Screening Cognitive Impairment in Older Adults: An ICT-Based Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Screening Cognitive impairment in Older adults: an ICTs-based study

I congratulate the authors for their study on assessing the ICTs based methods in cognitive screening in older adults in Spain which was warranted during COVID-19 pandemic situation. However, the present study requires minor modifications and justifications for it to be fit for publication. The detailed review of the article is given below:

Comments to Authors:

1. The whole manuscript requires English language editing especially the introduction part is missing coherence.
2. The main aim of the study was to evaluate the differences between online and offline performance of individuals on multiple cognitive screening tests. However, the instroduction section does not talk about the review of available studies on the validation of ICT based assessment methods such as MoCA.
3. The paragraph on incidence and prevalence from line numbers 70 to 76 seems irrelevant in introduction section. Similarly the criteria of DSM-5 and other approaches to cognitive screening is not apt in this section.
4. The authors may elaborate information about the selection of relevant variables to determine the risk for developing cognitive impairments as given in line number 137 to 140.
5. The authors may elaborate on the sample size calculation as the population-based studies require large samples to generalize the results. Provide a justification for inclusion of only 148 participants which may limit the generalizability of results.
6. The present study included only normal participants without any subjective memory complaints. It would be more appropriate if the same study was carried out on patients with complaints of memory impairments as there would be significant differences between normal individuals and individuals with memory impairments either in offline or online modes. Hence, the results cannot be generalized to individuals with memory impairments.
7. The authors mentioned the mean, SD and range for low and high education levels. However, in the demographic details, it is given as postgraduate, primary, secondary and high school levels rather than low and high. Can you please elaborate on these? Further, are these differences between low and high statistically significant?
8. In line number 409, it is mentioned that high education includes both university and high school. This is debatable as in many countries, the high education level is only considered for studies at graduate level and above. This is going to be important as there are quite a large number of studies proving the relation between education level and onset of MCI in older adults.
9. Discussion section requires English language editing as there is no continuity between the paragraphs.

Comments on the Quality of English Language

The whole manuscript requires English language editing especially the introduction & discussion parts are missing coherence.

Author Response

On behalf of the co-authors of our research, we want to thank you for your contributions and comments on the research article submitted to the journal; we will be pleased to receive further comments once you receive our text modified with your suggestions, which undoubtedly has improved and enriched the final version.

We have made the changes described below, which are included and marked in the final version of the document, based on the publisher's instructions. We have responded to your observations, comments, or suggested changes in the order in which they appear.

1. Reviewer inquiry: The whole manuscript requires English language editing; especially the introduction part is missing coherence.

Response: We would like to point out that the whole text has been revised by a professional translation service to improve the final wording.  Therefore, we believe that this improved version meets the quality standards of a high-impact publication.

2. Reviewer inquiry: The main aim of the study was to evaluate the differences between the online and offline performance of individuals on multiple cognitive screening tests. However, the introduction section does not talk about the review of available studies on the validation of ICT-based assessment methods such as MoCA.

Response: We agree with you in the sense that, since the aim of our study was to explore cognitive variables in the elderly during COVID-19 confinement using both online and face-to-face assessment modalities, we failed to mention any reference of validations of the instruments used in the online modality, so we have added two references of previously conducted studies in this regard in lines 41-59 of the final version of the text.

3. Reviewer inquiry: The paragraph on incidence and prevalence from line numbers 70 to 76 seems irrelevant in the introduction section. Similarly, the criteria of DSM-5 and other approaches to cognitive screening is not apt in this section.

Response: After re-reading the document, we have found that lines 70-87 of the original document are redundant because they are aspects that are mentioned in other parts of the text, so we have deleted them.

4. Reviewer inquiry: The authors may elaborate information about the selection of relevant variables to determine the risk for developing cognitive impairments as given in lines number 137 to 140.

Response: A paragraph has been added to this line to further explain that as a research team we can't know the risk of developing cognitive impairment beforehand, hence the importance of the screening protocol, whether online or face to face, because once it's applied, we can detect indicators and early signs of cognitive impairment and therefore take action to address them (lines 115-130)

5. Reviewer inquiry: The authors may elaborate on the sample size calculation, as population-based studies require large samples to generalize the results. Justify the inclusion of only 148 participants, which may limit the generalizability of results.

6. Reviewer inquiry: The present study included only normal participants without any subjective memory complaints. It would be more appropriate if the same study was carried out on patients with complaints of memory impairments as there would be significant differences between normal individuals and individuals with memory impairments either in offline or online modes. Hence, the results cannot be generalized to individuals with memory impairments.

Response to 5 and 6 inquiries: We are aware that a larger sample would help us to obtain more meaningful results and avoid the risk of bias or generalisation in the results. However, as explained in the modified version of our text, the population served is based on requests from residents of the city over the age of 60.

Therefore, it is a random sample (it is not stratified because there are no previous categories and it is formed until the number of participants in each of them is found) in this process, the research team does not have any capacity to select the sample or knowing which of them have cognitive impairment risk at the first moment. For this reason, we have expanded this consideration in the conclusions as one of the limitations of our study, but despite this, we believe that being an exploratory study with this sample, we have obtained important results that may be useful for people interested in this topic (lines 493-503).

7. Reviewer inquiry: The authors mentioned the mean, SD, and range for low and high education levels. However, in the demographic details, it is given as postgraduate, primary, secondary, and high school levels rather than low and high. Can you please elaborate on these? Further, are these differences between low and high statistically significant?

Response: Thanks for this important observation. For a better understanding following previous literature, we have redesigned Table 1 by organizing the groups into two blocks: higher and lower education.  (Lines 286-287). Also, we comment on the differences between low and high education, which are statistically significant in the response to the bellow inquiry.

8. Reviewer inquiry: In line number 409, it is mentioned that high education includes both university and high school. This is debatable as in many countries, the high education level is only considered for studies at graduate level and above. This is going to be important as there are quite a large number of studies proving the relation between education level and onset of MCI in older adults.

Response: First, we confirm that the differences between participants with low and high educational levels were statistically significant across several cognitive and informant-based measures. These results are reported in the Results section and have been further clarified to highlight the statistical relevance (lines 311-314).

Regarding the classification of educational level, our approach grouped participants into two categories: low education (primary and secondary education) and higher education (high school and university). This classification follows the structure of the Spanish educational system, in which high school is considered a post-compulsory academic track that prepares students for university and differs substantially from lower secondary education.

We fully acknowledge that this grouping may not align with international standards, where "higher education" often refers exclusively to post-secondary or university-level studies. To address this, we have added a paragraph in the Discussion section explaining this methodological choice and encouraging caution when comparing results across studies conducted in different educational systems. We also emphasize that previous literature has found cognitive reserve to be an important preventive factor to Mild Cognitive Impairment (Lines 437-444).

9. Reviewer inquiry: The Discussion section requires English language editing as there is no continuity between the paragraphs.

Response: As mentioned before, the whole text has been revised by a professional translation service to improve the final wording.  Therefore, we believe that this improved version meets the quality standards of a high-impact publication.

We hope we have answered all your questions and welcome any further comments you may have.

Thank you for taking the time to analyse our work and for suggesting improvements.

Best Regards,

Correspondent author.

Reviewer 2 Report

Comments and Suggestions for Authors

This is an interesting manuscript that addresses a very important topic such as cognitive aging and related pathologies.

The proposed article is well formulated, with a clear structure, well argued and complete. However, I propose the following modifications in order to improve it for its publication.

Introduction: why in line 127 is it written "...mild cognitive impairment (MCI)...". This specification is already done before in the introduction section. Then, it should be written directly MCI or mild cognitive impairment. The rest of the acronyms should be reviewed to be consistent, and if what they mean has already been specified previously, either use the acronym directly or put the full word, but do not repeat it both.

Method is very clear and well explained, it does not need any modifications in my opinion.

Results: In Table 4, the correlation between IDDD abd WAT has a p-value of .000 but the asterisks do not appear as it is a statistically significant correlation, nor have I found any comments on this. Is this correct? I recommend reviewing this table and its explanation.

In relation to the previous comment, although the SPSS output gives p-values of 0.000, we know that this is not correct or exact, rather that these are very small p-values but not equal to 0. These p-values must be modified and set to <0.001.

I am not sure I understant the explanation of Figure 4. What does each different colored column refer to? The information regarding this figure needs to be expanded and improved.

Discussion and conclusions are well explained. The most important limitation, in my opinion, is the lack of formal diagnosis of the participants prior to the study, but is already explained by the authors. Perhaps it could be added the implications of this limitation on the obtained results in the present study.

Author Response

Dear Reviewer 2:

On behalf of the co-authors of our research, we want to thank you for your positive feedback towards our work, also for the contributions and comments to the research article submitted to the journal; we will be pleased to receive further comments once you receive our text modified with your suggestions, which undoubtedly has improved and enriched the final version.

We have made the changes described below, which are included and marked in the final version of the document, based on the publisher's instructions. We have responded to your observations and comments and suggested changes in the order in which they appear.

RESPONSE JUSTIFICATION

1. Reviewer inquiry: Introduction: why in line 127 is it written "...mild cognitive impairment (MCI)...". This specification is already done before in the introduction section. Then, it should be written directly as MCI or mild cognitive impairment. The rest of the acronyms should be reviewed to be consistent, and if what they mean has already been specified previously, either use the acronym directly or put the full word, but do not repeat both.

Response: Thank you for this important observation. We have thoroughly checked that the acronyms used are not repeated.

2. Reviewer inquiry: Results: In Table 4, the correlation between IDDD and WAT has a p-value of .000, but the asterisks do not appear as it is a statistically significant correlation, nor have I found any comments on this. Is this correct? I recommend reviewing this table and its explanation.

Response: We have highlighted the significant correlation between IDDD and WAT, as well as a paragraph detailing why a negative correlation between autonomy level and cognitive decline and the patient's cognitive status exists (Lines 333-337).

3. Reviewer inquiry: In relation to the previous comment, although the SPSS output gives p-values of 0.000, we know that this is not correct or exact, rather that these are very small p-values but not equal to 0. These p-values must be modified and set to <0.001.

Response: It’s already been changed in the table (line 326).

4. Reviewer inquiry: I am not sure I understand the explanation of Figure 4. What does each different colored column refer to? The information regarding this figure needs to be expanded and improved.

Response: In the graph provided, the blue bars represent individuals with higher educational levels, such as those who have completed high school or university, while the orange bars correspond to individuals with lower educational levels, such as those with only primary education or secondary school. The chart compares performance on various cognitive and functional measures, highlighting differences potentially related to cognitive reserve based on educational attainment. We added a note to the graph for better understanding.

5. Reviewer inquiry: Discussion and conclusions are well explained. The most important limitation, in my opinion, is the lack of formal diagnosis of the participants prior to the study, but is already explained by the authors. Perhaps it could be added the implications of this limitation on the obtained results in the present study.

Response: We have included a paragraph reflecting this aspect as a limitation of our study, with the intention that it can be improved in future research. (493-502)

We hope we have answered all your questions and welcome any further comments you may have.

Thank you for taking the time to analyse our work and for suggesting improvements.

Best Regards,

Correspondent author.

Reviewer 3 Report

Comments and Suggestions for Authors

It seems that the authors are late to the study's objectives, which is to validate digital psychometric assessment to deal with cases where face-to-face assessment is complicated or not possible, as was the case during the COVID pandemic. Currently there are digital versions of a large majority of psychometric tools even with their own websites, such as MoCA, IQCODE and the electronic applications of the MMSE, CDR-SB, Clock Test, ADAS cog and other psychometric assessment packages that include multiple tests and a final assessment such as the Neuropsychological Test Battery by ePROVIDE.

Furthermore, they do not justify many aspects that support the study:

1. Why do they choose 60 years as the cut-off age when in Western countries population studies come referring to over 65 or even 75 years old which is when cognitive impairment is more prevalent?
2. The objectives of the study are not clear although they come in the summary of a general way of analysing the cognitive variables of the elderly during confinement by COVID using tele-neuropsychology and comparing it with face-to-face assessment.
3. The authors confuse terms such as MCI and dementia which are well defined by the new NIA-AA (https://doi.org/10.1002/alz.13859) and the IWT criteria (https://jamanetwork.com/journals/jamaneurology/fullarticle/2825806) and which are not subject to the detailed and overly detailed interpretation of the various psychometric tests. The difference between MCI and mild dementia is the need or not for supervision in ADLs, it is as simple as that. The DSM-V is not used as an element to differentiate these two entities (line 107-108).
4. They seem to have to justify the use of the tools they have in their Memory Unit despite the redundancy in using MoCA and the Cacho clock test or two ADL scales, IQCODE and IDDD.
5. The references do not correspond to what is written:

Lines 63-65: where the use of both MoCA and the clock test is supported.

Line 66: Reference 15: No wording the Mini-Clock.

6. Lines 200-203: Reference 39 to endorse the MoCA cut-off points to distinguish MCI and dementia, does not justify it. The cut-off point should be 26 (Montreal Cognitive Assessment for the detection of dementia https://doi.org/10.1002/14651858.CD010775.pub3).
7. Lines 246-248. Reference 42 where there is no mention to the IQCODE neither to the clock test. The recommended reference is: The screening of mild dementia with a shortened Spanish version of the ‘Informant Questionnaire on Cognitive Decline in the Elderly’. doi: 10.1097/00002093-199509020-00008Montreal Cognitive Assessment for the detection of dementia. https://doi.org/10.1002/14651858.CD010775.pub3
8. Line 255-260: Reference 44 is not Peres et al, but Cattaneo G, Solana-Sánchez J, Abellaneda-Pérez K, Portellano-Ortiz C, Delgado-Gallén S, Alviarez Schulze V, et 757 al. Sense of Coherence Mediates the Relationship Between Cognitive Reserve and Cognition in Middle-Aged 758 Adults. Frontiers in Psychology. 2022 Mar 28;13. Where no mention is made to the Spanish version of the IDDD or to the advantage of using it together with the IQCODE.
9. Lines 299-300 They do not explain in which cases they apply the significance level of p ≤ 0.05 or p≤ 0.01.
10. Lines 543-607 The conclusion section is excessive, it should have no more than two well-supported paragraphs.

Author Response

Dear Reviewer 3

On behalf of the co-authors of our research, we want to thank you for the contributions and comments to our paper; we will be pleased to receive further comments once you receive our text modified with your suggestions, which undoubtedly has improved and enriched the final version.

We have made the changes described below, which are included and marked in the final version of the document, based on the publisher's instructions. We have responded to your observations, comments, and suggested changes in the order in which they appear.

Reviewer inquiry (RI)

Response (R)

1. RI: It seems that the authors are late to the study's objectives, which is to validate digital psychometric assessment to deal with cases where face-to-face assessment is complicated or not possible, as was the case during the COVID pandemic. Currently there are digital versions of a large majority of psychometric tools even with their own websites, such as MoCA, IQCODE and the electronic applications of the MMSE, CDR-SB, Clock Test, ADAS cog and other psychometric assessment packages that include multiple tests and a final assessment such as the Neuropsychological Test Battery by ePROVIDE.

R: The aim of our study is not to validate the psychometric instruments, as they have already been validated in both modalities. The aim is to find early indicators of MCI in the COVID contingency by assessing the differences between the online/face-to-face modalities. To make this clearer, we have added a paragraph to further explain this issue. (Lines 115-130).

2. RI: Why do they choose 60 years as the cut-off age when in Western countries, population studies come referring to over 65 or even 75 years old, which is when cognitive impairment is more prevalent?

R: The participants in this study are the population served by the Cognitive Impairment Unit of Salamanca City Council, whose users are people aged 60 and over. However, we did not mention this at the beginning, offer a better explanation concerning this in the modified version of our text, the population served is based on requests from residents of the city over the age of 60. (Lines 115-130).

Therefore, it is a random sample (it is not stratified because there are no previous categories and it is formed until the number of participants in each of them is found) in this process, the research team does not have any capacity to select the sample or knowing which of them have cognitive impairment risk at the first moment. For this reason, we have expanded this consideration in the conclusions as one of the limitations of our study, but despite this, we believe that being an exploratory study with this sample, we have obtained important results that may be useful for people interested in this topic (lines 493-503).

3. RI The objectives of the study are not clear although they come in the summary of a general way of analysing the cognitive variables of the elderly during confinement by COVID using tele-neuropsychology and comparing it with face-to-face assessment.

R: The aim of our study arose from the need to care for the elderly population, which at that time was vulnerable due to their confinement. Once we identified the need to provide cognitive stimulation based on knowledge of their cognitive status to define the most appropriate strategies to implement, our objective emerged: to find indicators of MCI in the elderly in the context of the COVID contingency, assessing the differences found in the application of online/in-person modalities. We have expanded the information in the new versión of the text for better understanding.

4. RI: The authors confuse terms such as MCI and dementia which are well defined by the new NIA-AA (https://doi.org/10.1002/alz.13859) and the IWT criteria (https://jamanetwork.com/journals/jamaneurology/fullarticle/2825806) and which are not subject to the detailed and overly detailed interpretation of the various psychometric tests. The difference between MCI and mild dementia is the need or not for supervision in ADLs, it is as simple as that. The DSM-V is not used as an element to differentiate these two entities (lines 51-66).

R: We thank the reviewer for this accurate and important comment. We acknowledge that the original version of the manuscript did not differentiate between mild cognitive impairment (MCI) and mild dementia, potentially leading to conceptual confusion.

As suggested, we have revised the section around lines 107-108 to align with the updated criteria proposed by the National Institute on Aging and Alzheimer's Association (NIA-AA, 2023) and the International Working Group (IWG, 2024). According to these frameworks, the core distinction between MCI and mild dementia is not based on detailed psychometric interpretation but rather on the presence or absence of functional impairment, particularly in instrumental activities of daily living (IADLs).

MCI is characterised by objective cognitive decline that does not significantly interfere with activities of daily living - such as managing money, preparing meals, using transportation, or managing medications - whereas mild dementia involves loss of functional capacity and requires some supervision or assistance with these tasks.

We have also removed the reference to the DSM-5 as a source for this distinction, as it does not define these clinical categories in the same terms as current diagnostic frameworks.

5. RI: They seem to have to justify the use of the tools they have in their Memory Unit despite the redundancy in using MoCA and the Cacho clock test or two ADL scales, IQCODE and IDDD.

R: We appreciate this comment and the opportunity to clarify the rationale for our selection of the instruments. The tools included in the protocol were not chosen for convenience or availability but were carefully selected to provide a multidimensional and complementary screening approach for identifying cognitive vulnerability in older adults.

In terms of cognitive assessment, the MoCA provides a global yet detailed overview of cognitive performance, including attention, memory, and executive function. The Cacho Clock Test, administered in both command and copy versions, allows the calculation of the Improvement Pattern (IP) - defined as the difference in performance between spontaneous and guided tasks. This index has been shown to be sensitive to subtle early cognitive changes, particularly those associated with early AD, even when global cognitive scores remain within normal limits (Cacho et al., 1999; Cacho et al., 2010). Thus, rather than being redundant, these instruments provide different but complementary information. (Lines 177-191).

6. RI: The references do not correspond to what is written:

RI: Lines 63-65: where the use of both MoCA and the clock test is supported.

R: Lines 63-65: These lines were eliminated and changed for a better understanding of the use of both MoCA and the Cacho Clock Drawing Test.

RI: Line 66: Reference 15: No wording the Mini-Clock.

R: This line was eliminated.

RI: Lines 200-203: Reference 39 to endorse the MoCA cut-off points to distinguish MCI and dementia does not justify it. The cut-off point should be 26 (Montreal Cognitive Assessment for the detection of dementia https://doi.org/10.1002/14651858.CD010775.pub3).

R: For our study, we used as a reference a previous study on the Spanish population in which the cut-off point was the same as that used in our paper. We have added the reference.

Gallego, M. L., Ferrándiz, M. H., Garriga, O. T., Nierga, I. P., López-Pousa, S., & Franch, J.V. (2009). Validación del Montreal Cognitive Assessment (MoCA): test de cribado para el deterioro cognitivo leve. Datos preliminares. Alzheimer Real Invest. Demencia, 43(4), 11.

RI: Lines 246-248. Reference 42 where there is no mention to the IQCODE neither to the clock test. The recommended reference is: The screening of mild dementia with a shortened Spanish version of the ‘Informant Questionnaire on Cognitive Decline in the Elderly’. doi: 10.1097/00002093-199509020-00008Montreal Cognitive Assessment for the detection of dementia. https://doi.org/10.1002/14651858.CD010775.pub3

R: We thank you for your contribution to our paper. In this regard, we have also added the following reference:

Davis, D. H., Creavin, S. T., Yip, J. L., Noel-Storr, A. H., Brayne, C., & Cullum, S. (2021). Montreal Cognitive Assessment for the detection of dementia. Cochrane database of systematic reviews, 1(7). https://doi.org/10.1002/14651858.CD010775.pub3

RI: Line 255-260: Reference 44 is not Peres et al, but Cattaneo G, Solana-Sánchez J, Abellaneda-Pérez K, Portellano-Ortiz C, Delgado-Gallén S, Alviarez Schulze V, et 757 al. Sense of Coherence Mediates the Relationship Between Cognitive Reserve and Cognition in Middle-Aged 758 Adults. Frontiers in Psychology. 2022 Mar 28;13.

Where no mention is made to the Spanish version of the IDDD or to the advantage of using it together with the IQCODE.

R: We thank you for this important observation. The reference has been corrected, and we also clarified the description of the functional instruments used in the study. The IDDD is a validated instrument that directly assesses both basic and instrumental daily functioning. The IQCODE, although not a functional scale per se, provides valuable longitudinal information from an informant's perspective on perceived cognitive decline and its impact on daily functioning. This use is consistent with the interpretation of Lezak et al. (2012), who emphasise the complementary value of informant-based insights in neuropsychological assessment.

7. RI: Lines 543-607 The conclusion section is excessive; it should have no more than two well-supported paragraphs.

R: We have consciously revised the section of the conclusions and removed redundant information for better understanding.

We hope we have answered all your questions and welcome any further comments you may have.

Thank you for taking the time to analyse our work and for suggesting improvements.

Best Regards,

Correspondent author.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I would like to thank the authors for their willingness to follow my advice. Thanks to this, the article is ready for publication.

Author Response

Dear Reviewer 3

On behalf of the co-authors of our research, we would like to thank you for your contributions and comments on the research article submitted to the journal, which have undoubtedly improved and enriched the final version.

We greatly appreciate your perspective and professionalism in suggesting very pertinent changes to improve our work.

Best Regards,

Correspondent author.