Clinical Registries
3 min read

Key Takeaways
NCDB abstraction follows a single cancer patient across months or years of care, spanning medical oncology, surgery, radiation oncology and pathology.
Oncology data specialists spend over an hour on straightforward cases and up to two and a half hours on complex ones, before quality assurance and follow-up are counted.
Commission on Cancer accreditation requires complete reporting, so a team under pressure falls behind instead of sampling less, stretching the reporting window further from the care it's meant to inform.
AI-guided abstraction assembles pathology, staging and treatment evidence across the full timeline, so registrars review a finished record instead of reconstructing one.
NCDB Tracks a Cancer Case for Years
The National Cancer Database, run jointly by the American College of Surgeons and the American Cancer Society, holds nearly 40 million patient records from more than 1,500 hospitals and underlies more than 1,500 published studies.
NCDB captures roughly 1.5 million newly diagnosed cancer cases each year, which represent about 72% of US cancer cases, through Commission on Cancer–accredited programs.
Most registries reconstruct a single encounter. But cancer cases span years, from diagnosis, through treatment and often lengthy follow-up care. NCDB abstraction has to track all of it.
This is the fifth post in a series looking at what registry-specific abstraction actually demands. For the bigger picture on why registries are more valuable than ever and where the strain is coming from, see Clinical Registries Are More Valuable Than Ever. The Work Behind Them Is Under Strain.
One Case Spans Months or Years of Care
NCDB requires registrars to build a single, longitudinal account of a patient's cancer, covering every phase of the disease from diagnosis forward:
Diagnosis
Staging
Pathology
First course of treatment
Follow-up
That account gets assembled across multiple services, each documenting in its own system and its own language:
Medical oncology
Surgery
Radiation oncology
Pathology
A registrat working a single case moves across encounters separated by months, following the sequence closely enough to determine what was actually intended as treatment.
Registrars spend more than an hour on straightforward cases and closer to two and a half hours on complex ones, before quality assurance, staging updates and follow-up are counted.
Staging and Pathology Coding Require Reading the Full Report
Coding a case for the NCDB Registry requires extensive interpretation:
Staging means reconciling clinical and pathologic evidence that may appear in several documents at different levels of specificity, then applying the current staging manual to the result
Histology and behavior codes come from the pathologist's own report, which means the registrar reads the narrative and interprets it
First course of treatment means determining clinical intent, which is documented inconsistently and sometimes not at all
Commission on Cancer Accreditation Requires Complete Reporting
Cancer registry reporting is required for Commission on Cancer accredited programs.
Other registries can sample when abstraction capacity is not available. A cancer program cannot. Every case has to be abstracted, so a team under pressure falls behind instead of covering less, and the backlog pushes the reporting window further from the care it's meant to inform.
Treatment Complexity Is Growing Faster Than the Registrar Workforce
Cancer incidence is rising, and treatment has grown more complex on nearly every front:
More lines of therapy
More targeted agents
More multi-specialty coordination
More clinical trial participation
Each of those adds documentation a registrar has to work through on every case. The credentialed registrar workforce isn't growing at the same pace, and registrar expertise takes years of practice to build. That gap won't close quickly.
AI Brings the Timeline Together for the Registrar
Layer Health reads the complete longitudinal record and surfaces the evidence for each data element, with its source, alongside a suggested answer. Pathology, staging documentation and treatment records across the full timeline show up together, so the registrar isn't reconstructing the sequence by hand.
The registrar still decides what counts as the intended treatment and what stage the case actually meets. The hours a registrar would otherwise spend moving between oncology, surgery, radiation and pathology systems to piece that sequence together are no longer part of the job.
Time Saved Returns Registrars to Cancer Program Work
Cancer registry data supports essential work including:
Quality measure tracking
Care gap identification
Trial matching
Registrars are the people best positioned to act on all three. When abstraction takes less of the week, that expertise goes back to the parts of the cancer program that need it most.

