One importer
This is the same screen used by people who choose “Import your search” in stage 2. If you created the search with revisia (path 1), add results from other databases here. If you imported your own search (path 2), the import already happened in stage 2 and this stage does not appear in the rail.
When you need it
revisia searches PubMed, Europe PMC, OpenAlex and ClinicalTrials.gov. If your review needs no other databases, click “Continue without adding” and move to criteria. If you also use Scopus, Web of Science, Embase, CINAHL or Cochrane, this stage makes sure none are left out.
At the bottom of the screen, “Copy your search to other databases” translates your query into each database's syntax and lists the limits and filters you must set manually. Copy it, paste it into that database, export the results and return here.
Upload your files
- Export from each database. The screen includes instructions: in PubMed, “Send to → Citation manager (.nbib)”; in Web of Science, “Export → Plain text file → Full record”, with no more than 1,000 records per file; in Scopus, RIS or CSV with abstracts; in Embase and Cochrane, RIS; and in Zotero or EndNote, export the collection as RIS.
- Drag in all files at once. You can upload several files from several databases: RIS, BibTeX, CSV, PubMed (.nbib), Web of Science (.txt) and WHO ICTRP (XML), up to 5 MB each. The app detects each file's source and groups them. Three Web of Science files count as one Web of Science search.
- Confirm the source if needed. If a file does not identify its source, select the database or choose “Other methods”. This choice is later shown in the PRISMA diagram.
- Document each database search. Every database has a “Your search” panel for the exact query, platform, date, limits and number of results. This is optional, but PRISMA-S asks for it and it goes into the appendix. If you enter the database result count, the app checks it against the files: “✓ Matches”, or “340 missing” with a hint that a Web of Science batch may be missing.
- Click “Import N files”. They are processed in sequence and retain their source. The summary shows imported records, duplicates removed, records outside the plan and records ready to screen. If a batch is missing or you want to add an updated search, the drop zone stays available and only new records are added.
What happens next
- Deduplication. Imported references are compared with existing ones by identifier and title. Every merge records what was merged and why in the audit log.
- The same screening. An imported article is screened exactly like one found through a search. There are no separate paths or rules.
- Its PRISMA branch. Imports from a recognised database are counted under “identification via databases and registers”. Items marked “Other methods” use their own branch. The diagram handles this automatically.
- The cap is cumulative. Records removed from the project still count toward the plan's article cap. If an import exceeds it, the app warns you before importing.
Troubleshooting
- The file cannot be read. Check its format and the 5 MB limit. A Web of Science .txt must use “Plain text file → Full record”; the old UTF-16 tab-delimited format is not supported.
- My reference manager only exports another format. Zotero, Mendeley and EndNote can export RIS or BibTeX from their export menu. That is the safest route.
- The file is over 5 MB. Split it into several exports and upload them together. They are grouped by database and duplicates are merged as usual.
- I do not know which database to select. If the file did not come from a specific database, such as a supervisor's list or another review's references, select “Other methods”. That is the correct PRISMA category.
- It says “N missing”. The result count you entered is larger than the number of records in the files. A batch is usually missing, because Web of Science exports 1,000 at a time. Upload it and the warning changes to “✓ Matches”.