Records Reviewed on a Calendar Are Always Late

Most systems I’ve worked inside have an official record that lags reality. In healthcare the lag can kill people. For two decades I treated that as a fact about institutions: paperwork is slow, committees are slower. It’s a fact about a design choice. The record gets reviewed on a calendar, and reality doesn’t happen on one.

A record is late by exactly its review interval. Reality moves on events: a trial reads out, a label changes, a patient’s phone number stops working, a scheduler goes on leave. The record moves when someone is scheduled to look. Everything between the event and the scheduled look is the gap, and the gap is the interval itself, picked by whoever set the cadence and paid for by whoever the record is about while it’s wrong.1

Structure decides what information can do

At nineteen I was in Roger Wartell’s lab at Georgia Tech, modeling the thermodynamic barriers that decide whether a gene-regulation reaction proceeds or stalls. A strand of RNA could carry the right nucleotides and still fail, if the folding produced a barrier that blocked the interaction. The sequence was correct. The shape was wrong. Structure decided what the information could actually do.

I co-wrote a book chapter on sRNA-mRNA interactions and Hfq, spent a summer on thoracic surgery outcomes at Sloan Kettering, and absorbed something I couldn’t say yet. Most consequential failures live in the distance between a system’s official design and its actual behavior. Who can act on information, under what constraints, with what review path, on what timeline: those decide whether the information matters at all.

Building for the event, not the schedule

After Georgia Tech I implemented EHR configurations at Epic and met the calendar’s opposite. Decision-support alerts built for patient safety didn’t wait for anyone to schedule a look. They fired on every event, with no triage between binding and advisory, and clinicians learned to dismiss all of them. A trigger with no threshold is as useless as a calendar with no trigger: one is always late, the other is always ringing. The official record said “alert delivered,” which was true, and the delivery had been engineered into noise. Waiting is the arbitrage: the institution collects the value of your time for the whole time the record says nothing is happening.

At Doximity I was early product lead for Dialer, which let doctors call patients from personal devices while the patient’s phone showed the office number. The technical problems were carrier spoofing and HIPAA. The identity signal also had to work for a doctor running between rooms and a patient wary of spam. A missed connection could mean a missed diagnosis. Whatever we built had to survive the worst day, because the worst day is an event and it doesn’t consult the release schedule. Doximity has since described Dialer as enabling over 110 million video and audio visits from more than 300,000 active clinicians.

In oncology navigation serving 30 million annual visitors, and in care coordination across surgery, urgent care, behavioral health and oncology, fragmented guidance left people to reconcile the steps themselves. Metrics counted app opens when we needed to know whether someone had finished a care pathway. The dashboard was reviewed monthly. A patient who fell out of a pathway fell out on a Tuesday.

Co-founding Andwise, a physician financial wellness company, brought the pattern to a new surface. Physicians dealing with contracts, debt, home buying and the transition out of training faced the same structural problem I had seen in clinical systems: fragmented guidance, trust-sensitive information, no review path when something went wrong. We raised $240,000, grew to 1,200 physician users, and built a medical advisory board. Our constraint map became the operating manual: which decisions were binding, which could be adjusted, which needed escalation, what the response window was for each. Nobody had to remember to check. The trigger fired on its own, and only when a defined thing had moved.

The interval is the design

Now the object is medical knowledge itself. NextConsensus produces source-traced evidence briefs for disputed healthcare claims. Coverage decisions, formulary positions and market-access claims depend on evidence that shifts continuously: new trials, safety signals, label updates, guideline revisions, payer actions. Most teams review episodically. By the time a committee convenes, the evidence has moved. Whatever position they’re confirming may already be stale, and it will stay stale until the next meeting, by design.

We monitor continuously. When evidence shifts, the brief surfaces what changed, what still holds, where the claim is vulnerable, what could change the conclusion, and whether the review threshold has been crossed. Each step traces to its sources and the limits of those sources are explicit, and re-review fires when the evidence moves.

What works is neither the calendar nor the untriaged trigger but a trigger with a threshold: a defined movement in the evidence that opens a review, and a defined size below which it does not.2

This holds for institutions in general, with or without software. Institutions keep three clocks, harm in weeks, review in years, memory in decades. They’re three calendars, and the interval between the first two is where the injured pay for the postponement.3 A diagnostic tripwire is a threshold trigger under another name: a cluster of exceptions that opens the rule for review whether or not anyone had a meeting on the books.4 An institution that reviews itself on a calendar has decided, in advance, how long it is willing to be wrong. The number is rarely written down. It is always the same as the interval.

If you work in payer, provider, pharma, policy, guideline or clinical review, I’d like to hear where the official record falls behind your work: what had moved, and how long the calendar kept the decision from moving with it. kanav@kanav.net.

Kanav Jain · kanav.net · Healthcare product leader. Previously: Doximity Dialer (110M+ visits), Andwise (co-founder/CEO), Transcarent, CTCA. Currently building NextConsensus, source-traced evidence briefs for disputed healthcare claims, and writing at The Crumple Zone.