[Skip to Navigation]
June 1989

Reliability of Clinical Criteria for the Diagnosis of Dementia: A Longitudinal Multicenter Study

Author Affiliations

From the Hôpital Broca (Drs Forette and Henry) and the U 324INSERM Centre Paul Broca (Dr Boller), Paris, France; the Hôpital Pellegrin (Drs Orgogozo, Dartigues, and Pere), Bordeaux, France; the CHRU de Grenoble (France) (Drs Hugonot and Israel); the Laboratoires Sandoz, Rueil Malmaison, France (Drs Loria, Goulley, and Lallemand); and the Alzheimer Disease Research Center, University of Pittsburgh (Pa) Medical School (Dr Boller).

Arch Neurol. 1989;46(6):646-648. doi:10.1001/archneur.1989.00520420066024

• The reliability of the clinical diagnosis of dementia was estimated by comparing the diagnosis made at 1-year intervals on 55 consecutive subjects with suspected cognitive impairment seen at three different centers by neurologists and gerontologists. The diagnosis was based on history and clinical examination, the criteria of the Diagnostic and Statistical Manual of Mental Disorders (revised ed 3), the Modified lschemic Score, and a computed tomographic scan. Fifty-two of 55 subjects were given the same diagnosis a year later indicating a reliability of 95%. The study shows that a diagnosis of dementia established by simple clinical criteria comparable to the NINCDS/ADRDA criteria affords sufficient reliability to allow the comparability of groups at different centers for purposes of research, including research on the evaluation of the efficacy of pharmacologic treatment.