FHC #223: How GenAI & World Cup strategy can help fix healthcare

0:11Dr. Robert Pearl: Hello and welcome to Diving Deep, part of the Fixing Healthcare podcast series. I'm one of your hosts, Jeremy Corr, also host of the Popular Books of Medicine podcast and CEO of Executive Podcast Solutions. With me is Dr. Robert Pearl. For 18 years, Robert was the CEO of the Permanente Group, the nation's largest physician group. He's currently a Forbes contributor, a professor at both the Stanford University School of Medicine and Business, and author of the best-selling book, Mistreated, Why We Think We're Getting Good Healthcare and Why We're Usually Wrong, and Uncaring, How the Culture of Medicine Kills Doctors and Patients.

0:38Dr. Robert Pearl: His new book, ChatGBTMD, How AI-Empowered Patients and Doctors Can Take Back Control of American Medicine, remains a bestseller. If you want information on a broad range of healthcare topics, you can visit his website, robertperlmd.com. Robbie, listeners continue to be very interested in generative AI. They're unclear exactly how it can be best used in medicine and how its introduction will lower costs. Can you dive deep into what you see?

1:02Jeremy Corr: Jeremy, let me begin by saying that I don't believe that the primary way that generative AI will make health care more affordable for Americans is by replacing doctors and nurses. Of course, in every major transformation, regardless of the industry, work does change. What some people were doing before, that's accomplished by a technological tool. But at the same time, what we see is that other jobs spring up in its place. Having said that, in earnings reports and internal communications, the CEOs of Amazon, Salesforce, IBM, and Shopify, alongside with dozens of other publicly traded companies, they link the introduction of generative AI to reduced headcount and decreased costs.

1:43Jeremy Corr: The World Economic Forum now estimates that AI will displace 92 million jobs by 2030, while Goldman Sachs reports that AI is already erasing a net 16 ,000 U.S. jobs per month. Given these trends, the logical question is why does reducing the number of employees make sense in these other industries, but not in medicine? The concept of saving dollars by replacing humans with technology, it's easy to understand. A generator of AI can allow a software engineer to write faster code, a lawyer to review more documents, or a customer service agent to resolve more cases each hour. Companies can produce the same output with fewer people.

2:30Jeremy Corr: For centuries, technology has lowered costs by substituting mechanical power for human labor. Agriculture is the classic example. Tractors and other machines allowed the United States to produce more food, with far fewer farmers. Over time, the share of the U.S. labor force working and farming fell from roughly 41 % in 1900 to less than 2 % today. Meanwhile, food has become far more affordable relative to income. Americans spend roughly 20 % of disposable income on food, and that was in the 1960s, compared to less than 10 % today. The difference between these industries and medicine is that these other industries, the primary driver of demand, that's price.

3:16Jeremy Corr: But in medicine, it isn't the case. Even if hospitalization and surgery were free, there would be fewer people wanting more of either. In medicine, keep people healthy, demand falls, and costs plummet.

3:32Dr. Robert Pearl: Robert, can you dive deeper into these differences between healthcare and other industries? Sure.

3:37Jeremy Corr: Jeremy, in healthcare, pushing doctors and nurses to care for more patients in less time, that results in very limited savings. Yes, it might lower costs slightly, at least at first. But the major expenses in medicine, they arise when preventable and controllable diseases, hypertension, diabetes, progress to life-threatening problems like heart attacks, strokes, and kidney failures. avoid a serious medical condition, you prevent 100 % of what otherwise have been a massive expense. This transition from relatively easy-to-manage chronic disease to life-threatening acute problems is common. It's when people develop severe medical challenges that they usually require surgery, hospitalization, and expensive medications.

4:28Jeremy Corr: As an example, poorly controlled hypertension that accounts for 40 % of strokes. Unmanaged diabetes is the leading cause of kidney failure and lower limb amputation, major expenses for Medicare. That's why preventing these dangerous, life-threatening medical problems will have a far greater impact on healthcare affordability than reducing headcount. Just look at the math. National health care spending is currently$5.7 trillion, and it's projected to reach$9 trillion by 2034. Of that, physician and clinical service would account for only a fifth of total medical costs. That's according to the Kaiser Family Foundation estimates.

5:13Jeremy Corr: Direct wages for doctors and nurses make up an even smaller share. The season of generative AI reduced clinician labor expenditures by 10%. Total medical spending, that would fall by only 2 % or 3%, and that assumes that every dollar saved from reducing staffing would flow back to patients, employers, or taxpayers. In contrast, and this is based on CDC data, effective control of chronic disease can reduce life-threatening complications, heart attacks, strokes, kidney failures by 30 % to 50%. And it would save upward$1 trillion annually, nearly 20 % of total medical expense. Staying healthy is what every American wants for themselves.

6:03Jeremy Corr: It's what they want for their families. And it is what broad application of generative AI now makes possible.

6:10Dr. Robert Pearl: Robbie, can you provide details on the three biggest opportunities application of generative AI makes possible?

6:16Jeremy Corr: rise to the top of the list are using generative AI to, number one, improve chronic disease control, two, increase access to medical care and expertise at night and on weekends, and three, maximize the coordination of care to avoid medical errors and complications. All three opportunities to improve quality. They've been recognized for decades, but the expense of doing so has been so great. that the major investments that would be required have never been made. But now with the availability of generative AI, each could be accomplished at an affordable cost by empowering patients. Let's begin with chronic disease.

7:00Jeremy Corr: Long-term illnesses like hypertension, diabetes, heart failure, and kidney disease affect an estimated 76 % of American adults, and they drive a disproportionate share of medical spending. CDC says that chronic diseases account for most illnesses, disabilities, and deaths in the United States. Yet despite enormous spending, chronic disease remains poorly controlled. Less than half of U.S. adults who have high blood pressure have the problem under control. Millions of patients with diabetes and chronic heart failure experience similar gaps. It's what medical science makes possible or what the health system reliably delivers.

7:41Jeremy Corr: generative AI can help close the gap. It can explain diagnoses and treatments in language that patients understand. And this will increase adherence to evidence-based medical recommendations. It could connect with bedside and wearable monitors to continuously track blood pressure in patients with hypertension, blood glucose in people with diabetes, and early signs of deterioration in those with chronic heart failure. Rather than waiting months until the next scheduled doctor's visit, Gen. AI could alert patients when a chronic condition is not under control, and it would allow clinicians to make necessary medication changes months earlier.

8:20Dr. Robert Pearl: Robbie, what's the second biggest opportunity?

8:23Jeremy Corr: The second area is relative to access. Remember, the typical medical offices open only about a third of the total hours in a week. The medical problems, they arise regardless of time or day. At night or on the weekend, patients are often forced to do an impossible choice. Should they wait until the doctor's office opens and risk getting worse? Or should they go to the emergency department and face hours of waiting and far higher costs for care that may not have required an emergency room visit in the first place? It's 10 p.m. Parents of young child see that their infant has spiked the fever to 103.

9:05Jeremy Corr: Go now to the ER, they'll sit in a waiting room surrounded by people with colds and infectious diseases. Wait until morning, their child could be dead. For individuals and families, generally I can assess new symptoms, evaluate a patient's problem, and alert people if the symptoms require immediate attention, but also inform them when it's safe to wait. Clinicians worry that AI tools make mistakes. But compared to a patient happy to guess what to do or consult the internet for information, the risks of obtaining expertise from generative AI, they should prove far lower. It makes no sense to believe that nothing is better than this information, even if the information is not perfect.

9:51Jeremy Corr: With recent studies demonstrating that large language models can answer medical questions at a level comparable to, if not better than most clinicians, a new day has dawned.

10:04Dr. Robert Pearl: Robbie, what's the third way generative AI application could improve quality and make medical care more affordable?

10:10Jeremy Corr: The third way that generative AI could be powerful would be coordinating medical treatment among clinicians and helping patients to safely manage their medication prescriptions. This is a growing challenge for our nation's seniors. Currently, more than 90 % of adults who are 65 and above live with at least one chronic condition. Nearly all take prescription medications. On average, Medicare patients with multiple chronic conditions are treated by 14 different physicians annually. With this population care coordination, it's not a minor issue. It's central to safety, to quality, and to cost control.

10:51Jeremy Corr: Many of the medications that doctors prescribe, they're life-saving. But over time, as the patient ages, many of these same drugs become unnecessary. And they increase the risk from falls, fractured hips, and provide excess sedation, negatively impacting people's lives, rather than improving their health. Yet despite these risks, most prescriptions are refilled automatically for years. A second problem occurs when a different physician from the patient's usual clinician describes a new medication without knowing the person's full drug list. This can produce a dangerous drug-drug interaction with life-threatening complications.

11:35Jeremy Corr: Generva can assist in resolving these challenges. It can provide oversight by identifying medications that no longer appear indicated based on national guidelines and warning patients when two drugs they are taking may interact. Genitive AI tools could serve as a repository for the patient's medical history and help resolve the lack of interoperability of today's electronic health records. By reducing pharmaceutical errors, Genitive AI tools could increase patient safety

12:05Dr. Robert Pearl: and lower medical expenses. Robby, the opportunity is clearly massive. What stands in the way?

12:13Jeremy Corr: Jeremy, the barriers that restruct the use of Gen.AI in medical practice, they're not technological. They're human. The first barrier is financial. As long as doctors and hospitals are paid primarily for the volume of work they do, Gen.AI tools designed to empower patients and provide medical expertise in the patient's home will be seen as a threat. A decapitation, a single payment, made to a group of doctors to provide the totality of medical care for a population of patients. This approach aligns incentives. The second barrier is systemic. Gen.I. will not lower costs simply because patients download an app, because doctors added another tool to the electronic health record.

12:57Jeremy Corr: The model of medical care delivery will have to change in parallel. Medical care today remains two-fragmented. It's too uncoordinated, and it's too dependent on calendar-based office visits. To maximize value, clinicians will need to use generative AI to provide patients with expertise continuously and do so in the patient's home, identify medical problems earlier, and coordinate care delivery across clinicians.

13:28Dr. Robert Pearl: Robbie, is there anything else?

13:32Jeremy Corr: Yes, and Jeremy, it might be the most difficult barrier to overcome. It's the culture of medicine. Doctors still prize intervention more than prevention. The specialist who opens a blocked artery, that individual celebrated more than a primary care team that prevents the heart attack in the first place. and any technology that performs work that was once reserved for physicians will be perceived by doctors as a threat. If healthcare leaders rate the current medical care system as excellent, then they will fail to pursue medicine's biggest opportunities and they'll view replacing people with technology as the only path to lower costs.

14:17Jeremy Corr: But once they acknowledge the many ways that medical care today fails, Then cultural change can begin. Clinicians view the clinical care provided in the United States as excellent. What they don't focus on are the more than half a million people who die annually from complications of poorly controlled chronic diseases. They die from heart attacks and strokes and kidney failures. Doctors don't recognize the 400 million Americans who die annually from misdiagnoses. and they don't realize the problems for patients and families when they can't obtain medical expertise at night and on weekends. They think that the recorded message on their office phone telling individuals to go to the ER if they're concerned about a medical issue solves the dilemma.

15:06Jeremy Corr: But for millions of Americans, it doesn't. And the poor communication among doctors, they don't see that as a problem since all clinicians are confident about the medical care they individually provide. and they can see and they don't see, they fail to see the cracks through which individuals fall every day when care isn't coordinated. Once clinicians acknowledge the problems, then they'll be able to recognize the mass of opportunities to increase quality, to improve access into lower costs that Gen. AI provides. But until then, the medical culture will block advances. The good news is that once clinicians implement Gen.

15:47Jeremy Corr: AI solutions, they will begin a virtuous cycle. As avoidable demand falls, they'll gain more time for complex patients, provide better care to these complex individuals, and there'll be fewer complications from their diseases. Fewer complications will further lower costs, and lower costs will make it possible for the healthcare system to invest even more in prevention, access, and coordination. As I said at the beginning, replacing doctors and nurses, that's not the future of American medicine. Helping them keep people healthier, that is.

16:25Dr. Robert Pearl: Robert, you teach at the Stanford Graduate School of Business. What do you tell these future entrepreneurs about how best to create generative AI tools to maximize patient use and clinical outcomes?

16:35Jeremy Corr: Jeremy, the first thing I tell students, and by the way, I say the same thing to CEOs of dozens of startups every year, is that having a great product isn't enough. At least once a month, I receive a request to help an innovative company figure out why hospitals and doctors and insurers aren't willing to provide or pay for their product. Both in the classroom and in consulting, I emphasize that 90 % of startups fail. Not because their products aren't great, but because the founders don't understand and they haven't figured out how to monetize their products or services. In the past, the path to success was clear.

17:14Jeremy Corr: You create technological tools, you design them for doctors, and then you sell them to hospitals and health systems at high prices. There are many examples, MRI machines, CT scanners, surgical robots, and nearly all generative AI companies today are trying to follow that playbook. They're building narrow generative AI tools that are built for specific clinical tasks and designed solely for physician use. But a new study challenges the assumption that entrepreneurs today should follow the traditional approach of the past. Can you tell listeners the details of the research? This research was published June 12th in Nature Medicine.

17:59Jeremy Corr: The study compared two specialized physician-facing AI tools. One's called Open Evidence and the other's called Up to Date. with three widely available general-purpose large-language models, BOD Opus 4.6, GPT 5.2, and Gemini 3.1 Pro. The researchers tested the five models. They gave it 100 medical knowledge questions and multiple real-life clinical scenarios. Then they had practicing clinicians grade the answers without knowing which system had produced them. The results were striking. On the medical knowledge portion of the study, the three general purpose consumer models outperformed the more expensive physician-only models.

18:47Jeremy Corr: Gemini scored 97.4%, ChatGPT 94.2%, Claude 90.2%. These three led the pack. Open evidence, it trailed at 89.6%, and up to date, 88.4%. These unexpected findings raise two important questions for entrepreneurs. If doctors and patients could obtain comparable expertise from large language models like ChatGPT, Claude Gemini, for$20 a month or less, why build these expensive AI systems instead of applying the currently available large language models? And if these models are excellent, why should we focus only on physician use? entrepreneurs should consider the larger opportunity, helping the 330 million Americans use the tools that are already available on their computers and smartphones to manage and improve their health.

19:45Dr. Robert Pearl: Could you provide more details about the difference between the generative AI models now available and the AI tools that have been used for the past 20 years?

19:53Jeremy Corr: Before Chash APT launched publicly in November of 2022, the AI tools used in medicine were task-specific, machine learning applications, called narrow AI. These tools are trained in large sets of data. As an example, they might take 10 ,000 mammograms, half of which show cancer, and the other half either normal or with a benign problem. In these examples, the model identifies dozens of differences between images, the ones labeled cancer and those labeled normal or benign. It then uses statistical analysis to provide a risk score or a likelihood that cancer is present in a mammogram that it has never seen.

20:35Jeremy Corr: In a large international evaluation published in Nature, an AI system outperformed the average radiologist and reduced false positives by 5.7 % and false negatives by 9.4%. However, narrow AI has major limitations. The tool built to evaluate mammograms, it can't interpret chest X-rays or identify a rib fracture, even though the lung and ribcage are a few inches away. Moreover, although the FDA's list of authorized AI-enabled medical devices now includes more than 1 ,500 applications, adoption in day-to-day medical practice, that remains scant. It's limited by high price and liability concerns.

21:19Dr. Robert Pearl: How are current general AI applications different?

21:26Jeremy Corr: Jeremy, generative AI applications, in contrast to narrow AI tools, are trained on enormous bodies of text and data. And these include the majority of medical information from textbooks and research studies. With that massive foundation, it can respond to nearly all medical questions. As examples, they can explain imaging reports. They can assess clinical symptoms, provide prime disease management assistance, details medication side effects, and they can help patients understand their acute and chronic conditions. As a result, these powerful applications can, and already are being used, not only by clinicians, but also by patients.

22:07Jeremy Corr: Currently, research from OpenAI shows that 55 % of U.S. adults turn to chat GPT for medical assistance, both to check on their current problems or to explore new clinical symptoms. Yet, despite the broad application generative AI provides, its growing popularity among patients, its commercial use in medical settings has been limited. Doctors mostly use it for administrative tasks. We're talking billing and claims. Currently, the main applications of generative AI in clinical practice are limited to ambient listening tools like Microsoft's Dragon Copilot. These tools listen to doctor-patient conversations.

22:48Jeremy Corr: They convert the information into clinical notes. This has freed up doctors from being data entry clerks and allowed them to look at their patients during the visit rather than, as in the past, having to stare at the computer screen. However, to date, there's almost no general application that entrepreneurs or companies have built for independent patient use. How do we reach this point? Jeremy, looking back, it didn't make economic sense to create new tools for patients at a time when medical AI mostly consisted of narrow AI models. Obtaining the data sets, validating performance, and securing FDA approval, all of that is time-consuming and very expensive.

23:30Jeremy Corr: That forces companies to charge prices beyond what nearly all patients can afford. Furthermore, selling narrow AI tools designed to address a single specific medical problem and selling them directly to patients, that would create legal risk if someone were harmed. By selling to hospitals and health systems, manufacturers can place physicians in the role of ultimate decision makers and make them liable should anything go wrong. That's why the NYU study matters so much. It opens an entirely new pathway for AI to improve clinical outcomes and lower costs. As I pointed out, in comparing the five alternatives, the specialized clinical tools carried the medical branding.

24:14It had the professional positioning and implied a promise of greater reliability.

24:20Jeremy Corr: Yet the inexpensive, the widely available large language models, these performed just as well or better than the clinicians. When I look at the two options of developing new tools for patients versus helping them to use the models they already can access for clinical expertise, I'm impressed by the opportunities that are currently available from tools like Chachi, BT, Colorado, Gemini. Unlike a medical device built for a defined clinical task, these general AI tools aren't marketed to diagnose or treat a specific condition. Instead, the applications could answer medical questions in the same way they provide expertise in response to any prompt they're given, all of question that they're asked.

25:05Jeremy Corr: And that makes these large language models almost impossible to regulate. And consequently, they don't require FDA review at this time. and they've led to only minimal medical legal risk. With a narrow AI application, the data upon which the tool is trained and the accuracy of the answers it provides, that can be evaluated and calculated. With a large language model, the expertise depends on the person asking the questions, giving the prompts. That's also the reason why large language models can provide personalized answers based on the individual's medical history, medications, and individual personalized preferences.

25:45Jeremy Corr: That flexibility and broad availability makes these applications ripe for entrepreneurial innovation. Creating tools that teach Americans about how to obtain better information shouldn't require FDA approval, since they're being broadly educational, not disease-specific, even if they're used by patients for that purpose. And the examples we talked about, including both chronic disease management and improving access to medical expertise at night and on weekends, they would be powerful examples of how patients might apply them. Individuals could learn how effectively they were controlling their blood pressure and their blood sugar and whether they would benefit from a change in medication.

26:27Jeremy Corr: So many people could learn whether their new symptoms were potentially problematic and whether it was safe to wait until the doctor's office was available the next day or the following Monday or even weeks in the future. when the next office visit and appointment was available. In the era of generative AI, it's illogical to believe that no information is better than a tool that matches doctors' accuracy and is readily available 24 by 7.

26:55Dr. Robert Pearl: What advice do you have for the next generation of entrepreneurs? If inexpensive large language models are as good as expensive medical AI,

27:03Jeremy Corr: that entrepreneurs have an opportunity to create the next healthcare unicorn by helping more Americans apply current generative AI models for medical care rather than building high-priced, specialized applications. Within this opportunity, many difficult decisions await. The next generation of entrepreneurs will have to decide if the product will be an application to distill expertise from a current large language model whether it will be a set of agents to obtain the needed information on behalf of the person or whether it's going to be an educational tool to teach the patient how to obtain this information directly.

27:46Jeremy Corr: Entrepreneurs will need to decide whether to sell directly to consumers or charge a monthly subscription fee or contract with employers and insurers to provide the tools to a subgroup of individuals or simply be rewarded based upon measurable clinical improvements in chronic disease management and the prevention of heart attack, strokes, and kidney failures in all of a company's employees. Before broadly promoting their tools, they'll need to find a way to test their product for safety and reliability, even though ultimately accuracy depends on how patients enter information and how they ask the follow-up questions.

28:23Jeremy Corr: Finally, entrepreneurs could offer, in conjunction with the Genitive AI tools, telemedicine services, services that can provide clinical and clinician expertise when patients have a problem that the large language model identifies and their doctor isn't available. But one thing is clear. Inexpensive Genitive AI solutions designed for patients remain a wide open opportunity. The financial and medical logic for pursuing it is simple. Treat patients in a doctor's office, and you help them once. Teach patients to use generative AI effectively. You can improve their ability to manage their health for the rest of their lives.

29:07Dr. Robert Pearl: What else will be needed if our nation wants to apply generative AI to American health care? The big answer is leadership.

29:15Jeremy Corr: Leaders make things happen, otherwise we're not. And we're talking about transforming 20 % of U.S. gross national product. and that's the totality of products and services that Americans spend their money on each year. This magnitude change just doesn't happen on its own. Leadership is a topic I cover in the classes at the Stanford Medical School and the Stanford Graduate School of Business. And in watching the recent World Cup, I realized how much overlap there was between winning in sports and the challenges of American medicine today. The beauty of his sporting competition is that almost always there's a clear winner and a clear loser.

29:56Jeremy Corr: In this case, 48 nations entered the 2026 World Cup, with only one Spain left its champion at the end. The other 47 teams, they had brilliant players, experienced coaches, and enormous financial support. France fielded the tournament's leading scorer, Kylian Mbappe. Argentina had Lionel Messi, arguably the greatest player in history. In England, reached the semifinal with one of the most talented teams in decades and its own hero, Harry Kane. But none of these won the championship. American medicine is similarly loaded with talent, loaded with money. The United States has exceptional doctors, world-class hospitals, groundbreaking technologies, and per capita healthcare spending nearly twice that of most other nations.

30:42Jeremy Corr: Yet the results continue to fall short. Care remains unaffordable for millions of families. Access is unreliable. Crime and disease are worsening. Clinicians are increasingly frustrated. And the reasons that healthcare leaders fail, the reasons that many of these immensely talented soccer teams fail, are eerily similar.

31:03Dr. Robert Pearl: Robbie, like you, I am too a sports fan. I'm fascinated to learn what caught your eye.

31:08Jeremy Corr: Jeremy, the first reason why immensely talented World Cup teams don't win is the absence of an effective and clear strategy. So the medicine lacks one today. Players need a game plan that tells them how to attack, how to defend, and when to adjust as circumstances change. But American health care doesn't have a clear strategy today. Doctors only have a vague idea about ways to improve quality, expand access, and make care more affordable for patients. They tried their best, but the outcomes were inconsistent and far from optimal. They asked hospital administrators, physicians, and insurance CEOs how to achieve those quality access and cost goals.

31:50Jeremy Corr: You'll get dozens of answers, many of which will be contradictory, and most will blame someone else. In short, American health care is an abundance of self-serving corporate strategies, but no shared clinical strategy for winning on behalf of patients and the nation as a whole. Medical care remains fragmented across independent practices, across hospitals and competing medical groups. Daily all doctors are paid through a FIFA service to pay for volume model. The more visits and tests procedures you provide, the more you earn, regardless of whether those interventions improve outcomes. Care, it's rarely coordinated, and physicians have few financial incentives to prevent disease or avoid complications.

32:37Jeremy Corr: Why would our country expect better outcomes than the failed ones of today? During the World Cup, superstar Lionel Messi demonstrated the importance of strategy. At 39, he could not. And he didn't try to match younger players stride for stride. In fact, he spent long stretches of each match, estimated maybe two-thirds of the time, walking, conserving energy, and studying how defenders position themselves. But then, when the space opened, even a crack, he was ready to accelerate. A strategic approach helped Argentina produce repeatedly late-match heroics and come within one goal of winning a second consecutive World Cup.

33:21Jeremy Corr: Healthcare leaders need to recognize that periods of disruption similarly create opportunities for action. What was impossible in the past suddenly becomes possible today or in that moment. When it arrives, timing proves vital. Not too soon, but never too late.

33:41Dr. Robert Pearl: I understand the problem in the general direction, but can you be a bit more specific from the strategic front? Doctors have a massive strategic opportunity to shift how medicine is practiced.

33:53Jeremy Corr: Using generative AI, they can move medical expertise into patients' homes. They can make medical care continuous rather than episodic. As we've discussed larger models, Chachabit, Cod, and Gemini can provide patients with immediate clinical guidance and help them achieve effective control of their chronic diseases. But strategy is far more than great opportunities. It requires translating strategy into action. Without that, even the most powerful technology becomes relatively powerless. As we said, today's physicians and healthcare organizations are using Gen.AI almost exclusively to improve documentation, capture billing codes, and reduce administrative work.

34:36Jeremy Corr: Those applications, they can increase revenue, but they will not transform care. The greatest opportunity is to use this readily available technology to improve clinical outcomes and make medical care affordable by keeping people healthy and preventing life-threatening complications. Less administrative work, that's positive. It's a tactic that busy clinicians can use. But to solve the massive access and cost challenges of the nation, like a World Cup team, medicine will need a winning and well-listed strategy for success. And today, that remains elusive.

35:17Dr. Robert Pearl: What's the second problem that you have observed in the tournament and in medicine?

35:21Jeremy Corr: Both on the pitch and in day-to-day medical practice, individual excellence cannot compensate for poor teamwork. As an example, France arrived at the 2026 World Cup with extraordinary individual talent. In the 70s, Spain defeated them with a more cohesive and collaborative approach. Spain exemplified team excellence. Its players dominated time of possession through coordination and cooperation. Throughout the World Cup, Spain achieved world-setting, record-setting passing efficiency. And by maintaining possession, it limited their opponents' opportunities to score. In eight matches, Spain allowed a total of one goal.

36:05Jeremy Corr: U.S. Medicine, by contrast, it celebrates individual excellence and it organizes care as a series of separate encounters. Multiple specialists offer orders for hospitalized patients, yet communication among those physicians is rare and frequently fragmented. And this is a recognized threat to care coordination and patient safety. The traditional office-based model, it was adequate for the types of acute problems that physicians treated. in the last century. But today, three-fourths of American adults have at least one chronic disease and over half have two or more. Managing illnesses such as diabetes, hypertension, and heart disease requires continuous monitoring, optimal medication adjustments, and coordination among clinicians, not a series of isolated appointments with individual doctors who don't share a common electronic health record and fail to work as one.

37:02Jeremy Corr: Teams are essential to achieve the best clinical outcomes. Without effective coordination, patients fall through the cracks. Chronic diseases remain uncontrolled. Preventable complications, including the heart attacks, strokes, and kidney failures you mentioned earlier, they ensue. Doctors value autonomy and individual excellence over group performance, but as medicine has grown more complex, no individual physician, regardless of skill, can match the capabilities of a well-coordinated team, whether we're talking about longevity or maximize the quality of life. To improve current clinical outcomes, doctors will need to form highly effective medical groups, need to share information in real time, and they'll need to accept collective responsibility for clinical outcomes.

37:51Jeremy Corr: Like Spain, they must function as one team rather than a roster of individual superstars. What's the third problem you've observed in both? Leaders, whether in soccer or medicine, they must avoid becoming overly conservative and letting someone else take control of their team as their organization's destiny. The coach of England, he failed to do this. His team let Argentina win nothing early in the second half. But rather than continuing to attack England, they became increasingly defensive. Suddenly, Argentina could find openings, could create opportunities, and scored twice to advance to the finals.

38:28Jeremy Corr: American physicians have made this similar retreat. For decades, doctors helped shape the structure and direction of healthcare. Today, insurers determine which treatments require authorization. Hospitals and private system administrators establish productivity expectations and referral requirements. Private equity firms increasingly influence how medical practices operate. Physicians complain about these decisions and their loss of control, but they've not organized at sufficient scale to present a worthy counterattack. Retreating further into employment and joining hospitals or insurance companies that may be at least initially positive from an income perspective, but doing so will not restore professional autonomy.

39:14Jeremy Corr: Regaining that autonomy, the authority will, however, require more than signing agreements to form a medical group. Physicians must also designate and empower skilled clinician leaders need to reach consensus on evidence-based practices and accept financial accountability for capitation. Under capitation, doctors would receive a fixed payment to care for a group of patients rather than being paid on the number of visits, tests, and procedures they do. This model would require them to accept financial risk when care costs exceed that amount, but also reward them when they achieve superior outcomes and lower expenses for patients for prevention, collaboration, and excellent chronic disease control.

40:01Jeremy Corr: This more aggressive approach, I think it offers the best path for preserving the profession's longstanding values while providing patients with clinical excellence, timely access, and greater affordability.

40:14Dr. Robert Pearl: Robbie, when it comes to medicine, what do you see for the future?

40:18Jeremy Corr: Jeremy, the United States has all the clinical talent and the resources required to succeed. What its medical system lacks is a clear strategy, coordinated teamwork, and effective clinician leadership. Like the 47 World Cup teams that fell short, American medicine must improve. To win for their patients and themselves, clinicians will have to come together. You'll have to agree on a strategic plan, embrace new technology. You have to be willing to empower leaders and accept responsibility for the medical system's performance. As the teams who lost learned, massive opportunities don't come along very often.

40:59Jeremy Corr: And if you fail to seize them, when they do come along, you inevitably regret it.

41:05Dr. Robert Pearl: We hope you enjoyed this podcast and will tell your friends and colleagues about it. Please follow Fixing Healthcare on Apple Podcasts, Spotify, or your favorite podcast app. If you like the show, please leave it at five stars and leave a review. Visit our website at FixingHealthcarePodcast.com and follow us on LinkedIn, Facebook, and Twitter at FixingHCPodcast. Thank you for listening and have a great day.