Cultural Adaptation, Validation and Evaluation of the Psychometric Properties of an Obstetric Violence Scale in the Spanish Context

Round 1

Reviewer 1 Report

Dear Authors,

Thank you for the opportunity to review this paper. Methodological papers like this are scarce but very much needed. I think the study conducted in two steps is nicely presented. The analyses build on each other and it all seems well thought out. For instance, in phase one experts assess validity and relevance while the user experts (target population) assess understandability and acceptability. I think “poorly related to” very related to” are errors that should be omitted (Line 151 and 152). Item-level content validity index (I-CVI) scores were calculated, but the scale-level content validity index/average (S-CVI/Ave) and scale-level content validity index/universal acceptance (S-CVI/UA) could also be given.

I find the methodology updated and e.g., INFLESZ was new to me. Having said that I find the analyses very advanced and that the authors might engage more readers if they take the time to add a few simple explanations and motivations e.g., why conduct a factor analysis on a one-dimension scale?

Given that a validation is valid for the studied population i.e., the results are difficult to generalize. I wonder how the authors were reasoning when they included both high and low risk pregnancies/births as well as public and private hospitals. Is the scale automatically valid for all these groups? Could you look into any potential differences in the current dataset? I think this needs to be addressed at least in a methodological discussion.

My Spanish is a bit rusty, does this scale have a name? In headlines e.g. in tables it is referred to with different names such as OV, Violence Obstetric Scale, Violence scale etc.

The authors stated that they evaluated “…as many as possible psychometric properties with an as robust as possible methodological approach.” I agree and I salute that they continued the work on an already developed tool. Having said that I find the statistics too advanced for me. My recommendation is acceptance after minor revision and a consultation with an independent statistician.

For instance, I wonder if the reliability is assessed in a correct way and if it could be calculated and expressed in an simpler way? Line 528-540 doesn’t make sense to me. Sensitivity is usually a measure for concurrent validity and reliability can easily be expressed in %.

About the known group assessment (Line 543), not giving consent is part of the definition of OV that is used in the introduction, so I find it difficult to understand the meaning in this sentence.

I want to emphasis how important this work is and I wish the research group all the best on their journey!

Author Response

Dear Authors,

Thank you for the opportunity to review this paper. Methodological papers like this are scarce but very much needed. I think the study conducted in two steps is nicely presented. The analyses build on each other and it all seems well thought out. For instance, in phase one experts assess validity and relevance while the user experts (target population) assess understandability and acceptability. I think “poorly related to” very related to” are errors that should be omitted (Line 151 and 152). Item-level content validity index (I-CVI) scores were calculated, but the scale-level content validity index/average (S-CVI/Ave) and scale-level content validity index/universal acceptance (S-CVI/UA) could also be given.

Thank you for your kind comments, we agree that this type of studies, beyond the subject under study, are useful to improve the validation processes of the instruments. We consider appropriate your consideration about content validity, so we have included in method and results the calculation of scale-level content validity index/universal acceptance (S-CVI/UA).

However, we have not modified the face validity. Content validity should be distinguished from face validity. Face validity responds to the acceptability and the relationship with the construct to be measured that the items of an instrument have from the point of view of the target population, and therefore it makes sense to establish differences in what is asked from the participants with respect to what is asked from the experts.

I find the methodology updated and e.g., INFLESZ was new to me. Having said that I find the analyses very advanced and that the authors might engage more readers if they take the time to add a few simple explanations and motivations e.g., why conduct a factor analysis on a one-dimension scale?

Thank you for your comments. INFLESZ is a scale developed to assess the comprehension of healthcare texts for users. It is very useful for assessing whether an instrument is understandable by a person from a purely linguistic approach. Our research team has the custom of using it as an additional mechanism to ensure the comprehensibility of the instruments we validate or develop, especially if the scale is going to be used by population with educational levels that may be low.

Our team always tries to conduct analyses based on up-to-date methodological recommendations. Although we often include explanations in the discussion of our work, we cannot always explain everything (this is an article, not a book; excessive length may be contrary). We thank you for your comment, as it indicates that our line of research (validation and evaluation of psychometric properties of instruments) is of interest to you.

In this case, the CFA with one dimension was performed to test the assumption of unidimensionality, necessary to perform the Rasch analysis and to confirm the original model proposed by Cardenas and Salinero. In addition, the authors of the original scale, Cardenas and Salinero provided very little information on key aspects of the factor analysis performed by them (linear or nonlinear approximation, factor retention method, rotation, etc). They did not even provide sample adequacy tests (e.g. KMO). We do not want to criticize other researchers, but the factor analysis performed by these authors had aspects and deficiencies that we could not assume from our methodological rigor, so we did an advanced CFA on our sample to check the viability of the model (also performing the analysis of unidimensionality with other indexes Unidimensional Congruence (UniCo), Explained Common Variance (ECV) and Mean of Item REsidual Absolute Loadings (MIREAL)).

Given that a validation is valid for the studied population i.e., the results are difficult to generalize. I wonder how the authors were reasoning when they included both high and low risk pregnancies/births as well as public and private hospitals. Is the scale automatically valid for all these groups? Could you look into any potential differences in the current dataset? I think this needs to be addressed at least in a methodological discussion.

Indeed, we included low-risk and high-risk pregnancies/deliveries (although we did not collect this variable), as well as those from public and private hospitals (we did not find significant differences between centers), because it is a validation study and we believe that the first thing to do is to evaluate whether this tool is useful and valid for measuring OV. From a theoretical point of view, in our opinion, it should not matter whether a woman has a high-risk or low-risk pregnancy; her delivery should be respected in the same way and she should not perceive obstetric violence.

The inferential analyses we have performed seems to support the idea that what is important is not the characteristics of the population per se, but the way in which the different obstetrical procedures are carried out. For example, if we assume that there are more risk pregnancies in the Inducing labor yes group, we can observe that there are no differences when compared to Inducing labor no. But within the Inducing labor yes group, there is much greater perceived OV when there is no consent. Theoretically, we believe that perhaps the variable high obstetric risk could be associated with greater perception of obstetric violence, since these women receive more interventions, but what is important is how the care is provided, whether it is with consent, respect and professionalism. Our interest is that this scale can be used to evaluate this aspect. If this work is finally published, we will try to explore these aspects in new studies with different populations (as explained in the discussion).

My Spanish is a bit rusty, does this scale have a name? In headlines e.g. in tables it is referred to with different names such as OV, Violence Obstetric Scale, Violence scale etc.

Surely your Spanish is better than our English. The scale has no name, and it is a problem, but not being the authors of the original scale, we wanted to be respectful and did not propose any specific name. We have revised the nomenclature of the scale to unify its denomination throughout the manuscript. Thank you, this will improve the consistency of the work.

The authors stated that they evaluated “…as many as possible psychometric properties with an as robust as possible methodological approach.” I agree and I salute that they continued the work on an already developed tool. Having said that I find the statistics too advanced for me. My recommendation is acceptance after minor revision and a consultation with an independent statistician.

Thank you for your comment. We respect your opinion and are receptive to the revisions that the editor considers necessary. You can also consult other works published by the authors with this methodology.

For instance, I wonder if the reliability is assessed in a correct way and if it could be calculated and expressed in an simpler way? Line 528-540 doesn’t make sense to me. Sensitivity is usually a measure for concurrent validity and reliability can easily be expressed in %.

There are several methodological approaches to evaluate the psychometric properties of an instrument (see COSMIN standards). The paragraph alluded to refers to the concept of reliability, as conceived from the Rasch approach. Rasch model analysis produces evidence of person-separation reliability. It describes the separation of persons by their pattern of scores and better separation further reflects a more precise measurement.

We understand that the Rasch methodology is complex, but it is adequate when there are certain assumptions such as those presented in this paper (independence between items, unidimensionality). We believe that it is a strength of this work to combine two approaches (advanced factor analysis and Rasch); it is not common to be able to consult them in the same work because it cannot be done if the assumptions are not met (and because it is not a widely known methodology). We recommend the reviewer to read the following article (Open Access) that may help him/her to better understand this methodology if he/she is not familiar with it:

Stolt, M.; Kottorp, A.; Suhonen, R. The use and quality of reporting of Rasch analysis in nursing research: A methodological scoping review. Int J Nurs Stud. 2022,132,104244. doi: 10.1016/j.ijnurstu.2022.104244.

About the known group assessment (Line 543), not giving consent is part of the definition of OV that is used in the introduction, so I find it difficult to understand the meaning in this sentence.

We have explained before, there are differences between women depending on whether or not they have given their consent to an obstetric procedure. This scale evaluates the perception from the women's point of view. Incredible as it may seem, there are women who still do not perceive certain actions as OV.

I want to emphasis how important this work is and I wish the research group all the best on their journey!

Thank you very much for your time and comments. Of course we will continue working on these two lines of research: validation of measurement instruments and study of obstetric violence in our country. Best regards

Reviewer 2 Report

Although I find the text interesting and the scale quite enriching, I am a bit concerned about some aspects of the study that may limit it:

a. I am a bit concerned that the research seems to have been done only for Spain or scales related to Spain. I would like them to delve theoretically into more references to see if there are other scales in other parts of the world, as there may be elements that are being left out due to limited research. At this moment practically all the precedents are Spanish or Latin American.

b. I am concerned that there is no theoretical validation stage. In scale validation processes it is suggested that each question be associated with relevant theoretical aspects that are associated, in this case with the OV. In addition, we need a table that presents all the itesms and argues them theoretically before the validation with experts.

c. I am left with the doubt... is it your own scale? or are you validating someone else's? Why the Cardenas scale is mentioned? .... could you explain it better?

d. More information is needed from the experts. Area of expertise and years of experience. In order to be able to really argue that they are individuals who contribute to the validation, even to argue the selection of each one.

e. We return to a point mentioned earlier. We need to know more information about the participants. Are they all Spanish? Because I feel that the scale is very focused on Spain and this may become a limitation that does not allow us to generalize the results.

f. Put with XXX the region where the participants are from. We need to know where they are from to identify if there is a common regional pattern that influences their responses.

g. It is not clear what was the sample of women to whom the survey was applied in the pilot phase.

h. I am concerned about the gender imbalance of the experts. Why are there more women? Please either balance the experts or argue it, as it seems to be somewhat biased.

i. In the conclusions, it is important to consider a section on the limitations of the study, as well as the theoretical and practical implications of these results.

Author Response

Although I find the text interesting and the scale quite enriching, I am a bit concerned about some aspects of the study that may limit it:

1. I am a bit concerned that the research seems to have been done only for Spain or scales related to Spain. I would like them to delve theoretically into more references to see if there are other scales in other parts of the world, as there may be elements that are being left out due to limited research. At this moment practically all the precedents are Spanish or Latin American.

Indeed, obstetric violence is a subject that has been studied mostly in Latin America, and more recently in Spain. In other countries, as discussed in the introduction (lines 43-45), obstetric violence is not so much discussed and other terms are preferred. We have not found other scales in non-Spanish speaking countries that refer to obstetric violence, although in similar terms with disrespect, abuse, etc., as described in the introduction (lines 96-99). The construct is complex and is undoubtedly greatly influenced by the environment-context-country. We believe that in English-speaking countries obstetric care is generally more respectful. But we believe that this manuscript provides methodological aspects that can help other researchers in other countries with interest in the topic of obstetric violence, as well as in the validation of instruments.

2-I am concerned that there is no theoretical validation stage. In scale validation processes it is suggested that each question be associated with relevant theoretical aspects that are associated, in this case with the OV. In addition, we need a table that presents all the items and argues them theoretically before the validation with experts.

You are right, but the phase you allude to is when a new scale or instrument is developed, which is not the case here. We cannot present this since we have started from an already developed scale.

3-I am left with the doubt... is it your own scale? or are you validating someone else's? Why is the Cardenas scale mentioned? .... could you explain it better?

In relation to the above, this is explained in the introduction and the objective: There is a specific scale to evaluate OV as perceived by women, called Escala de Violencia Obstétrica (Obstetric Violence Scale), developed by Cárdenas and Salinero [40] based on the Test de violencia obstétrica (Obstetric violence test) developed by the association El Parto es Nuestro. (lines 109-117)

“The objective of this study was to carry out a cultural adaptation and validation of the Obstetric Violence Scale developed by Cárdenas and Salinero [40] to the Spanish context” (lines 120-122)

The test was born as a proposal of the association of women "El parto es nuestro" (in english “The childbirth is ours”) for women to express feelings and emotions related to obstetric violence. This test was converted into an OV measurement scale by Cardenas and Salinero and validated in Chilean population. This validation study is published in Spanish: Cárdenas, M.; Salinero, S. Validación de la escala de violencia obstétrica y pruebas de la invarianza factorial en una muestra de mujeres chilenas. Interdisciplinaria. 2021, 38, 209-223, doi:10.16888/interd.2021.38.2.14.

Answering therefore to your question, it is not our own scale, it is a scale developed by these authors. What we have done is to submit this scale to a stricter validation process, evaluating psychometric properties not considered in the original study and testing its validity in the Spanish population.

4.More information is needed from the experts. Area of expertise and years of experience. In order to be able to really argue that they are individuals who contribute to the validation, even to argue the selection of each one.

We accept your comment, but we are sincerely surprised (few studies include a complete profile of the experts as we have done). You can consult the experts' profiles in supplementary material S1. We have added the years of experience and reason for choice. We believe that by consulting the document you can assess that they are professionals with experience in childbirth care for women, combining professional and academic experience. Even the greatest expert in obstetric violence in Spain, Dr. D. Mena de Tudela, creator of the PercOV-S, was recruited. If the editor considers it appropriate, we can name the experts with names and surnames in the acknowledgements, since they are recognized professionals in our country.

5 We return to a point mentioned earlier. We need to know more information about the participants. Are they all Spanish? Because I feel that the scale is very focused on Spain and this may become a limitation that does not allow us to generalize the results.

All participants are Spanish, both experts and women. We accept that this is a limitation, and that our results cannot be generalized to other settings. We have added this limitation in discussion. We understand that this may perhaps be of less interest to readers from other settings or countries, although in our opinion the methodological design is rigorous, thorough, and well conducted and presented. We believe it may be of interest to readers interested around instrument validation.

6. Put with XXX the region where the participants are from. We need to know where they are from to identify if there is a common regional pattern that influences their responses.

Done. It was set to blind the review. All these data have been introduced.

7. It is not clear what was the sample of women to whom the survey was applied in the pilot phase.

Although it was there, we have marked it in red. Line 146: This pilot test was carried out in a sample of 20 puerperal women. Line 322: The scores assigned by the 20 participants in the pilot study.

8. I am concerned about the gender imbalance of the experts. Why are there more women? Please either balance the experts or argue it, as it seems to be somewhat biased.

While respecting your opinion, we do not believe that there is a gender bias in the choice of experts. Obstetric violence is an issue where women must be the protagonists, since they are the ones who suffer it. This is a topical issue because it is a demand of all women in the world, whose rights are violated. Most authors and experts on this subject are logically women (this is true all over the world, not only in Spain). On the contrary, we believe that the bias would have occurred if there were more men than women in the election. In fact, the choice of the male participant was very well thought out, as he was also an obstetrician (sometimes physicians have a very different view from midwives on this subject).

As the lead author of this study, although I am a midwife and consider myself to have acceptable knowledge of methodology, I believe that I would not have been able to lead this research without the help of my two female colleagues, whose vision as women has been fundamental to this work.

9. In the conclusions, it is important to consider a section on the limitations of the study, as well as the theoretical and practical implications of these results.

Thank you. Based on this comment we have introduced more limitations and we have tried to improve the conclusions, trying to insert the practical implications of this study. Best regards

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear Authors,

Thank you for this interesting article.

The article is fairly written. The article is fairly written and gives an elaborate description of obstetric violence (OV) and the relevance of the Obstetric Violence Scale in measuring OV. There is a need to attend to the minor corrections. I would be glad to read this when finally published (Subject to the Editorial board's decision).

My specific comments:

In the title, the Authors should add the context upon which the validation and evaluation is done. For example, the authors should add 'Spanish context'

On abstract, lines 20-21 can be shortened into one word: 'manifested through the attitudes of health professionals or the performance of unjustified or outdated practices without maternal consent.'......can read..."manifested in different ways". I mean, leave the details for the main manuscript.

The authors also to consider deleting redundant or unnecessarily repeated sentences. For instance, on line 576, "It has been described that the hospital setting is directly related to OV in childbirth." It can be deleted as the preceding sentence already addresses it.

The conclusion section has to be re-looked. The conclusion needs to give the readers take-home messages, and we should see the authors fulfilling the promises made by/in the title.

Regards

English is fairly good.

Author Response

Reviewer 3

Dear Authors,

Thank you for this interesting article.

The article is fairly written. The article is fairly written and gives an elaborate description of obstetric violence (OV) and the relevance of the Obstetric Violence Scale in measuring OV. There is a need to attend to the minor corrections. I would be glad to read this when finally published (Subject to the Editorial board's decision).

Thank you for your kind comments. We hope too that the manuscript will be published.

My specific comments:

In the title, the Authors should add the context upon which the validation and evaluation is done. For example, the authors should add 'Spanish context.'

Thank you. We appreciate this suggestion and accept it. We believe it improves the contextualization of the study, in line with the comments of the other reviewers.

On abstract, lines 20-21 can be shortened into one word: 'manifested through the attitudes of health professionals or the performance of unjustified or outdated practices without maternal consent.'......can read..."manifested in different ways". I mean, leave the details for the main manuscript.

Done. We believe that the abstract is better now.

The authors also to consider deleting redundant or unnecessarily repeated sentences. For instance, on line 576, "It has been described that the hospital setting is directly related to OV in childbirth." It can be deleted as the preceding sentence already addresses it.

Done. We appreciate very much these comments that improve the style of the manuscript. Many thanks.

The conclusion section has to be re-looked. The conclusion needs to give the readers take-home messages, and we should see the authors fulfilling the promises made by/in the title.

We have tried to improve the conclusions by trying to convey take-home messages to readers, as you’ve suggested. Thank you very much for your comments. We believe they improve the manuscript. Best regards.

Round 2

Reviewer 2 Report

Many thanks to the authors for their improvements. 

It is important that you consider that what we request as referees is not a matter of annoyance, but something that we consider relevant to add to your article for a better experience for the readers.

I do not doubt that you have all the knowledge on the subject and that you are specialists in this, however, at the level of academic writing, we must strive to make the articles interesting and clear for the readers. It is in this sense that we ask for changes. We are not questioning you or your work.

I am still concerned about the gender gap in the experts. I understand what you point out, but the fact that obstetric violence is primarily a female topic does not limit the fact that there are specialists on the subject who are men, and this allows us to have a better appreciation of the subject... above all, with the aim of having greater objectivity in the methodological process. 

Furthermore, it is important to point out that we are not asking you to give us the names of the experts, but rather more information, such as their area of expertise or simply why you have considered them as experts. This may even help to argue the lack of representation of male specialists.

Please, we as reviewers do not have access to the additional material 1, so I would appreciate a small table where you point out.... No. of expert, years of experience, specialty or why you are considered an expert.  It would really help the readers to understand the selection and why the majority are women.

In general, the change made in the rest of the article is much appreciated, which is very interesting and contributes a lot to this topic.

Author Response

Dear Reviewer:

Thank you for your time and comments. As we commented, we believe that there is no gender bias since the experts are prestigious professionals. Additionally, this paper is signed by three men and two women.
We must apologize as we thought that the reviewers had access to the supplementary material. Please find enclosed the profiles of the 8 participating experts. All this information will be available as supplementary material S1 for readers in table format.

Expert 1 Midwife Woman 25 years of experience.Coordinator of the Low Intervention Childbirth group of the Andalusian Association of Midwives. Lecturer in Perinatal Mental Health training. Choice: Women's rights activist in childbirth

Expert 2 Nurse Woman 15 years of experience.PhD. Associate Professor at the Universitat Jaume I (Castelló). Collaborator in nursing degree studies and with the advanced breastfeeding program at Blanquerna University. Published multiple studies on Obstetric Violence and creator of the PercOV-S questionnaire. Choice: In our opinion, the leading expert on obstetrical violence in our country. Creator of the PercOV-S questionnaire

Expert 3 Midwife Woman 30 years of experience.Midwife in Primary Care in Gran Canaria. Teaching collaborator. Responsible for the area of midwifery in the College of Nursing of Las Palmas for 20 years. Choice: After her long professional experience, she knows the testimony of thousands of women.

Expert 4 Midwife Woman 20 years of experience.PhD. Part-time lecturer at the University of La Laguna and works in assistance programs in the Canary Health Service over the area of women. She is a member of the group “Obstetric Violence Observatory” of the Ministry of Health. Choice: Knows the legal institutional environment of obstetric violence as a member of the group Obstetric Violence Observatory of the Ministry of Health

Expert 5 Obstetrian Man 35 years os experience.PhD. Part-time Lecturer at the University of Las Palmas de Gran Canaria. Head of Obstetrics and Gynecology Section of the Complejo Hospitalario Universitario Insular Materno Infantil de Canarias. Master in Bioethics. Member of the Provincial Ethics Committee. Choice: Expert in bioethics and patients' rights, in addition to being an obstetrician with a long professional career

Expert 6 Midwife Woman 20 years of experience.Midwife attached to the Multiprofessional Teaching Unit of Obstetrics and Gynecology. Speaker at national and international congresses and conferences related to maternity, childbirth, lactation and pelvic floor. Author of research papers published in national and international scientific journals. IP research study in collaboration with the Pelvic Floor Rehabilitation Unit of the Negrín Hospital. Choice: Responsible for the training of midwives in the study environment, in addition to training as a midwife in a country (UK) where there is a different sensitivity in childbirth care

Expert 7 Midwife Woman 10 years of experience.Midwife at Complejo Hospitalario Universitario Insular Materno Infantil de Canarias. Teaching collaborator, assistance and TFE tutor in the Multidisciplinary Unit of Obstetrics and Gynecology. President of ACAMAT (Canarian Association of Midwifes). Doctoral thesis on Obstetric Violence. Choice: She is doing a doctoral thesis on obstetric violence. She has a deep knowledge of the theoretical framework

Expert 8 Obstetrician Woman 8 years of experience.Gynecologist and Obstetrician at the General Hospital of Fuerteventura Virgen de la Peña. Choice: Having worked in two of the hospitals where the study has been carried out (HUMIC-Gran Canaria and Fuerteventura), she can provide a double vision, as they are two very different obstetric care contexts

We hope that with this information your comments will be satisfactorily answered. Our apologies again.
Best regards