DRS-R-98: Delirium Rating Scale-Revised-98

The original validation assessed 68 participants across delirium, dementia, schizophrenia, depression and other groups using the 16-item total score and 13-item severity score. Depending on threshold and comparator, total-score sensitivity ranged from 91% to 100% and specificity from 85% to 100%; PubMed lists a published erratum in the following issue.

This Portuguese validation included 64 participants across five diagnostic groups and used two independent blinded raters. A total-score threshold of 20 yielded 92.6% sensitivity and 94.6% specificity, with an area under the curve of 0.99.

The Japanese validation included 81 participants, of whom 48 had delirium and 33 did not, and found high inter-rater reliability. At a total-score threshold of 14.5 sensitivity was 98% and specificity 94% overall, although specificity was 75% when delirium was compared specifically with dementia.

This Korean validation included 157 participants: 69 with delirium, 29 with dementia, 32 with schizophrenia and 27 with other disorders. The total-score area under the curve was 0.948 overall and 0.873 against dementia, with high internal consistency and inter-rater reliability but threshold-dependent classification.

This study tested DRS-R-98 in 125 participants, 85 with dementia, and found that only 19 of 36 delirium cases met every diagnostic system examined. Areas under the curve ranged from 0.905 to 0.929 and optimal total-score thresholds varied slightly by diagnostic system, demonstrating reference-standard effects in dementia-rich populations.

This review screened 9,409 records and evaluated 42 delirium-severity instruments, identifying six leading measures that included the DRS family. It assessed psychometric and clinical utility rather than pooling DRS-R-98 diagnostic sensitivity and specificity, so it supports instrument standing rather than a single accuracy estimate.

This non-intensive-care review examined 75 reviews covering 30 instruments and identified DRS-R-98 among four notable tools based on citations, validation methods and DSM-5 coverage. The review did not calculate a DRS-R-98-specific pooled sensitivity or specificity and therefore does not remove the need to consider population and threshold.

This preliminary critical-care study assessed severity in 22 patients who already had delirium, comparing clinical scoring with scoring reconstructed from records. Correlations were about 0.63 and internal consistency was high, but the small sample, limited inter-rater subset and lengthy information gathering make this severity evidence rather than a diagnostic-accuracy validation.