The Revised Memory and Behavior Problems Checklist for Nursing Homes: Validation among Different Spanish Territories

Round 1

Reviewer 1 Report

Thank you for the opportunity of review this study entitled “Validation of the Revised Memory and Behavior Problems Checklist-Nursing Homes from different territories in a Spain” (land-1499939). The study presents the Spanish validation of the Revised Memory and Behavior Problems Checklist- Nursing Homes (RMBPC-NH), by involving a sample of 312 formal caregivers working in nursing homes from different territories in Spain.

In my opinion, the research topic is relevant, and the study is interesting. Parallelly, there are some minor issues that need to be addressed before the paper will be suitable for publication.

1. In my opinion, the title is misleading. It would be correct to use this term "The Spanish validation". Furthermore, I'm not sure it's correct in English "from different territories in a Spain". Please revise.
2. The abstract is complete and well written. A small suggestion to improve it further is to immediately clarify some characteristics of the two samples (Mean age and standard deviation, percentage of males and females) to provide a clear snapshot of the subjects involved in the study.
3. Lines 77 -81: this part should be deepened to allow the reader to understand the characteristics of the other measures and, consequently, the need for the new instrument presented by the authors.
4. A "Procedure" section should be included in "Materials and methods", where lines 100-107 should be moved.
5. Lines 100-101. Please provide more information about the recruitment procedure. The only information is "We contacted with managers of nursing homes ".
6. A "Data analysis" section should be included in "Materials and methods", where lines 108-126 should be moved.
7. Please, add the limitations.

Author Response

Dear reviser,

We appreciate your comments and the invitation to respond to their suggestions, revise, and resubmit our manuscript. Thank you for your observations and your time. We appreciate that you consider an important topic to make available tools. We think your commentaries have helped us to improve the manuscript. This new version of our paper includes the following changes that the reviewer´s comments suggested. 

In my opinion, the title is misleading. It would be correct to use this term "The Spanish validation". Furthermore, I'm not sure it's correct in English "from different territories in a Spain". Please revise.

• We have eliminated the “a” in the title

The abstract is complete and well written. A small suggestion to improve it further is to immediately clarify some characteristics of the two samples (Mean age and standard deviation, percentage of males and females) to provide a clear snapshot of the subjects involved in the study.

• We have included mean age (SD) and percentage of males and females sample.

Lines 77 -81: this part should be deepened to allow the reader to understand the characteristics of the other measures and, consequently, the need for the new instrument presented by the authors.

• It was explain in more detail the Nursing Homes Behavior Problems Scale (NHBPS) [Ray et al.] in line 81.

A "Procedure" section should be included in "Materials and methods", where lines 100-107 should be moved.

• Materials and Methods section has been moved considering the procedure section. Line 139-171

Lines 100-101. Please provide more information about the recruitment procedure. The only information is "We contacted with managers of nursing homes ".

• We have added extra information of the recruitment procedure. Line 142-144

A "Data analysis" section should be included in "Materials and methods", where lines 108-126 should be moved.

• Materials and Methods section has been moved considering the data analysis section. Line 139-171

Please, add the limitations.

• We have added some limitations at the end of discussion section. Line 355 -358

We hope that all these changes have improved the manuscript and that it has now reached the standards for its publication in The Land.

Thank you for reading and reviewing our work.

Cristina Velasco

Author Response File: Author Response.docx

Reviewer 2 Report

The present study developed a Spanish version of measurement scale for memory and behavioral assessment of nursing home residents. It would be recommended to elaborate Methods and Results sections, which in current version remains several questions. 

1. Introduction

Overall, it remains to be questioned why memory and behavior should be assessed in one scale. There have been established scales to assess cognitive impairment (memory) or behavior (neuropsychiatric symptoms). Furthermore, the RMBPC-NH seems to include physical impairment or ADL (subscale named "Functional"). The rationale of this scale should be given in 1.2. section.

First three paragraphs of Section 1.1. may be difficult for readers to understand the association with behavior and memory problems in nursing home residents.

2. Materials and Methods

Explanation of the mixed-effect models is not clear: e.g. the aim of random intercepts for participants (participant means caregiver but not resident? One caregiver evaluated two or more residents?), and variables of fixed effects (caregiver characteristics in Table 1?).

Subscales of PMBPC-NH, i.e. frequency, care-related burden, and socio-emotional burden, should be described in 2.2. Instruments section. It should be also mentioned who assessed BMP of nursing home residents.

The Materials and Methods section may lack information on how questionnaire was distributed and answers were collected.

3. Results

How many were nursing home residents assessed? On average, how many residents did one caregiver evaluated using the RMBPC-NH?

Table 1: the title appears to be a template and should be corrected.

Table 5: Does "Sex" mean "Female"? Other variables (care experience, technical equipment, and level of training) may have been continuous variables, but it was not clarified in Materials and Methods section. It is also questioned why other continuous variables in Table 4 were not used in the mixed-effect models.

In the area highlighted with yellow marker, I did not understand "AACC" and what this paragraph meant to present.

4. Discussion

Overall, there was a gap between the hypothesis (BMP negatively impacts formal caregivers) and the mixed-effect models (BMP differs according to caregiver characteristics). hence, discussing any causal relationships may cause confusion and misleading.

Author Response

Dear reviser,

We appreciate your comments and the invitation to respond to their suggestions, revise, and resubmit our manuscript. Thank you for your observations and your time. We appreciate that you consider an important topic to make available tools. We think your commentaries have helped us to improve the manuscript. This new version of our paper includes the following changes that the reviewer´s comments suggested. 

The present study developed a Spanish version of measurement scale for memory and behavioral assessment of nursing home residents. It would be recommended to elaborate Methods and Results sections, which in current version remains several questions. 

- Materials and Methods section has been moved considering the procedure section. Line 139-171

1. Introduction

Overall, it remains to be questioned why memory and behavior should be assessed in one scale. There have been established scales to assess cognitive impairment (memory) or behavior (neuropsychiatric symptoms). Furthermore, the RMBPC-NH seems to include physical impairment or ADL (subscale named "Functional"). The rationale of this scale should be given in 1.2. section.

• It be considered the functional problems because it can cause stress in caregivers. It has been included a commentary about this in line 58

First three paragraphs of Section 1.1. may be difficult for readers to understand the association with behavior and memory problems in nursing home residents.

2. Materials and Methods

Explanation of the mixed-effect models is not clear: e.g. the aim of random intercepts for participants (participant means caregiver but not resident? One caregiver evaluated two or more residents?), and variables of fixed effects (caregiver characteristics in Table 1?).

• Mixed-effects models are statistical models that consider different clusters to explain the variability of the fixed effects under study (see for example the reference below for an example of the use of different clusters to study fixed effects). In the specific case of this study, all the intercepts are related to the levels of the variables and, given that the variables were measured in caregiver, then they represent the variability of means of participants (caregivers). A brief sentence was added in the “Data analysis” section regarding to the random intercepts.

• Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects modeling with crossed random effects for subjects and items. Journal of Memory and Language, 59(4), 390-412. https://doi.org/10.1016/j.jml.2007.12.005.

Subscales of PMBPC-NH, i.e. frequency, care-related burden, and socio-emotional burden, should be described in 2.2. Instruments section. It should be also mentioned who assessed BMP of nursing home residents.

The Materials and Methods section may lack information on how questionnaire was distributed and answers were collected.

- We have included and extra information of recruitment process in line 107.

3. Results

How many were nursing home residents assessed? On average, how many residents did one caregiver evaluated using the RMBPC-NH?

Table 1: the title appears to be a template and should be corrected.

• We changed it. Thank you.

Table 5: Does "Sex" mean "Female"? Other variables (care experience, technical equipment, and level of training) may have been continuous variables, but it was not clarified in Materials and Methods section. It is also questioned why other continuous variables in Table 4 were not used in the mixed-effect models.

• Thank you for this comment. “Sex” means “Female” as we used “Male” as reference. We clarified the nature of the variables in the manuscript. It is worth mentioning here that the variables were ordinal as it was presented in Table 1.
• Basically, we used correlation coefficients to test the relationship between the instrument and other continuous variables, and then we used mixed-effects models for testing the relationship between the instrument and other categorical and ordinal variables.

In the area highlighted with yellow marker, I did not understand "AACC" and what this paragraph meant to present.

• It was referred to “autonomous communities”

4. Discussion

Overall, there was a gap between the hypothesis (BMP negatively impacts formal caregivers) and the mixed-effect models (BMP differs according to caregiver characteristics). hence, discussing any causal relationships may cause confusion and misleading.

• The first thing is to avoid any type of causal interpretation of the data. In the second place we have consider that you can try to emphasize more that these variables can be indicators of the risk of suffering and being affected by problems in residences.

We hope that all these changes have improved the manuscript and that it has now reached the standards for its publication in The Land.

Thank you for reading and reviewing our work.

Cristina Velasco

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear the authors,

Thank you for the revisions in response to my pervious comments. I confirmed that all comments are addressed now, and I have no further questions.