Development of a Cognitive Screening Tool for Adults to Detect Early Dementia at Primary Care Level: A Pilot Study in...

Dementia is an emerging public health challenge in India, particularly among older adults in rural and underserved regions. Early detection is crucial for timely intervention and care planning. However, existing screening tools, such as the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA), have limitations in low-literacy populations due to their reliance on reading, writing, and numeracy skills. In India's primary healthcare (PHC) settings-where time, training, and resources are limited-there is a critical need for a culturally appropriate, easy-to-administer cognitive screening tool.

This study aimed to develop and validate a brief, culturally relevant, and literacy-independent tool-the Primary Healthcare Cognitive Screening (PHC-CS) tool-for the early detection of cognitive impairment among adults aged 50 years and above in Indian primary healthcare settings.

The PHC-CS tool was developed through a multistep process, including a literature review, expert consultation, and field testing. The final 21-item tool assessed eight cognitive domains (memory, attention, language, visuospatial ability, executive function, orientation, constructional ability, and mental flexibility) using orally delivered tasks supported by visual aids. The tool was administered to 172 participants at a rural PHC in Karnataka. MoCA scores and neurologist-confirmed ICD-10 diagnoses served as reference standards. Psychometric validation included ROC curve analysis, internal consistency (Cronbach's alpha), test-retest reliability, and interrater agreement.

At an optimal cutoff score of <45, the PHC-CS tool demonstrated an area under the ROC curve of 0.957 (95% CI: 0.922-0.992), with a sensitivity of 95%, a specificity of 92%, a positive predictive value of 96%, and a negative predictive value of 92%. Internal consistency was strong (Cronbach's alpha = 0.90), with good test-retest reliability (r = 0.88) and interrater agreement (κ = 0.88). The average administration time was 15 minutes.

The PHC-CS tool demonstrates promising preliminary validity and feasibility for routine cognitive screening in Indian PHC settings, particularly for low-literate populations. Further multicentric validation is recommended.