A guide to the directory
How much validation evidence is there?
There are more than 60 named tools and variants. Their evidence bases differ greatly. The directory now lets you see which have more than one identified primary validation paper.
What the count means
The filter counts distinct primary publications with eligible diagnostic-accuracy evaluations in the site’s current evidence audit. Reviews are not counted as primary papers. Several evaluations within one publication count once for that tool.
This is a working list, not a completed systematic census. Translations and adaptations may be included. We have not fully reconciled overlapping cohorts across publications. A count of three does not necessarily mean three independent studies.
Of the 82 core profiles, 34 currently have one such paper and 8 have two. This helps explain why a long directory can contain a much smaller group with repeated validation. A zero count means no eligible paper is included in this audit, not proof that no validation exists.
Groups with three or more identified primary papers
Numbers in brackets are publications. Some tools appear in more than one row. This table describes diagnostic-accuracy evidence. It does not provide a comparable count of all validation for severity, arousal or experience measures.
| Group | Tools and identified papers |
|---|---|
| Episodic assessment | CAM-ICU (49), CAM (36), 4AT (32), ICDSC (26), NuDESC (24), DOSS (12), CAPD (10), 3D-CAM (9), FAM-CAM (8), DDT-Pro (7), NEECHAM (7), SQiD (5), bCAM (5), psCAM-ICU (5), SOS-PD (4), pCAM-ICU (4), DSI (3), I-AGeD (3), RADAR (3), S-PTD (3), Sour Seven (3), UB-2 (3) |
| Monitoring | ICDSC (26), NuDESC (24), DOSS (12), CAPD (10), NEECHAM (7), SOS-PD (4), RADAR (3), S-PTD (3) |
| Severity | MDAS (15), DRS-R-98 (10), DRS (5), DDS (4) |
| Ultra-brief tools | UB-2 (3) |
| Arousal | None identified under this definition. |
| Distress and experience | None identified under this definition. |
| Cognitive tests | DelApp (3) |
Validation includes more than diagnostic accuracy
A severity or distress scale needs evidence about what it measures, reliability, responsiveness and interpretation. It should not be judged solely by sensitivity and specificity. CAM-S, for example, has published validation as a severity measure even though this diagnostic-accuracy count does not represent that work.
Each profile links its original or source paper and, where identified, further studies. The primary-paper list will grow as checking continues. Send a missing study or correction.
A separate question: routine care
A tool can perform well with a research team and less well in clinical care. Look at who administered it, who was included, the assessment procedure, completion and what happened after a positive result.
See tools with large-scale implementation reports · How the existing evidence labels work