A medical records index for litigation is a table that lists every document in a paginated record set with its Bates range, provider, document type, date range and page count. Build it from the numbered pages in order, one row per document, so anyone on the case can find any page fast and see what the set contains.
Columns a litigation records index needs
A litigation records index needs six columns: Bates range, provider, document type, date range, page count and notes. Together they answer the two questions every reader brings to a record set: where is a given document, and what is in this set at all. Extra columns are fine, but these six are the core.
Each column has a specific job:
| Column | What it holds | Why it matters |
|---|---|---|
| Bates range | First and last Bates number of the document | Lets anyone jump straight to the pages |
| Provider | The facility or practice that created the record | Groups records by source and shows who is missing |
| Document type | Emergency record, office notes, imaging report, bill, and so on | Lets a reader filter to the kind of record they need |
| Date range | First and last date of service in the document | Places the document in time without opening it |
| Page count | Number of pages in the range | Checks that the range and the pages agree |
| Notes | Undated pages, illegible pages, duplicates kept, missing pages | Keeps every problem visible in one place |
The page count column looks redundant, since you can work it out from the Bates range. It earns its place as a check. If the range says 26 pages and the PDF has 25, something was dropped or misnumbered.
Optional columns depend on the case. A "received" column with the date each production arrived helps when supplements keep coming. A "source file" column with the working file name links the index to the folder. A "request" column naming the authorization or subpoena behind each production helps answer the question of how a record was obtained. Add these only when someone will use them, because every extra column is one more thing to keep current when the set changes.
A sample index for eight documents
This index is fictional. The providers, dates and the Bates prefix "SMPL" are invented for the example.
| Bates range | Provider | Document type | Date range | Pages | Notes |
|---|---|---|---|---|---|
| SMPL-000001 to 000009 | Lakeview Example Hospital | Emergency department record | 2026-03-14 | 9 | |
| SMPL-000010 | Lakeview Example Hospital | Radiology report | 2026-03-14 | 1 | |
| SMPL-000011 to 000018 | Lakeview Example Hospital | Itemized bill | 2026-03-14 | 8 | |
| SMPL-000019 to 000030 | Riverbend Example Family Practice | Office visit notes | 2026-03-19 to 2026-05-02 | 12 | |
| SMPL-000031 | Riverbend Example Family Practice | Fax cover sheet | Undated | 1 | Fax header shows 2026-05-12 |
| SMPL-000032 to 000057 | Northside Example Physical Therapy | Therapy notes | 2026-03-26 to 2026-05-20 | 26 | Page 000044 partly illegible |
| SMPL-000058 to 000061 | Northside Example Physical Therapy | Billing statement | 2026-03-26 to 2026-05-20 | 4 | |
| SMPL-000062 to 000064 | Corner Example Pharmacy | Dispensing history | 2026-03-14 to 2026-04-30 | 3 |
The page counts add up to 64, and the last Bates number is 000064. That match is the first check on a finished index. The Bates ranges come from the numbering step, whose conventions are explained in the guide to paginating and Bates numbering medical records.
Litigation index versus medical chronology: which do you need?
A litigation index locates documents; a medical chronology tells the treatment story. The index has one row per document in page order and says nothing about what the care meant. The chronology has one row per encounter in date order and summarizes what each record says. Most cases with substantial records use both, and the index always comes first.
The two differ in almost every way:
| Litigation index | Medical chronology | |
|---|---|---|
| Row | One document (or visit, or provider) | One encounter |
| Order | Bates order | Date order |
| Content | Where the document is and what type it is | What happened, quoted or summarized |
| Who can build it | Anyone careful with page numbers | Someone who can read clinical notes |
| Main use | Finding pages, checking completeness | Understanding treatment over time |
You need only an index when the case turns on whether certain documents exist, or when someone else will do the medical review. You need a chronology when the treatment history itself matters, which in a personal injury case it usually does. Writing chronology rows from an index is the order the guide on building the injury chronology entry by entry follows.
The index is also faster to build and easier to check. It can be finished, reviewed and handed over while the chronology is still being written.
Building the litigation index from Bates ranges
Building the litigation index from Bates ranges starts with a record set that is already sorted, deduplicated and numbered. Work through the pages in Bates order, start a new row each time a new document begins, and close each row with the last page of that document. Then check the totals.
The steps:
- Confirm the set is final. Duplicates should already be removed and every page stamped. Indexing before deduplication means editing ranges later. The guide on deduplicating pages in medical records covers that step.
- Open the numbered set in Bates order. Use the working copies, in the folder order they were numbered in.
- Mark document boundaries. A new document starts where the provider, the document type or the date run changes. A cover letter, a new header or a change in page layout are the usual signs.
- Write the row. Record the first and last Bates numbers, provider, document type and date range. Count the pages.
- Log problems immediately. An undated page, an illegible page or a page that looks out of place goes in the notes column on the spot.
- Check the totals. Add the page counts. The sum should equal the last Bates number minus the first plus one. Any difference means a skipped page or a typo in a range.
- Check for gaps in the ranges. Each row should start one number after the previous row ended. A jump means pages were never indexed.
- Have a second person spot-check. Pick a handful of rows and open the pages they point to.
Document types should come from a fixed list, not from free text. "Office visit notes", "Office notes" and "Progress notes" for the same kind of record make filtering unreliable. Keep the list short and add to it only when a new kind of record appears.
Folder structure decides the order the pages were numbered in, so it decides the order of the index. The recommended layout is in the guide to which folders a legal case record set needs.
How detailed should each litigation index entry be?
Each litigation index entry can describe one document, one visit or one provider. One row per document suits most cases. One row per visit suits long, mixed records where a reader needs to find individual encounters. One row per provider suits small sets or a first-pass inventory before full indexing.
One row per document. A document is whatever the provider sent as one unit: an emergency department record, a set of office notes, a bill. This is the default because it follows how the records arrived and keeps the index short enough to scan.
One row per visit. A family practice that sends four years of notes as one 300-page file produces one useless row at document level. Splitting it into one row per visit makes each encounter findable. The cost is a much longer index and more boundary decisions.
One row per provider. For a set of a few dozen pages from three providers, one row per provider may be all anyone needs. It also works as a quick inventory while records are still arriving.
You can mix levels in one index if you are consistent within each provider. A hospital's emergency record might get one row, while its long admission gets one row per day. Say which level you used for each provider in a short header note.
Description length matters too. The document type column should be a few words from your fixed list. Anything longer belongs in the notes column, and anything that interprets the content, such as what a test showed, belongs in the chronology rather than the index.
Reviewing a sample of entries against the pages is a human review task. The human review category covers how review queues, second checks and correction logs work in general.
Undated and illegible pages in a litigation index
Undated and illegible pages are the ones most likely to get lost, because they do not fit a date range or a clear document type. A litigation index should give each one its own row or a clear note, so that it stays visible to everyone who uses the index. Hiding them inside a larger range is how they disappear.
Undated pages. Write "Undated" in the date range column rather than guessing. In the notes, record any clue: a fax header date, a date on the page before or after, or the visit it arrived with. If the page sits inside a document, you can keep it in that row and add "page 000031 undated" to the notes.
Illegible pages. Index them like any other page, with a note such as "000044 partly illegible, handwriting". Do not skip them or drop them from the range. If a clearer copy is needed, the note tells whoever writes the follow-up request exactly which page to ask about.
Fax cover sheets and blank pages. These carry a Bates number like every other page, so they appear in the index. A fax cover sheet can be its own one-page row, as in the sample above. Blank pages inside a document stay in that document's range.
Pages from another patient. Occasionally a provider sends a page that belongs to someone else. Log it with a neutral note, such as "page 000052 appears to belong to a different patient; flagged to attorney", and let the attorney decide what happens next. Do not remove it from the numbered set yourself.
Duplicates kept on purpose. Some near-duplicate pages are kept because they differ, for example by a signature or a handwritten note. Mark them "near-duplicate of 000021, kept" so the next reader does not remove them.
For typed records, software can help with the dates. Schema Care reads encounter dates and problem lists from medical records and scores its confidence in each value, so a date it is unsure of goes to a person to confirm against the page; the threshold demo shows how that review works on a fictional page. The index itself, and every judgment in its notes column, stays with the people on the case.