Two Narrative Summaries, Same Records: Which One Holds Up

Two Narrative Summaries, Same Records: Which One Holds Up

Icon representing a calendar or date selection interface.
Published Date :

August 13, 2026

Icon representing a calendar or date selection interface.
Modified Date :

August 13, 2026

Home
>
Blog
>
>
Two Narrative Summaries, Same Records: Which One Holds Up

Give the same records to two summarizers and you can get two very different documents. Here is what separates the one that holds up:

  • One characterizes, one reports – Summary A calls the pain severe; Summary B says the record states the patient reported pain, rated 7 of 10. Only the second survives a challenge.
  • One is traceable, one asks you to trust it – Every line in the defensible summary points to a page or Bates number; the other reads well and cannot be checked without redoing the work.
  • One flags the gaps, one smooths them – A prior condition or a treatment gap is surfaced in the defensible version, not written around.
  • Both read fine on page one – The difference only appears when someone checks the document against the record, which is exactly what the other side does.

Read on for how the same file produces two summaries, and which one you want to be holding at deposition.

Hand the same 1,200-page file to two summarizers and ask each for a narrative summary. Both come back readable, both tell the treatment story in plain prose, and on a first read both look done. Then someone checks them against the record, and the two documents stop being equal. One holds up line by line. The other says more than the file does, and it becomes the exhibit the defense reads back to your witness.

A narrative summary is a readable, chronological account of the medical treatment documented in the records. That readability is its value and its trap: the same smooth prose that makes it useful can slide past the record without anyone noticing. So rather than list the abstract qualities of a defensible summary, it is worth watching what the same records do in two different hands. Call them Summary A and Summary B.

Summary A: the one that reads well and characterizes

Summary A is fluent. It describes a plaintiff with constant, disabling pain, a serious injury, and a clearly worsening course. It flows, an adjuster could read it in minutes, and it makes the case sound strong. The problem is under the sentences. The underlying note says the pain was worse on some days. No provider wrote disabling, and no record used clearly. Summary A did not lie, it characterized, reaching for words that make the prose persuasive and carry a judgment the record never made. On page one it is the more impressive document. Under a careful read it is the more dangerous one.

The version that holds up is the one that traces to the record
LezDo TechMed runs every narrative summary through a three-layer quality-control process and holds a published 99.8% accuracy rate, so the version you attach is the one that traces back to the record rather than past it.

Summary B: the one that cites the page

Summary B reads a little flatter, and that flatness is the point. Where Summary A wrote disabling pain, Summary B writes that the record states the patient reported pain, rated 7 out of 10, and cites the page. Where Summary A called the injury serious, Summary B reports the documented diagnosis in the provider's own words and leaves the weight to the reader. Every material statement points to a page or Bates number, so any line can be checked in seconds. The qualifiers the providers used, reports, denies, some days, are preserved on the facts that matter. Summary B is the document you can read into the record and defend, because each line resolves to the source. This is the difference source-linked narrative summaries make in case prep: a claim you can trace is a claim you can use.

What each summary does to your expert

The difference compounds when the summary reaches your expert. An expert who builds an opinion on Summary A inherits its overreach, so the opinion rests on a fact the record does not support, and a careful cross will find it. An expert working from Summary B builds on statements that each trace to the chart, so the foundation holds when it is tested. LezDo TechMed's medical narrative summary services are built to be Summary B: readable, but sourced and objective, so what your expert stands on is the record itself.

Want the narrative summary that reads well and still survives cross?

What happens to each one at deposition

Picture both summaries at a deposition. Opposing counsel takes Summary A, reads the line about disabling pain, hands your witness the underlying note that says worse on some days, and asks which is correct. The witness has to walk back the summary, and once one line is shown to overstate the record, every other line is suspect, including the ones that were accurate. Now take Summary B. Counsel challenges the line about pain, your witness turns to the cited page, and the line matches the record exactly. The challenge dies there, and the cross has to move on. Same records, same case, and the only difference is which document was built to be checked.

How to tell which one you are holding, before you rely on it

You can judge a narrative summary the way the other side will. Look for adjectives with no source, a claim that pain was severe or an injury serious when the cited page does not say so. Look for statements you cannot trace to a page. Look for a timeline that disagrees with the file, and for conclusions no provider wrote. A summary that fails these is a Summary A wearing good grammar, and judging the accuracy of a narrative summary before you attach it is cheaper than discovering the problem at deposition.

Two summaries from one file can read the same on page one. Only the one that traces to the record survives being checked against it.

quotes-icon

The line both summaries are supposed to hold

Here is the boundary that Summary B keeps and Summary A crosses. A narrative summary organizes and reports what the records document, sources each statement, and flags gaps and prior conditions. It does not decide whether the injury is serious, whether the pain is disabling, what caused a condition, or what the case is worth. Those are the treating physicians', the retained experts', and the attorneys' calls. The moment a summary reaches for them, it stops being evidence and starts being argument, and argument in the summarizer's voice is exactly what opposing counsel uses to call the whole document biased. Summary B does the persuading you need by staying neutral, so the persuasion you add on top rests on a record no one can accuse of leaning.

AI sits behind this too. It drafts a fast first pass and indexes a long record faster than a person alone, which is part of what makes a same-week summary possible. But unverified AI output is exactly how a Summary A gets made, a fluent draft that adds a fact or a tone the record does not support. That is why a defensible summary pairs AI drafting with a trained reviewer who checks every line against the source, rather than shipping the draft as written.

A gut-check before you attach your next narrative summary: pick three of its strongest sentences and find the page each one cites. If all three resolve to the record and nothing was added, you are holding Summary B. If one of them is an adjective with no source, you are holding Summary A, and the other side will find it before you do.

What separates the summary that holds up

Sourced

Traces to the page

Every statement tied to a page or Bates number, so any line can be checked in seconds.

Reported

Not characterized

The record's facts and its qualifiers kept, with no adjective the chart did not use.

Complete

Gaps flagged

Prior conditions and inconsistencies surfaced, not smoothed into a cleaner story.

Frequently asked questions

How can two narrative summaries of the same records be so different?

Orange downward pointing arrow icon.

Because a narrative summary is prose, and prose invites characterization. One summarizer reports what the record documents and sources each line; another reaches for adjectives that read well but say more than the chart does. Both look fine on a first read, and only one holds up when checked against the file.

What is the single biggest sign a narrative summary will not hold up?

Orange downward pointing arrow icon.

An adjective with no source. If the summary calls pain severe or an injury serious and the cited page does not use those words, the summary is characterizing rather than reporting, and that gap is the first thing opposing counsel uses to argue the document is biased.

Why does one overstated line put the whole summary at risk?

Orange downward pointing arrow icon.

Because once a single line is shown to overstate the record, the reader stops trusting the rest. At deposition, an accurate summary with one exaggerated sentence can have all of its accurate sentences treated as suspect, which is why objectivity on every line matters, not just most of them.

Does a defensible narrative summary have to read flat or boring?

Orange downward pointing arrow icon.

It reads a little flatter than a characterized one, and that is by design. The summary stays neutral so the persuasion happens in your hands, built on a record no one can accuse of leaning. A flat, sourced summary is far more useful than a vivid one you have to walk back.

How do I check whether a summary is the defensible kind before I use it?

Orange downward pointing arrow icon.

Pick its strongest sentences and trace each to the cited page. Confirm the tone reports rather than characterizes, the dates match the file, and gaps and prior conditions are flagged. If the strong lines resolve to the record and nothing was added, the summary is built to hold up.

Does a narrative summary decide whether an injury is serious or what caused it?

Orange downward pointing arrow icon.

No. A narrative summary organizes and reports what the records document and sources it. Whether an injury is serious, what caused a condition, and what the case is worth are determinations for the treating physicians, retained experts, and attorneys, not for the summary.

Orange downward pointing arrow icon.

Orange downward pointing arrow icon.

Orange downward pointing arrow icon.

Orange downward pointing arrow icon.

Which one do you want to be holding

Two narrative summaries from the same records can look identical on page one and end up on opposite sides of a case. Summary A reads beautifully and adds risk with every characterization. Summary B reads a little flatter and holds up every time someone checks it against the file. The test is not which one is more persuasive on its own, it is which one survives being read against the record by someone who wants it to fail.

So order and read your next narrative summary the way opposing counsel will. If every line traces to the record and nothing was added, you have a document you can build on. If it reads well but says more than the file does, you have a Summary A, and a Summary A is a liability with good grammar.

Ready for the narrative summary that reads well and still survives cross? Partner with LezDo TechMed, and see what makes a narrative summary defensible before you rely on the next one.

Source Credit :  All metrics derived from LezDo TechMed’s internal project data.
Anjana Devi Vijay

Anjana Devi Vijay

Anjana Devi Vijay is a Certified Legal Nurse Consultant (CLNC) and Medical–Legal Research Analyst with 9+ years of experience in medical record review, deposition summary analysis, and medico-legal research. She specializes in transforming complex healthcare documentation into accurate, actionable insights that support attorneys, insurers, and medical evaluators. With expertise in clinical documentation analysis and legal case support, she creates research-driven content focused on improving decision-making and case outcomes.