How Primary Care is Changing

Primary care is the hardest job in medicine to do well. And it’s the least acknowledged for being hard. Mic drop. 🎤 I say that as an ER doc — and yes, we do chaos for a living — but primary care is a different kind of pain: Relentless cognitive load and relentless competing expectations from everyone in the healthcare system, plus a customer service overlay and nowadays, an inbox that never sleeps. Last year I did a full week on the Acquired Epic episode because you can’t really understand U.S. healthcare without understanding Epic. This week is the same idea: you can’t understand healthcare without understanding primary care. If you want to understand why medicine feels impossible right now, start with Lisa Rosenbaum’s NEJM podcast series (link below). It is simply outstanding. Here’s what we quietly expect PCPs to be: 𝗙𝗼𝗿 𝗽𝗮𝘁𝗶𝗲𝗻𝘁𝘀: — Diagnose everything and miss nothing in 15 minutes — Refill meds, explain labs, answer portal messages instantly — Be lifelong continuity + therapist + navigator 𝗙𝗼𝗿 𝘀𝗽𝗲𝗰𝗶𝗮𝗹𝗶𝘀𝘁𝘀: — Know when a referral is warranted — Implement consult plans — Manage the “everything else” forever 𝗙𝗼𝗿 𝘁𝗵𝗲 𝘀𝘆𝘀𝘁𝗲𝗺 / 𝗽𝗮𝘆𝗲𝗿𝘀: — Close care gaps, hit quality metrics, keep satisfaction scores high — Do prior auth paperwork — Coordinate, supervise, and oversee a small village: pharmacy, social work, dietitians, care managers, PT. That’s not a doctor. That’s an air traffic controller for human lives. And most PCPs are expected to do that (plus way more, I had to chop down my list for character limit) for 2,000 human lives at a time. And then we act surprised primary care is collapsing while spending on it stays tiny — Lisa notes it's ~5% of health spending, despite >35% of visits running through primary care. As goes primary care, so goes the rest of medicine. Cannot recommend the podcast and Lisa's writings in the NEJM enough. Podcast: https://lnkd.in/gUxrG55C

Mark Cuban I’ve been following your posts about transparency and how distorted incentives quietly reshape healthcare. Here’s the math that shows exactly how those incentives push physicians out of independent practice and into hospital consolidation. For a Level 3 office visit, hospital-owned clinics get paid $155.1 for new patients and $108.4 for established patients. Independent practices doing the exact same work are paid $129.1 and $91.4. Nothing about the care is different. But the paycheck is—by $26 per new patient and $17 per established patient. I see 24 patients a day. Those price differences add up to over $100,000 a year that independent physicians don’t receive simply because we refuse to sell our practice to a hospital system. This is the quiet force driving consolidation. The message is clear: stay independent and get financially punished, or join a hospital system where you are able to make a living. And once consolidation happens, everything changes. Independent physicians promote health, prevention, and long-term patient relationships. Hospital-owned physicians—through no fault of their own—are pulled into a system that needs to promote profit centers, facility fees, imaging volumes, specialty referrals, and downstream revenue to justify the acquisition. The incentives shift from doing what’s best for the patient to doing what’s best for the system. And patients pay more. Employers pay more. The entire market gets less competitive. And the core mission of primary care gets diluted into a revenue pipeline. This is exactly the type of misaligned incentive you call out: the system rewards power, not value. Independent primary care is one of the most cost-effective assets in healthcare, yet the market structure pushes it toward extinction. Mark, your voice matters here. If we want a healthcare system built around health—not hospital profits—we have to expose the pricing distortions that push doctors into consolidation. If you want to see what real, transparent, patient-first primary care looks like outside the hospital machine, my door is open. — Adam McCall, MD

AI scribing and intake are the warm-up act.  The real shift is coming next: the AI Consult. What is the AI Consult? It is an end-to-end clinical encounter, conducted autonomously by an AI system, before a patient sees a clinician. This isn’t just data gathering. An AI Consult is a clinical-grade visit - including history, preliminary assessment, and initial plan - between a patient and an AI system. The output becomes a progress note the clinician reviews before the visit, then uses to ask clarifying questions and finalize the plan. Think of it another way - it’s like the best Chief Resident you ever worked with prepping every case. Why is the AI Consult transformative? Because overnight we will take 60-70% of the visit time that used to be spent in synchronous data gathering and chart review and free that time up for more discussion on the major issues and the plan. And the benefits are huge and immediate. Visit quality goes up because doctors and patients are now spending most of their time focused on making decisions, not just exchanging data. And efficiency improves because what used to be a 30 minute visit comprised of mostly low-quality interaction is now a 15 minute visit comprised of almost all high-quality interaction. Patients feel like their doctor finally has the time and space to pay attention to them. Because they do. For practices, the implications are straightforward and dramatic: throughput doubles while operating expenses only increase by ~20%. Overnight, primary care turns from a loss leader to a profit center, and major capacity issues get resolved. For patients, there will be more opportunity to share a full set of concerns before every visit - so conversations with doctors are richer and more informed. And for clinicians, we will have more bandwidth than ever to focus on our patients, not the chart. It would be naive to think that such a profound shift would happen quickly, but it will be impossible to ignore - and it will come faster than anyone is expecting. We already see the positive impact every day at Doctronic where the AI Consult is a routine part of our care delivery model. If the visit starts before the visit, what changes next in the system around it?

A patient handed me a printed summary of her symptoms this week. Organized chronologically. Differential diagnoses ranked by likelihood. She had used ChatGPT to prepare for the visit. She was not challenging my training. She was making sense of months of uncertainty using the most accessible tool she had. This is happening more often now. Over 40 million people ask ChatGPT health questions daily. When patients arrive with AI-generated answers instead of questions, the encounter changes. The conversation shifts from exploration to negotiation. The clinical challenge is not disagreement. It is slowing down. Creating space to examine what the AI got right, what it missed, and what the patient is worried about underneath the language. AI gives patients a framework. It does not give them context. In primary care, context is the diagnosis. The patients relying most on free AI tools between visits are the ones with the least access to follow-up care. That makes this an equity pattern, not just a technology pattern. I started asking a new question: "Before we start, did you look anything up that shaped what you are expecting today?" The answer tells me what they believe and what they need. I wrote more about this anchoring pattern and how it is showing up in primary care and emergency settings in this week's MedTechxHeart. How often are your patients arriving with AI-generated information? Has it changed how you start visits? Read the full analysis: https://lnkd.in/gDdAPucT If this reflects what you are seeing, share it with a colleague navigating the same shift. #DigitalHealth #HealthEquity #PrimaryCare #AIinPrimaryCare

There is a word for when the packaging stays the same but there is less inside. In groceries we call it shrinkflation. In family medicine, the same thing has been happening for years and we do not have a proper name for it. So I will use that one. When I trained as a family medicine resident, the expectation was clear. You delivered your own patients' babies. You sutured their lacerations, did their biopsies, inserted their IUDs, managed their mental health, saw their children, provided their prenatal care. You treated every condition as though there were no specialists available, because in Canada, there often are not. That was the point of family medicine. It was the foundation the system was built on. Since the pandemic, I have watched that foundation contract. Procedures referred out. Prenatal care moved elsewhere. Mental health reduced to a prescription. Children's visits, wellness checks, counselling, contraceptive management, all migrating out of the family practice setting and into referral queues, specialist offices, and emergency departments that were not designed to absorb them. I do not think family physicians decided en masse to de-skill because they got bored or lost interest. I think there is a structural problem that is not being discussed with the seriousness it deserves. The VDoFP benchmarking data makes the economics plain. Average clinic expenses run approximately $110,000 per physician per year. At 25% overhead on a $450,000 billing, a clinic generates $112,500. There is no room in that arithmetic for the additional time, equipment, and staff that comprehensive practice requires. So clinics on constrained margins shed services. One at a time. Rationally. And the load transfers to every other part of the system. Every laceration that goes to emergency. Every biopsy sitting in a specialist queue for months. Every mental health crisis with nowhere to land. Every diagnostic lag that turns weeks into months for conditions where timing matters. We say we have a primary care-based health system in Canada. But what happens to a building when the foundation develops cracks and stops bearing the load it was designed to carry? We are watching that happen. I have written a longer piece on what is driving it, what it is costing us, and what would actually help. Link in the comments. (Source: Financial Review of Primary Care Clinics in Vancouver, VDoFP, 2025) #PrimaryCare #FamilyMedicine #ComprehensiveCare #BCHealth #HealthPolicy #ClinicSustainability #LongitudinalMedicine

The era of preventive medicine has arrived. For decades, healthcare has been reactive. You went to the doctor when something was wrong. Symptoms were treated. Then life continued, until the next issue appeared. That model is breaking. ⛓️💥 We’re moving into a world where health is continuously measured, monitored, and improved not repaired after the fact. People don’t wait to get sick. They track biomarkers, do regular check-ups, and use data to actively optimize their health. The signals are everywhere: • One Medical (backed by Amazon) scaling membership-based preventive primary care • Function Health enabling 50M+ biomarker tests, raising ~$350M, and reaching ~$2.5B valuation to scale proactive lab monitoring • WHOOP and ŌURA Health turning continuous physiological monitoring into everyday behavior • Neko Health founded by Daniel Ek raised $260M, reached unicorn status, and delivers AI full-body preventive scans …. and many more! Globally, prevention- and biomarker-focused startups are emerging fast. ⚡️ People are doing proactive check-ups. They monitor sleep, recovery, and metabolic health. They track biomarkers. They try to improve, not just recover. This marks a fundamental shift: 👉 From episodic care → continuous health management 👉 From treating illness → optimizing longevity & performance 👉 From “sickcare” → real healthcare Diagnostics are getting cheaper. Wearables smarter. Consumers more engaged. And prevention is becoming the entry point into the health system, not the afterthought. Preventive medicine is no longer a vision. It’s already happening. 🚀

AI is rapidly entering primary care, offering real opportunities to reduce workload and expand access, but its deployment is outpacing robust evaluation, regulation, and safeguards for quality and equity. 1️⃣ Primary care is under global strain from workforce shortages, burnout, and rising complexity, creating strong pressure to adopt AI-enabled solutions. 2️⃣ AI is already being used to support clinicians with clinical knowledge queries, documentation, messaging, and care coordination, mainly through LLM-based tools. 3️⃣ AI scribes can reduce cognitive load and improve clinician satisfaction, but may degrade medical records, promote automation bias, and omit socially meaningful patient context. 4️⃣ Evidence for AI improving clinical outcomes in primary care remains limited, with few RCTs and mixed real-world results despite strong in silico performance. 5️⃣ Algorithms trained outside primary care settings often perform poorly when deployed in community practice, highlighting the importance of context-specific data and validation. 6️⃣ Patient-facing AI tools (eg, symptom checkers, chatbots) are increasingly used as first points of contact, but show variable diagnostic and triage accuracy with safety risks. 7️⃣ LLMs can generate patient-friendly summaries and support shared decision making, yet hallucinations and omissions require close clinician oversight. 8️⃣ AI has the potential to improve patient-centred care by incorporating PROs and PREMs, but these data are rarely used in current AI systems. 9️⃣ Bias remains a major concern, arising from unrepresentative datasets, algorithm design, and real-world implementation, with risks of widening health inequities. 🔟 AI has environmental costs due to high energy and water use, but could also support planetary health by reducing unnecessary care and strengthening prevention in primary care. Liliana Laranjo, Lorainne Tudor Car, Rebecca Elizabeth Payne, Ana Luísa Neves, Michael Kidd, J. Jaime Miranda. Artificial intelligence in primary care: innovation at a crossroads. The Lancet Group Primary Care. 2025. DOI: 10.1016/j.lanprc.2025.100078 ❗ Healthcare AI in 2025 - a reflection: https://lnkd.in/dgFCgBAC

We are trying to solve the primary care shortage with AI, but we are ignoring the absurdity of the rules we force human doctors to play by. The industry’s reflexive response to the fact that 100 million Americans lack a primary care provider is to rush toward automation. We are racing to build solutions that make each unit of supply—the physician—more productive. But applying AI to a fundamentally broken system is like putting a high-performance engine in a car with four flat tires. Before we automate doctors, we need to address the friction we have artificially introduced into the supply and demand equation. We are leaving the lowest-hanging fruit entirely untouched. First, we must expand the pipeline. We need to incentivize more people to enter the profession, build more medical schools, and unblock the artificial cap on residency slots that leaves thousands of qualified medical graduates unmatched every year. We have to make the job of being a doctor attractive again. Second, we must protect the supply we already have. Today, 43% of U.S. primary care physicians report feeling burned out, and more than a third plan to stop seeing patients. If we make their jobs more enjoyable and reduce the administrative burden that causes moral injury, we reduce churn and keep a larger, more experienced supply in the workforce. But the most glaring absurdity is how inefficiently we utilize the supply we do have. There are doctors across this country who are willing to work more hours. They sit idle with openings on their calendars, hoping a patient will show up. Yet, patients cannot access them because of geography, distance, affordability, and insurance networks. The structural friction points are staggering. Consider state licensing laws: a doctor located just across the state border, who went to medical school in your state and underwent the exact same rigorous training as your local physician, cannot legally treat you simply because they hold a license from a neighboring medical board. Furthermore, health plans frequently refuse to admit doctors into their networks unless those doctors maintain a physical brick-and-mortar location in the state or hold admitting privileges at a local hospital. How does this make sense? We know there is a severe shortage of up to 86,000 physicians projected by 2036. We know the profound return on investment that primary care delivers for both healthcare outcomes and financial savings. And we know that 80% to 90% of a primary care doctor’s job can be done virtually. Why, then, do we continue to enforce archaic rules that make it impossible for doctors to deliver that 80% to 90% of value to the patients who desperately need it? Automation is a powerful tool, but it is not a substitute for structural reform. Before we ask AI to save primary care, we should probably stop making it so difficult for human doctors to do their jobs.

Primary care was not built for isolated symptoms anymore. Not really. The patient does not walk in as one clean diagnosis. They walk in tired. Bloated. Foggy. Not sleeping. Reacting to foods they used to tolerate. Maybe their joints hurt. Maybe their mood changed. Maybe their labs look “normal,” but they know something is off. In the chart, those symptoms get separated quickly. GI. Endocrine. Rheumatology. Psychiatry. Neurology. But the body does not practice medicine by department. The gut talks to the immune system. The immune system talks to the brain. Inflammation changes energy. Microbial imbalance can change mood, cognition, pain, sleep, and tolerance. Primary care physicians see this every day. The problem is not that they are missing the patient. The problem is that the system often gives them fragments and asks them to make sense of the whole person in minutes. That is where functional medicine needs to mature. Not as an alternative universe. Not as a supplement list. Not as protocol medicine with different branding. As a practical framework for seeing connections earlier. Because many of the patients filling primary care schedules are not simple cases. They are system cases. The exhausted patient with bloating and brain fog. The anxious patient with reflux and poor sleep. The autoimmune patient whose symptoms flare after food reactions. The “normal labs” patient who keeps coming back because they still do not feel well. These are not separate stories. Very often, they are connected stories. And the earlier we see the connection, the better chance we have to change the trajectory. That is why I believe the future of primary care will need better integration between allopathic medicine and functional medicine. Allopathic medicine is essential. It identifies danger. It treats acute disease. It saves lives. But functional medicine can help answer a different question: Why is this patient’s system behaving this way in the first place? That question belongs closer to primary care. Not after years of referrals. Not after the patient has been told everything is fine. Not after the disease has become harder to reverse. Earlier. When the symptoms are still signals. When the story is still forming. When the body is still giving us a chance to intervene. The goal is not to make primary care more complicated. It is to give physicians better tools to see the complexity that is already there. That is the version of functional medicine worth building. For physicians: where do you see this most often — GI symptoms, fatigue, brain fog, autoimmune flares, or “normal labs” patients who still feel unwell? #PrimaryCare #FunctionalMedicine #ClinicalReasoning #SystemsMedicine #GutHealth #HealthcareInnovation

Much of primary care as we know it will be replaced by AI. I say this as a primary care physician with 25 years of global practice experience, as well one who's building in AI. I'm not catastrophizing. I'm making a prediction. 32% of Americans now use AI chatbots for health information. 64% of those do it weekly. Among millennials and Gen Z, usage is approaching 50%. These are the patients of tomorrow — they're not waiting. AI is in its infancy. It hallucinates. It gets things wrong. People still use it. Imperfections eventually get sorted out, an imperative for healthcare. Primary care was already on life support. Only 24.4% of U.S. physicians practice in primary care, well below the 50% benchmark. As of December 2025, 92 million Americans live in federally designated primary care shortage areas. We don't have enough primary care doctors, and we're not going to catch up. It's not like AI is displacing a thriving PCP workforce. It's filling an access gap patients have suffered for years. Procedures will migrate to specialists. Pap smears to gynecologists, lacerations to urgent care, skin biopsies to dermatologists. Human needs — anxiety, grief, chronic loneliness — will go to nurses, therapists, and social workers. Those who spend more than 15 minutes with you. Primary care was never about procedures. It was about the "relationship". That relationship was destroyed by the system long before AI arrived. Trust in doctors dropped from 71.5% in 2020 to 40.1% in 2024. Trust in one's own doctor fell from 93% to 85% in just two years. People don't distrust science. They distrust being dismissed. AI doesn't dismiss you. It doesn't judge you. It has infinite patience. It doesn't get up after 15 minutes. The healthcare AI companies being built today, including the ones called "chatbots", are the beginning of something massive. The global healthcare AI market was $32 billion in 2024. It's projected to reach $431 billion by 2032. Some of these companies will become the largest in the world. The ones that win will be built by clinicians. Technical brilliance without clinical insight produces falsely-impressive tools that give dubious or dangerous answers. Regulation is coming. Guardrails are coming. Malpractice frameworks are coming. That's not a bad thing, it's legitimacy. Primary care won't disappear. It will transform. The doctor who triages 30 patients in a morning, refills metformin, orders routine labs, and manages hypertension by protocol? AI will do that. Better. Faster. At all times. In any language. Doctors won't disappear, but their roles will change. They'll become diagnosticians. Guides. The necessary human presence. When auto-pilot fails, pilots are needed. The question isn't whether primary care will change. The question is who builds what people want, not what doctors and payers want. A new dawn approaches. I'm a primary care physician and the founder of ASTRID, a global AI health platform now in 81 countries.