Evaluating the Cultural Validity of the Montreal Cognitive Assessment: A Comparison of Performance in Bengali and English tests in Bangladeshi adults with Parkinson's disease in East London.

This study demonstrates that while the Bengali version of the Montreal Cognitive Assessment offers improved detection of cognitive differences in Bangladeshi adults with Parkinson's disease compared to the English version in East London, it still overestimates impairment, highlighting the critical need for culturally adapted tools and population-specific cut-offs.

The Gist

Imagine you are trying to measure the height of a group of people, but you only have one ruler: a ruler made for people who grew up in a specific climate. If you try to use that same ruler on people from a completely different climate, the measurements might look "off," not because the people are shorter or taller, but because the ruler itself doesn't fit their reality.

This paper is about doing exactly that, but with brain health instead of height.

The Problem: A One-Size-Fits-All Ruler

The MoCA (Montreal Cognitive Assessment) is like a popular, standard ruler used by doctors worldwide to check if people with Parkinson's disease are having trouble with their memory or thinking. It's usually given in English.

The researchers in East London noticed a problem: many of their patients with Parkinson's were Bangladeshi adults who spoke Bengali as their first language. They wondered: Is this English ruler fair for them? Maybe the test is too hard because of the language, not because their brains are actually struggling.

The Experiment: Two Rulers, One Group

To find out, the researchers played a game of "compare and contrast." They asked 50 people with Parkinson's and 22 healthy friends to take the test twice:

1. Once in English (the standard ruler).
2. Once in Bengali (a ruler made for their specific culture).

Think of it like asking someone to solve a math problem written in a language they barely know, and then asking them to solve the same problem in their native tongue. You'd expect them to do better in the second round, right?

The Results: The "Language Gap"

The results were clear and dramatic:

• The English Test: When the patients took the test in English, they scored very low. It looked like they had severe memory problems.
• The Bengali Test: When they took the same test in Bengali, their scores jumped up by about 4 points on average.

The Metaphor: Imagine a fish trying to climb a tree. If you judge the fish by its ability to climb, you'll say, "This fish is terrible at climbing!" But if you judge the fish by its ability to swim, it's an expert. The English test was asking the fish to climb a tree; the Bengali test let them swim.

The Twist: The Ruler Still Needs Adjusting

Here is the most important part: Even though the Bengali test was much better at spotting the difference between sick patients and healthy people, the scores were still a bit too low.

Even in their native language, the patients scored below the "passing grade" that doctors usually use. This suggests that while the language barrier was a huge part of the problem, the test itself might still be too strict for this specific group of people. It's like having a ruler that is now the right material (wood instead of plastic), but the markings are still a little too far apart.

The Takeaway

The paper concludes that:

1. Language matters: Giving a test in a patient's native language is crucial. It gives a much truer picture of their brain health.
2. Culture matters too: Even with the right language, we can't just copy-paste the "passing score" from the original English test. We need to create customized rules (cut-offs) specifically for Bengali speakers in London.

In short: You can't judge a book by its cover, and you can't judge a brain by a test written in a language it doesn't speak. To get a fair diagnosis, we need tools that fit the person, not just the disease.

Technical Summary

Technical Summary: Evaluating the Cultural Validity of the Montreal Cognitive Assessment in Bangladeshi PD Patients

1. Problem Statement
The Montreal Cognitive Assessment (MoCA) is a widely recommended brief screening tool for detecting cognitive impairment in individuals with Parkinson's disease (PD). However, its validity is often compromised when applied to non-native English speakers or culturally distinct populations without adaptation. This study addresses the critical gap in assessing cognitive function among Bangladeshi adults with PD residing in East London. Specifically, it investigates whether the standard English version of the MoCA is culturally and linguistically appropriate for this demographic or if the Bengali version offers superior diagnostic accuracy. The core issue is the potential for misdiagnosis—either false positives (overestimating impairment) or false negatives (missing deficits)—due to linguistic barriers and cultural mismatches in test items.

2. Methodology
The research employed a cross-sectional study design involving two distinct groups:

• Participants: 50 individuals diagnosed with Parkinson's disease (PD) and 22 healthy controls (HC), all of Bangladeshi origin residing in East London.
• Procedure: Participants underwent administration of both the English and Bengali versions of the MoCA.
• Statistical Analysis: The study utilized a robust statistical framework including:
  • ANCOVA (Analysis of Covariance): To adjust for potential confounding variables.
  • Paired and Unpaired t-tests: To compare scores between the two language versions within groups and between the PD and HC groups, respectively.
  • Bland-Altman Analysis: To assess the agreement and bias between the English and Bengali scores.
  • Stratification: Analyses were conducted on the full sample and a subset of age-matched samples (n=29 PD and 22 HC) to control for the confounding effects of age on cognitive performance.
• Validation: Bengali MoCA scores were cross-referenced with multidisciplinary clinical assessments to determine clinical alignment.

3. Key Results

• Language Effect: Both the PD and HC groups achieved significantly higher scores on the Bengali version of the MoCA compared to the English version. The mean difference was approximately 4 points ($p < 0.001$), indicating a substantial linguistic barrier in the English administration.
• Discriminative Validity:
  • In the English version, the study failed to detect a statistically significant difference between PD patients and healthy controls in the age-matched analysis.
  • In contrast, the Bengali version successfully detected significant differences between the PD and HC groups in the age-matched analysis ($p = 0.02$).
• Clinical Alignment: Scores obtained via the Bengali MoCA aligned more closely with independent multidisciplinary clinical assessments than the English scores did.
• Normative Limitations: Despite the improved performance in Bengali, the mean scores for the PD group remained below standard normative cut-offs, suggesting that even the translated version may still overestimate the severity of impairment when using standard thresholds.

4. Key Contributions

• Empirical Evidence of Linguistic Bias: The study provides quantitative evidence that the English MoCA is an inadequate tool for screening cognitive impairment in Bangladeshi PD patients, leading to a significant underestimation of their actual cognitive abilities.
• Validation of Translated Tools: It demonstrates that the Bengali MoCA offers superior discriminative validity in this specific population, correctly distinguishing between PD patients and healthy controls where the English version failed.
• Methodological Rigor: By employing age-matched controls and Bland-Altman analysis, the study isolates the effect of language from the natural cognitive decline associated with aging and PD.

5. Significance and Implications
The findings underscore the critical necessity of culturally adapted assessment tools and population-specific normative cut-offs in clinical practice.

• Clinical Practice: Relying on the English MoCA for Bangladeshi patients in the UK likely results in the over-diagnosis of cognitive impairment. Clinicians should prioritize the use of the native language version (Bengali) to ensure accurate screening.
• Policy and Research: The study highlights that translation alone is insufficient; the "cut-off" scores used to define impairment must be recalibrated for specific cultural and linguistic groups to avoid overestimating pathology. This supports a shift toward personalized, culturally competent neurology care for diverse immigrant populations.