Advanced search

• Year rangeRestrict to a specific window, useful for "last five years" or "since the 2018 guideline update" kinds of queries.
• Venue qualityA slider that ranks results by the venue’s indexed quality score, so higher-impact journals rise first.
• Study typeFilter to randomized controlled trials, systematic reviews, meta-analyses, longitudinal studies, or other design types.
• Open access onlyShow only the papers Academe can pull the full text of without authentication.
• Peer-review statusInclude or exclude preprints and grey literature alongside peer-reviewed journals.
• Abstract keywordsRequire (or forbid) specific terms to appear in the abstract, combined with AND logic.

Most searches are best written as a natural-language question. Academe's semantic matching ranks results by meaning, not exact tokens. When you need finer control, the search bar also accepts a handful of query operators:

• +term - require the term in the result.
• -term - exclude results containing the term.
• journal:"..." - restrict to a specific venue (quotes preserve spaces).
• author:"Last, First" - match an author name exactly.
• citations:>N - only results with more than N citations.
• year:YYYY or year:YYYY-YYYY - single year or inclusive range.
• doi:..., arxiv:..., pmid:... - jump straight to a known identifier.

Operators combine with parentheses, so (+RCT -observational) journal:"NEJM" citations:>50 is a valid query.

Trials live in a separate registry, not the journal corpus. Switch the source picker from Literature to Clinical trials to query registry entries directly. The filters shift to match:

• Trial phaseEarly Phase 1 through Phase 4, or any combination.
• Recruitment statusRecruiting, active not recruiting, completed, terminated, suspended, or withdrawn.
• Has resultsOnly include trials with published outcome data.
• Date rangesStart date, completion date, and results-posted date can all be bounded independently.

For questions that need more than a single search pass - "find every trial on X since 2015 that measured Y and reported a null result" - start a systematic review instead. The systematic-review workflow runs multiple queries in parallel, deduplicates across them, and preserves the full audit trail a filter chain cannot.