We conducted a literature search to identify all relevant publications on the psychometric properties of the Aphasic Depression Rating Scale (ADRS). We identified five studies. More studies are required before definitive conclusions can be drawn regarding the reliability and validity of the ADRS.
Benaim et al. (2004) examined the test-retest reliability of the ADRS in 15 subacute patients with aphasia due to stroke admitted to a neurorehabilitation unit. Patients were assessed twice at a 2-week interval by the same rehabilitation team. Agreement between items were assessed with kappa coefficients, and agreement for global scores was assessed with correlation coefficients. Kappa coefficients over the 9 items was adequate (kappa =0.58) (ranging from kappa = 0.33 to 1.00). For the global ADRS score, the correlation was excellent (r = 0.89).
Benaim et al. (2004) examined the inter-rater reliability of the ADRS in 15 subacute patients with aphasia due to stroke admitted to rehabilitation unit. Patients were assessed twice within 24 hours by 2 different rehabilitation teams. Kappa coefficients over the 9 items was excellent (kappa = 0.69) (ranging from 0.37 to 1.00). For the global ADRS score, the correlation was excellent (r = 0.89).
A team of 18 neurorehabilitation clinicians were interviewed regarding the most frequently reported depressive behaviours observed in patients with aphasia. Six experts then analyzed 3 existing depression scales that contained items on observable behaivour: the Hamilton Depression Rating Scale (HDRS); the Montgomery and Asberg Depression Rating Scale; and the Salpetriere Retardation Rating Scale (SRRS). Only items that could be completed without interviewing, that described depressive behaviours reported by the team, and that were selected by at least 4 experts were retained. A total of 15 items were selected by the experts (Benaim et al., 2004).
Benaim et al. (2004) examined the concurrent validity of the ADRS. Both dep-psy (psychiatrist rating of depression) and dep-rehab (ratings made by members of the rehabilitation team) were used as the ‘gold standard‘. Wilcoxon test was calculated to find the best model. The Wilcoxon test value was 0.121 for the 7-item model and 0.116 for the 8-item model which was a minor increase so the 7-item model was selected: Apparent Sadness; Insomnia-Middle; Anxiety-Psychological; Somatic Symptoms-Gastrointestinal; Mimic-Slowness of Facial Mobility; Loss of Weight; and Anxiety-Somatic.
Not yet examined.
Benaim et al. (2004) examined the construct validity of the ADRS by comparing it to the dep-psy (psychiatrist rating of depression – 50 patients), dep-rehab (ratings made by members of the rehabilitation team – 50 patients), and the Hamilton Depression Rating Scale (HDRS – 25 patients). The correlations between ADRS and dep-psy, dep-rehab, and HRDS were excellent (r = 0.60; r = 0.78; r = 0.77, respectively). Correlation coefficients in right hemisphere stroke (RHS) patients only and in left hemisphere stroke (LHS) patients only ranged from adequate to excellent (r = 0.58; r = 0.70; r = 0.84, for RHS; r = 0.60; r = 0.86; r = 0.64, for LHS). The ADRS correlated better with dep-psy and dep-rehab (r = 0.59; r = 0.85, respectively) than did HDRS (r = 0.40; r = 0.59, respectively).
A principal component analysis was conducted to analyze the structure of the original 15 items selected during content validity. Six items were eliminated to avoid redundancies and the remaining 9 items were selected to make up the final ADRS (Benaim et al., 2004).
Benaim et al. (2010) examined the responsiveness of the ADRS and the Visual Analog Mood Scale (VAMS) in 49 patients with aphasia due to stroke admitted to rehabilitation units. A trained psychologist evaluated the patients at baseline, rating the severity of their depression on a scale from 0 (no symptoms of depression) to 10 (extremely severe depression); and at the 30 day follow-up, classifying their status as ‘deteriorated’, ‘stable’ or ‘improved’ where changes greater than 1-point/10 were considered to be the minimal clinically important difference. The ADRS and VAMS were also administered at baseline and at the 30-day follow-up; and ADRS scores were converted to a 10-point scale for comparison. The ADRS was found to be more sensitive than the VAMS for detecting change in patients, demonstrating a large effect size for detecting deterioration and improvement (1.18 and -0.89 respectively) compared to the moderate and small effect size demonstrated by the VAMS (0.42 and -0.50 respectively). Changes in ADRS scores also showed excellent correlation (r=0.72) with severity of depression as rated by the trained psychologist on a scale from 0 to 10.
Benaim et al. (2004) examined the sensitivity and specificity of the ADRS in patients with stroke. The threshold for the diagnosis of depression was calculated by comparing ADRS scores with the diagnosis made by the psychiatrist (depression vs. no depression). With a score of less than or equal to 9/32 as a threshold, compared with the diagnosis made by the psychiatrist, sensitivity of the ADRS was 0.83 and specificity was 0.71.
Benaim, C., Cailly, B., Perennou, D., Pelissier, J. (2004). Validation of the aphasic depression rating scale. Stroke, 35, 1696.
Benaim, C., Decavel, P., Bentabet, M., Froger, J., Pelissier, J. & Perennou, D. (2010). Sensitivity to change of two depression rating scales for stroke patients. Clinical Rehabilitation, 24, 251-257.
Dantchev, N., Widlocher, D. (1998). The measurement of retardation in depression. J Clin Psychiatry, 59, 19-25.
Hamilton, M. (1967). Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol, 6, 278-296.
Montgomery, S. A., Asberg, M. (1979). A new depression scale designed to be sensitive to change. Br J Psychiatry, 134, 382-389.