THE RELIABILITY AND VALIDITY OF THE COLLATERAL SOURCE VERSION OF THE GERIATRIC DEPRESSION RATING-SCALE ADMINISTERED BY TELEPHONE

Objective. To prospectively evaluate the reliability and validity of t he Collateral Source Geriatric Depression Scale (CS-GDS) administered by telephone (T-CS-GDS) in patients undergoing outpatient comprehensiv e geriatric assessment. Subjects. Eighty-three geriatric patients eval uated in a I-year period at the outpatient Geriatric Assessment Center of the University of Nebraska Medical Center. Methods. The 30-item CS -GDS was completed by the collateral source of all patients on three o ccasions: by telephone several days before their assessment, face-to-f ace during their assessment visit and several days later, again by pho ne. During their assessment, all patients were evaluated by one of thr ee geriatric psychiatrists who were blinded to CS-GDS results. The tes t-retest reliability of the T-CS-GDS was measured by comparing the res ults of the two phone interviews. The construct validity of the T-CS-G DS was estimated by comparing the results of the initial T-CS-GDS with the CS-GDS obtained during the comprehensive assessment, The criterio n validity of the T-CS-GDS was estimated by comparing the results of t he T-CS-GDS with the clinical diagnosis of depression assigned by the psychiatrists. Results. The individual items of the initial T-CS-GDS s howed substantial concordance with the second T-CS-GDS (kappa range 0. 41-0.8, mean = 0.61) and with the assessment GDS (kappa range 0.33-0.8 5, mean = 0.61). Twelve items showed evidence of bias when comparing t he two T-CS-GDSs and four items when comparing the initial T-CS-GDS wi th the CS-GDS done during the assessment. The mean number of symptomat ic responses was not significantly different for the T-CS-GDS vs asses sment administration but did decline slightly when comparing the two T -CS-GDSs. ROC curve analysis showed good agreement between the clinica l diagnosis and the T-CS-GDS. Conclusion. The CS-GDS appears to mainta in its reliability and validity when administered via telephone and th us may be useful for a variety of epidemiologic and clinical purposes. ((C) 1997 by John Wiley & Sons, Ltd.)