Early Detection of Post Stroke Depression: psychometric properties of the Signs of Depression Scale
Summary
Title: Early detection of Post Stroke Depression: psychometric properties of the Signs of Depression Scale
Background Post stroke depression (PSD) is a common consequence after stroke. Depression has been negatively associated with functional outcome, length of stay in hospital, cognitive impairment, and social activities. One third of the patients post stroke suffer from communicative impairments. For assessing PSD, irrespective of the fact that patients have communicative impairments, an observational screening tool could be useful. Of the available tools the Signs of Depression Scale is adapted. This revised version (SODS-R) needs validation.
Aim and research question(s) This study evaluates the internal consistency, interrater and intrarater reliability, criterion validity against the nine-items Patient Health Questionnaire (PHQ-9), construct validity against the Barthel Index (BI), diagnostic accuracy and clinical utility of the SODS-R.
Method A quantitative, observational, cross-sectional cohort study in a Dutch nursing home from April 2013 till June 2013.
Results 18 residents, 14 proxies and 23 nurses were included. 4 residents had a major depression. Cronbach’s alpha for internal consistency was 0.91, ICC for interrater and intrarater reliability was 0.96 (95% CI 0.87 – 0.98) and 0.88 (95% CI 0.70 – 0.96), respectively. Correlation between SODS-R and the PHQ-9 and BI, respectively, was rp = 0.25, p 0.344 and -0.18, p 0.474.The area under the curve for diagnostic performance of SODS-R was 0.68 (95% CI 0.32-1.00), and the optimum cut-off point was 5. Sensitivity and specificity were 75% and 69%, respectively. The positive and negative predicted value was 0.43 and 0.90, respectively.
Conclusion Although the sample size of this study is too small to provide evidence for psychometric of the SODS-R, the tool seems to be reliable and accurate and useful, but should be further validated.