Dementia is a syndrome that can be caused by a number of diseases which over time destroy nerve cells and damage the brain, typically leading to deterioration in cognitive function beyond what might be expected from the usual consequences of biological ageing. While consciousness is not affected, the impairment in cognitive function is commonly accompanied, and occasionally preceded, by changes in mood, emotional control, behaviour, or motivation ^{[1] [2]}. It has physical, psychological, social and economic impacts. There is often a lack of awareness and understanding of dementia, resulting in stigmatisation and barriers to diagnosis and care ^{[1] [3]}. The anticipated growth rate in dementia cases, expected to double every 20 years, underscores the urgency for effective diagnostic tools tailored to diverse populations and varying literacy levels, cultural contexts, and health challenges ^[4].

Early diagnosis of dementia is crucial for implementing strategies that may decelerate disease progression and improve quality of life for affected individuals ^[5]. However, diagnosing dementia, particularly in its nascent stages, remains a nuanced challenge, as initial symptoms are often subtle and easily misinterpreted as standard aging phenomena ^[6]. Cultural and linguistic barriers can further complicate the diagnosis, particularly within non-English speaking populations where traditional tools like the Mini-Mental State Examination (MMSE) may not fully translate or apply ^[7]. This inadequacy in tools necessitates a closer examination of alternatives, prompting the development of culturally and linguistically appropriate screening instruments.

In the context of dementia screening, several cognitive assessment tools have gained prominence, including the Hindi Mental State Examination (HMSE), the Montreal Cognitive Assessment (MoCA), and the Addenbrooke’s Cognitive Examination III (ACE III). These instruments differ significantly in their methodologies, the cognitive domains they evaluate, and their efficacy in detecting dementia and distinguishing between normal cognitive aging, MCI, and more severe dementia types ^[8]. For instance, the HMSE, a culturally adapted derivative of the MMSE, has shown promise in Indian populations ^[9], but its ability to detect early cognitive decline remains controversial ^[10]. Conversely, the MoCA is widely recognised for its sensitivity, particularly in identifying MCI, with studies indicating it can outperform the MMSE in this regard ^[11]. Lastly, the ACE III is an expansive tool that assesses various cognitive domains, providing a more thorough analysis of conditions like Alzheimer’s and frontotemporal dementia, albeit at the expense of longer administration times ^[12][13].

Only few studies compare HMSE, MoCA, and ACE III in the Indian context. This study exams the diagnostic accuracy of these instruments within a Hindi-speaking demographic^[14], so we planned in order to facilitate earlier interventions and better patient outcomes by analysing their sensitivity and specificity in detecting dementia ^[15].

Study Design

This prospective observational study was conducted to compare the diagnostic accuracy of three cognitive screening tools the HMSE, the MoCA, and the ACE III in assessing dementia. The study was conducted at a tertiary care centre, from January 2024 to January 2025. Ethical clearance was obtained from the Institutional Ethics Committee and all procedures followed the ethical principles outlined in the Declaration of Helsinki.

The sample size was determined using the expected AUC values for HMSE (0.79), MoCA (0.90), and ACE III (0.88) in detecting dementia ^[16]. Assuming a 95% confidence level, an alpha error of 0.05, and an absolute allowable error of 5%, the minimum required sample size was 182 participants. The final sample size was adjusted to 190 participants, ensuring sufficient statistical power.

Study Population

The study included individuals who attended the Psychiatry Outpatient Department at tertiary care centre, between January 2024 and January 2025.

Inclusion criteria-

Subjects aged 60 years or older, who fulfilled the DSM 5 criteria for Major Neurocognitive Disorder and provided written informed consent and who were fluent in Hindi were taken for this study.

Exclusion criteria-

Subjects with a history of major psychiatric disorders (eg. schizophrenia, bipolar disorder, major depressive disorder with psychotic symptoms, or other severe neuropsychiatric conditions) , recent history (past five years) of traumatic brain injury, stroke, epilepsy, or other major neurological disorders (eg. Parkinson’s disease or multiple sclerosis ), severe substance or alcohol use disorders, those on psychoactive medications with significant cognitive side effects, and with severe visual or hearing impairments. Subjects with significant uncontrolled chronic medical conditions(eg. decompensated liver disease, end-stage renal disease, severe cardiovascular disease) were not included. Eligible subject were recruited consecutively from outpatient clinics, ensuring representativeness while excluding confounding cases to maintain a focused population.

Assessment Tools

Hindi Mental State Examination (HMSE)

The HMSE is an adapted version of the Mini-Mental State Examination (MMSE) that has been translated and culturally modified for Hindi-speaking populations ^[17] which retains the core components of the MMSE while incorporating culturally appropriate modifications to improve its validity for Indian populations. The MMSE was originally developed by Folstein et al. ^[18], which consists of 30 items assessing cognitive domains such as orientation, registration, attention, calculation, recall, and language abilities.

Montreal Cognitive Assessment (MoCA)

The MoCA is a 30-point cognitive screening tool designed to detect early cognitive impairment, particularly mild cognitive impairment ^[16]. It evaluates multiple domains, including attention, memory, visuospatial ability, executive function, and language. The Hindi MoCA has been validated for use among Indian populations, with a suggested cutoff of <26 for detecting cognitive impairment ^[19]. Studies indicate that MoCA has higher sensitivity (90%) for early-stage dementia detection compared to the MMSE/ HMSE (62%).

 Addenbrooke’s Cognitive Examination III (ACE III)

The ACE III is a 100-point cognitive screening tool designed to detect mild to moderate cognitive impairment with a focus on attention, memory, language, verbal fluency, and visuospatial function ^[20]. The Hindi version has been validated in Indian populations, making it a useful screening tool. The ACE III has been found to have a high diagnostic sensitivity and specificity ^[20], making it a valuable tool for differentiating between various types of dementia, including Alzheimer’s disease, frontotemporal dementia, and other cognitive impairments.

Assessment Procedure

The assessment procedure involved a detailed clinical evaluation that included a comprehensive medical and neurological examination. Basic demographic information was collected, including age, gender, education level, and occupation. The participants’ medical history was also assessed, focusing on cognitive decline, family history of dementia, vascular risk factors.

The Hindi version of the HMSE was administered in a quiet environment, with scores ranging from 0 to 30. Score 25 to 30 = normal cognition, score 20 to 25 = mild cognitive impairment, score 10 to 18 = moderate cognitive impairment and score <10 suggestive of severe cognitive impairment. The MoCA was also administered in Hindi, focusing on assessing executive function, attention, and memory, with adjustments made for low educational levels, and taking approximately 10-15 minutes to complete. MoCA score 18 to 25 is suggestive of mild cognitive impairment, score 10 to 17 is suggestive of moderate cognitive impairment and score <10 suggestive of severe cognitive impairment. Lastly, the ACE III, a comprehensive 100-point test, was used to assess orientation, memory, verbal fluency, language, and visuospatial abilities. The established cut-off scores for probable dementia and mild cognitive impairment (MCI) were 82 and 88, respectively, as per Hsieh et al. ^[20].

Statistical Analysis

The dataset's normality was assessed using the Shapiro-Wilk test. Descriptive statistics are presented as means ± standard deviations or as frequencies (percentages). For comparisons, the independent samples t-test or the chi-squared (χ²) test was utilised as appropriate. The diagnostic accuracy of the three tools was evaluated through sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Receiver Operating Characteristic (ROC) curve analysis. The Area Under the Curve (AUC) was calculated to compare the tools' effectiveness in detecting mild cognitive impairment (MCI) and dementia. The correlation among HMSE, MoCA, and ACE III scores was assessed using Pearson’s correlation coefficient (r). Data analysis was conducted using SPSS version 23.0.0 (IBM, New York, USA), and statistical significance was determined at p < 0.05.

The study included 190 participants, with 100 patients diagnosed with dementia and 90 cognitively healthy controls. The demographic and cognitive characteristics of both groups are presented in Table 1. The mean age of participants was similar in the dementia group (72.5 ± 7.2 years) and the control group (71.3 ± 6.5 years; p > 0.05). Education levels were also similar in the dementia group (6.8 ± 3.2 years) and control group (10.7 ± 4.5 years; p <0.001). However, the cognitive scores were significantly lower in the dementia group across all three screening tests (ACE-III: 49.2 ± 11.5 vs. 87.3 ± 7.8, p < 0.001; HMSE: 17.8 ± 4.2 vs. 27.9 ± 2.1, p < 0.001; MoCA: 17.1 ± 3.9 vs. 26.4 ± 2.8, p < 0.001), confirming their diagnostic value in distinguishing individuals with dementia from cognitively intact controls.

To assess the diagnostic accuracy of the three cognitive screening tools in identifying dementia, ROC curve analysis was conducted. The optimal cutoff points for ACE-III, HMSE, and MoCA were determined using Youden index. The sensitivity, specificity, and effect sizes for each test are presented in Table 2.

In table 3, The optimal cutoff for ACE-III was found to be 71, with a sensitivity of 91% and specificity of 89% (AUC = 0.93, Cohen’s d=2.8), making it the most sensitive tool among the three for detecting dementia. The MoCA demonstrated a sensitivity of 87% and specificity of 85% at a cutoff score of 23 (AUC = 0.90, Cohen’s d=1.83), indicating its utility for early dementia detection. The HMSE had a sensitivity of 83% and specificity of 82% at a cutoff score of 20 (AUC = 0.86, Cohen’s d=1.6), making it a reliable, yet slightly less sensitive, screening tool compared to the ACE-III and MoCA.

Correlation analysis was performed to examine the relationship between the three cognitive assessment tools. The results showed that all three tests were significantly correlated (p < 0.001) with each other. The highest correlation was observed between ACE-III and MoCA (r = 0.87, p < 0.001), indicating strong agreement between these two tools. The correlation between HMSE and ACE-III was also high (r = 0.80, p < 0.001), while the correlation between HMSE and MoCA was slightly lower (r = 0.78, p < 0.001). These findings suggest that all three tests measure overlapping cognitive domains, with ACE-III and MoCA showing the strongest correlation.

The results indicated that the ACE-III was the most accurate tool for detecting dementia, followed by MoCA and then HMSE. Based on the ROC curve analysis, the optimal cutoff scores for each test were determined to maximise both sensitivity and specificity. The high AUC values (all above 0.80) confirm the diagnostic validity of the tests, with the ACE-III showing the strongest discriminative ability (AUC = 0.93), followed by MoCA (AUC = 0.90) and HMSE (AUC = 0.85). The study established cutoff scores for differentiating dementia cases from healthy controls, suggesting that an ACE-III score ≤ 71, HMSE ≤ 20, and MoCA ≤ 23 provided the best balance of sensitivity and specificity in distinguishing dementia from normal aging.

The primary objective of this study was to compare the diagnostic accuracy of three commonly used cognitive screening tools the HMSE, the MoCA, and the ACE-III in detecting dementia. The study included 190 participants, comprising 100 patients diagnosed with dementia and 90 cognitively healthy controls. The main findings indicate that ACE-III demonstrated the highest sensitivity (91%) and specificity (89%) at a cutoff score of 71. Previous studies, such as Hsieh et al. ^[20], have also reported high sensitivity (93%) and specificity (89%) for ACE-III, making it a valuable tool for early and accurate detection . The study revealed that ACE-III had the highest diagnostic accuracy with an AUC of 0.93, demonstrating excellent discriminatory power in detecting dementia. This finding aligns with previous research indicating that ACE-III provides a comprehensive assessment of cognitive domains, making it a superior tool in clinical practice ^[20][6].  Matías-Guiu et al. ^[6] reported an AUC score of 0.906 for ACE-III, while we found an AUC score of 0.930 for ACE-III which shows the highest diagnostic accuracy.

MoCA also showed strong diagnostic performance (sensitivity 87%, specificity 85% with cutoff score of 23. The cutoff score in our study is consistent with that reported in previous studies ^[16], which found optimal MoCA cutoffs of 22–24 for dementia detection.

Meanwhile, HMSE had the lowest diagnostic accuracy (sensitivity 83%, specificity 80%) at a cutoff of 24 and AUC of 0.85 ,consistent with studies suggesting its limited ability to detect early cognitive impairment ^[23]. The poor sensitivity of HMSE in detecting early dementia suggests, despite of being widely used in clinical setting ,it may not be as effective as MoCA or ACE-III for identifying mild cognitive impairment ^[21]. HMSE primarily focuses on orientation, memory, and attention but lacks a comprehensive assessment of executive function and visuospatial abilities ^[23].

When considering memory-specific performance, ACE-III demonstrated a clear advantage over HMSE and MoCA in discriminating dementia patients from cognitively healthy controls. This is in accordance with Noone ^[22], who also reported that ACE-III had higher sensitivity for detecting early cognitive decline. The ACE-III score for dementia patients (49.2 ± 11.5) was significantly lower than that of controls (87.3 ± 7.8, p < 0.001), highlighting its superior ability to distinguish between cognitive impairment and normal aging. In comparison, MoCA (dementia: 17.1 ± 3.9; controls: 26.4 ± 2.8, p < 0.001) and HMSE (dementia: 17.1 ± 3.9; controls: 27.4 ± 2.8, p < 0.001) scores also demonstrated significant differences. However, the higher effect size (2.8) of ACE-III suggests its superiority in detecting cognitive impairment across multiple domains, compared to MoCA (1.83) and HMSE (1.6). In this study the observed effect sizes are slightly deviated from previous research, which observed 1.80, 1.53, and 1.60 for ACE III, MoCA and HMSE, respectively ^[6] which could be because of relatively moderate sample size.

The correlation analysis revealed a strong positive correlation between ACE-III and MoCA (r = 0.88, p < 0.001), consistent with findings from previous research, which reported r = 0.91 ^[10]. This suggests that while both tests assess multiple cognitive domains, ACE-III provides a more detailed evaluation. We also observed a moderate correlation between the HMSE and the ACE-III with a correlation coefficient of r = 0.76. This correlation is slightly lower than what was reported previously ^[6], where they found r = 0.87 which could be because the representative population was largely from  the rural background.

The correlation between HMSE and MoCA (r = 0.72) is slightly lower than previously reported (r = 0.82) ^[6], supporting previous studies that MoCA is more effective in detecting early cognitive impairment compared to HMSE ^[24].

Our findings are consistent with past research that has indicated the higher sensitivity of MoCA and ACE-III compared to HMSE. A study by Matias-Guiu et al. ^[6] demonstrated ACE-III’s effectiveness in detecting mild to moderate dementia, making it a preferred clinical tool. Similarly, Freitas et al. ^[25] found MoCA more sensitive than HMSE, especially in highly educated individuals, reinforcing HMSE’s limitations in early-stage dementia detection. HMSE primarily focuses on orientation, memory, and attention but lacks a comprehensive assessment of executive function and visuospatial abilities ^[23].

Despite the study’s significant contributions there are some limitations. The sample size (n = 190) being relatively moderate might affect the generalisability of our findings. Additionally, the study was conducted in a single tertiary care centre, potentially limiting external validity.  While ACE-III showed superior sensitivity and specificity, its longer administration time compared to MoCA and HMSE could limit its feasibility in busy clinical settings. Future research should evaluate the practicality of implementing ACE-III in routine clinical practice while exploring additional longitudinal studies to assess the predictive accuracy of these tools over time.

The findings suggest that ACE-III is the most effective screening tool for diagnosing dementia in Hindi-speaking older adults, followed closely by MoCA, while HMSE, although useful, demonstrated slightly lower sensitivity and specificity. Given its superior diagnostic accuracy and comprehensive cognitive assessment, ACE-III may be the preferred tool in clinical settings. The results also reinforce the need for education-adjusted cutoff scores, as cognitive test performance was significantly influenced by literacy levels. Future research is needed to validate these findings in larger populations in diverse settings.

Acknowledgements

None.

Financial Support and Sponsorship

Nil.

Declaration of Interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

1. World Health Organization. Dementia [Internet]. Geneva: World Health Organization; 2022 Sep 20 [cited 2025 Mar 18]. Available from: https://www.who.int/news-room/fact-sheets/detail/dementia
2. Kan K, Subramaniam P, Shahrizaila N, Kamaruzzaman S, Razali R, Ghazali S. Validation of the Malay version of Addenbrooke’s Cognitive Examination III in detecting mild cognitive impairment and dementia. Dement Geriatr Cogn Disord Extra. 2019;9(1):66-76. doi: 10.1159/000495107.
3. Wang B, Zheng H, Xu C, Sun Y, Zhang Y, Shi J. Comparative diagnostic accuracy of ACE-III and MoCA for detecting mild cognitive impairment. Neuropsychiatr Dis Treat. 2019;15:2647-2653. doi: 10.2147/ndt.s212328.
4. Peixoto B, Baeta É, Machado M, Rocha P, Macedo C, Machado A, et al. Diagnostic utility of the Portuguese version of the Mini-Addenbrooke’s Cognitive Examination in early dementia. Geropsych. 2019;32(4):175-180. doi: 10.1024/1662-9647/a000214.
5. Patel D, Patel R, Arya V, Shah R, Sood S. Evaluation of cognitive function in patients with type 2 diabetes mellitus: comparison with age-matched controls: a prospective study. Apollo Med. 2023;20(4):329-335. doi: 10.4103/am.am_52_23.
6. Matías-Guiu J, Cortés-Martínez A, Valles-Salgado M, Rognoni T, Fernández-Matarrubia M, Moreno-Ramos T, et al. Addenbrooke’s Cognitive Examination III: diagnostic utility for mild cognitive impairment and dementia and correlation with standardized neuropsychological tests. Int Psychogeriatr. 2017;29(1):105-113. doi: 10.1017/s1041610216001496.
7. Ghosh A, Monisha S, Sunny A, Diwakar L, Issac T. Prevalence and patterns of vitamin D deficiency and its role in cognitive functioning in a cohort from South India. Sci Rep. 2024;14(1). doi: 10.1038/s41598-024-62010-5.
8. Alirezaei Z, Amouheidari A, Hassanpour M, Davanian F, Iraji S, Shokrani P, et al. Early detection of radiation-induced injury and prediction of cognitive deficit by MRS metabolites in radiotherapy of low-grade glioma. Biomed Res Int. 2021;2021:6616992. doi: 10.1155/2021/6616992.
9. Qassem T, Khater M, Emara T, Rasheedy D, Tawfik H, Mohammedin A, et al. Validation of the Egyptian-Arabic version of the Addenbrooke’s Cognitive Examination III (ACE-III) in diagnosing dementia. Dement Geriatr Cogn Disord. 2020;49(2):179-184. doi: 10.1159/000507758.
10. Yu R, Lai J, Hui E, Mukadam N, Kapur N, Stott J, et al. Systematic review and meta-analysis of brief cognitive instruments to evaluate suspected dementia in Chinese-speaking populations. J Alzheimers Dis Rep. 2023;7(1):973-987. doi: 10.3233/adr-230024.
11. Charernboon T. Diagnostic accuracy of the Thai version of the Mini-Addenbrooke’s Cognitive Examination as a mild cognitive impairment and dementia screening test. Psychogeriatrics. 2019;19(4):340-344. doi: 10.1111/psyg.12417.
12. Senda M, Terada S, Takenoshita S, Hayashi S, Yabe M, Imai N, et al. Diagnostic utility of the Addenbrooke’s Cognitive Examination-III (ACE-III), Mini-ACE, Mini-Mental State Examination, Montreal Cognitive Assessment, and Hasegawa Dementia Scale-Revised for detecting mild cognitive impairment and dementia. Psychogeriatrics. 2019;20(2):156-162. doi: 10.1111/psyg.12480.
13. Gogisetti Y, Pathania M, Mittal S, Yadav P, Kharibam P, Kant R. Assessment of cognition in hypertensives and normotensives: a comparative P300 study. Cureus. 2022. doi: 10.7759/cureus.28397.
14. North C, Heatley M, Utoomprurkporn N, Bamiou D, Costafreda S, Stott J. Adaptation and preliminary validation of the Addenbrooke’s Cognitive Examination-III as a screening test for mild cognitive impairment and dementia in hearing-impaired individuals. Eur J Neurol. 2021;28(6):1820-1828. doi: 10.1111/ene.14753.
15. Kant S, Poddar K, Kamle M, Patil C. Assessment of cognitive impairment by using Addenbrooke’s Cognitive Examination (ACE) and Montreal Cognitive Assessment (MoCA) amongst type 2 diabetes mellitus patients in Eastern Uttar Pradesh, India. Indian J Neurosci. 2020;4(4):185-190. doi: 10.18231/2455-8451.2018.0044.
16. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695-699.
17. Ganguli M, Ratcliff G, Chandra V, Sharma S, Gilby J, Pandav R, Belle S, Ryan C, Baker C, Seaberg E, Dekosky S. A Hindi version of the MMSE: The development of a cognitive screening instrument for a largely illiterate rural elderly population in India. Int J Geriatr Psychiatry. 1995;10(5):367– doi: 10.1002/gps.930100505.
18. Folstein MF, Folstein SE, McHugh PR. "Mini-Mental State." A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–198. doi: 10.1016/0022-3956(75)90026-6.
19. Gupta M, Gupta V, Nagar Buckshee R, Sharma V. Validity and reliability of Hindi translated version of Montreal Cognitive Assessment in older adults. Asian J Psychiatry. 2019;45:125–128. doi: 10.1016/j.ajp.2019.09.022.
20. Hsieh S, Schubert S, Hoon C, Mioshi E, Hodges JR. Validation of the Addenbrooke’s Cognitive Examination III in frontotemporal dementia and Alzheimer's disease. Dement Geriatr Cogn Disord. 2013;36(3-4):242– doi: 10.1159/000351671.
21. Ismail Z, Rajji TK, Shulman KI. Brief cognitive screening instruments: An update. Int J Geriatr Psychiatry. 2010;25(2):111-120.
22. Noone P. Addenbrooke's Cognitive Examination-III. Occup Med (Lond). 2015;65(5):418– doi: 10.1093/occmed/kqv041.
23. Arevalo-Rodriguez I, Smailagic N, i Figuls MR, Ciapponi A, Sanchez-Perez E, Giannakou A, Pedraza OL, Bonfill Cosp X, Cullum S. Mini-Mental State Examination (MMSE) for the detection of Alzheimer's disease and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev. 2015;(3):CD010783.
24. Luis CA, Keegan AP, Mullan M. Cross-validation of the Montreal Cognitive Assessment in community-dwelling older adults with mild cognitive impairment. Aging Clin Exp Res. 2009;21(2):129-134.
25. Freitas S, Simões MR, Marôco J, Alves L, Santana I. Construct validity of the Montreal Cognitive Assessment (MoCA). J Int Neuropsychol Soc. 2012;18(2):242–250. doi: 10.1017/S1355617711001573.