Dementia Prevention Self-Management in Older Thai Adults with Type 2 Diabetes: Development and Psychometric Properties of Two Questionnaires

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Congratulations to the authors for their efforts in putting together two validated instruments. It would be worth reflecting on the relevance of publishing the validation of each of the instruments separately. In this respect, the title is too long, which goes against the editorial guidelines; the introduction is too generic, and the relevance of the instrument's validation in the target population is not understood. 

The methodological choices are congruent with the aim of the research and the analysis sections present the main results in detail. For the reader, since this is a validation of two instruments, I find the analysis confusing. There should be an individual reflection demarcating the two stages, rather than a global presentation. The results are thus generalised, and the article lacks details that should be highlighted: the authorisation, translation and cultural adaptation procedures; the selection of participants and the role of the research team in the data collection process.

The tables and figures take up a considerable amount of text in the article, suggesting that the authors select those that can be left in the appendix / supplementary material.

Authors are encouraged to resubmit, considering the autonomy of the two validated data collection instruments.

Author Response

Congratulations to the authors for their efforts in putting together two validated instruments. It would be worth reflecting on the relevance of publishing the validation of each of the instruments separately. In this respect, the title is too long, which goes against the editorial guidelines; the introduction is too generic, and the relevance of the instrument's validation in the target population is not understood. 

response:

We would like to thank the reviewer for the suggestions. The development of both questionnaires in this study is based on the IFSMT theory, which includes the context dimension, self-management processes dimension, and outcomes dimension. The self-management process dimension can be enhanced and directly influences the outcomes dimension. In this study, which focuses on older adults with T2DM for dementia prevention, the context is both specific and interconnected. Therefore, the researchers have integrated the development of both instruments, which will provide readers with a more comprehensive understanding and greater benefits. Due to this we respectively assert that the two measures should be reported together.   

We have shortened the title to make it clearer. The journal’s author guidelines only state that the title should be “concise, specific and relevant”; therefore, the revised title does not seem to be in contravention of the guidelines

The methodological choices are congruent with the aim of the research and the analysis sections present the main results in detail. For the reader, since this is a validation of two instruments, I find the analysis confusing. There should be an individual reflection demarcating the two stages, rather than a global presentation. The results are thus generalised, and the article lacks details that should be highlighted: the authorisation, translation and cultural adaptation procedures; the selection of participants and the role of the research team in the data collection process.

response:

We appreciate your observations, which highlight key areas for enhancing clarity and detail in our study. However, we would like to clarify that the methodological approach for the two questionnaires was intentionally designed to be parallel in structure, leading to a consolidated and summarized presentation of the findings. We also believe that given the same methods are used for the development, validation and testing of the two questionnaires it is sensible to present the two closely related questionnaires in one study.

Authorisation, translation and cultural adaptation are not relevant to this study, as the questionnaires are new and therefore were created in the Thai language with Thai people (i.e. they were not available in another language or culture).

We have added more information about the selection of participants and roles of the research team as suggested.

The tables and figures take up a considerable amount of text in the article, suggesting that the authors select those that can be left in the appendix / supplementary material.

response:

As suggested by the reviewer, we have moved Table 1 and 2 to the supplementary Table S1 and S2.

Authors are encouraged to resubmit, considering the autonomy of the two validated data collection instruments.

response:

Thank you for your encouragement. We have revised our manuscript according to the reviewers' suggestions and hope it is now suitable for publication.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors presented the development of a new questionnaire to assess dementia preventive behaviors among patients with diabetes and their family. But there some concerns raised during my reading.

1. Although some self-management approaches may be effective to prevent cognitive decline (line 65-68), the authors should give a clear definition between self-management behaviors and dementia preventive behaviors to link the section (line 69-80) to the next section.

2. Line 245-250 indicated that 300 and 150 samples (total 450) were enough for EFA and CFA. Please explain the reasons for extra number.

3. Please describe briefly the reasons that 150 was sufficient for CFA (line 249).

4. Total 565 participants completed the surveys, please explain how the number of sample being allocated to EFA and CFA. For example, the authors allocated 311 (311/300 = 103% of minimum requirement) participants in EFA, while 254 (254/150 = 169% of minimum requirement) in CFA.

5. For table 1, item 16 &17 were confusing. I was not sure if the participants understand what was dementia preventive behaviors.

6. Table 2, item 1, what are the five food groups? How did the authors develop the sub-group of the questionnaire?

Author Response

. Although some self-management approaches may be effective to prevent cognitive decline (line 65-68), the authors should give a clear definition between self-management behaviors and dementia preventive behaviors to link the section (line 69-80) to the next section.

response:

We would like to thank the reviewer for the useful suggestions.  We have revised the manuscript as suggested by the reviewer.  We have added the information regarding the definition of self-management behaviors in the revised manuscript as suggested by the reviewer.

Line 245-250 indicated that 300 and 150 samples (total 450) were enough for EFA and CFA. Please explain the reasons for extra number.

response: 

According to a timeline for data collection, with the EFA data being collected from April to July, and the CFA data being collected from September to December.
A minimum sample size has been defined, as in our view, having a sufficiently large sample size will positively impact the reliability of the test results.

References

1. Kline RB. (2011). Principles and Practice of Structural Equation Modeling(3rd ed.). Guilford Press.

2.  Hair JF, et al., (2010). Multivariate Data Analysis (7th ed.). Pearson.

Please describe briefly the reasons that 150 was sufficient for CFA (line 249).

response:

Total N = 150 for normal indicators with no missing values was sufficient for a CFA model with three factors and five continuous indicators per factors.

Reference

Muthén LK, and Muthén BO., Structural Equation Modeling. 2002;9:599-620.  

Total 565 participants completed the surveys, please explain how the number of sample being allocated to EFA and CFA. For example, the authors allocated 311 (311/300 = 103% of minimum requirement) participants in EFA, while 254 (254/150 = 169% of minimum requirement) in CFA.

response:

According to a timeline for data collection, with the EFA data being collected from April to July. At the end of this period, data were collected from 311 participants which exceeds minimum requirements.

      Regarding the CFA data collection, the data being collected from September to December.  Since we aimed to continuously collect the data until the end of this period, we collected 254 participants which exceeds minimum requirements.

      Although a minimum sample size has been defined, as in our view, having a sufficiently large sample size will positively impact the reliability of the test results^1,2.

References

1. Kline RB. (2011). Principles and Practice of Structural Equation Modeling(3rd ed.). Guilford Press.

2.  Hair JF, et al., (2010). Multivariate Data Analysis (7th ed.). Pearson.

For table 1, item 16 &17 were confusing. I was not sure if the participants understand what the dementia preventive behaviors was.

response:

According to the content in items 2 to 7 from table 1 (new table S1), the participants already understand what the dementia preventive behaviors are.

Table 2, item 1, what are the five food groups? How did the authors develop the sub-group of the questionnaire?

response:

Five food groups in item 1 of table 2 include macronutrients (carbohydrate, protein, and lipids) and micronutrients (minerals and vitamins). We have added this information in the item 1 of table 2 (new table S2).  In the present study, we did not develop sub-group of the questionnaire.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors:

The article aims to develop two scales on dementia preventive behaviours and process dimension and self-management in people with dementia and type 2 diabetes, and the authors explore their content in a clear and organised way. The topic is innovative, pertinent and focussed on a current public health problem. It provides detailed information on the development and validation of questionnaires, participant recruitment criteria, statistical analysis methods, and results from factor analysis and reliability assessments.

The state of the art/Introduction is comprehensive and relevant to the field; references are mostly recent publications and relevant in the field of this thematic. The structured presentation of the document, including tables and figures, enhances the clearness and relevance of the research findings. Discussion must be improved, it lacks limitations and future research directions. You do not address the ethical considerations of the study, such as requesting an ethical committee approval

Please evaluate the ethics statements and data availability statements to ensure they are adequate.

With regard to the content of the article, I have some suggestions for improvement and clarifications to ask for:

·         “Participants and Recruitment”, in line 156-157, it is not clear why you excluded persons with family history of dementia, brain infection, brain injury, or psychiatric illness.

·         “Participants and Recruitment”, in line 157-158 explain why you use de Thai Mental State Examination (TMSE) for dementia screening instead of other scales with better specificity for dementia and cognitive decline (e.g. Mini Mental State Examination, Montreal Cognitive Assessment).

·         “Materials and Methods” must be added the ethical considerations of the study.

·         “Item Generation”, in lines 176-177 and 197-198, you mentioned that you elaborated a “4-point Likert scale (1 = strongly disagree, 2=disagree, 3=agree, and 4= strongly agree)”. The reason why your scale as only 4-itens, instead of 5 should be clarified, as well as why you omitted the neutral option - neither agree nor disagree option.

·         “Construct validity”, in line 238-243 you say that “A cross-sectional study was carried out to test construct validity. (…) First, we evaluated 238 the exploratory factor analysis (EFA) with a varimax between April to July 2023. The Kaiser-Meyer-Olkin (KMO) and Bartlett’s test of sphericity were used to determine the appropriate sample size for factor analysis. Eigenvalues greater than or equal to 1.0 and factor loading graters greater than 0.5 were considered appropriate to verify the possible underlying factors. The internal consistency reliability was measured by Cronbach’s alpha coefficient with 30 participants.” Explain why you only tested 30 persons and how as this recruitment made.

·         Line 273, please correct the missing space: “of the 4,621invited older adults completed the survey”.

·         “Discussion”, in line 396, please correct the sentence “develop and evaluate the psychometric properties of two (..)”. Scales have metric properties not psychometric properties.

·         “Discussion” must have listed the study limitations, namely risk of bias, sample recruitment, non-random sampling, sample size, limitations of questionnaire application, between others.  

Author Response

The article aims to develop two scales on dementia preventive behaviours and process dimension and self-management in people with dementia and type 2 diabetes, and the authors explore their content in a clear and organised way. The topic is innovative, pertinent and focussed on a current public health problem. It provides detailed information on the development and validation of questionnaires, participant recruitment criteria, statistical analysis methods, and results from factor analysis and reliability assessments.

The state of the art/Introduction is comprehensive and relevant to the field; references are mostly recent publications and relevant in the field of this thematic. The structured presentation of the document, including tables and figures, enhances the clearness and relevance of the research findings. Discussion must be improved, it lacks limitations and future research directions.

You do not address the ethical considerations of the study, such as requesting an ethical committee approval. Please evaluate the ethics statements and data availability statements to ensure they are adequate.

response:

We would like to thank the reviewer for the useful suggestions.  We have added more information regarding the study limitations in the discussion section of the manuscript as suggested by the reviewer.  In addition, we have already added the information regarding the ethical considerations in the material and methods section of the revised manuscript as suggested by the reviewers. 

“Participants and Recruitment”, in line 156-157, it is not clear why you excluded persons with family history of dementia, brain infection, brain injury, or psychiatric illness.

response: 

We have excluded individuals who have these factors due to the facts that these conditions have been reported to be highly associated with the incidence of dementia and hence would confound the results.

References

1. Cannon-Albright LA, et al., Neurology. 2019;92(15):e1745-e1753.
2. Muzambi R, et al., J Alzheimer’s Dis. 2020;76(4):1609–1626.
3. Gu D, et al., Neuroepidemiology. 2022;56(1):4-16.
4. 4. Stafford J, et al., Int J Geriatr Psychiatry. 2022;37(5):10.1002/gps.5711

“Participants and Recruitment”, in line 157-158 explain why you use de Thai Mental State Examination (TMSE) for dementia screening instead of other scales with better specificity for dementia and cognitive decline (e.g. Mini Mental State Examination, Montreal Cognitive Assessment).

response:

The Montreal Cognitive Assessment (MoCA) is a cognitive screening test designed to evaluate mild cognitive impairment (MCI). Since we aim to exclude patients with dementia, individuals with MCI, which is a mild stage of cognitive impairment, are still included in the present study.

      Studies indicate that TMSE may have better psychometric properties for detecting mild cognitive impairment (MCI) compared to the MMSE in terms of several aspects including

1. Cultural and Linguistic Suitability
2. Validation
3. Education and Age Considerations
4. Healthcare Context and Integration
5. Reliability and Sensitivity

     Thereby, this TMSE may suggest better sensitivity in identifying cognitive decline among different demographic groups in Thailand.

References
1. Pongpirul K, et al. J Med Assoc Thai. 2007;90(3):568-574.

2. Wang L, et al., J Clin Neurol. 2006;2(2):75-81.
3. Laohapoonrungsee A, et al., J Med Assoc Thai. 2012; 95(4):479-485.
4. Sukonthasarn A, et al. J Med Assoc Thai. 2010;93(11):1384-1390.
5. Kittirattanapaiboon P, et al., Int J Geriatr Psychiatry. 2006;21(7): 640-646.

6. Cheng H, et al., Cogn Behav Neurol. 2011;24(2):64-70.

   7. Nantavith S, et al., J Med Assoc Thai 2010;93(6):618-625.

“Materials and Methods” must be added the ethical considerations of the study.

response:

We have added the ethical considerations of the study in the materials and methods section of the revised manuscript as suggested by the reviewer.

“Item Generation”, in lines 176-177 and 197-198, you mentioned that you elaborated a “4-point Likert scale (1 = strongly disagree, 2=disagree, 3=agree, and 4= strongly agree)”. The reason why your scale as only 4-itens, instead of 5 should be clarified, as well as why you omitted the neutral option - neither agree nor disagree option.

response:

In this study, we chose a 4-point Likert scale that omits the neutral option of 'neither agree nor disagree' in order to eliminate central tendency bias and enhance the reliability of decision-making without the neutral option.

References

1. Allen IE and Seaman CA, Quality Progress 2007;40(7):64–65.

2. Yu C, Int J Market Res. 2010; 52(4):547-556.
3. Revilla M, et al., Sociol Methods Res. 2014;43(1):73-97.

“Construct validity”, in line 238-243 you say that “A cross-sectional study was carried out to test construct validity. (…) First, we evaluated 238 the exploratory factor analysis (EFA) with a varimax between April to July 2023. The Kaiser-Meyer-Olkin (KMO) and Bartlett’s test of sphericity were used to determine the appropriate sample size for factor analysis. Eigenvalues greater than or equal to 1.0 and factor loading graters greater than 0.5 were considered appropriate to verify the possible underlying factors. The internal consistency reliability was measured by Cronbach’s alpha coefficient with 30 participants.” Explain why you only tested 30 persons and how as this recruitment made.

response:

Selecting 30 participants for a reliability test is common in research ^1,2. A sample of 30 offers a practical compromise that’s small enough to manage efficiently but large enough to provide meaningful reliability estimates.  Moreover, Bonett and colleagues suggest that samples as small as thirty subjects are sufficient to measure the reliability³. 
References
1. Funk SM, et al., J Thromb Haemost. 2020;18(2):285-294. 
2. Izci B, et al., 2008;12(2):161-8

3. Bonett DG and Wright TA. (2014). Cronbach’s alpha reliability: Interval estimation, hypothesis testing, and sample size. J Organ Behav. 2014 36(1)

Line 273, please correct the missing space: “of the 4,621 invited older adults completed the survey”.

response: 

corrected.

 “Discussion”, in line 396, please correct the sentence “develop and evaluate the psychometric properties of two (..)”. Scales have metric properties not psychometric properties.

We have revised this sentence as suggested by the reviewer.

We have retained the term “psychometric properties” throughout the manuscript because this term is appropriate- it is defined as: “Psychometrics is the construction and validation of measurement instruments and assessing if these instruments are reliable and valid forms of measurement. In behavioral medicine, psychometrics is usually concerned with measuring individual’s knowledge, ability, personality, and types of behaviors” 

(Ginty, A.T. (2013). Psychometric Properties. In: Gellman, M.D., Turner, J.R. (eds) Encyclopedia of Behavioral Medicine. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-1005-9_480) 

“Discussion” must have listed the study limitations, namely risk of bias, sample recruitment, non-random sampling, sample size, limitations of questionnaire application, between others.  

Response:

We have added the study limitations in the discussion section of the revised manuscripts as suggested by the reviewer.

Reviewer 4 Report

Comments and Suggestions for Authors

This article is appropriate methodology and statistical analysis. However, please check the percentage in Table 3 again. 

Author Response

This article is appropriate methodology and statistical analysis. However, please check the percentage in Table 3 again

response:

We would like to thank the reviewer for this suggestion.  We have checked the percentage in table 3 (new table 1) as suggested by the reviewer.  All data are correct. 

Reviewer 5 Report

Comments and Suggestions for Authors

The authors develop this study with the aim of construct and testing two new psychometric instruments that assess self-management behaviours related to dementia prevention in older adults with T2DM, based on the theory of individual and family self-management.

To do this, the authors first presented the importance of the prevalence of dementia and type 2 diabetes (T2DM) and then presented the study that saw the creation of the items through literature review and subsequently the statistical analyses. This is an interesting idea that can increase the knowledge of methodologies that can help prevent dementia.

The organization of the paragraphs is clear and smooth. They are detailed, well-constructed and focuses on the theme of the paper with adequate bibliographic references. The data are well-presented and represented through the figures.

However, I ask the authors to consider the following comments to improve and implement the quality and robustness of the manuscript.

Comment 1: The quality of English needs to be improved. Sometimes the complexity in the construction of the sentences makes reading not very smooth and fluid. Authors are encouraged to do a minor review of the manuscript. The use of appropriate terminology is critical to the context. Authors are encouraged to rephrase some sentences and review the use of some terms.

Comment 2: The introduction, although well-constructed and detailed, seems excessively long. The presence of different themes and digressions, makes the focus on the objective of the study vague and not very focused. The authors could concentrate their attention on the arguments strictly related to the objective of the study; they can present bibliographical citations that help the reader to adequately project himself into the context in which the study takes place; the themes related to the objective but not necessary to focus on it, could be better argued, if considered essential, in the discussion paragraph.

Comment 3: Authors are advised to pay attention to citations in the text. There are several statements in the text that do not follow the set bibliographic numbering. Authors are advised to edit the content, inserting relevant and adequate references. Here are the affected lines: line 62, "According to a comprehensive narrative review (Rakesh et al., 2017)"; line 73-74 "systematic literature review by Li et al. (2022)"; line 85-86 "focused on dementia preventive behaviours in the older adult (Caro-Bautista et al., 2014)".

Comment 4: The "Discussion" paragraph is well structured; however, in this section it seems that the authors limit themselves to presenting and discussing their data, without many references to other literature. Although it is understandable that this is the creation of new psychometric instruments, perhaps it would be more appropriate to compare their data more thoroughly with other similar ones already present in the literature and referable to other instruments for the assessment of similar factors; in this way perhaps it could be presented more clearly how and if the achievement of their objective affects the state of currently known research, regardless of whether or not it brings elements of originality. Furthermore, very few lines are dedicated to the usefulness and relevance of these new tools in the field of dementia. The authors are invited to argue in a more in-depth and detailed manner, also with additional bibliographical entries, how their tools can be an aid to the existing literature and to the clinical context of dementia.

Comments on the Quality of English Language

Comment 1: The quality of English needs to be improved. Sometimes the complexity in the construction of the sentences makes reading not very smooth and fluid. Authors are encouraged to do a minor review of the manuscript. The use of appropriate terminology is critical to the context. Authors are encouraged to rephrase some sentences and review the use of some terms.

Author Response

The authors develop this study with the aim of construct and testing two new psychometric instruments that assess self-management behaviours related to dementia prevention in older adults with T2DM, based on the theory of individual and family self-management.

To do this, the authors first presented the importance of the prevalence of dementia and type 2 diabetes (T2DM) and then presented the study that saw the creation of the items through literature review and subsequently the statistical analyses. This is an interesting idea that can increase the knowledge of methodologies that can help prevent dementia.

The organization of the paragraphs is clear and smooth. They are detailed, well-constructed and focuses on the theme of the paper with adequate bibliographic references. The data are well-presented and represented through the figures.

However, I ask the authors to consider the following comments to improve and implement the quality and robustness of the manuscript.

Comment 1: The quality of English needs to be improved. Sometimes the complexity in the construction of the sentences makes reading not very smooth and fluid. Authors are encouraged to do a minor review of the manuscript. The use of appropriate terminology is critical to the context. Authors are encouraged to rephrase some sentences and review the use of some terms.

response:

We would like to thank the reviewer for the useful suggestions.  After being advised by an English native co-investigator we have revised the manuscript as suggested by the reviewer. 

Comment 2: The introduction, although well-constructed and detailed, seems excessively long. The presence of different themes and digressions, makes the focus on the objective of the study vague and not very focused. The authors could concentrate their attention on the arguments strictly related to the objective of the study; they can present bibliographical citations that help the reader to adequately project himself into the context in which the study takes place; the themes related to the objective but not necessary to focus on it, could be better argued, if considered essential, in the discussion paragraph.

response:

Thank you for your useful suggestion. We have revised the introduction section in the revised manuscript as suggested by the reviewer for a better focus on the objectives of the present study. 

Comment 3: Authors are advised to pay attention to citations in the text. There are several statements in the text that do not follow the set bibliographic numbering. Authors are advised to edit the content, inserting relevant and adequate references. Here are the affected lines: line 62, "According to a comprehensive narrative review (Rakesh et al., 2017)"; line 73-74 "systematic literature review by Li et al. (2022)"; line 85-86 "focused on dementia preventive behaviours in the older adult (Caro-Bautista et al., 2014)".

response:

We apologize to the reviewer for our mistake. We have now revised and rechecked all citations in the revised manuscript as suggested by the reviewer.

Comment 4: The "Discussion" paragraph is well structured; however, in this section it seems that the authors limit themselves to presenting and discussing their data, without many references to other literature. Although it is understandable that this is the creation of new psychometric instruments, perhaps it would be more appropriate to compare their data more thoroughly with other similar ones already present in the literature and referable to other instruments for the assessment of similar factors;
in this way perhaps it could be presented more clearly how and if the achievement of their objective affects the state of currently known research, regardless of whether or not it brings elements of originality. Furthermore, very few lines are dedicated to the usefulness and relevance of these new tools in the field of dementia.

The authors are invited to argue in a more in-depth and detailed manner, also with additional bibliographical entries, how their tools can be an aid to the existing literature and to the clinical context of dementia.

response:

      Thank you for your valuable suggestions. We have expanded our discussion in the revised manuscript as recommended by the reviewer.     

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Most of my comments were addressed. However, the sample size it questioned especially CFA. For CFA, the required number of subjects was 150. But the authors included 254. Please show the formula to support 254 is a sufficiently large sample size. 

Author Response

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

thank you to address my concerns.