Premium
A comparison of GMS‐A/AGECAT, DSM‐III‐R for dementia and depression, including subthreshold depression (SD)—results from the Berlin Aging Study (BASE)
Author(s) -
Schaub R. T.,
Linden M.,
Copeland J. R. M.
Publication year - 2003
Publication title -
international journal of geriatric psychiatry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.28
H-Index - 129
eISSN - 1099-1166
pISSN - 0885-6230
DOI - 10.1002/gps.799
Subject(s) - dementia , depression (economics) , medical diagnosis , epidemiology , psychology , geriatrics , psychiatry , medicine , clinical psychology , disease , pathology , economics , macroeconomics
Background Empirical evaluation of the agreement between different diagnostic approaches is crucial for the understanding of epidemiological results in geriatric psychiatry. Objectives In this paper, we analyse differences between widely used diagnostic approaches of dementia and depression and offer evidence that diagnostic thresholds vary substantially on quantitative dimensions, but that conceptual and other differences between approaches must also been taken into account. Methods In an epidemiological study of n = 516 persons, aged 70–103 years, we compared psychiatric diagnoses of dementia and depression obtained by GMS‐A/HAS‐AGECAT, DSM‐III‐R and clinician's diagnoses of subthreshold depression (SD). Results For depression, cumulative prevalence of clinician's diagnosis (including SD, GMS‐A/HAS‐AGECAT and DSM‐III‐R defined forms) was highest, followed by GMS‐A/HAS‐AGECAT‐diagnosis and DSM‐III‐R, while for dementia DSM‐III‐R was followed by GMS‐A/HAS‐AGECAT. Overall agreement between DSM‐III‐R and GMS‐A/HAS‐AGECAT was moderate. Adapting thresholds for AGECAT resulted in slightly better diagnostic efficiency. Diagnostic disagreement was found predominantly for cases with intermediate symptom severity, supporting the hypothesis of differing thresholds between DSM‐III‐R and GMS‐A/HAS‐AGECAT, while cases with lower or higher symptom severity were similarily seen as cases or non‐cases. Conclusion Disagreement is not only caused by conceptual differences, but also different thresholds of diagnostic algorithms. Adaptation of threshold levels should be feasible, depending on the purpose of the analysis. Copyright © 2003 John Wiley & Sons, Ltd.