Categories

Tag: Stephen Farber

Big Bucks, No Whammies, Part 2: So You Won. Now What?

By STEPHEN FARBER

Part one argued that patient access was never the finish line; it was the foundation. This is about what we build now that we are standing on it.

Every good game show has a second round, and the second round is always harder. The first round was about getting health information out of the buildings that held it. Regulation, standards, and a decade of unglamorous plumbing have brought us to an important threshold. Individuals can increasingly pull records from multiple institutions into one place. The board paid out. The audience cheered.

Then a quieter and stranger question arrives. You are holding years of visits, laboratory results, medications, and discharge summaries, and you realize that having your history in one place is not the same as understanding it. Assembly was the first hard problem, but it was not the last.

The record is finally readable

This is where artificial intelligence walks onto the stage, and the promise is real. Few clinicians have time to read a patient’s entire history every time a decision must be made. Appropriately safeguarded AI can summarize years of encounters, support medication reconciliation, explain a laboratory trend in ordinary language, and surface a pattern buried across multiple providers. Large consumer platforms have already begun connecting AI assistants to medical records; Amazon’s Health AI, for example, draws on a person’s One Medical history to provide personalized explanations and guidance. Once the record can be assembled, a capable assistant can finally make more of it legible.

Legibility, however, is not reliability. The farther AI moves from explaining a record toward acting on it, the thinner the evidence becomes. A 2026 scoping review of agentic AI in healthcare found that most of the available studies were exploratory and lacked robust clinical validation, and that only one trial involved patients. An assistant working from an incomplete, duplicated, or subtly wrong record can produce a fluent and plausible conclusion that is still wrong. AI does not make data quality less important. It makes the consequences of poor data quality harder to ignore.

AI cannot read what was never written

There is a more fundamental limitation, and it has less to do with the model than with the record itself. A longitudinal record can tell us what happened, but it rarely tells us enough about why. It preserves the encounter, the diagnosis, the result, and the prescription while losing much of the reasoning that surrounded them. Why was one treatment chosen over another? What was the person hoping to preserve at that point in life? Who else was helping? What changed after the visit, and what did a family member notice before anyone thought to schedule another one?

Continue reading…

Big Bucks, No Whammies: Patient Access Was Never the Finish Line

By STEPHEN FARBER

If you grew up in the 1980s, you probably remember Press Your Luck. Contestants stood in front of a flashing game board chanting, “Big bucks… no Whammies… STOP!” hoping the next square held cash instead of the mischievous cartoon character that erased everything they had won.

For years, assembling your own medical history felt remarkably similar.

You knew you had been treated at one hospital years ago, had imaging performed somewhere else, saw specialists who later retired, and changed insurance more than once along the way. One patient portal showed part of the story. Another required a password you hadn’t used in years. A medical records department offered to mail you a PDF after you completed a release form. Somewhere there was still a CD that no computer in your house could read. None of this meant your information had disappeared. It simply meant that putting it together depended almost as much on persistence as process.

That has always been one of healthcare’s great ironies. The industry became exceptionally good at documenting encounters. Every office visit, laboratory result, prescription, imaging study, discharge summary, and insurance claim was carefully preserved somewhere. What it never became particularly good at was helping individuals assemble those encounters into a coherent picture of their health over time.

For years, most of us accepted that as an unavoidable consequence of a fragmented healthcare system. In reality, it reflected a combination of technical limitations, business incentives, and regulatory uncertainty that made sharing information far more difficult than creating it. Patients often became the courier between organizations because there were few practical alternatives.

Quietly, that has begun to change.

While much of the industry’s attention has shifted toward artificial intelligence, another transformation has been taking place beneath the surface. The 21st Century Cures Act, Information Blocking regulations, standardized FHIR APIs, CMS interoperability requirements, and the continued evolution of TEFCA have collectively changed the trajectory of patient access. None of those developments solved the problem on their own, but together they have created an environment in which individuals can retrieve far more of their own health information electronically than was practical only a few years ago.

Although we’re still in the early stages, the technology has matured to the point where individuals can increasingly assemble and steward their own longitudinal health record. That changes the conversation. For much of the past decade, patient access was largely a policy discussion centered on whether people should have meaningful electronic access to their own information. Increasingly, the more interesting question is what becomes possible once they do.

As the market has evolved, complementary approaches have emerged. Some companies focus on helping individuals retrieve records directly from the organizations that hold them. Fasten Health is an interesting example because it begins with a simple premise: individuals should be able to assemble a record they control. Other organizations focus on discovering where records exist through exchange networks, while infrastructure companies simplify connectivity so developers can build applications without creating thousands of individual integrations. These approaches solve different problems, but they increasingly reinforce one another instead of competing.

That convergence has quietly moved the industry across an important threshold. For years, success was measured by our ability to collect fragmented data. Standards had to mature, regulations had to evolve, organizations had to expose information electronically, and software developers had to build practical ways of retrieving it. Much of that work occurred outside public view, but together it has made something increasingly realistic that once felt aspirational.

Bringing information together is a significant achievement because it gives people, often for the first time, a more complete picture of their interactions with the healthcare system over many years. A longitudinal health record tells us where someone received care, what diagnoses were made, which medications were prescribed, and what procedures were performed. Those are essential building blocks, but they rarely explain why decisions were made, what alternatives were considered, who participated in those conversations, or what mattered most to the individual at that point in life. Records preserve information extraordinarily well. Context has always been more difficult to preserve.

Healthcare has good reasons for operating around encounters because that is how care is delivered, documented, and reimbursed. People, however, experience their lives as a continuous story in which one decision influences the next, often over decades.

Continue reading…