Context: Mild cognitive impairment (MCI) is an under-diagnosed health problem in the community. Cognitive screening tools are widely used for MCI detection, but many of them lack sensitivity and specificity in MCI detection. On the basis of literature review, Short test of mental status (STMS) was selected for the present purpose. Aims: The present study purports to standardize STMS for using as a cognitive screening tool in MCI detection in the community-living elderly people in Kolkata. Materials and Methods: Data were collected from 102 community-living elderly people from the city of Kolkata using the purposive method of sampling. MCI was diagnosed using the Peterson's criteria of MCI as the gold standard. A semi-structured demographic proforma, clock-drawing test (CDT), and the Groningen Activity Restriction Scale (GARS) were used for the purpose. Finally, STMS was administered. Statistical Analysis: Statistical computation was done using the SPSS 21. Descriptive statistics, receiver operating curve analysis, and binary logistic regression were used for statistical analysis of the data. Results and Conclusion: STMS emerged as a sensitive and specific cognitive screening tool in the detection of MCI in the current cultural setting. It was also found to be more suited for the purpose than CDT. A score of 34.5 with a sensitivity of 97.5 and a specificity of 90.3% were selected for the optimum cutoff score for the detection of MCI in the present population. With a unit increase in STMS score, the odds of getting diagnosed with MCI were found to be reduced by 48.5%. © 2019 Indian Journal of Psychiatry Published by Wolters Kluwer - Medknow.