How to Write a Medical Chronology (With an Example)

Nothing here is legal advice, and the example below is illustrative rather than a real file. What follows is the working method: the order the steps go in, the fields each row needs, and the four or five judgment calls that separate a chronology an attorney can rely on from a list of dates nobody trusts.

Start by proving the set is complete

The most expensive mistake in this work is reading first. Records arrive in waves, from different custodians, with different production formats, and the set you have on day one is almost never the set you will argue from. So before anyone opens page 1, reconcile what came back against what you asked for: every authorization you sent, every provider named at intake, every facility referenced inside the records you already have. Log the Bates range from each provider on a single sheet.

That sheet does two jobs. It tells you when to start, and it becomes the "known gaps" section of the finished cover page, which is the part defense counsel reads first. A chronology built on a set that turns out to be missing eight months of orthopedic follow up does not get amended. It gets rebuilt.

The seven steps, in order

The order matters more than it looks. Steps 1 through 3 are mechanical and cheap. Steps 4 through 6 are the expensive part, and every hour you save by doing the cheap steps properly is an hour you do not spend re-reading.

1. 1. Confirm the record set is complete. Reconcile pages received against authorizations sent. Note the Bates range per provider and the date range each set covers.
2. 2. Make every page searchable. Scanned and faxed pages have no text layer until you OCR them. Do this before you split anything, and confirm the page numbers in your file line up with the stamps you are going to cite.
3. 3. Split by provider, then sort by date. One 3,000 page production is usually 30 or 40 separate documents from a dozen providers, concatenated in whatever order the copy service felt like. Split first, then interleave everything into one date ordered stream.
4. 4. Extract one row per encounter. Not one row per page, and not one row per provider. The unit is the clinical encounter.
5. 5. Flag the entries that decide the case. First report of each complaint, treatment gaps, prior injuries, imaging findings, anything touching causation.
6. 6. Verify every citation. Open the cited page. Confirm the row says what the record says.
7. 7. Write the cover summary last. Case type, date range, providers, total pages reviewed, known gaps. One page.

What every entry has to contain

Six fields, and the citation is not optional. A row without a page number is an assertion, and an assertion is exactly what opposing counsel will ask you to support in a deposition.

What does a medical chronology look like?

Like this. The example is a rear end collision with a soft tissue presentation that later develops a lumbar finding, which is the most common shape a personal injury record set takes. Names, dates and Bates numbers are invented for illustration.

Eight rows carry a case that lives in several hundred pages. The gap row is doing as much work as any treatment row, which is the point of the next section.

How do you flag a gap in treatment?

Put the gap in the chronology as its own row, with the number of days and an explicit note that you have not yet established the reason. Adjusters and defense counsel treat any break longer than about 30 days as an argument that the plaintiff recovered, so a gap you did not surface is a gap that gets used against you at the worst moment. Surfacing it yourself turns it into a task: call the client, check for outstanding authorizations, document the reason.

Most gaps have ordinary explanations. Insurance lapsed, a referral took eleven weeks, the client was caring for a parent, the provider never produced records in the first place. The distinction that matters is between a gap in treatment and a gap in the production, and you cannot tell those apart from the pages alone. That is why step 1 exists.

Do you include every page of the medical records?

You read every page, and you summarize the ones with clinical content. A produced set is mostly billing ledgers, insurance correspondence, duplicate faxes, blank consent forms and repeated demographic sheets. Those do not get rows. What always gets a row: every encounter with a clinician, every imaging study with its stated findings, every medication change, every work restriction, and every note that mentions a prior or unrelated injury.

The exception is duplicates, and they deserve a rule of their own. The same ED note will often appear four times because four providers requested it. Deduplicate before you extract, or you will produce a chronology with the same encounter listed on four different Bates ranges, which reads as carelessness even though it is really a production artifact.

How do you handle handwritten and illegible records?

Slow down and mark what you cannot read. Handwritten notes cut review speed by more than half, from roughly 50 pages per hour on clean typed records to around 20 on handwritten ones, and that is the single biggest driver of what a chronology costs. When a word is genuinely illegible, write "illegible" in the row rather than guessing, and cite the page anyway so someone else can try.

Two practical moves help. Pull the corresponding billing record, because CPT and diagnosis codes are typed and will often tell you what an unreadable note was about. And request a typed transcription from the provider for the handful of pages that actually carry the case. Extraction tools have improved a great deal on typed and scanned text, but handwriting remains the place where automated output needs the most human checking. The same document abstraction problem shows up well outside litigation: commercial real estate teams spend the same hours pulling the key dates and obligations out of long lease documents, and the tooling in both fields has converged on the same answer, which is machine extraction with a human verifying the citations.

Should a medical chronology include your opinions?

No. The chronology states what the record says and cites where it says it. The moment a row contains an inference, the whole document becomes arguable, and its value comes from being the one artifact in the file that both sides can check. Keep interpretation in a separate memo or in the medical summary, which is the narrative document that explains mechanism of injury, course of treatment and prognosis in prose.

Flags are the allowed exception, because a flag is a pointer, not a conclusion. "First report of low back pain" is a fact about the record. "Low back pain was caused by the collision" is an opinion, and it belongs in the demand letter where it can be argued.

How long should a medical chronology be?

As long as the encounters require and no longer, which in practice is one page of chronology for roughly every 100 to 150 pages of records, plus the one page cover summary. A 2,000 page file usually lands somewhere between 14 and 20 pages. If yours is running much longer, you are almost certainly writing rows for pages instead of encounters, or copying note text wholesale rather than summarizing it.

Where the hours actually go

At roughly 50 pages per hour, a 2,000 page file is 40 hours of reading before anyone drafts a demand. That is five full working days for one case, and it is why chronology work is the first thing most personal injury firms try to move off a paralegal's desk. Reported preparation time for an average chronology runs about 8 to 10 hours and past 20 on complex files, which is consistent with those page rates once you account for the small files that make up most of a docket.

The arithmetic is worth running on your own numbers before you buy anything, because the answer changes completely between a firm closing 20 files a year and one closing 200. Our guide to medical chronology software compares 2026 per page, per seat and per chronology pricing and includes a calculator that takes your page count and hourly rate and returns the hours, the working days and the cost. For how the rest of the file fits around it, see personal injury case management, and if your matter is heading into discovery volumes rather than record volumes, legal document review software is priced by the gigabyte and is a different category of tool entirely.

One organizational note that has nothing to do with software. Six paralegals will produce six formats unless somebody writes the standard down: the field list, the flag vocabulary, the citation format, and what counts as an encounter. Firms that skip that step buy a tool, get inconsistent output anyway, and blame the tool. Store the standard where the records live, which for most firms means their legal document management system rather than somebody's desktop.