Psychometric evaluation of the Signs of Depression Scale with a revised scoring mechanism in stroke patients with communicative impairment.

The diagnostic accuracy analysis included 52 patients with communicative impairment, and only 19 had depression, so the estimates rest on a small number of cases. For patients with communication impairment, depression was diagnosed from relatives' interview ratings, but agreement between patients and relatives was tested only in patients who could communicate. The two-week symptom duration needed for a depression diagnosis could not be met because the mean length of hospital stay after stroke in the Netherlands was less than 14 days. Nurses mostly used the lower Likert response options, with 85%-96% of ratings at 'not at all' or 'several days', so the revised scale behaved like the original yes/no scale. The study was done in hospital settings only, so results may not apply to rehabilitation or longer-stay care. The scale observes behavioural signs and cannot assess symptoms such as guilt or thoughts of death or suicide.

The authors declared no potential conflicts of interest. Funding was from the Foundation of Innovation Alliance, Regional Attention and Action for Knowledge circulation (SIA RAAK) and an internal grant of the University of Applied Sciences Utrecht.

Do not read the nurses' high agreement on usability as evidence that the Likert scale is better at detecting depression. It did not improve diagnostic accuracy compared with the original scale, and the authors did not recommend it.