# Mindfulness intervention RCT search: worked example

## Search audit

- Source: PubMed (NCBI), accessed through the PubMed database connector.
- Search date: 2026-09-16.
- Exact submitted query: `("mindfulness"[Title/Abstract] OR "mindfulness-based"[Title/Abstract]) AND ("JAMA Psychiatry"[Journal] OR "Behaviour Research and Therapy"[Journal])`
- Date filter: publication date (pdat), 2019-01-01 through 2025-12-31.
- PubMed query translation: `("mindfulness"[Title/Abstract] OR "mindfulness-based"[Title/Abstract]) AND ("JAMA Psychiatry"[Journal] OR "Behaviour Research and Therapy"[Journal]) AND 2019/01/01:2025/12/31[Date - Publication]`
- Total returned: 62.
- Records retrieved: 62.
- Truncation: none (has_more = false).

## Operational inclusion rule

Include a publication when its PubMed metadata and complete abstract describe a primary randomized report in which a mindfulness intervention is a defined study arm and participant outcomes are reported. A mechanistic outcome is not by itself grounds for exclusion. Exclude clearly identified secondary, pooled, moderator, predictor, review, observational, or correction publications, and randomized studies in which mindfulness is only a measured construct rather than the intervention. Mark uncertain when a multicomponent intervention only partly comprises mindfulness, when mindful attention is embedded in a broader manipulation, when primary-versus-secondary publication status is unclear, or when date evidence conflicts.

## Revised results

- Included: 20
- Excluded: 37
- Uncertain: 5
- Total accounted for: 62

Uncertain records are PMID 41418645 (mixed intervention plus 2026 print-date discrepancy), PMID 38837133 (SEKT includes MBCT elements within a broader psychotherapy), PMID 36116229 (mindfulness bundled in a multicomponent fibromyalgia program), PMID 34700258 (primary-versus-secondary publication status unclear from the abstract), and PMID 30579131 (mindful attention embedded in exposure therapy).

## Evidence boundary and limitations

Decisions are based on PubMed metadata and complete returned abstracts, not full text. The unchanged JSON preserves the exact query and all returned metadata/abstracts. This remains a focused worked example, not an exhaustive systematic search: no additional databases, citation chasing, registry search, full-text verification, or duplicate independent screening were performed. Search results may enter the literature inbox; acceptance remains for user review.