botminds.aiRequest a demo
WorkflowRegulatory & Clinical Evidence

Clinical Evaluation Report (CER) Update

A clinical evaluation report update that finds the sentences a new study changes, redoes the pooled figures where you can check them, and leaves approval and signature to the qualified evaluator and the clinical affairs manager.

See the demoBook a demo
Follow one randomized study into revision 8 ↓
Watch with transcript →
The problem

A new study changes one sentence somewhere in a report of hundreds of pages

For an implant, the report is opened again each year. Across that year the literature team appraises new articles and R&D revises the technical file of the device you claim equivalence to. Someone then has to find, by hand, each sentence the new evidence moves: a pooled score, a survivorship figure, a variant result, a line in the conclusion.

The pooled figure gets redone in a spreadsheet and retyped. It changes in the analysis and stays old in the summary. A source the journal has since retracted keeps propping up a claim. The requirement checklist cites a section that moved, and the notified body is the first to notice.

Built forClinical evaluation specialistsMedical writersQualified clinical evaluatorsClinical affairs managersRegulatory affairs reviewers

The Problem Today

A new study can change one sentence in a long report, and a specialist has to find it by hand.

4 months
At most, a year's upkeep of clinical documents, most makers say published
50%
Of makers had a clinical evaluation significantly challenged published

Done by hand today

  • Searching
  • Pooling
  • Tracing
  • Retyping
  • One figure, three placesThe conclusion and summary keep the old number
  • Retracted, still citedA withdrawn source stays behind a claim

The Solution

  1. Each study meets its claimsAgents read the results and find the claims a study touches
  2. The sum, shownPooled figures redone from each study, with the arithmetic on screen
  3. Nothing kept unsupportedA claim whose only source was retracted shows as unsupported
  4. Approved after opening, then signedThe evaluator approves each claim after opening it; the manager signs

Built-in controls

  • Post-market data is held back from the agents
  • Nobody approves their own wording
  • Clinical safety questions go to a named person

What You Get

Clinical affairs sign each yearly update with each claim traced to the source behind it.

60 hrs
Of mapping and drafting saved per update, half of the work our estimate
  • Requirement rows kept in stepEach change names the sections and checklist rows it feeds
  • A pack reviewers can followChange log, claim trace and signatures in one file
  • Each report on one clockDue dates by class, owner and the reviewer's month
One study, two claims, one signature

A CER update for a cementless hip stem, the day the randomized study lands

Twelve of the cycle's thirteen evidence items are applied and approved. The thirteenth, a randomized trial of two surgical approaches, arrived yesterday, and revision 8 is due in 23 days.

demoThe hip stem, the studies, the people and the figures are sample data.

  1. 01Morning

    46 claims on the deck, and E-03 is the only item not applied

    Clinical evaluation specialist

    She opens the report as its 46 claims in groups, each card one sentence with its sources and requirement rows. The tray above holds the cycle's evidence. E-03 is marked new: 312 patients, 288 assessed at five years. The panel says what will be read, results and function scores, and what will not, the complications section.

  2. 02Morning

    Four agents run in place: two claims matched, the other 44 untouched

    Clinical evaluation specialist

    She applies E-03 and the run shows each step beside the study. The study's results are read from the cited pages and tables. Two claims match: the five-year function claim at 0.94 and the performance conclusion at 0.88. All 238 references are re-checked with no new retraction, and the pooled figure is redone with the sum written out.

  3. 03Morning

    The pooled five-year score drops from 91.4 to 90.5 across 1,474 hips

    Clinical evaluation specialist

    The claim page puts revision 7 beside the draft and shows the arithmetic, study by study. Adding 288 hips at 86.9 turns one pooled value into a range of 86.9 to 93.2 across five studies. A box lists the two studies that report the same endpoint but are left out of the pool, and why.

  4. 04Morning

    She writes the conclusion's new line herself, and it goes over uncompared

    Clinical evaluation specialist

    She signs her acceptance of the five-year claim, then opens the performance conclusion. She rewrites the drafted line in her own words. Typed wording cannot be checked against the sources by the agents, so the page marks it plainly: a person must read this, word for word. She cannot approve it herself.

  5. 05Late morning

    Approved after opening each claim with its sources for three seconds

    Qualified clinical evaluator

    The orthopedic surgeon who checks the report opens the five-year claim, then the conclusion, each with its quoted passages. The approve button only works from the claim's own page once it has been read. On the conclusion she also confirms the benefit-risk conclusion still stands, and both signatures go into the audit trail.

  6. 06Midday

    The summary is drafted from approved claims only, and she signs it

    Medical writer

    With the claims settled, the writer opens revision 8 against revision 7 section by section. The summary is drafted from approved claims: the range, the survivorship, the narrower equivalence and the removed claim. She reads each figure against its claim before she accepts it under her own signature.

  7. 07Afternoon

    Nothing waits for anyone, and the sign button sits under the summary

    Clinical affairs manager

    The manager's screen lists the four stages and the 20 approved claims. Each line of the final checklist is ticked, including that he read the summary on this screen. He made none of the edits, so he may sign. The panel also states plainly who agreed to what before him.

  8. 08Afternoon

    Revision 8 signed: 20 of 46 claims touched this cycle

    Clinical affairs manager

    Back on the deck, the ledger reads 20 touched: 7 changed, 9 strengthened, 3 reopened and 1 unsupported, all approved. The touched filter leaves only those cards. The next step names the pack for regulatory affairs, so nobody asks what happens now.

  9. 09Afternoon

    The pack is assembled only after the signature

    Regulatory affairs reviewer

    The pack holds revision 8 marked and clean, the change log, the claim-to-source trace, the checklist change rows and the signature page. The surveillance summary is named by number only. Regulatory affairs send it through the notified body's portal; this screen sends nothing on its own.

How 288 new hips move the pooled five-year score from 91.4 to 90.5

The same arithmetic the claim page shows, so the evaluator can redo it with a pencil.

Four studies in revision 7, hips times meanOver 1,186 hips, for a pooled 91.4108,391.6
The randomized study, 288 hips at 86.925,027.2
Five studies togetherOver 1,474 hips133,418.8
Pooled five-year score in revision 8The claim now reads as a range of 86.9 to 93.2, above the plan's level of 8090.5

demoSample data, from the working solution.

Around the revision

The evidence and requirement rows behind a clinical evaluation report

What the evaluator, the writer and regulatory affairs open on the other days of the cycle.

An unsupported claimThe return-to-driving claim had one source, and the appraisal log shows a retraction notice. The evaluator approved its removal from revision 8.
The equivalent deviceRevision K of the equivalent stem's file changed its coating thickness. Three claims reopened, and the evaluator limited equivalence to stems made with the earlier layer.
13 evidence itemsEach item with who signed it, what was read and what was not. The surveillance summary sits on top as one referenced document the agents do not open.
Requirement rowsThe claims behind each requirement, colored by what changed, with a drafted checklist change row regulatory affairs can copy.
404 reportsEach report by class, business, update clock, owner and the reviewer's month. Two are late, each with the reason on its row.
Audit trailWho did what and why, person or agent, with the meaning of each signature and the signing session. The trail exports to a file.

demoSample data, from the working solution.

Three things a page-by-page review of the CER tends to pass

Each one is on the screens above, in this cycle's sample.

A claim resting on a retracted paper

The return-to-driving claim had one source, and the appraisal log now carries its retraction notice. It shows as unsupported until someone decides, and here it was removed.

A coating change on the equivalent device

The layer moved from 80 ± 20 µm to 60 ± 15 µm in revision K. Three claims reopened for the evaluator before anyone outside asked.

A result pooled where it does not belong

The short-neck cohort feeds the short-neck claims only, and the over-75 study feeds its own claim. Each claim page lists what it leaves out of the pool.

demoSample data, from the working solution.

Who prepares, who approves and who signs the new revision

Makers never approve, the checker never edits, and the signer made none of the changes.

Clinical evaluation specialist
Sees
Each touched claim with its draft, its sum and its sources
Signs
Accepts or rewrites wording; never approves it
Medical writer
Sees
The new revision against the signed one, section by section
Signs
The summary, drafted from approved claims
Qualified clinical evaluator
Sees
Each changed claim, opened with its quoted passages
Signs
Approves claims, rules on equivalence and confirms the benefit-risk conclusion
Clinical affairs manager
Sees
The sign-off queue, the clock and the signed summary
Signs
The revision, once each touched claim is approved
Regulatory affairs reviewer
Sees
The signed revision, the trace and the pack
Signs
Nothing; reads and downloads
Who reads the evidence

Seven agents on the report, and none writes the benefit-risk conclusion

They read, match, redo sums and draft. People word, approve and sign.

Evidence reader

Reads each new item's performance and benefit results, citing page and table.

  • Never reads safety or adverse-event sections
  • Never opens the surveillance summary
Claim matcher

Matches each result to claims by variant, endpoint and time point, with a confidence.

  • Never changes a claim
  • Below its confidence line, hands to the specialist
Source keeper

Re-checks each claim's sources for retraction, replacement and age.

  • Never keeps an unsupported claim quietly
Claim drafter

Redoes pooled figures with the sum shown and drafts before and after wording.

  • Never writes a safety claim or the benefit-risk conclusion
Equivalence watcher

Compares the equivalent device's file revisions line by line.

  • Never decides whether equivalence holds
Requirement tracer

Lists the sections and checklist rows each change feeds.

  • Never marks a requirement met
Pack builder

Assembles the signed revision, change log, claim trace and signatures.

  • Never sends anything to the notified body

What clinical affairs teams ask about a clinical evaluation report update

What is a clinical evaluation report update?

A clinical evaluation report shows with clinical data that a device does what it claims and that its benefits outweigh its risks. An update applies the year's new studies and technical changes to it. Here the update works claim by claim: agents find the sentences each study changes and redo the sums, and people approve and sign.

Do the agents read post-market surveillance or follow-up data?

No. Post-market surveillance and follow-up results arrive as one signed summary from the surveillance team. The agents do not open it, and the claims resting on it stay with the qualified evaluator. Safety and adverse-event sections of articles are left unread too, and listed as such.

Who decides whether equivalence still holds?

The qualified clinical evaluator. The equivalence watcher shows what changed between revisions of the equivalent device's file and reopens the claims that rest on it. Keeping, limiting or dropping equivalence is the evaluator's ruling.

Can someone approve a claim without opening it?

No. Approval works only from the claim's own page, after it has been open with its sources. Bulk approval of strengthened claims needs the sampled ones opened first, and nobody approves their own wording.

What happens to a claim whose source was retracted?

It shows as unsupported and stays on the deck until a person decides. The specialist proposes the change and the evaluator approves it. Nothing unsupported is carried into the next revision quietly.

Does it replace our document system or regulatory system?

No. The report stays a controlled document in your document management system, and the checklist stays with regulatory affairs. The solution reads them, keeps the claims and their trace, and hands back a signed revision and a pack.

Regulatory & Clinical Evidence

More solutions for regulatory and clinical teams

All 5 solutions →
IVDR Performance Evaluation Report

Every IFU claim stands on its own studies, with each gap owned before the notified body asks.

1-min demoGlobal Device Registration Dossier Assembly

Every country's dossier from one core file, and every change traced to its filings.

1-min demo510(k) Submission Consistency Review

FDA's questions found and fixed before FDA asks them.

1-min demo

Send us one report
and this year's new studies.

We will show you the claims each study touches, with the pooled sums redone.

Book a demoBrowse solutions