is.
- [27:06]
B
Good because it's not covered.
- [27:07]
A
You and I, I think you were there too at the HLTH conference recently. And so much of what is advancing in, in the, for the public, for the general audience is based on a profit model. Right. It's all, there's very few doctors involved in all of this. The people who day in, day out take care of patients. It is much more for profit models with what can we do to make money. And oftentimes it comes from a good place. Right? Like some entrepreneur has a family history of cancer and decides to make this whole body MRI thing or whatever. Right. It comes from a personal experience, it comes from a good place. But there's a, the limitation in terms of like what is what, there's a reason behind everything and that's, and there's a reason we don't endorse. It's not because the research is not there. The research is there. Right? There is research saying this is home harmful and we're ignoring that.
- [28:00]
B
Yeah, yeah. And it's important to me to kind of emphasize, I think You've alluded to this also. There's positive intent in a lot of this. This is not a question of like throwing shade on this company or that company or a lot of people are coming to this with positive intent. And I think it's our responsibility who are in the field to say, I love the positive intent, I respect it. Here are the unintended harms that is being generated from this. And the tragedy that will definitely occur is, you know, five years, 10 years from now, when people look back at the patients who opted into this, it is almost certainly to be the case that they will not have improved cancer specific survival. It's almost certainly to be the case they'll have suffered significant personal harm to themselves and their health from doing this kind of stuff. Because what's being detected is this kind of lower risk disease. But it creates enough uncertainty that it promotes a lot of activity. One thing that, you know, the viewer may not be aware of is that on an imaging test I mentioned before, we have these indolent disease, we kind of have the intermediate risk disease and we have the aggressive ones. A lot of times on imaging test, you can't figure out which one of those that it is. If you could, this would be less of a problem because I could then on the imaging test say, oh, this is definitely aggressive, this is definitely indolent. You can ignore this, I could say that. And then you as a patient can make an informed judgment about it. The problem is a lot of these things don't distinguish themselves. I'll give you a great example that's in your space urology. When I look at a small renal mass, let's say it's 3cm, I can make some general guesses about what maybe what kind of subtype of the mass it is, but I can't tell you for certainty if it's benign or malignant. And I can't tell you for certainty what the level of aggressiveness of that mass is. And because of that, because of that uncertainty that's been introduced in it, it leads to activity like biopsy or resection or in some cases, active surveillance. The other day I was. I don't want to focus on this specific case because I don't know the details of it. But I was watching a video the other day of a politician who had one of these whole body MRI exams done, and they found a finding in the kidney. And they were telling the story about what happened and they said they weren't sure what it was, it might have been cancer. They were planning to do a Partial nephrectomy. They started to do a partial nephrectomy and there was significant bleeding. That happens. And because the bleeding was bad, they ended up taking the whole kidney out. And I don't know the details of the story, so I'm not commenting on whether it's a right or wrong decision. I have no idea. But that's the kind of outcome that you can end up with. This uncertainty that gets generated results in activities. There's a really interesting paper that was published, I think it was in JAMA internal medicine. It was in the mid 2010s. And what they looked at is 2017. What they looked at was Medicare. Beneficiaries across the United States have heterogeneous levels of access to care because that's the way things are in the US and what they did is they said, what is the. What is your risk of getting a CT scan? Which is kind of a cool idea. What is your risk of getting a CT scan? And based on that, they played it off against what's your risk of having an intervention on your kidney? And what they found is that by doing more scanning in a population, it meaningfully increased your risk of having someone do surgery on your kidney. Image more, find more. So this is what we see over and over again is that every time we take a picture of your body, there's maybe not direct arms from the MRI because there's no ionizing radiation and someone's not giving contrast, but it's not riskless because the complications that come or what follow. It's someone with well intentioned says, well, gee, I don't know what that is, so I better biopsy it or do an endoscopy or take it out. And then you suffer as a patient the complications from that. Now, if treating that thing was resulting in meaningful mortality benefit in the population, we take the risks. We're happy with that. It's okay, I'm happy. But we're okay with it, right? We're okay with the risks. But when you're not seeing mortality benefit and all you're seeing is the harms from the procedures, that's when it becomes a huge problem. And that's exactly what we're going to be dealing with in whole body mri. When people start looking back at what's.
- [31:55]
A
Happening, you just can't predict. You just can't predict. Do you see a future where we'll be able to have imaging tests that can tell us, hey, this is indolent. Hey, this is aggressive. Is there anything on the horizon?
- [32:09]
B
I would say yes. You know our conversation so far, I've had some pessimism about whole body MRI being used in the general population. I think that's based on evidence. However, I could certainly see a possible future where some of this testing is used for useful purposes. For example, some of the information you get in a whole body MRI gives you very rich, detailed information about your cardiovascular risk profile. Like, for example, the size of your muscles, how much fat is within the muscle, how much visceral fat, which is the fat that kind of outlines your organs, subcutaneous fat, which is the fat kind of around your, outside your muscles, outside your muscle layer, how much calcification is on the aorta, how much calcification is on your heart arteries. This information is being shown increasingly to show very strong risk prediction for cardiovascular outcomes. In fact, in some studies are showing better risk prediction than the Framingham risk criteria, which are like well established guidelines.
- [33:03]
A
You talk about the calcium score?
- [33:05]
B
Yeah, yeah, absolutely. We're also seeing the ability to detect risk of osteoporosis in patients just who have a CT scan for other purposes. So in other words, being able to risk model or risk profile a patient related to things that we know hurt people could have potential value. I also think what could be coming is, let's say that some future time, 20 years from now, we've done the trials, we know what incidental findings should be intervened on and which ones shouldn't we rethink about how we're describing and characterizing these things. Because some of this is a branding problem. Maybe we shouldn't call everything a cancer, a cancer, a cancer. Maybe we should be more specific. Is this indolent? Is this not indolent? You know, so yes, I could see a future where some of this information can be used before we have evidence showing that it works. Right now, if you're an average risk patient, you're 52 years old, you're a high net wealth individual, and you say, you know what, that's 2500 bucks. I don't care, I'm happy to pay for it. You are more likely to be hurt than helped by undergoing the test. And that's almost certainly to be the case. And that's the twist in this. That's kind of hard to wrap your head around.
- [34:11]
A
It's interesting. It's like we always worry about social determinants of health, right. And this is the one case where having more is actually going to harm you potentially.
- [34:19]
B
That's right.
- [34:19]
A
Which is the opposite of what we normally see. So you mentioned some tests if right now, if someone had access to, you know, any radiologic test they wanted to, let's say, based on what you've told me, what are the tests they should get? That they should, that if they could have the disposable income, it's not yet covered. Because it's not. We don't have the studies yet. But you would predict that these are going to help risk, help risk model you and help you decide what you need to do for your health.
- [34:44]
B
What I would say is right now, and this is going to be such a lame answer, but I apologize for that, is that you should follow the advice of evidence based guidelines you should follow. That's that should drive the decision making around which imaging tests are a good or bad idea to do. Even though I'm optimistic and hopeful about what might happen in the future, a decision to undergo the test now creates a ton of uncertainty around whether I'm right or not. I mean, maybe in the effort to, for example, measure my visceral fat and quantify my muscle density and so on, I end up incurring a bunch of harm from over detection of other things. And maybe on balance, even though I've given myself better risk prediction, I've also hurt myself because I ended up detecting other things that I shouldn't have detected. And without doing a formal trial, it's really hard to guess. And we are really bad at guessing which tests are going to end up creating value and which ones aren't, unless we've done a formal trial to kind of look at that and show the evidence behind it. So my answer is a little bit lame, which is to follow evidence based guidance and listen to your provider, assuming that you trust them, that's good.
- [35:49]
A
And then what about a calcium score? So I know that's not recommended widely for every single person, but do you feel like that is something which is a CT scan right of your heart, do you think that has value in terms of risk stratification?
- [36:05]
B
Yes, in certain patient populations. It makes a ton of sense to do in certain patient populations. And if you're listener and you say, well, am I one of those people? You know, you can just go online and say, you know, what is the indication for a coronary calcium screening test? And you just type it in and it'll come back and say, well, do you fit the criteria or do you not fit the criteria? Now one of the things about imaging that's interesting is just because you've had a test done doesn't mean it inherently creates value back to you. There's this really famous hierarchy that was published in the 1990s called the Freiback and Thornbury hierarchy, which I really like because it helps clarify some of these issues around diagnostic testing. And it kind of goes like this. And I'm going to simplify a little bit. You have this pyramid, and at the bottom of the pyramid you have, can I make a picture? Like, do I have the physics that can do it? Okay, if I can make a picture, the next layer up is, am I accurate in making a diagnosis? Okay, let's say I am. If I'm accurate making a diagnosis, the next level up is, did I change your mind about what's actually happening inside you? Did I change your diagnostic opinion, whether it was you as a patient or you as a provider? Okay, let's say I've done that. The next level up is let's say I've changed your mind. But did I change your behavior? Did it result in some different outcome to you than you know or. Not outcome, but a different behavior. So coronary calcium, be like this. I know I can make a picture that can take a picture of your coronary arteries. Okay, I know that's actually very accurate. I can give you a calcium score that says how likely you are to have important, you know, cardiovascular risk to your heart. Great. You've changed my mind. I didn't realize before I had heart disease, and now I do. And am I going to do something different? That's the trick. Am I going to do something different now with this? Yes. I mean, you probably would do blood pressure control more aggressively. You'd probably control your cholesterol more aggressively, maybe would change your behavior. You start eating differently or you'd exercise differently. Some behavior change. Okay, all those things I mentioned, make a picture, make a diagnosis, change your mind, change what you do. All those things are necessary to create value, but they don't actually intrinsically do it. In order for you to actually create value, you've got to go to the next level is, was there an outcome difference to the patient? So let's say, for example, I took a picture, I changed your mind about what was inside of you. I changed my opinion. I've done something different. But even though I've done something different, there was no outcome difference to the patient. All that extra stuff I've done is just churn. I've not actually created any value. It's the outcome that matters. And that's why, for these whole body tests, the key is to measure what's the outcome to the patient, which, again, to reiterate is do I reduce my cancer specific mortality, Do I reduce all cause mortality, or do I reduce the likelihood I'm going to develop T3 or T4 disease? Just like advanced stage cancers inside your body, that's what the measurement has to be on, not whether I found something or not. But in answer to your question of coronary artery calcification screening, yes, in certain patient populations it definitely adds value. And we talked before about how so much of our conditions we have are actually self inflicted in some part. I mean, in the US right now, 37% of our population is obese and something like 70 or 75% could be categorized as obese or overweight. And to some extent that's a somewhat controllable factor. Very high rates of hypertension, high cholesterol, high rates of diabetes, we have high rates of sedentary lifestyle. These kinds of things do meaningfully predict cardiovascular outcomes. So I could take a picture and show coronary recalcification. But the challenge then is we have all these modifiable risk factors that we know are problematic and yet still we struggle with them as a population, regardless of whether we have a test or not. The last data I saw on smoking, for example, I think we still have a rate of smoking in the U.S. somewhere in the neighborhood of 12 to 15% smoking, which obviously increases your risk of all sorts of cancers, including lung and others.
- [39:52]
A
No, the reason I bring it up is I think heart disease is one of the things that you can make meaningful difference on in terms of lifestyle. And there are some people who would argue that you should have a baseline calcium score and then you should check it periodically to see where you know where you're headed so that you can sort of find out that you're still in good shape or you need to be more vigilant on your diet or more vigilant on your exercise. And I think that maybe for a small subset of the population who's very eager to figure that out, it may make a difference. But as you said, we have a lot of other things that we can look at for the average risk person, like cholesterol, high blood pressure, other measures that your primary care doctor will be checking on a regular basis.
- [40:31]
B
Yeah. So in an appropriate patient who meets the criteria for coronary calcium screening, it's a great test. It works really well, actually. So you might say, well, what's the harm in doing it at all? Why doesn't everyone get it? Why doesn't everyone in their 30s get it or 20s? Why not in their teens? Why doesn't everybody get it? Well, when you do a coronary calcium screening test, you're using ionizing radiation and you're scanning the chest with something that has a very small be very clear, very small, but on a population base, meaningful potential risk of cancer development. And in addition, you're not just taking a picture only of the arteries, you're also taking a picture of everything else that happens to be in the field of view. Let me give you an example. So in the early 2000s, there was an interest in doing whole body CT screening. Same idea as now. Just they're using CT instead. Yeah, this is in the early 2000s, very kind of same idea. They would put these things in, like, you know, shopping malls and stuff. You go and you get your scan or whatever. There is a story, this is written in the, in the medical literature as kind of a cautionary tale. This individual got a CT scan for some reason, I don't remember what the reason was. And they found a lung nodule in their lung. And the person had no particular risk of disease, but it created uncertainty. They weren't sure what to do about it. What ended up happening in this person's tail, which is published in the early 2000s, is they ended up doing essentially a lung resection for this nodule. And they went in and they cut a piece of the person's lung out. Ended up with a protracted course in the hospital. I don't know if you're an ICU or not. Part of their lung was removed. They had pretty severe pain for multiple weeks. So they had to kind of open the ribs up to take this thing out. Meanwhile, the lung nodule was benign. It was like a post infectious, like histoplasmosis pulmonary nodule. All of this is to say that nothing is free. If you take a test and you find stuff, that stuff has consequences. So coronary calcium screening is perfect in the right patient population. And in others, be careful.
- [42:27]
A
Yeah. And it is radiation.
- [42:29]
B
Yes.
- [42:30]
A
Let's talk about MRI a little bit. Is there a number of MRIs? Is there a threshold where you can't get any more? Like there's two. Like, we know that CTs, we want to minimize how many we get because of the radiation. But is there the same sort of relevance for mri?
- [42:43]
B
The short answer is no. A slightly more nuanced answer is that in patients who are getting regular doses of contrast. There's a small consideration here. So let's say that I'm someone undergoing breast cancer screening because I have high risk for breast cancer and I meet the criteria for regular MRI based screening, or let's say I'm somebody who's a young person in their 20s who has inflammatory bowel disease and I'm routinely getting an MRI of the abdomen and pelvis to look at my bowels and stuff with contrast. If you're getting lots of scans, 10, 20, 30 scans with contrast. One of the additional considerations is that that contrast, which with MRI is usually called gadolinium based contrast material, that gadolinium is known to stick around and the meaningfulness of that is not clear. There's no direct evidence of harm that we know of. And the last statement by the FDA on this was that we know it's happening, but be aware, but there's no scientific evidence of harm. So in someone who's getting repeated scans over a long period of time, one of the considerations is to make sure that the agent that you're getting is a very stable agent and to be thoughtful about giving it. We don't want to indiscriminately be giving the contrast, we want to be giving it a thoughtful way. If no contrast is being given, I would say no. There's no evidence of harm from repeated scans over a period of time directly related to the test. The harm comes from all the downstream stuff that is provoked when you find things you maybe didn't want to find.
- [44:12]
A
Yeah. The other thing is, you know, we've talked a lot about the sort of harm like physically from the biopsies or the surgeries, but what about the psychological harm? Has that been quantified? I mean, we talk about it all the time. I talk about it to all my patients when I screen them for PSA and when they're a little bit older, we talk about the risk versus benefit versus continuing screening because there's a psychological stress with knowing you might have a cancer or then getting a biopsy and so on and so forth. Has that been quantified with radiology?
- [44:41]
B
Yes, in different domains you'd mentioned. I know in your practice in urology, a lot of research has looked into the anxiety of living with active surveillance. And this is a bunch of different domains and it's meaningful. It's double digit percentages of patients. It can psychologically torture you in a way to be dealing with this uncertainty. You end up getting converted from a walkie talkie person into a patient overnight. And once you become a patient, you can't undo that. You're stuck with it. There's this phrase called scanxiety where, you know, when people are about to get their test and they're waiting for their imaging test results to come back. It creates a lot of fear and anxiety. I tend to not focus on that element. Maybe I should talk about it more often. But I tend not to focus on it because a lot of people will say, you know, I'm a stoic individual, I can handle it. Don't tell me about how I'm going to feel about myself. So I tend to focus on the physical harms. But you're exactly right, there's a huge fraction of patients. I'd say it's double digit. If I was to put a number on it. It's probably in the 10 to 30% range. Some of the work we've done around anxiety and test results and stuff puts in about the 30% range in terms of meaningful anxiety.
- [45:43]
A
That's really high. It's high. Yeah. That's three in 10 people. That's really high.
- [45:47]
B
It's high. So yes, it's real, it's meaningful. And I can tell you as a radiologist, sometimes people say, what's a radiologist? I interpret imaging tests for a living. So I basically interpret the imaging test and I talk to referring doctors like yourself and others and say, this is what I think is going on. And here's maybe a potential best next step. Because I'm a radiologist, I get asked all the time by friends, others, family members, loved ones. I mean, constantly being asked, hey, I've got this test, can you tell me what this means? It's all the time. And that place that's coming from is curiosity. Yes, but it's also coming from fear. When they see the imaging test result and they read it in a stark way, maybe it's on the patient portal, maybe they haven't talked to their doctor yet. It's scary. You read something and some radiologist says there's a 2 centimeter mass in the kidney. Renal cell carcinoma not excluded. That's scary. I go online, I start typing in renal cell carcinoma and I start seeing mortality statistics and I'm like, what is happening? It's scary.
- [46:48]
A
Yeah. I mean, I don't know how it is at Michigan, but when I was at Maryland, they would release the records right away. Right. You'd get the results and they'd see it as soon as the radiologist read it. And I hadn't had time to read it yet. So it created a lot more anxiety in the patients. And it's good. We want you to have your records, but it should be until the doctor has read it and talked to you about it rather than you seeing it first.
- [47:09]
B
Yeah, there's a, there's a federal law that was passed last five years or so which is now required for any test that you have. Radiology test, lab test, clinical note has to immediately be sent out. Which pros and cons there obviously. But most patients say, great, I get to see, I get to see my results. There's no gatekeeping happening. And I think people do value that. However, there's a period of time where you're looking at something, you're like, what does this mean? And oh my goodness, this is scary. We actually did a multi center study around this topic and we tried to measure, we tried to measure people's feelings of anxiety and doubt and fear around an imaging test of a potential new cancer diagnosis. And we traded off a bunch of different examples. We used something called a conjoint analysis. What we basically found is that if someone gets their results before they've talked to someone, they otherwise would receive the results within three days. It actually creates net harm. So if you get your result immediately and it's a new cancer diagnosis and otherwise you would be speaking to someone in three days. The general anxiety is less if you just wait three days to get your results back. And once it starts to get longer and longer and longer, then it starts to invert. But it's complex. Yes, anxiety is really common with test results and that's certainly a piece of this.
- [48:27]
A
I think it's really hard to quantify. Like when I tell patients I'm going to give you anxiety, their eyes start to gloss over. Like they're like, oh no, I'm fine, I got this. But it's until you're actually in the moment and feeling the anxiety that you really can't quantify it. It's sort of challenging.
- [48:41]
B
That's right.
- [48:43]
A
You mentioned this comment about the gadolinium based contrast and using stable contrast agents. So how as a consumer would I know that my MRI technician is using stable contrast agents?
- [48:57]
B
Yeah. So gadonium based contrast media been around for a long time and you can put them into a couple different baskets in terms of how they arrange themselves. The groups that are the most stable. There's a word called macrocyclic and this is a crude description, but you can imagine that gadolinium all by itself is toxic. So to prevent gadonium from getting away and getting into your body and interacting with your tissues, it's attached to a chelate. That chelate holds that gadolinium really, really tightly and does not let it break away to interact with your tissue. These macrocyclic agents, what they basically do is you have the gadolinium kind of in the middle and there's a cage of bonds that surround that gadolinium to protect it from interacting with tissue. And they're very stable. The other kinds of agents there are called linear agents. You could imagine that like the gadolonema is in the end of a line, so it's more likely that something can come in and interact with it. If you're going in to get scanned and you're concerned about the stability of the contrast agent, you could say, is this a macrocyclic contrast agent? That'd be a question you could ask. And they look at you like really you're asking that question. Another way to say it is, is this a low risk contrast agent? Now, nowadays, most of the agents that folks are using are pretty stable and have great safety profiles. So I don't want this to come off like gadolinibased contrast media are some existential threat because they're not. And in fact, if you compare them to other drugs that are on, I mean, any drug, frankly, gadolinium based contrast media is some of the safest thing that you can receive. The risk profile is exceptional. The risk of a severe allergic reaction is like one in a million. You know, it's just, it's not likely to cause harm with a modern agent. Another way you could phrase this to your radiology teams, you could say, is this one of the more modern low risk agents? And they can answer the question, yeah.
- [50:46]
A
And how is gadolinium based contrast, which is used for mri, different than contrast you're getting for CT scans?
- [50:53]
B
Yep. So I mentioned that for mri, gadolinium based contrast medium uses gadolinium as the ion that changes the local magnetic field. So you can see stuff better with CT scan. Use an iodine based contrast agent and that you have iodine, which is causing attenuation of X rays, which makes it see things a little bit better in both cases, whether it's with MRI or with ct. Essentially what contrast material is doing is it's creating an image contrast between normal and abnormal. It's also showing you what is vascularized and what is not. So it can make things, it makes the blood vessels fill up with, make them bright. You can look for blood clots, for example. You can tell there's something solid or cystic. It lets you see what's vascularized and what's not and creates image contrast between normal and abnormal. So I'll give you an example with the whole body MRI stuff. So let's say you get a whole body MRI without contrast and something is found in your kidney, a small little kidney lesion without contrast. In many cases it's not possible to say whether it's solid or complex, cystic or a cyst, unless you give contrast. There are some edge cases where you can do that, but in many cases you cannot. So that provokes additional testing.
- [52:11]
A
Another CT scan.
- [52:12]
B
CT or mri? Yep. With contrast? Yes.
- [52:16]
A
And then in terms of one of our audience members asked, why do you feel that warmth when you get that CT contrast?
- [52:24]
B
That's a great question. I can give you a superficial answer, but not a biochemical answer. The superficial answer is that there's different kinds of reactions that one can get from contrast. Some of them are called physiologic, where you're having a direct reaction to the osmolality or the chemical composition of what it is. It's not an allergic reaction. It's just a sensation that you're having from this foreign substance. Examples of physiologic reactions are a warmth, a metallic taste in your mouth. Sometimes someone's heart could speed up some. You can get something called a vasovagal reaction where your heart rate slows down and your blood pressure goes down. You can feel faint or lightheaded. Some people can have nausea or vomiting. These are unusual. But these are all physiologic reactions and not allergies. Usually it's considered to be a direct chemotoxic type of reaction where you're being exposed to the osmolality of the contrast and it's causing this sensation in your body. Separate from that are allergic reactions, and those are managed differently. Most of these physiologic reactions, they're temporary, they're transient, they go away, and they don't cause any significant harm to you.
- [53:34]
A
Yeah, that's good. So nothing to worry about. Just sort of a known thing that happens.
- [53:38]
B
If I'll give One of my things in life is I've been involved a lot in the diagnosis and treatment of severe anaphylaxis type reactions from contrast material. So the one thing I would say is if you're getting contrast and you start feeling very strange, speak up and say something and say something's not right, I think something's wrong this contrast. So they are alerted and they can start taking the necessary steps. But don't minimize anything. But yeah, in general, if you just feel that little brief warmth, it's nothing, and it'll go away.
- [54:08]
A
Cool. Let's talk a little bit about prostate mri. Prostate imaging. So prostate MRI has really changed the way that we take care of men with prostate screening for prostate cancer. Where does prostate MRI truly add value for the patient who is getting screened for prostate cancer?
- [54:27]
B
Yeah, I mean, that's. Prostate MRI is one of those innovations that has really taken off. I mean, wow. In the last 10 years, it's completely transformed how a lot of us think about doing prostate cancer screening. And if I try to put my head around why, I think it comes from the following. I'm interested in your thoughts on this. If you go back to, like, the mid 2010s or so, prostate cancer screening was having a little bit of a moment. It was having an existential crisis. And in fact, the United States Preventive Services Task Force recommended against doing it. And they said, you know what? We think the harms of doing this are actually outweighing the benefits. There's just not enough people that are being benefited by it. There's a significant amount of harm from over treatment. Maybe we should back this off. And what they did is when they made that recommendation, the rates of prostate cancer screening declined. And what we saw was increased incidence of people developing metastatic disease and dying from prostate cancer. So it was a mistake, actually, it was a mistake. And they said, oops, sorry, can we put that back into place? But I think what prostate MRI is doing is it's meeting the call, the need, which is recognizing that we know prostate cancer screening saves lives. And even as recently I mentioned before, over the last month or so, another landmark study comes out kind of confirming this. We also know that detecting the bad cancers also detects a lot of indolent cancers we don't want to find. So I'll give you a small data point in patients who have their bladder taken out for bladder cancer. And if you're a man, you also have your prostate taken out too. It's called a cystoprostatectomy. And in a relatively famous study of this, in which they took out about 2,000 men's prostate with their bladder, the rates of incidental prostate cancer is very high. It's like 4 in 10 men in this population had prostate cancer just hanging out, doing its thing. And in most of those patients, it wouldn't have actually harmed them.
- [56:17]
A
Right.
- [56:18]
B
So what prostate MRI promises is prostate MRI actually does a bad job detecting low risk prostate cancer. So if I said I want to find all prostate cancer that possibly exists Prostate MRI is not your test. Why? Because it actually misses all these low risk diseases. But it's pretty good at finding Gleason 7 or higher disease, also called grade group 2, higher disease. And in an average risk patient population, the negative predictive value, which means if I get a negative test, how likely is that to be true? A real negative test, it's about 95% when it's done in a place that knows what they're doing and the image quality is good and we'll get to that in a second. So it's promising to decrease detection of low risk disease and improve detection of clinically important disease that otherwise would be missed by random biopsy. In the past, when people were diagnosing prostate cancer, we do a psa, maybe we do a digital rectal exam, we say, well, gee, maybe there's something there. And then we kind of randomly biopsy the prostate, literally randomly, using like a sextant based approach or kind of a template based approach. What that would do is it would miss a lot of cancers that are really far anterior or really far at the apex, the bottom of the prostate, really high at the base, the top of the prostate. Prostate lets you see all those things you otherwise would miss and it will avoid detection of lower risk disease. That's the promise of prostate mri. It has some problems, it's not perfect. Some of the problems of prostate MRI is it is it has too high of a false positive rate in my opinion. So right now, about four to six men out of 10 who get a prostate MRI that's positive, it will actually be a false positive.
- [57:53]
A
Well, are you defining positive as any lesion or are you defining it as a pirad, which is the grading scale, like 4 or 5 lesion?
- [58:00]
B
Yeah, great question. If we consider a PI rads 3, 4 or 5, and for the audience, PI rads is basically a five point scale that says what's your risk of having important disease? One being very low risk, five being high risk. Many times patients who have a PI Rads score of 3, 4 or 5 end up getting a targeted biopsy. Although those who have a Pyreds 3, it's kind of debatable. If we consider PI Rads 3, 4 or 5 as a positive test, about 4 to 6 people out of 10 are going to have a so called false positive. Has an excellent specificity though, excellent sensitivity, which means that the negative predictive value is really excellent. The other thing that's kind of a weakness of prostate MRI is that we don't have long term outcome data. From it. So it promises that we're going to detect less, lower risk disease and detect more clinically important disease, but we don't actually have it in the context of a randomized trial showing that it's improving outcomes for patients. So we think it's likely to. We're not 100% sure.
- [59:00]
A
Well, I would venture to guess that at the very least people are getting less biopsies.
- [59:05]
B
Yes, absolutely.
- [59:06]
A
In that way it is probably at least improving the risk of the biopsy. Right?
- [59:12]
B
100%.
- [59:12]
A
Reducing the risk of biopsy.
- [59:13]
B
Completely agree. Yes.
- [59:15]
A
Do you feel like in terms of you mentioned there's sort of like technical things you want to look out for when you're getting your Mr. Prostate? At least when we started doing prostate MRIs, I know it was very few centers that were pretty good at prostate mri. Do you feel like that's less of an issue now? And if so, what should people be asking?
- [59:35]
B
This is a really important point. So there's a recent trial that came out called the prime trial. This came out this year and prime trial was asking the following question. We know prostate MRI is really good at what it does. Can we take away contrast material and still get the same result? So can we do a non contrast prostate MRI and get the same result? And what they basically showed, the short answer is yes, actually, in a biopsy naive population, you can take away contrast material and have almost very similar performance characteristics. What that means for the average person listening is that if you are eligible to get a prostate MRI to help in detection of prostate cancer, increasingly sites will stop giving contrast material and just do the other sequences. A key caveat to all of this is that before that trial was done, they went to all the sites that were participating and made sure the image quality was excellent first, and before they did that, it wasn't. So across the country right now, there's too much heterogeneity in how the image quality is created and there's too much heterogeneity in how radiologists interpret the scans. Both of those are problems. And if you want to get the maximum benefit from prostate mri, you need to have it done at a place that's doing quality imaging and where the people are specialized, they know how to do it. Now because prostate MRI is so important, people have called it the manogram because it's like, you know, so many men are eligible for prostate mri. It needs to be done diffusely across the country. Like lots of sites have to be able to do it. If I was a consumer and I Wanted to make sure that my prostate MRI was done. Well, I would say, are the people who read your prostate MRIs here specialized in interpreting prostate MRI? Do you measure your outcomes? Do you measure your positive and negative predictive value? Are you monitoring your outcomes? Do you do radiology, pathology, correlation to make sure that you're accurate? And then are you using modern equipment? I think those would be some of the questions that I would ask. The sad reality is that today it's very hard to know if you're a consumer, whether you're getting good quality or bad quality. That's a longer conversation.
- [61:38]
A
Well, yeah, I want to touch on that, but I do want to make a comment. So I think a lot of these freestanding imaging centers. So I got a. I, my mother had breast cancer. I got a breast mri and I went to the, you know, the, the institutional mri. I got a message from my insurance. Well, you overpaid for your mri. You could have gone to this imaging center and paid $200 less or whatever. But I wonder, these, these freestanding imaging centers, are they really able to do the pathology and correlation? Because where would they have. They would have to go and find from every place where their patients go to find out if they actually found a cancer. So is that what you find in real practice?
- [62:17]
B
Yes. What you're referring to these freestanding imaging centers, and there's many kinds of imaging standards that fit into that description. I don't want to cast too broad of a brush here, but in general, if you're receiving imaging studies from maybe 30 different practices that are all disconnected from each other, the task of collecting meaningful quality data is a lot harder in that context. In the state of Michigan, there's something called the Michigan Urological Surgery Improvement Collaborative, or music. It's a beautiful thing. Actually, a very large fraction of urologists across the state participate in this. It's essentially funded by Blue Cross Blue Shield. It's a quality collaborative. And one of the things they drive towards, which I tried to help them to create this, was to create quality benchmarks that says, if you call a PI rads 3 or a 4 or a 5, what's your positive predictive value in doing that? And now across the state, they're getting scorecards that says, based on your practice, here's your outcome. Now, it's hard to scale that outside of the state of Michigan because this is a funded initiative and they're being essentially paid to abstract the data manually and collect this information. But in the absence of something like that, they have to take it on themselves and say, this is important to me. I want to measure it, I want to know how I'm doing. Otherwise you're flying blinds. It's like being a pilot and having no idea, like what your altitude is like. You have to be able to measure what your outcomes are to say, yep, I'm doing a good job and I'm stable and I'm not drifting.
- [63:42]
A
Yeah. And I think the challenge is. So I worked in Maryland. There was a very large freestanding radiologist everywhere. Like you could get an appointment the next day. It was very convenient for patients. And the same exists in most, I think, most places around the US if you live in a metro metropolitan city, you're going to find imaging centers that are much easier to navigate as a patient. Whereas if you go to the academic center, you might be waiting a month, two months to get your imaging test. So how do people sort of identify, hey, what is, what is the right place for me to go to?
- [64:13]
B
Yeah, gosh, that's the million dollar question that I can't easily answer, unfortunately, because right now we don't have publicly available quality data for radiology practices. If I was to say right now, let's say I'm just going to be an honest broker, I'm going to be super neutral about it and say, is practice A or practice B better? I have no way to know. Actually, in radiology there's not like national quality scores. You can say, is the site ACR accredited, American College of Radiology accredited, which gives some basic ideas about quality, but really substantively, you don't actually know the answer to that, unfortunately.
- [64:47]
A
Yeah, it's tough. Yeah, it's really tough. I want to talk about artificial intelligence.
- [64:51]
B
Yeah, of course.
- [64:52]
A
I think that a lot of people, so many people, when I asked for questions, were like, well, is artificial intelligence going to take over radiology? Is artificial radiology? Is artificial intelligence going to be reading radiology better than radiologists?
- [65:04]
B
Yeah, those are all super provocative questions. And I'm, I'm going to try to predict the future. And that's always fraught. Right. I think it was 2015 or 2016, where the individual who won the Nobel Prize for artificial intelligence predicted that radiologists will be out of a job and we should just stop training them. Now, I think, was his decision. And right now, 10 years later, the job market for radiologists is out of control. There's such a shortage of radiologists right now. The demand is insatiable for imaging tests, even appropriately ordered imaging tests. I'm not talking about whole body mri, for low risk patients right now, I'm just saying in general for imaging tests is so high. And almost every center that I'm familiar with has a pretty significant shortage of radiologists and it's contributing to a lot of delays. So I would say in the next 10 to 20 years, you will not have wholesale replacement of radiologists, but what you will have is a fundamental change in what radiologists do and how they interact with health. And I would even further that to say that's going to happen in every discipline. I mean, I'm going to take radiology off the table for a second. Let's say I am a primary care doctor and my function is to engage, form a relationship, learn of everything, make sure evidence based guidelines and screenings are introduced. I've got a kiosk. The patient walks in and speaks to the kiosk. It uses voice recognition to listen to the history. It compares it to all the medical records the person's ever done. It says, here's the recommended algorithm so it never forgets anything. It says, here's all the things you need to do. It spits out a thing and says here's what you have to do next. Take this over to your pharmacy or take this over to your testing algorithm. Like, I could see this happening in all corners of medicine, but with radiology specifically, it's going to touch every piece of it. It's going to touch the ordering process to make sure that the orders are done appropriately so we don't have as much wasteful imaging. It's going to be done in image reconstruction. So. Right. It's happening right now actually. So if your MRI takes 40 minutes, maybe you can do it in 30 minutes using AI based reconstruction algorithms that speed up the acquisition of images. It's going to happen on the diagnostic side so radiologists are able to make more accurate diagnoses and it's going to raise the floor. So right now there's a lot of quality variation in people's interpretation skill. It's going to raise that floor and give us more of a consistent product on the back end once the image is done. I could see it also being used to auto insert information into reports and also track things at scale. I mean, even now at Michigan, we developed an AI tool that essentially auto mines prostate MRI report data and prostate pathology data and tells all of our providers, me and all my peers and my center, exactly what my performance metrics are. And it's done at scale with like a touch of a button using like a GPT plugin. So I Can now say in my reports, if you have a Pyrads 3 finding, your positive predictive value is 15%. And that's not a hypothetical. That's my actual positive predictive value.
- [67:55]
A
They send you me. Yeah.
- [67:57]
B
So these kinds of things are going to be kind of. And also I could see it being used to make sure that you don't fall through the cracks. There's a lot of people who fall through the cracks. Let's say that something important needs to be done and it isn't. I could see a tool being kind of monitoring in the background, agent monitoring the background, saying, ah, you know, we got to make sure this actually gets done. And following up on that, on a system level, I think AI is going to have a transformative effect in radiology. I am highly skeptical of the idea that radiologists are going to be completely replaced the next 10 to 20 years.
- [68:24]
A
Well, it'll need to have oversight. Right. So the radiologist will need to view the image and agree with, with AI. I, I worry that it's. We're going to push so much to do so many imaging tests by AI that there won't be enough radiologists to oversee the AI. Right. Like radiologists will get pushed to see more, read more and more images because AI has done the pre work. But it doesn't take less time to look at an image then. Right. It might take less time to do the report y but it won't take less time to actually look at the image.
- [68:54]
B
Yeah, that's well said. I agree with that. And there's this paradox where when something becomes, when the barrier to entry, the barrier to do it becomes less, the demand goes up. And I think that's definitely going to happen in radiology because imaging provides such useful information in many, many different contexts. So I think as AI makes it easier to do, makes it faster, makes it more accurate, makes it more consistent, the demand is going to just keep escalating and escalating. There's an interesting parallel. This is a silly example, but, you know, pathologists who make diagnosis on slides and so on with special tests. There's something called a red blood cell count where you figure out how many red blood cells you have. In the olden days, someone was literally counting on a slide how many red blood cells there are. No one does that anymore. There's a tool that does it on their behalf. Right now EKG measures kind of like your heart rhythm. Right now there's tools that automatically tell you a rough idea of what the EKG shows to speed up diagnosis I think those models are kind of what you're going to see in radiology, and frankly, every other discipline, too, is to try and do more with less and to be more accurate and improve the quality floor. So we're doing a more consistent job.
- [70:07]
A
Yeah. And I do think that as medicine has evolved. Right. We have so many more imaging tests for every ailment that comes in. Right. And so there's going to be more of a demand for radiology because we're going to continue doing more testing that involves imaging.
- [70:23]
B
Right.
- [70:24]
A
And so I think that. I think your job is safe. Obviously. Let's.
- [70:29]
B
And here's. I want to add to that. Let's say that it's not. Yeah. Let's say that the AI tools come out and do a better job for patients than I could do. Bring it on. Honestly, like, my job as a physician is to take care of people. And if the AI tool is doing a better job than me, I want it doing the job.
- [70:44]
A
Yeah.
- [70:45]
B
So I. The reason why I'm skeptical about it is not from a protectionist kind of viewpoint. It's because it's a lot more complicated than the average person may realize. The same imaging finding can mean completely different things depending on the clinical context. We started this conversation talking about how the same finding on a whole body MRI in an average risk population has different meaning depending on what population you're talking about. The same thing is true for everything you could find in radiology. I am a specialist in the abdomen and pelvis, which means from the diaphragm to the pubic symphysis. That's my zone. And. And the amount of complexity, the amount of permutations, the amount of things that offshoots, the amount of algorithms, the clinical context involved, it's just a lot more complicated than most people appreciate. And I think it will be solved. It's a technological hurdle that will be solved, but it's complicated. The other piece of it, too, which I want to throw in as a wrinkle, is that a lot of this has to do not just with technological capability, but also the relationship between human and tech. The way that we interact with technology is going to have a huge role to play in the uptake of this. And this is on the regulatory side, like with licensing and credentialing and ownership and malpractice insurance, that kind of stuff. And also just how we interact with it as humans. We did a study at Michigan where we looked at detection of residual viable bladder cancer after treatment with versus without AI when the humans interpreted it without AI and then we're told what the AI thoughts. They dragged the performance of the AI machine down. So in other words, humans behaving like themselves performed at some level the AI was better than them for this particular task. Narrow AI, which commonly is true, better. And then when the humans were told what the AI thought, they said, yeah, I don't trust that. And they actually dragged that machine's performance down. So this relationship of like how we interact with AI when we're going to trust it or not, trust is a huge part of the puzzle actually in this equation. And it's not between doctors, it's also on the patient side.
- [72:45]
A
Yeah, well, exactly. That was going to be. My next point is like as a patient, right. If I go to a kiosk and my kiosk says, hey, put in all this stuff and you give me what I need to do, how am I going to feel about that? Right. And I think initially there will always be hesitation. But at some point will there be acceptance? Probably, if that's the way it is. Right. But also, like, who's liable then? Right. Say we do go to a model where AI is your front line. You don't even have a doctor overseeing that. Like AI is your front line. Who's liable.
- [73:13]
B
Yeah, Ultimately the physician is.
- [73:16]
A
Yeah.
- [73:16]
B
And the current, in the current framework.
- [73:17]
A
No, I know, but I'm saying if there's no physician involved, like let's say healthcare models say, okay, there is. To save on costs, we're going to implement AI as your first line. You're going to see the AI. And when AI says you need a doctor, then you see a doctor. But then, so does it go to the institution? Is it the institution's problem? Who's. Who's in trouble?
- [73:35]
B
Yeah, that's right. At the end of the day, someone or some entity needs to be responsible for what care gets delivered. I'll give you another example. So in Europe, it's common for two physicians to read a mammogram. You get the original reader and a backup reader to make sure nothing was missed. So they have a two read model. So this is a great population to test AI in. So what they did is they took one of the readers and said, no, no, no, we're going to replace you with an AI tool. And what they basically found is that they could do that safely, effectively and improve performance. Great. In the States, we only have like a single reader typically. So the idea then is, well, why don't we just replace the reader with this tool? We know it works why don't we just replace them? And the brief answer is it's illegal. So it's like, even if it were to work, even if people believed in it, even if they thought the test was outperforming, then you have all of like the regulatory, legal, political, diplomatic stuff that all comes downstream from that. So it's, it's just a wickedly complex problem and not so simple. I think if we're projecting forward 50 years from now, I mean, gosh, completely transformative.
- [74:40]
A
I mean, 50 years ago we didn't even have cell phones, right? I mean like life has evolved. So.
- [74:46]
B
Yes.
- [74:48]
A
What do you think? I mean AI has been, there's been a lot of studies on AI and radiology. Like I can think of 10 I've looked at in neurology. There's more in every use case. What do you think is the most promising AI use case for the near, like the near next 5 years or so?
- [75:04]
B
I think the most promising use case is around image reconstruction, not on diagnosis. And let me tell you why. In the States, as you know, we don't get paid for quality outcomes. Usually we get paid to do stuff. And why does that matter? Because if you're a health system and you don't have a capitated payment model and you're paid based on doing things, you actually weirdly have a financial incentive to improve diagnostic accuracy. You have a financial incentive to improve operational performance. And because of that, most big places that are looking at AI tools are really focused on how is this going to affect my operational performance. So in terms of near term transformative effect, I think because of the incentive structures that are there, that's where you're going to see the biggest change is around doing faster imaging, more appropriate imaging, making sure imaging is done more quickly. I think that's where we're going to see the near term changes. The tools that you're describing around better performance of an AI tool versus a radiologist are numerous. And in fact in most narrow AI tasks, narrow AI, meaning I have a specific diagnosis like is prostate cancer present or not? And this has been shown actually in some studies looking at AI models for prostate mri. Generally the outcome that you see even after you've done appropriate validation and testing is that the AI can perform at least as well as experts and can definitely raise the floor in terms of the quality and performance, performance and consistency. You see this over and over and over and over again. The problem is that radiologists don't do narrow AI. That's not the job. The job is complex AI. It's general AI, it's understanding hundreds or thousands of permutations of what you're being seen, what you're being shown. Rather, it's not the narrow task. So the appetite at big health systems to acquire narrow AI tools is pretty low. Unless you can show, and this is very dystopian, I apologize, unless you can show a financial return on investment. And that's just because of the nature of the set of structures we have in place. So if you had an AI tool that found better detection of something that allowed you to do a high margin intervention, they'd be interested in that. But if it's something that's just preventive long term population health benefit, there's not as much of a financial incentive for it. So I think that transformation is not likely to happen soon.
- [77:17]
A
Sort of a bummer.
- [77:18]
B
It's a bummer, right?
- [77:19]
A
Yeah, yeah. I mean, I can think of, I think in terms of like, where can they put an AI in where they can take a person away? Because then they save on the financials of hiring a human is always very lucrative for systems. But that's difficult to do in large systems because it's not. Not. There's a lot of protections in place for employees. But it'll be interesting to see. It'll really be interesting to see. I think, you know, you've obviously had a long career from your early days in residency to now. What is the one thing that you may have changed your mind about and something that has really challenged what you believed in?
- [77:58]
B
I thought that it was enough to make a diagnosis, it was enough to write the diagnosis on my radiology report. That was enough and that is not enough. And I think when I look back on that, it was naive actually creating a diagnosis or making a diagnosis and writing it on a piece of paper. I mentioned before, that hierarchy, that's like level two of the hierarchy. What actually matters is are we delivering health value to patients, are we improving the quality of life, are we improving the quantity of life? And unless we're doing those things, the rest of it's noise. The rest of it's just a mirage. It's not actually creating value. And a huge number of radiology tests are actually deemed to be low value care. The estimates vary depending on your judgment and opinion about what low value means, but it's a lot. It's probably in the range of 20 to 30%, conservatively here in the United States, for a variety of incentive based reasons. So when a test is done and you find something, you make a diagnosis and you feel that rush of I've done good, I've helped, I've made a diagnosis, I, there's many cases in which I do that. And it actually doesn't create a health delta for the patient. And although it made me feel good in the moment, it didn't actually help the patient. So ultimately it was like a waste.
- [79:16]
A
Can you give an example?
- [79:18]
B
Yeah. So let's say for example, I'm looking at a patient who's a post operative patient, they've had a complex surgery and I make a subtle detection that there is early bowel ischemia. And it looks like the person's bowel is compromised, which is a big deal in the post surgical setting. And I say, okay, I think the bowel is ischemic. And I call the floor and say, I think the person's bowel is ischemic. And they say, thanks for letting us know. We just got done coding the patient, they had PEA arrest and it looks like they're not going to make it the rest of the day. And we've put it on, put the patient on hospice care. This is a dramatic example, but this kind of stuff happens all the time. I gave some examples earlier in our discussion about the over detection of certain things which technically have the name cancer attached to them but don't create long term value for patients. A good example of this would be in the urology space. This is a good example. When we characterize cystic masses in the kidney, we use a classification called the Bosniak classification and it assigns the risk of there being cancer in the finding. And if you have a Bosniak 3, it goes on a 5 point scale that stops at 4, 5 point scale that stops at 4. If a Bosniak 3, historically we still quote people a 50, 50 chance of this being cancer. And when I was early in my career I would say I have done a good deed because I found something that has a 50% chance of representing cancer. And for a long time, for many years, the management of a Bosniak 3 was surgical resection. And I mean thousands of patients had their kidney operated on, they had complete nephrectomy, they had a partial nephrectomy for this. And what we've since learned is that many, many Bosniak 3s don't actually result in metastatic disease or death. They don't locally progress, they just kind of hang out. And now active surveillance is a very commonly accepted kind of approach for a Bosniak 3 Mass. So early in my career I thought I made a great diagnosis and I helped this patient. And what I've started to realize, I guess now towards the middle of my career, is that not so fast. And many times when we find something, we think we're creating value, we're actually not. Both from a low value care perspective and also because the outcomes we're getting aren't what we want them to be. It's a little complicated.
- [81:32]
A
It is. So how do you define high value care when it comes to radiology?
- [81:36]
B
So when people talk about healthcare value, what they're basically saying is outcomes, meaning longer life, better quality of life, more enriched life, happier, healthier, longer. That's the value, that's the outcome that you're looking for relative to what is the expense generated to create that value. So let's say that I could extend someone's life by a year, which is great. That's a great outcome. And it costs $1 trillion to achieve that outcome. That will be low value care. Even though I've prolonged life by one year. That $1 trillion is not scalable. We can't actually do that as a population, even if you're a high net worth individual most of the time. Maybe we're getting there now or some people can afford that. But whereas something that's maybe very inexpensive, like asking someone to five days a week do 30 minutes of vigorous exercise, that's a fairly low cost intervention. From a financial standpoint, the outcome is amazing. I mean, the study's looking at people who are eating right, are not obese, are not smoking, are engaged in regular exercise, are paying attention to their mental health. I mean, those people, they live longer and they live happier and they live healthier lives. And we're talking about years and years of extra life gain and quality life gain from doing those interventions. And they're relatively low cost. That's a very high value outcome. If you think about value in the context of screening, which kind of started our conversation today, we know that in high risk patients who've got a significant smoking history, doing lung cancer screening is absolutely the right thing to do. And by the way, take a small moment to talk to your audience and say this. Most of the screening programs we have, we have about a 70% rate of appropriate screening, like colorectal cancer screening, prostate cancer screening, if you wish to opt into that, cervical cancer screening, they have about a 70% uptake of patients who are eligible, say, yeah, I'm going to do that. Which means 3 out of 10 people aren't. It's not so good. Yeah, not so good. And in general, these Screening which are approved have a 20 to 40% mortality reduction by doing them. That's a big deal. Lung cancer screening is, will give you about a 20% mortality reduction by opting into it if you have a significant smoking history and the rate of people doing it. Right now, people who qualify is only about 15 to 20%, which means there's lots of people who are prior smokers or who are current smokers who qualify for this test and could gain a 20% mortality reduction and are choosing for whatever reason not to do it. And that's very sad.
- [84:09]
A
Yeah.
- [84:10]
B
So value is created by outcomes and what it takes to get that outcome and it's the relationship between those two things.
- [84:16]
A
Yeah. And so basically all the evidence based screening that we offer is high value. Essentially.
- [84:23]
B
Yes.
- [84:23]
A
It's always been interesting to me. A year in life is considered a very good outcome. So a lot of the prostate cancer drugs that have come out for metastatic prostate cancer extend life by eight months, nine months, a year at most. I mean, so much money goes into those trials. As a layperson, I feel like a year doesn't seem that long, but it can, it depends on where you are in life, obviously. Like I had a good friend whose wife had breast cancer and she had an baby and a year was a long time for her to live and be with that baby.
- [84:53]
B
Absolutely.
- [84:54]
A
So I think it's very, it's, it's very interesting to think about, but I think as a person it's important to realize like what does that year mean to you?
- [85:01]
B
Yes.
- [85:01]
A
Right. And what is, what is, what are you going to do with that time too?
- [85:04]
B
That is very well said. And I think it's up to the individual to decide what the value they're creating is. Is this outcome something that I think is important or not important? If this extends my life by six months or this makes me healthier for an extra two years, what's that worth? To me, I think that's a very rational conversation. Everyone's going to be different. That's why kind of those informed discussions with your healthcare provider are so important because you can work on those risks and benefits of testing. In fact, this is explicitly recommended for prostate cancer screening. Have a discussion with a provider about the pros and the cons, trading off the over diagnosis and over treatment against the potential mortality benefit which we know exists, and making an informed choice of where you want to err. Kind of where do you want to sit on that? That's a judgment and that's a person's judgment.
- [85:47]
A
You know, I think the hard part about this discussion though is that it's a very emotional discussion. Right. And so even the most well meaning, and I have a lot of very intelligent patients, as I'm sure you've met, many very intelligent people who struggle with this. Right. And like, no, just screen me. Yeah, right. I just want to know.
- [86:04]
B
Right.
- [86:05]
A
And even though you can tell them all the data, all the information is so emotional. No, I need to know if I can, I need to know if I can treat it. And that's a reality that I think we cannot contend with in any. Like, there's no way to take that out of it.
- [86:16]
B
Yeah, you're spot on. And I think my pitch to that mindset is the following. I believe strongly in autonomy, independent decision making, understanding your risks and benefits, informed consent. We live in a free country, right? If someone wants to do something, God bless them, I mean, they're welcome to do that. I think our job as healthcare providers, at least as I see it, is to do our best to inform people about what they're signing themselves up for so that they can make an appropriate choice. It's always for me about the informed consent. If I was doing informed consent for a whole body MRI and a general risk patient, I'd say the following. You are more likely to be hurt than helped by doing this. You believe there's an opportunity for you to save your life by detecting something early. But what is much more likely to happen is you will be harmed by interventions on you, biopsies, surgeries, complications from those things that create negative health value for you. Currently, there is not a single medical society that recommends doing this. Not the American College of Radiology, not the American College of Preventive Medicine, not the American Academy of Family Practice, not Health Canada, not the New Zealand radiologist or Australian radiologist, no major cancer center recommends doing it. And on and on. None of the major guidelines, nccn, American Cancer Society, like, pick your organization. None of them recommend doing this. This includes, by the way, organizations that have a financial conflict of interest in which doing more imaging is good for their members, like the American College of Radiology, don't recommend doing it. So why would that be that? If this test is so great and it creates such great health care value, why do we see virtually, not virtually zero societies recommend doing it? And the only groups that are recommending doing it are those that have a financial conflict of interest on the for profit side to motivate someone to do it. That would be my challenge. So, and I think ultimately once someone has that information, then they should make the choice that's that's right for them.
- [88:08]
A
Yeah, absolutely. What is something now, as you look forward specifically in the men's health space, that you're really excited about in terms of imaging or AI and imaging or anything in the future that you're sort of either working on or things that are coming along?
- [88:23]
B
One thing I'm working on right now is I mentioned before kind of a detailed example of a cystic kidney mass and trying to figure out is this important or not. I mentioned that that cystic kidney mass historically had a 50% chance of being a cancer. And what we've recently learned from a multi center study that our group helped participate in is the risk of aggressive disease. Within that is 10% or less. And so there's an initiative happening right now with the American College of Radiology trying to explore development of something called Chirads, which is an analogous to PI rads for prostate. And the focus of this would be on trying to identify aggressive phenotypes within a mass, a small renal mass that's identified on imaging. The point of it would be to say to a patient who has a small renal mass identified imaging test, yeah, there might be 80% chance of this being a cancer in a small solid renal mass. But I'm going to go one step further and say, what's the probability this is an aggressive cancer that you need to be concerned about? And the reason why this motivates me is because it will likely dramatically decrease the over treatment of these small renal masses in kidneys because people will be more likely to opt into more active surveillance based methods. Right now in our current state, like what's the motivation to do this? Of all the partial nephrectomies and nephrectomies that happen right now, about 30% of them are still for benign masses, which blows my mind. And of the ones that are resected, a huge fraction of them are actually indolent masses or low risk masses. Patients who have a life expectancy of less than 10 years are getting no benefit from that treatment in many cases. So this Kyrads idea, this kind of focusing on aggressive cancer rather than just any cancer, this is the most exciting thing to me because we want to be more precise with what we do and not just randomly take stuff out.
- [90:16]
A
Yeah, no, I agree. I think that is really exciting. I think again, the thought that we'll be able to differentiate more aggressive cancers is really valuable. I think in the long term it really is. Where can people find more about you Your research, what you're working on, some of the guidelines and things.
- [90:36]
B
I am not that interesting.
- [90:38]
A
Yes, you are. I've already heard about all your interests.
- [90:42]
B
So let's see. So if you're interested in things around contrast media, we talked about that. The American College of Radiology puts out something called the Manual on Contrast Media. That's a good go to source. If you're interested in more about the idea of low value care and incidental findings, I published a paper a couple years back called Incidental Findings and Low Value Care, I think is what it's called. And it goes into some more detail to some of these reasons of why this is kind of a fraught area and why it's so complex. And then recently I published a paper with some colleagues at UCSF around what kind of evidence do you need to justify doing additional medical imaging testing? And it was published in the New England Journal of Medicine. Evidence. And that's a worthy thing to read also. And then I'm a co director of the Weiser center for Prostate Cancer, which is all about patient centered care. And I love the center. It's amazing. And if you're interested in checking it out, just Google Weiser center for Prostate Cancer.
- [91:35]
A
Amazing. Amazing. So we end our podcast with questions we ask everyone. I've already been told you have a lot of hobbies, a lot of interesting hobbies outside of medicine. So you are not a boring person. You're a very interesting person. But tell me something you know now in life that you wish you knew earlier. Could be about radiology, could be about anything, really.
- [91:53]
B
I'll tell you a personal story. So when I was younger, in college and in medical school, my primary motivation to get into medicine was, it seemed to me to be a very stable job. And I say that because when I was younger, growing up, my dad lost his job very often. We're constantly moving around, constantly kind of struggling with that. And it appealed to me because it seemed stable and I'd always gotten decent grades. And so people said, oh, you should be a doctor or a lawyer. It was so trivial and silly. But that's the reason why I went into it, frankly. And when I was a resident, my mom got really sick when I was in my second year of residency and went from basically totally normal and healthy to developing a very aggressive cancer and dying within six months. And it was horrible. And going through the process on the patient side, watching her sit in the icu, watching her undergo her paracentesis, and she would say, I'm pregnant. And she was Delirious and she didn't realize it was just malignant ascites. And seeing like all the stuff that we did to her to help her and to comfort her, the things she prioritized and focused on, it completely changed my perspective about what it is that healthcare providers do. It's not technocratic, actually. It's very emotionally connecting. It's deep, it's meaningful. You're affecting people's lives in profound ways. And I honor the profession, actually. I honor what we try to do and I try my best to push it and push it to do better. And there's so many things we could do better in health and especially in this country and trying to do a better job. And it's because at the end of the day, it comes back to the patient and their family and like what's important to them and what's going to make them happier and healthier and live a better life. And we're doing our best to do that in a partnership with them and what their needs are in the moment. So I guess, you know, compared, you know, 25 years ago to today, I think about this much more as a calling and an honor and a responsibility than I actually ironically did when I was younger. And a lot of people have the opposite. They start off with this kind of like inspired self and then they lose it over time because of burnout. And that was the opposite, actually. I kind of like developed this after that event.
- [94:00]
A
Well, that's, that's amazing and thank you for sharing that. I'm sorry about your mom. That's really sad and hard to go through as a medical student.
- [94:06]
B
It was a second year resident at.
- [94:08]
A
Radiology, I feel like even worse. But yeah, that's, that's so tough. What is a non negotiable? Something you have to do every day.
- [94:17]
B
I like to think about big problems and big things and try to move the ball on those. And our lives are so busy and they're so frantic that we can get caught up in the moment of what's happening today, what's the email that is happening today, what's the phone call I gotta deal with today? And unless we kind of take a step back and say no, I have to focus on these medium and long term goals and I have to do that on the regular. I have to do it on a daily basis. I won't actually deliver on any of those long term projects that I'm interested in delivering on. So a non negotiable for me would be I have to look up, I have to figure out what is waiting for me in six months or 12 months, what am I going to deliver in six or 12 months that's going to create value. And it's not just what's happening today. It's like a long term kind of thing. So I think that's my answer. And it changes from year to year depending on what I'm working on. But my answer is the non negotiable is I have to think big picture. I can't keep my head down.
- [95:13]
A
I think that's so valuable. I mean, I do that all the time. I try to think about what my year is gonna look like, what my month is gonna look like and what is the goal, like what is the ultimate long term thing that I can do that can provide the most impact. And you know, I was a full time practicing physician and decided that my impact was gonna be greater doing this and doing social media and creating content and educating the public. And I still practice, obviously, but I think that if I didn't take that time, I would have never like even had the foresight to take those steps. Right. And I think it's so valuable for anyone listening is like, just like take a minute and think about what you're doing and why you're doing it. Because sometimes we just do it because we have to. Right? Yeah. I remember someone told me, like, there are no real deadlines and there's a few but like there's no real deadlines. And that gave me so much freedom in terms of like, okay, if I can't get it done by the exact due date and it's not that serious, like it's okay. I need to make sure my priorities are taken care of.
- [96:09]
B
Yeah, that's very well said. I agree.
- [96:11]
A
Yeah. And then what is a life hack or health hack you'd share with our audience?
- [96:15]
B
I pack my lunch every day. Such a minor thing. I pack my lunch every day in the morning when I wake up, I put together something for myself that I'm going to be eating at lunch. And I know that my future self doesn't get a lot of control over it if my present self makes them eat something healthy. So I'll put my piece of fruit in there, I'll put some nuts in there, I'll put in something, I'll put my bottle of water together for my lunch. And I do that because if I don't, when lunch comes, what am I going to do? I'm going to walk up to the cafeteria and I'm going to buy some unhealthy thing, whatever looks good in the moment. So in other words, I kind of set my future self up for success by doing something simple in the morning. The second small little life hack thing that I do, this is also silly. But physicians can choose where they want to park. They can get the fancy expensive parking pass and park right next to the building, or they can get the less expensive pass that's kind of a distant walk. And for years I've always bought the less expensive one. So I every day bake into my day, like a 1 1/2 mile trek back and forth to the hospital.
- [97:13]
A
In Michigan.
- [97:14]
B
In Michigan. And the reason why is because again, if I say on my daily basis I have to do exercise, I force myself to do it because I know that if I don't come in early enough in the morning, I can't get my parking space and I can't walk to my building. These are dumb things, but I think they have a common element to them which is it's hard to trust your present self because your present self is going to make a bunch of decisions that are based on the immediacy of the moment. Whatever feels good in that moment. Whereas if you kind of plan ahead and say, I'm going to set the conditions such that my future self has no hope but to eat healthy food or do one and a half miles of exercise every day or whatever it is for yourself, set yourself up for success.
- [97:54]
A
I love that. You know, I think we've come into a culture of comfort, right? And I had a friend, a good friend, she was like, you have to make things a little bit hard. Like you have to just choose the hard because otherwise you will just choose comfort. I mean, we are sitting in an air conditioned room, we're very comfortable. Like life is comfortable. You can make it as comfortable as you want. And so I think, I think it is really important and that those are great tips, really good. If you couldn't be a radiologist, this is where I want to hear about all your hobbies. What would you be?
- [98:24]
B
Oh, goodness. So in the medicine world, I originally was thinking about being a urologist of all things, so probably would have done that instead. But outside of medicine, I do have a lot of hobbies. So I'm a amateur singer songwriter. So I have a couple albums and about three or four years ago I was, I went to like a pizza parlor and there was this dude who was like sitting on a chair, strumming along, singing songs and stuff. I was like, I would love to do that because the whole Audience was. It wasn't like he was capturing the whole room, but you could feel it was a vibe. And I was like, I would love to do that. It's like. But I don't know how to play guitar. I don't know how to sing. I don't know how to write music. I can try. So for a couple years, like, that was my project, so I learned how to play. I worked with a professional musician, and we kind of put a couple albums together. And that's been such a rewarding part of my life. Now, obviously, I'm never gonna make a living in that for a bunch of reasons, but it's a very joyful thing for me.
- [99:22]
A
Is it under your name? Can we listen to them?
- [99:24]
B
It's on Spotify, Amazon Music, whatever. Yeah, it's under Matthew Davenport. So I've written a variety of children's books and novels, and then I do fiction. Fiction?
- [99:34]
A
Yeah. What kind of genre? For the adult novels.
- [99:37]
B
So for the adult, I'm currently writing one right now about an older person who develops Alzheimer's disease and the effect that that has on the people around them. And then for some of the children's novels, they kind of are written like, they're sort of written for adults, but in the children's style.
- [99:53]
A
Okay.
- [99:53]
B
So they're picture books, but they have adult themes, so.
- [99:57]
A
Interesting.
- [99:57]
B
Yeah.
- [99:58]
A
So you already have a lot of other jobs, basically, which, if you could do one of them full time or something else, what would it be?
- [100:06]
B
I love what I do. I really do. I love what I do because I do clinical care, I do research, I do public policy. I try to make a difference in the world. And when I think back on the stuff that I've done in my life that I have the most excitement about, it's when I've moved the needle on something that I think is that benefited millions of people. And that's happened a few moments in my research career and a few moments in my public advocacy career. And that's where I'm happy is doing those things. I think if I turn my hobby into my job, it probably would stop being so fun.
- [100:40]
A
That's probably true. Well, thank you so much. It's been an honor to have you.
- [100:43]
B
Likewise. Thank you. Appreciate it.
- [100:45]
A
Thank you guys so much for joining me on today's episode of the Rename Malik, M.D. podcast. If you enjoyed this episode, guys, please do me one solid. Subscribe to the podcast either on YouTube or on your favorite podcast platform. And if you have an extra minute, leave me a rating or review this helps people see that this podcast is worth listening to. And as always, we're going to take care of yourself because is your worth Get a jump on next Summer with vrbo's Early Booking Deals don't wait to claim your dream summer spot, whether that includes a good porch swing or a poolside lounger. When you book early, you get the best places at the best prices. But back to the poolside loungers. With vrbo, you don't have to reserve any loungers. They're all yours. In fact, the whole private home is yours. Book with early booking deals and you can lounge around all summer long however you please@vervo.com this episode is brought to.
- [101:39]
B
You by Athletic Brewing Co. No matter how you do game day, on the couch, in the crowd, or manning the snack table, Athletic Brewing fits right in with a full lineup of non alcoholic beer styles. You can enjoy bold flavors all game long. No hangovers, no buzz, no subbing out for water in the fourth quarter. Stock the fridge for kickoff with a variety of non alcoholic craft styles available at your local grocery store or online at athleticbrewing. Com. Near Beer Fit for All Times.