INTRODUCTION

Accurate dementia diagnosis remains challenging due

to overlapping clinical manifestations and diverse

etiologies. Misdiagnosis may delay appropriate care

and increase the burden on patients and caregivers.

This study examined factors associated with diagnostic

errors and determined which dementia subtypes were

most frequently misclassified in a specialized memory

clinic.

METHODS

In this ambispective cohort study, patients with

dementia misdiagnosis referred to a cognitive

outpatient clinic between May and November 2025

were evaluated. Initial diagnoses were compared with

final diagnoses established after clinical follow-up and

diagnostic review. Clinical and demographic data were

collected, and final diagnoses were classified as

Alzheimer’s disease (AD) or non-AD dementia.

Diagnostic performance metrics were calculated, and

binary logistic regression was used to identify

predictors of the final diagnosis, adjusting for age, sex,

and education.

RESULTS

Of the 36 patients included in the study, 7 received a

final diagnosis of AD and 29 were classified as non-AD.

The initial clinical suspicion of AD showed 71.4%

sensitivity, 44.8% specificity, 23.8% positive predictive

value, and 94.1% overall accuracy (Table 1).

Diagnostic reclassification was frequent. Among

patients initially diagnosed with AD, only 14%

retained a diagnosis of pure AD after specialized

reassessment. Most were reclassified as mixed

dementia, increasing its prevalence from 2.8% at the

initial assessment to 38.9% at the final diagnosis

(Table 2). Compared with patients with non-AD

dementias, those with confirmed AD were younger (p

= 0.003), more physically inactive (p = 0.046), and

differed in the frequency of visual impairment (p =

0.039). No significant differences were observed for

sex, education, vascular risk factors, depression,

hearing impairment, obesity, alcohol consumption,

smoking, or social isolation (Table 3).

CONCLUSION

Pure AD was frequently overdiagnosed, and mixed dementia

underrecognized. AD biomarkers improved diagnostic accuracy,

whereas structural neuroimaging had limited value. Standardized

biomarker-based approaches may enhance dementia diagnosis.