Background: Delirium is a common psychiatric disorder in general hospital elderly patients. Several delirium screening tests exist. Few nurse based delirium severity measures are available. The aim of this study was to evaluate the Delirium-O-Meter, a new nurses' behavioural rating scale that is an efficient and sensitive measure of delirium severity.
Methods: Analysis of cross sectional and repeated assessments data. Participants were 92 elderly general hospital patients; 56 with delirium, 24 with dementia or other cognitive disturbances (no delirium) and 12 with other psychiatric disorders or no mental disorder. Measures were the Delirium-O-Meter (DOM), Delirium Rating Scale-Revised version (DRS-R-98), Delirium Observation Scale (DOS), Behavioural observation scale for geriatric inpatients (GIP) and Mini Mental State Examination (MMSE).
Results: The majority of DOM items show a (near-) normal score distribution. Reliability of the DOM was high; Cronbach's alpha values ranged from 0.87-0.92; Intra Class Correlation (ICC) range was 0.84-0.91 for total scores and 0.40-0.97 for item scores. Factor analysis produced a 'Cognitive/Motivational' factor explaining almost half of variance and a smaller 'Psychotic/Behavioural' factor. The two-factor model results support the conceptual distinction between hyperactive and hypoactive delirium. DOM observations differentiated delirium from non delirium patients. DOM total scores were highly related to the DRS-R-98, DOS, MMSE and GIP apathy and cognitive sub scales, but less so to the GIP affective disturbances subscale, indicating convergent and divergent validity. Temporal difference scores calculated for DRS-R-98 and DOM assessments on subsequent days were also highly related (rho = 0.80-0.95).
Conclusions: The newly constructed DOM is a brief and valid nurses' behavioural rating scale that can be useful for measuring different aspects of delirium and for efficiently monitoring delirium severity in elderly patients.
Copyright (c) 2005 John Wiley & Sons, Ltd.