Normal ageing, impaired cognitive functioning and senile dementia--a mixture distribution analysis

Area of Science:

• Gerontology
• Cognitive Neuroscience
• Biostatistics

Background:

• Cognitive decline is a significant concern in aging populations.
• Understanding dementia score distributions is crucial for early detection and intervention.
• Previous studies on very old adults' cognitive function have yielded varied results.

Purpose of the Study:

• To investigate the distribution of dementia scores in individuals aged 70 and above.
• To determine if dementia scores exhibit a bimodal distribution, suggesting distinct normal and pathological groups.
• To explore the relationship between age and the shift between these score distributions.

Main Methods:

• Statistical analysis of dementia scores from a cohort of very old subjects (70 to >95 years).
• Application of heterogeneity analysis to compare one-distribution and two-distribution models.
• Examination of score distribution shifts in relation to increasing age.

Main Results:

• Dementia scores in very old adults did not show an obvious bimodal distribution.
• Heterogeneity analysis revealed that a two-distribution model significantly better explained the data than a one-distribution model.
• The two identified distributions likely represent normal and pathological cognitive ranges.
• A notable finding was the observed shift of the normal range distribution towards the pathological range with advancing age.

Conclusions:

• The cognitive status of very old adults can be better understood by considering two distinct score distributions: normal and pathological.
• Increasing age is associated with a progression from normal cognitive functioning towards pathological dementia scores.
• These findings highlight the dynamic nature of cognitive aging and the importance of age-specific assessments.