OnCo
ideasIdea

Point-of-care randomisation built into the oncology record

When two accepted treatments are equally reasonable, the computer system would offer to randomise the choice and track the result, turning ordinary care into a continuous trial.

Learning health systems embed randomisation into workflow where genuine equipoise exists (dose schedules, supportive care, sequencing). The US Veterans Affairs system and the NHS have run point-of-care trials; oncology has few. The proposal builds a reusable EHR module (consent, randomisation, outcome capture via registry linkage) so that any cancer network can run a low-cost comparative effectiveness trial in weeks.

Hypothesis
Embedded randomisation will enrol more than ten times as many patients per year as conventional comparative effectiveness trials at less than a tenth of the per-patient cost.
Rationale
Registry-based randomised trials in cardiology (TASTE, SWEDEHEART) enrolled thousands at roughly 50 dollars per patient; the design transfers to questions with routine outcomes.
What would test it
Run one embedded trial (for example, two accepted anti-emetic regimens or two follow-up imaging schedules) in one network; measure enrolment rate, cost and completeness of outcome capture.
Maturity
early clinical
Who has to act
clinic
Cost to try
Medium ($1M to $50M)
Years to first evidence
3
Bottlenecks it attacks
  • Data silos · Records, scans, genomes and outcomes sit in separate systems that cannot talk. Every patient's experience is lost to the next.
  • Trial design, endpoints and cost · A phase 3 trial takes years and hundreds of millions of dollars, and often answers a question that has already moved on.
  • Trials enrol too few, too slowly · Fewer than one in ten adults with cancer joins a trial. Trials close for lack of patients, not lack of ideas.

Connected

5top