Abstract
BACKGROUND: Subjective Cognitive Decline (SCD) is recognized as a risk stage for future cognitive impairment and dementia. The criteria for SCD include normal performance on neuropsychological testing; however, there is a lack of consensus regarding standard score cut-offs for neuropsychological tests to define cognitive impairment and to differentiate between SCD and Mild Cognitive Impairment (MCI). This study aimed to assess the frequency of SCD diagnosis using various neuropsychological definitions of cognitive normality and to characterize the sociodemographic and neuropsychological features of SCD patients diagnosed under these criteria. METHODS: The Cognitive Complaints Cohort (CCC) participants were diagnosed following Subjective Cognitive Decline Initiative (SCD-I) criteria. Normal cognitive performance was defined by the absence of Mild Cognitive Impairment (MCI) according to the five sets of MCI neuropsychologically based criteria defined by Jak and Bondi. Descriptive statistics were used to analyze sociodemographic, clinical, and neuropsychological data. A bootstrap methodology was employed to estimate the mean and 95% confidence intervals (CI) for specific parameters of interest, namely the SMC scale (subjective memory complaints scale), Mini-Mental State Examination (MMSE), Blessed Dementia Rating Scale - first part (BDRS first part), and Geriatric Depression Scale (GDS). RESULTS: Among the 1268 subjects included, the prevalence of SCD diagnosis exhibited substantial variation across SCD-I criteria using different neuropsychological definitions of cognitive normality (ranging from 16.4 to 81.3%). When using the most conservative criteria to define cognitive impairment (2 tests within a cognitive domain > 1.5 SD below age-adjusted means), the resulting Conservative SCD group had poorer global cognitive function (MMSE: mean 27.15, 95% CI 27.00-27.31), whereas when using the most liberal criteria to define cognitive impairment (only one test > 1 SD below age-adjusted means) the resulting Liberal SCD group had superior performance in daily-life functioning (BDRS first part: mean 0.30, 95% CI 0.23-0.38). However, subjective cognitive complaints and neuropsychiatric symptoms did not significantly differ among SCD diagnostic groups. CONCLUSIONS: The utilization of diagnostic criteria using distinct neuropsychological definitions of cognitive normality significantly impacts the frequency of SCD diagnosis and characterizes different patient populations. Consequently, it is essential to specify the criterion when diagnosing a SCD patient and to understand the risks and benefits of using different criteria to define cognitive impairment.