Predictors of agreement between general practitioner detection of dementia and the revised Cambridge Cognitive Assessment (CAMCOG-R)

BACKGROUND: Dementia is a complex and variable condition which makes recognition of it particularly difficult in a low prevalence primary care setting. This study examined the factors associated with agreement between an objective measure of cognitive function (the revised Cambridge Cognitive Assessment, CAMCOG-R) and general practitioner (GP) clinical judgment of dementia. METHODS: This was a cross-sectional study involving 165 GPs and 2,024 community-dwelling patients aged 75 years or older. GPs provided their clinical judgment in relation to each of their patient's dementia status. Each patient's cognitive function and depression status was measured by a research nurse using the CAMCOG-R and the 15-item Geriatric Depression Scale (GDS), respectively. RESULTS: GPs correctly identified 44.5% of patients with CAMCOG-R dementia and 90% of patients without CAMCOG-R dementia. In those patients with CAMCOG-R dementia, two patient-dependent factors were most important for predicting agreement between the CAMCOG-R and GP judgment: the CAMCOG-R score (p = 0.006) and patient's mention of subjective memory complaints (SMC) to the GP (p = 0.040). A higher CAMCOG-R (p < 0.001) score, female gender (p = 0.005), and larger practice size (p < 0.001) were positively associated with GP agreement that the patient did not have dementia. Subjective memory complaints (p < 0.001) were more likely to result in a false-positive diagnosis of dementia. CONCLUSIONS: Timely recognition of dementia is advocated for optimal dementia management, but early recognition of a possible dementia syndrome needs to be balanced with awareness of the likelihood of false positives in detection. Although GPs correctly agree with dimensions measured by the CAMCOG-R, improvements in sensitivity are required for earlier detection of dementia.