Objectives: To design and test a brief, efficient dementia-screening instrument for use by general practitioners (GPs).
Design: The General Practitioner Assessment of Cognition (GPCOG) consists of cognitive test items and historical questions asked of an informant. The validity of the measure was assessed by comparison with the criterion standard of diagnoses of dementia derived from the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition).
Setting: Primary care doctors' offices.
Participants: Sixty-seven GPs administered the GPCOG to 283 community-dwelling patients aged 50 to 74 with memory complaints or aged 75 and older.
Measurements: The Cambridge Mental Disorder of the Elderly Examination, the Abbreviated Mental Test (AMT), the Mini-Mental State Examination (MMSE), the 15-item Geriatric Depression Scale, and the 12-item Short-Form Health Survey.
Results: The GPCOG was reliable and superior to the AMT (and possibly to the MMSE) in detecting dementia. The two-stage method of administering the GPCOG (cognitive testing followed by informant questions if necessary)had a sensitivity of 0.85, a specificity of 0.86, a misclassification rate of 14%, and positive predictive value of 71.4%. Patient interviews took less than 4 minutes to administer and informant interviews less than 2 minutes. The instrument was reported by GPs to be practical to administer and was acceptable to patients.
Conclusion: The GPCOG is a valid, efficient, well-accepted instrument for dementia screening in primary care.