Search

Objective This study aims to test the accuracy of the Nijmegen Observer-Rated Depression (NORD) scale, a new short scale for screening of depression in nursing home (NH) residents with and without dementia. Methods This cross-sectional study with 103 residents with dementia ( N = 19 depressed) and 72 residents without dementia ( N = 10 depressed) was undertaken in 13 Dutch NH units. An elderly care physician and a psychologist of each unit assessed residents for the presence of clinical depression. Primary professional caregivers administered the NORD scale. Results Five of the six proposed items showed acceptable performance in screening for depression. Receiver operating characteristic analyses revealed significant areas under the empirical curve (AUC) for the 5-item NORD scale in the total sample ( AUC = 0.83, p < 0.001), as well as in residents with dementia ( AUC = 0.84, p < 0.001) and without dementia ( AUC = 0.84, p < 0.001). The cutoff score of >1 showed the highest sum of sensitivity (100) and specificity (69) in non-dementia and >2 the highest sum of sensitivity (79) and specificity (77) in dementia. The cutoff score of >1 showed the lowest negative likelihood ratio of 0.0 in non-dementia and of 0.2 in dementia. The highest positive likelihood ratios were found for the cutoff of >2 in non-dementia (3.4) and for >4 in dementia (26.5). Conclusion The 5-item NORD scale showed acceptable accuracy comparable with those of more extensive scales in other studies. It is easy and quick to administer and can be used for screening of depression in NH residents with or without dementia. Copyright © 2011 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]