This post explains our published research: A Two-Layer Deterministic Retrospective Computational Analysis: Dose Intensity, Decision Latency, and the Separation of Information Failure from Execution Failure (Rai & Jain, 2026, DOI: 10.5281/zenodo.22579527).
The question the study asked
When a cancer treatment goes worse than it should have, there are two very different explanations:
- An information failure — nobody could reasonably have known better at the time.
- An execution failure — the right move was known, even written down, but was not made, or was made too late.
These demand opposite remedies. Information failures call for better tests and research. Execution failures call for better process — checklists, triggers, accountability. Our study built a mathematical way to tell them apart, and applied it to the complete records of a young man with aggressive B-cell lymphoma.
What the numbers showed
- The information gap was 0.00 at all four major decision points. Not small — zero. Everything needed for the better decision was already documented in the patient’s own file when the decision was made.
- The decision gap at the very first treatment choice was the largest in the whole journey (+23.58 on the study’s scoring scale). The most expensive mistake was also the earliest.
- Three clinical recommendations written in the medical record were never executed, or were delayed by 47 to 145 days. The advice existed; the follow-through did not.
- Under 40,000 Monte-Carlo simulations, median survival was 27.0 months under the treatment actually given versus 43.4 months under the feasible optimal protocol — a difference of 16.4 months using only options that were realistically available.
How the simulation works, simply
The model ran the same patient through four parallel worlds: ACTUAL (what happened), FEASIBLE (best decisions using only drugs and timing genuinely available to this family), IDEAL (best decisions with everything on the market), and ORACLE (a theoretical ceiling). The gap that matters most is ACTUAL versus FEASIBLE — because that gap was choosable.
What this means for patients and families
The uncomfortable, empowering conclusion: in this case the system did not lack knowledge — it lacked execution. Which means families can help close the gap with process, today, for free:
- Ask for the escalation plan in writing: “if the next scan shows X, we do Y by date Z.”
- When a recommendation appears in a discharge summary, ask at the next visit: has this been done? Track it like a to-do list.
- Days matter. A 47-day delay is invisible while you live it and enormous in the survival mathematics.
Read the full open-access study: zenodo.org/records/22579527.