Objective:
To develop and evaluate the a brief, literacy-independent, culturally sensitive screening tool for detecting mild cognitive impairment (MCI) in Indian adults.
Background:
Common screeners (MMSE, SLUMS, Mini-Cog) and even adapted tools (MoCA-B) can be confounded by literacy and years of schooling, leading to misclassification in underserved populations. This is likely in part due to the nature of the content of questions. A practical, education-inclusive instrument is needed to improve early detection and equitable care.
Design/Methods:
We conducted a mixed-methods tool-development, preliminary analytic observational study at a tertiary care hospital in suburban India. Consecutive adults aged 60-80 years (N=120; 60 cognitively normal [CDR=0], 60 with MCI/very mild dementia [CDR=0.5]) completed the assessment (10 items; max score=30) in Marathi, Hindi, or English, followed by MoCA-B, MMSE, SLUMS, and Mini-Cog. Test-retest and inter-rater reliability were assessed ≥2 weeks apart. Diagnostic performance was examined via ROC analysis. MRI-based volumetry explored clinico-radiological correlations. Education-stratified ROC analyses and optimal cutoffs were derived.
Results:
Our assessment achieved accuracy 0.80 (95% CI 0.70-0.92), sensitivity 0.83 (0.70-0.92), and specificity 0.77 (0.68-0.90), outperforming MMSE (AUC 0.66), SLUMS (0.64), and Mini-Cog (0.64) and performing comparably to MoCA-B (AUC 0.79; p>0.99). DeLong tests favored our assessment over MMSE (p=0.018), SLUMS (p=0.028), and Mini-Cog (p=0.005). Reliability was robust (inter-rater kappa=0.48-0.74; highest for Memory/Recall kappa=0.74; ICCs up to 0.97; test-retest Spearman 0.62-0.96). It's performance was stable across education levels; optimal thresholds for cognitive impairment were 23 for 0-6 years and 24 for >6 years of schooling. Lower temporal/hippocampal volumes and higher deep white matter Fazekas scores were associated with poorer assessment scores; hippocampal volume correlated strongly with Memory/Recall (rho=0.77).
Conclusions:
Our assessment is a brief, reliable, and literacy-independent screener that detects MCI with higher accuracy than commonly used tools and performance comparable to MoCA-B, while minimizing educational bias. Prospective and cross-cultural validation studies are warranted.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.