Medical malpractice record review worksheet you can copy
Answer in brief: this worksheet turns a malpractice record review into 10 blocks you fill in order, from the conflict screen to the sign-off. It exists to stop 1 failure: an opinion written before the record has been read in sequence, by a reviewer who already knew how the case ended.
Limits: Prices, turnarounds and claims come from each provider's own pages or the cited survey and statute, read 2026-11-14. We bought no review from anyone listed and cannot vouch for the quality of any reviewer, and nothing here is legal advice on who may testify in your state.
Why the blocks are in this order
The blocks follow the 9 steps on the review process page, with the record inventory kept separate from the chronology because that is where reviews quietly go wrong. A blank you cannot fill in is not an inconvenience. It is a finding, and it belongs in the report.
The order is the point. Facts first, then the standard as it stood on the date of care, then the comparison, then causation. Reverse that order and you are judging a physician for an outcome, which is the error the evidence describes: in Caplan and colleagues' 1991 study of 112 anesthesiologists rating the same 21 cases, appropriate-care ratings fell 31 percentage points when the outcome was switched from temporary to permanent injury (read 2026-11-14).
The worksheet, in 10 blocks
Copy the block below into Word, a spreadsheet or your report template. Anything in square brackets is a prompt to fill, not text to keep. Blocks 1 to 3 are completed before the file is opened; blocks 4 to 9 are worked in order; block 10 is written last.
MEDICAL MALPRACTICE RECORD REVIEW WORKSHEET Reviewer: [name, credentials] Opened: [YYYY-MM-DD] Case ID: [your reference, not the patient name] File as received: [number of PDFs, total pages, date received] BLOCK 1 CASE AND CONFLICT SCREEN step 1 Specialty under review: [ ] Providers and their roles: [ ] Dates of care: [ ] to [ ] Questions, in the requester's words: Q1 [ ] Q2 [ ] Q3 [ ] Form of answer: [phone screen / letter / affidavit / full report] Recent practice in this setting: [years, unit or clinic type] Any party, counsel or facility known to me: [yes / no] Any relative or partner working there: [yes / no] Reviewed this case before for anyone: [yes / no, for whom] Decision: [accept / decline] Reason in 1 line: [ ] BLOCK 2 SCOPE AND LIMITS OF THE OPINION step 1 In scope: [standard of care / causation / both] Out of scope, and the report will say so: [ ] Records I need before I can answer Q1: [ ] Records that would settle Q2: [ ] Deadline and form the requester needs: [ ] Fee basis agreed in writing before I read: [yes / no] BLOCK 3 RETENTION TERMS step 2 Rate: [$ per hour] Retainer: [$] Estimate of hours: [ ] Billable time: [reading / research / calls / writing / travel] Payment terms: [ ] Cancellation: [ ] Work product held by: [ ] Who may disclose it: [ ] Compensation contingent on the outcome: [no, and the agreement says so] BLOCK 4 RECORDS RECEIVED, RECORDS MISSING step 3 Source Date range Pages Received on [facility or office] [ ] to [ ] [ ] [ ] [ ] [ ] to [ ] [ ] [ ] The chart mentions these and the file lacks them: [ ] MAR and nursing flowsheets [ ] orders with times [ ] monitor strips [ ] imaging, not just the report [ ] EHR audit trail [ ] policies in force on that date [ ] other: [ ] Asked for: [what, from whom, on what date] Received: [yes / no] BLOCK 5 CHRONOLOGY COLUMNS step 4 Date | Time | Event time or chart time | Provider | Record type | What the record says, quoted where it matters | Page | Flag Flags: prior / key event / conflict / copy-forward / gap / missing / undated 1 row per event. A row without a page is a note to yourself, not a citation. BLOCK 6 THE STANDARD ON THE DATE OF CARE step 5 Decision point: [ ] Source in force then: [guideline / label / institutional policy / text] Edition, version or policy number and its date: [ ] Still the current version today: [yes / no, and do not cite it if no] What it required, in its own words: [ ] Where I found it: [URL or page in the file] BLOCK 7 DEVIATION GRID step 6 Decision point | What was known then (page) | Standard (source) | What was done (page) | Departure: yes / no / cannot tell | Record that would settle it 1 row per decision point. Never 1 row for a whole admission. BLOCK 8 CAUSATION, TESTED SEPARATELY step 7 Departure: [ ] What would most likely have happened with care that met the standard: [ ] Basis for that view: [ ] Alternative cause, and the page that supports it: [ ] Degree of certainty, in the words this jurisdiction uses: [ ] BLOCK 9 HINDSIGHT TEST step 8 Finding: [ ] Would this finding stand if the outcome had been good: [yes / no] If no, the finding is dropped, not softened: [ ] Second reader: [name] Blind to my opinion: [yes / no] BLOCK 10 LIMITS AND SIGN-OFF step 9 Records I did not see: [ ] What would change my opinion: [ ] Opinions I am not offering: [ ] Qualified in this specialty by: [training, years, setting] Signed: [ ] Date: [ ]
Read block 4 before you read the medicine: the list of what the file lacks decides how far any opinion can go.
A completed deviation-grid row
Synthetic, no real patient. Block 7 filled in on a made-up post-operative case, 1 row from a longer grid. The policy it cites is fictional, which is the point: the standard cell quotes a source with a date, not a memory of how the case is usually run.
| Decision point | What was known then (page) | Standard and source | What was done (page) | Departure | Record that would settle it |
|---|---|---|---|---|---|
| Post-op day 1 hemoglobin drop | Hgb 6.8 g/dL at 05:40, HR 118, note records "no visible bleeding" (pp. 212 to 213) | Hospital policy CP-14, in force 2024-03: transfuse at Hgb under 7.0 g/dL with tachycardia, notify the attending within 30 minutes (p. 0412) | 08:15 note: "will recheck in the morning." No transfusion ordered. Transfused at 19:40 after a cardiac arrest (pp. 214, 226) | Yes | Nursing notification log for 05:40 to 08:15, requested and not in the file |
3 things make that row usable by someone who was not there. Each clinical claim carries a page. The standard carries a policy number and the date it took effect, so a reader can check it was in force that day. And the last column names the missing record instead of guessing, which is how a reviewer writes "cannot tell" without weakening the finding.
How to use it in 20 minutes, and what a full review adds
For a screen, fill blocks 1, 2 and 4 only, then read the notes around the event with the block 5 column headers in front of you. Answer 3 questions in 3 lines: is a departure plausible, which records would decide it, and what would change your mind. That is enough to decline a case, ask for a better record set or quote a full review. It is not enough for an affidavit.
A full review uses all 10 blocks, and the hours go where the rows are: 1 row per decision point, not per admission, each with its own page and its own dated standard. The hindsight pass in block 9 is cheap and it changes findings: re-read each row as if the patient had recovered, and drop the ones that only stand because of the outcome.
8 checks before you sign
- Every clinical fact carries a page, and the page is the file as you received it, not a page number from your own chronology.
- Every standard cited was in force on the date of care, with its edition, policy number or label date shown.
- Each decision point has its own row, and no row bundles 2 decisions into 1 judgment.
- "Cannot tell" is used wherever the record is silent, and the record that would settle it is named.
- Causation is written as a separate finding, with its own reasoning and its own certainty language.
- The hindsight pass is done, and every finding that failed it is gone rather than softened.
- Missing records are named, with the date you asked and the answer you got.
- Block 10 says what would change your opinion, which is the part opposing counsel reads first.
A worksheet does not make a review right. It makes the reasoning checkable, which is the only thing a peer reviewer, a committee or a deposition can test. The who reviews page covers which reviewer may sign what in your state, and the directory lists the outside options we read, with their published terms.