Adaptive cognitive testing in dementia.

Area of Science:

• Gerontology
• Neuropsychology
• Medical Informatics

Background:

• The Cambridge Cognitive Examination (CAMCOG) is a standard dementia screening tool.
• Current lengthy cognitive tests can be burdensome for elderly patients, leading to potential measurement errors.
• Computerized Adaptive Testing (CAT) offers a method for personalized and efficient assessment.

Purpose of the Study:

• To evaluate the feasibility of tailoring the CAMCOG using CAT for individual patient cognitive assessment.
• To determine if CAT-based CAMCOG maintains diagnostic accuracy for dementia while improving efficiency.

Main Methods:

• Utilized a two-step approach with existing CAMCOG data (n=797).
• Estimated patient abilities and item difficulties using the One Parameter Logistic Model (OPLM).
• Applied CAT to re-estimate patient abilities and compared results with the full CAMCOG.

Main Results:

• CAT achieved excellent agreement (ICC > 0.98) with full CAMCOG scores despite a 60% average test reduction.
• CAT demonstrated comparable accuracy (AUC 0.91) to the original CAMCOG for dementia diagnosis.
• Findings were validated in an independent sample (n=170).

Conclusions:

• CAT provides a highly efficient and accurate method for dementia screening.
• Tailored cognitive testing with CAT significantly reduces patient burden and improves assessment efficiency.
• CAT-adapted CAMCOG is a promising tool for both clinical practice and research in dementia detection.