[version 1; peer review: 2 approved with reservations, 1 not approved]
No competing interests were disclosed.
Data are one of the most important and crucial aspects of any research study. Scientific research is based on collecting and analysing data. Datasets are essential to promote the development of data analysis and sharing is an emerging practice of scientific communication and facilitates the advancement of science by making data accessible, verifiable and reproducible (
Participants were selected according to their genocide survival category. 309 Dyads of mothers and their daughters from genocide survivors, non-targeted and 1959 returnees were selected in rural provinces and in the city of Kigali. Data were collected by a team of 6 local clinical psychologists with clinical backgrounds and experience in data collection. Prior to data collection, they received 2 days of training that enabled them to make a clinical diagnosis of depression using MINI and ethical considerations in data collection.
To reach participants, data collectors were assisted by local authorities who referred them to households with participants who met the study criteria. Participants were approached and the interview was conducted with those who agreed to participate. The interview was conducted in Kinyarwanda, the local language, in a secure room prepared for a face-to-face clinical interview. The interview was conducted in the participant's home or in a nearby office of the local leader. Interview duration ranged from 15-20 min. The sample consisted of 309 dyads of mothers and their adult daughters, including 103 dyads of genocide survivors, 111 of non-targeted, and 95 dyads of 1959 returnees.
Mothers participants had to be Rwandan citizens and permanent residents of the country, a mother with an adult daughter who was able to complete a clinical interview. The daughter met the study criteria if she lived with her mother on a full-time basis or lived until she married.
Participants with communication difficulties, or who had experienced a mental health crisis at the time of the study, or who had recently experienced a traumatic event, or who refused voluntary participation were excluded. Ethical clearance for this study was obtained from the University of Rwanda' Institutional Review Board, College of Medicine and Health Sciences No 72/CHMS/IRB/2020. Written informed consent was obtained from the respondents before the interview.
The Mini International Neuropsychiatric Interview (MINI) questionnaire (
The dataset was created with 86 variables. Since the sample was composed of the grandmothers and their adult daughters who completed the questionnaires separately, we created one dataset. The first 309 cases are the data for the grandmothers while from 310 to 618 are the data for their daughters. The last four variables, which include depression, place of residence (urban or rural), generation (grandmother or daughter), and survivor status (genocide survivor, non-survivor, and 1959 returnee), were coded separately to facilitate the analysis (
Before the analysis was performed, relevant assumptions of the statistical analysis were checked. First, it was verified that there were no missing values or outliers. The exploratory data showed that there were no missing values or outliers. Examination of the independence of the independent variables revealed that there was no multicollinearity and all independent variables were not highly correlated. To obtain the socio-demographic characteristics of the participants, a descriptive statistical analysis was performed. To test the difference in trauma exposure between the three survival categories of mothers and daughters, the one-way test ANOVA was used. Since the results indicated significant differences, a Tukey HSD post-hoc test was conducted to find these specific differences. A chi-square test was performed to examine prevalence differences of depression in the sample of mothers and daughters by survival category and place of residence. To test associates of major depression, a four-level hierarchical logistic regression was performed for the sample of mothers and a six-level logistic regression for the sample of daughters. Depression score of the mothers and daughters, were the dependent variables. The common independent variables in both samples were survival status, residence, and trauma exposure, while mothers' depression and trauma exposure in the daughters' sample were additional independent variables. Each independent variable was entered into the model individually. Age, education level, marital status and occupation were entered together as covariates at the last level.
Data validation was performed using SPSS version 27 to identify suspect and invalid cases, variables, and data values. At the first level, basic checks were performed, including maximum percentage of missing values (100), maximum case in a single category (100), maximum percentage of categories with a count of 1 (90), maximum standard deviation (0), and maximum coefficient of variation (.001). At this level, all cases passed the test. At the second level, rules were defined for individual variables (Name of variable, type of variable, and specify the minimum and maximum value of each) to see if there is no variable that falls outside the defined rules and all cases passed the required test. At the third level, cross-variable validation rules were created, displayed and modified and no violation was found, only gender violated the rules, but this was because the participants were only female. At the last level we saved the output and it was found that all cases did not violate any rules and the data was validated for analysis (
Zenodo: Questionnaires used in the study rural-urban depression prevalence in Rwanda
This project contains the following underlying data:
Questionnaires: rural urban depression prevalence (Kinyarwanda/English)
Zenodo: ‘Rural-urban differences in the prevalence of depression among Rwandan grandmothers and their daughters 26 years after the 1994 genocide against the Tutsis’,
This project contains the following underlying data:
Data file 1. Depression prevalence.sav
Data file 2. Depression prevalence.jasp
Zenodo: ‘Output of analysis for Urban-rural depression prevalence in Rwanda’,
This project contains the following underlying data:
Data file 1. Depression prevalence analysis.spv
Data file 2. Output analysis excel format.xlsx
Data are available under the terms of the
Written informed consent was obtained from the participants before the interview. Ethical clearance for this study was obtained from the University of Rwanda' Institutional Review Board, College of Medicine and Health Sciences No 72/CHMS/IRB/2020.
The authors would like to thank the participants of this study as well as the data collectors including Jonas Bikorimana, Christelle Ishimwe, Munganyinka Claudine, Etienne Kayumba, Albert Nahishakiye and Emile Bambara. Thanks to Remy Twagirimana for his contribution in managing the online software system.
While reading the title of this work you think really on the trauma transmission from parent to the child especially from mother to daughter and the role or associations between depression observed with other variables like place of residence (urban/rural area).
Authors are changing terms mother and grandmother in different parts like in title mother-daughter but in abstract grandmother-daughter and this may confuse, the author has clarified the sample used because this may imply the involvement of grandchildren and think transmission mode (trauma related symptoms from grand-mother to daughter, daughter (mother) to daughter (child) or grand-mother to grand-daughter). If the grandmother, mother/daughter and grand-daughter/daughter has meaning, the methodology especially sample should be changed from dyad to triad and study the co-factors.
Authors didn't present in their work the prevalence of depression among target population or state it in previous studies to show the gap so that further studies have to tackle or work on, I will advise them to clarify the gap and the full clarification of the importance of creating dataset.
I suggest the authors to clarify the population where they drawn samples and the used techniques and I want them to show the results as they mentioned that data were analysed so that in the introduction readers may be pushed to go further for checking or not their replicability.
I want authors to show target group (that calls authors to conduct the study or creating dataset) among three groups if there was a control group(s) among three groups (survivors, non-targeted during genocide and returnees).
In the test administration, the setting was not common for participants (samples) as some were met at their home and others at state houses or offices this wouldn't affect the results or quality of data? I would like authors to clarify the data collection procedures, if possible, clarifying the measurements (define the test used, their goals and the reliability)
Authors has to present some findings and discuss a little bit on it and call for other researches to work on the study limits
Authors need to rethink on key words (what about grand-mothers)?
Are sufficient details of methods and materials provided to allow replication by others?
Partly
Is the rationale for creating the dataset(s) clearly described?
Partly
Are the datasets clearly presented in a useable and accessible format?
Partly
Are the protocols appropriate and is the work technically sound?
Partly
Reviewer Expertise:
Trauma related Disorders
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
Thank you for asking me to review this submission. The authors describe how mental health issues persist as a major challenge in Rwandan society, especially among women who survived the Tutsi genocide. There is a significant risk of these mental disorders being passed down to subsequent generations. The manuscript provides an explanation of on open access dataset's creation and its features, facilitating access and use by other researchers.
I have some suggestions to improve the manuscript.
It is not clear why the researchers have not reported on the prevalences and associations themselves, either in this or in prior papers. The authors should suggest how the data may be used by others, e.g. for cross-comparisons in other settings, to contribute to systematic review data, to conduct sub-analyses exploring particular variables etc. The terms ‘grandmothers and daughters’ and ‘mothers and daughters’ are used interchangeably – I suggest consistent use. Abstract: The sentence: ‘The risk of inheritance of mental disorders is eminent.’ The authors seem to be making the important point about the intergenerational impact of trauma on mental health status. However, I suggest the wording should be amended as ‘inheritance’ implies genetic mechanisms and the term ‘eminent’ is inappropriate here. ‘Data’ should consistently be used in the plural, e.g. Data were analyzed… Do not start a sentence with a number, rather e.g. We recruited 309 dyads of…. Recruitment of participants
there is unnecessary repetition of the numbers per category in the second sentence and at the end of the second paragraph. Third paragraph – who is grandmother vs mother vs daughter is confusing What is the rationale for why the ‘daughter’ needed to live with her mother? Measures
The first three lines are not constructed as a sentence. From this, it is not clear whether you used the MINI or the PHQ-9 or both. If both, why” Much more description, rationale and validation information are required for the tools described. I am very puzzled by the ‘Perception of Parents Questionnaire’ – it seems to have been developed in Brazil for adolescents – looking at the wording, I am not clear how this tool would be cross-culturally valid in the Rwandan setting. For the MINI usage – was only the depression module used? Information on translation processes should be included too. The section that starts ‘before the analysis was performed’ – should be labelled ‘Analysis’ and this section should follow ‘Data validation’. Again is it not clear why the results are not shared. I am not able to comment on the data validation and analysis sections and suggest a statistician be asked to review this.
I feel this work has potential to contribute importantly to a critically important field, but substantial improvements would be required first.
Are sufficient details of methods and materials provided to allow replication by others?
Partly
Is the rationale for creating the dataset(s) clearly described?
Partly
Are the datasets clearly presented in a useable and accessible format?
Partly
Are the protocols appropriate and is the work technically sound?
Partly
Reviewer Expertise:
Public health, mental health, perinatal mental health, implementation science, social determinants of mental health
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
This study addresses an important health topic (depression and generational inheritance of depression after genocide) and provide a relevant dataset to study this health problem.
The Introduction should start with addressing the health problem and gaps in knowledge. In the next paragraph in the introduction, the authors should make clear the (un)availability of data and the need for the creation of an accessible dataset. Thereafter the authors should state their aim – creation of a dataset and the opportunity to do so. Since this paper is a Data Note the authors should shift their focus on the creation and description of the created dataset – something like a data profile to inform and persuade researcher to use the dataset. The examination of prevalence and testing hypotheses should be addressed in separate research paper(s).
The sentences about the 3 groups – genocide survivors, non-target groups, and 1959 returnees are important but should be moved to the Materials and Method section.
In the Material/Methods section, more information is needed on:
-selection process of mothers and daughters: who were eligible and why; in the article the authors sometimes speak of grandmothers but at other times of mothers – consistency and clarity are needed here. Besides, what if a mother had multiple daughters, which daughter was selected and why (always eldest, for example).
-data collection, by whom, when, and how.
-non-response: number of refusals and reason if possible; this provides information about potential bias in the dataset.
-sample size of subgroups (genocide survivors, non-target groups, 1959 returnees)
In the section on Measures, much more information needs to be provided on the outcome variable or issue under study, namely depression. How is that measured and are other studies such as those on the Holocaust and post-WWII depression or related illnesses among Jews consulted. Is it a constructed variable and what are the underlying factors? The authors should provide some preliminary descriptive statistics on these variables.
One of the important explanatory factors is ‘who was targeted’ (survivor status). Provide the sample sizes of these 3 groups.
In this section the authors should also provide some descriptive statistics on sociodemographic of participants, preferable separately for mothers and daughters and/or by survivor status group such as on age, education, employment, marital status, religion, (other) heath issues. If available, also on living area such as ethnically composition (mixed or homogeneous). Is there some information collected on fathers/husbands during the interview such as their age, mental status, employment – if so, mention that. These factors could be (other) explanatory factors or be included in the analyses as confounders.
The section on data validation should come earlier and may be described after the data cleaning/inspection section (on missing info, etc).
Also, the Consent section should be moved forward – perhaps as a subsection in the Material section. More info should be given about anonymizing the data, where the data is stored, who is responsible for the dataset, who might have access and how researchers can get access.
In the final section – a conclusion or discussion section, the authors should discuss the opportunities for research, for example addressing a few testable hypotheses, and some other perspectives such as improving healthcare for victims with depression. Perhaps, follow-up data collection if that is relevant.
After restructuring/re-writing the paper, the authors should rethink their paper title, reflecting the building of a dataset.
There are many data or cohort profiles in epidemiological journals for inspiration to restructure this paper.
Are sufficient details of methods and materials provided to allow replication by others?
No
Is the rationale for creating the dataset(s) clearly described?
Partly
Are the datasets clearly presented in a useable and accessible format?
No
Are the protocols appropriate and is the work technically sound?
No
Reviewer Expertise:
Holocaust; public health
I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.