Beta
Podcast cover art for: When doctors don't know what's wrong
Unexplainable
Vox Media Podcast Network·15/07/2026

When doctors don't know what's wrong

This is a episode from podcasts.apple.com.
To find out more about the podcast go to When doctors don't know what's wrong.

Below is a short summary and detailed review of this podcast written by FutureFactual:

Diagnosing the Unseen: Louise Proctor, ACDC, and the Undiagnosed Diseases Network

Overview

This episode examines why diagnostic errors occur and how a pioneering NIH program tackles puzzling medical cases. Through the story of Louise Proctor and her family, the podcast highlights the human cost of long, uncertain journeys to diagnosis and what can change when doctors, researchers, and patients collaborate more closely.

Key insights

  • Diagnostic errors are common and often driven by time pressures and fragmented care.
  • The Undiagnosed Diseases Network provides a structured, multi-disciplinary approach to tackle hard cases.
  • Louise Proctor’s family finally receiving a diagnosis reveals both the relief of having a name and the limits of current treatment options.
  • Lessons from this model could inform care for more common conditions, emphasizing patient-centered, time-rich collaboration.

Introduction and framing

The podcast opens by exploring wellness culture and the cost of being well, segueing into the persistent problem of diagnostic error in health care. It centers on Alexandra Sifferlin, deputy science editor at the New York Times, whose investigative work on diagnostic errors became the backdrop for a book about doctors, patients, and the diagnosis crisis. The episode uses one deeply human case to illuminate systemic issues and potential improvements.

The story of Louise Proctor

Louise Proctor, the eldest of five siblings, grew up in rural Kentucky and experienced a sudden, severe pain that felt like her legs turning to stone. Over decades, pain and mobility issues persisted, while a cascade of “could be” explanations—ranging from mere lack of fitness to early arthritis—failed to help. Louise and her family faced disbelief, especially since Louise appeared healthy and needed accommodations (like wheelchairs) in ordinary settings. The absence of a clear diagnosis extended into many years during which Louise also raised two children.

Meanwhile, Louise’s daughter Suzanne was diagnosed with cystic fibrosis in 1984, a fatal disease at the time, which intensified the emotional weight of uncertain health for Louise and her family. Suzanne’s eventual death highlighted, paradoxically, the power and necessity of a diagnosis for community understanding and support, even when the prognosis was grim. Louise’s frustration culminated in a breakdown in her physician’s office, after which imaging finally revealed abnormal calcium buildup in her leg vessels—an extraordinary finding in someone so young and ostensibly healthy.

The discovery connected Louise’s condition to a broader familial pattern. Her siblings also reported similar leg pain and vascular issues, suggesting a possible genetic basis. The medical mystery prompted Louise’s case to be considered by a national, collaborative group at the National Institutes of Health (NIH). This set the stage for what would become a turning point in her medical journey and in the understanding of rare vascular diseases.

The Undiagnosed Diseases Network (UDN) and a team-based approach

In 2009 Louise’s case was reviewed by the NIH Undiagnosed Diseases Program, which later evolved into the Undiagnosed Diseases Network. The program is designed to tackle the most complicated medical mysteries by gathering experts from multiple disciplines and meticulously reviewing medical histories before patient visits and after test results. The goal is to avoid the handoff frustrations that patients often face when moving from one specialist to another. Instead, clinicians across fields convene to discuss the patient as a unified team, sometimes identifying connections that would not be apparent within a single specialty.

Louise and her sister Paula eventually joined the NIH process, undergoing an array of scans and genetic tests. The NIH team, which included a postdoctoral researcher whose PhD work on a specific gene revealed a phenotype mirroring Louise’s pattern of arterial calcification, connected Louise’s condition to a gene deficiency: CD73. This coincidence—an academic project informing a clinical diagnosis—illustrates the collaborative potential of the network, where cross-pollination across fields can illuminate rare diseases with wider implications for arterial disease more generally.

What diagnosis meant for Louise and her family

After about 30 years, Louise and her siblings finally received a formal diagnosis: arterial calcification due to deficiency of CD73, known as ACDC. The family described a sense of relief at having a concrete name, which allowed them to articulate symptoms more clearly to others and to plan accommodations and future care. The discovery also carried practical implications: it clarified that their children were unlikely to inherit the condition, offering some relief about genetic risk for the next generation. Yet the story also underscored a crucial reality: a diagnosis is not a cure. While Louise and her siblings could contextualize their experiences and pursue targeted research, effective treatments were not yet available, highlighting the ongoing gap between diagnosis and relief.

Systemic lessons and broader implications

The podcast uses Louise’s case to discuss broader themes in diagnostic medicine. It emphasizes that many diagnostic errors are not simply about missing a disease on a scan; they are rooted in systemic issues such as time constraints, fragmented care, and limited feedback loops for clinicians to learn from mistakes. In the NIH UD Network, time was redistributed: clinicians could spend more in initial consultations and follow-ups, and case reviews involved a coordinated, cross-disciplinary team that could synthesize complex data from diverse sources. The approach challenges typical hospital workflows that push clinicians to see many patients in a short period while juggling electronic health record demands. The episode suggests that adopting UD Network-inspired models in more clinics could improve diagnostic accuracy, even for non-rare diseases like long COVID or other complex conditions, where a few extra hours with a patient and coordinated input from multiple experts can alter care trajectories.

Beyond the specific case: implications for medicine

Alexandra Sifferlin notes that diagnostic error is a thorny, systemic problem tied to many factors beyond individual clinician performance. The UD Network demonstrates that sharing data, fostering collaboration, and maintaining a long-term engagement with patients can yield valuable results. The conversation also touches on the ethical and social benefits of giving patients a named condition: it improves communication with families, informs daily decision-making (for example about how to manage mobility or accommodations), and anchors research that could translate into future treatments. The host and guests discuss how the Undiagnosed Diseases Network ethos—staying with patients over time and pursuing deeply collaborative inquiry—could be adapted to more common medical scenarios. The episode ends by noting that even with a diagnosis, medicine remains uncertain, and the human demand for answers remains strong. Alexandra Sifferlin’s work, including her forthcoming book, is cited as a resource on the diagnosis crisis and the broader topic of how we approach health and disease with empathy, time, and rigor.

Conclusion

The podcast closes with reflections on the value of certainty in medicine: even when certainty is not immediately accompanied by a cure, a name and a plan can transform patients’ lives by offering explanations, social understanding, and a route toward better care. The episode invites listeners to consider how health systems can balance the need for thorough, time-intensive evaluation with the realities of modern medicine, and whether Undiagnosed Diseases Network-like models could be scaled to improve care for more patients facing difficult diagnoses.