Measuring Disrespect and Abuse During Childbirth in a High-Resource Country: Development and Validation of a German Self-Report Tool

Background. Increasing evidence on disrespect and abuse during childbirth has led to growing concern about the quality of care childbearing women are experiencing. To provide quantitative evidence of disrespect and abuse during childbirth services in Germany a validated measurement tool is needed. Research Aim. The aim of this research project was the development and psychometric validation of a survey tool in the German language that measures disrespect and abuse of women during childbirth. Methods. A survey tool was created including the following measures: German adaptations of the short and long form of the “Mothers on Respect” (MOR) index (MOR-7 and MOR-G); the “Mothers’ Autonomy in Decision Making” (MADM) scale; a mistreatment-index (MIST-I) comprising indicators of mistreatment during childbirth; and a set of items that measure experiences of discrimination during maternity care. Internal consistency reliability and construct validity of the scales were assessed using Cronbach’s alpha, unweighted least squares factor analysis and non-parametric correlation analysis with a scale that measures a related construct, the Posttraumatic Symptom Scale – Self Report (PSS-SR) scale. We distributed the survey online, recruiting through snowball sampling via social media. The nal sample of participants (n=2045) had given birth in Germany between 2009 and 2018. Findings. More than 77% of the study participants reported at least one form of mistreatment with non-consented care being the most commonly reported type of mistreatment, followed by physical violence, violation of physical privacy, verbal abuse and neglect. All included scales showed good psychometric properties with high Cronbach’s alphas (0.95 for both MOR versions and 0.96 for MADM). Factor analysis generated one factor scales with high factor loadings (0.75 to 0.92 for MOR-7; 0.37 to 0.90 for MOR-G and 0.83 to 0.92 for MADM). MOR-7, MOR-G, MADM and MIST-I scores were signicantly (p<0.001) correlated with PSS-SR scores (Spearman’s rho -0.70, -0.61 and 0.68 for MOR-G, MADM and the MIST-I, respectively). Conclusions. This study presents a valid and reliable instrument for the quantitative assessment of disrespect and abuse during childbirth in Germany. Childbearing women’s experiences of disrespect and abuse are a relevant phenomenon in German hospital based maternity care. Disrespect and abuse during childbirth appear to contribute to post-traumatic symptoms and may be associated with severe mental health problems postpartum. disabilities or chronic diseases and number of children. Perinatal questions asked about the date of childbirth, parity, multiple birth, stillbirth or disabled child, birth place, mode of birth and pregnancy risks. These questions were selected and translated from validated items in the “Giving Voice to Mothers” survey (18) and two other large national maternity care surveys, the British “Women’s Experiences of Maternity Care” (23) and “Listening to Mothers III” from the US (24). Scales and items measuring different manifestations of disrespect and abuse were chosen from the “Giving Voice to Mothers” survey (18): The 14 item “Mothers on Respect” (MOR) index (21) was included with three additional items. The seven item “Mother’s Autonomy in Decision Making” (MADM) scale (22) was used as is, and the seven “Mistreatment by Care Providers in Childbirth (MCPC) Indicators (18), items that measure experiences of mistreatment during birth, were also included in the survey. As the “Giving Voice and and Canada, was the items and scales were relevant in high resource settings and therefore adequate for use in Germany. The selected MOR, MADM and mistreatment items cover et proposed third-order themes. The items had a Likert format with answer options ranging The MCPC items measure distinct interactions with care providers rather than components of a scale to measure a single construct, hence the different types of mistreatment experiences could be indicated by checking a box. to review the instrument regarding relevance in the German context, comprehensibility and missing items. Engaging content experts in developing or reviewing items enhances the content validity of measures. disrespect and abuse during childbirth (D&A) in German maternity care. The results suggest that women in Germany experience disrespectful and abusive care similar to women in other high-income countries. The women’s reported experiences reected all third-order themes of Bohren et al.’s (2015) typology of mistreatment during childbirth (16).


Introduction
Increasing evidence of disrespect and abuse (D&A) of women during childbirth has caused growing concern worldwide about the quality of care childbearing women experience. The World Health Organization emphasises a positive childbirth experience as a human right and provides evidence for respectful maternity care as essential to improve women's experience of labour and birth, (1). A negative experience of the birth event can have serious adverse effects on maternal mental health and was found to be the most signi cant predictor for postpartum posttraumatic stress disorder (PP-PTSD). PP-PTSD affects between 4.6 and 6.3% of childbearing people (2), and experiences of traumatic childbirth seem to be strongly related to care provider actions and interactions (3)(4)(5).
Disrespect and abuse during childbirth was rst de ned and categorized by Bowser & Hill (6) in their landscape analysis. The authors identi ed different manifestations of D&A in facility-based childbirth and classi ed them in the following seven categories: physical abuse, non-consented care, non-con dential care, non-digni ed care, discrimination based on patient attributes, abandonment of care and detention in facilities. These categories have been used to inform tools to measure D&A, predominantly in low resource settings (7)(8)(9)(10)(11)(12). Other self-report measures have been developed and validated in Ethiopia (13), India (14) and Iran (15).
Based on the evidence of a systematic review of 65 predominantly qualitative studies conducted in 34 countries, the WHO Research Group on the Treatment of Women during Childbirth presented a typology of the mistreatment of women during childbirth in health facilities worldwide (16). The authors differentiate the themes physical abuse, sexual abuse, verbal abuse, stigma and discrimination, failure to meet professional standards of care, poor rapport between women and providers and health system conditions and constraints. To identify grievances in maternal care and improving its quality, the authors emphasise the need to develop evidence-based, validated and reliable tools for a quantitative assessment of mistreatment of women during childbirth, suggesting their typology to be used for this purpose. A recent cross-sectional study with labour observations and community-based surveys performed in four low-income and middle-income countries based on this typology showed a prevalence of D&A between 35 and 42 percent (17).
Quantitative evidence on disrespect and abuse during childbirth in high-income countries is still scarce. Some data are available from the USA: The "Giving Voice to Mothers" study (18), a nationwide online cross-sectional survey addressing service users of maternity care, revealed that 17.3% of the 2,138 participants who had given birth in the USA between 2010 and 2016 had experienced one or more types of mistreatment.
In Germany respectful maternity care is understudied. However, an increasing number of reports of women who have experienced disrespectful and abusive care spread on social media. Every year, on November 25th, the international day against violence towards women, the activists of "Roses Revolution", a global movement against obstetric violence, invite women to lay down a rose at the door of the labour ward where they had suffered disrespect and abuse. Following this call in 2017, women laid down roses at the doors of 171 labour wards and three birth centres in Germany (19).
When developing items and scales to measure disrespect and abuse during childbirth, Freedman et al. (20) emphasise that D&A needs to be understood in local contexts, because what is considered as D&A may differ from country to country, between cultures and maternity care settings. Thus, tools have to be developed, adapted and validated for their respective local contexts. To date, no validated measurement tools are available in German language and no studies have quantitatively assessed the phenomena of D&A in German speaking countries. This paper presents the development and psychometric validation of a survey tool that measures disrespect and abuse of women during childbirth in the German language. Methods A German language survey instrument was developed, based on existing tools that measure common aspects of respectful maternity care and mistreatment during childbirth in high-income countries. The tools, originally developed for use in North America (21,22), were translated and adapted to German maternity care conditions. A cross-sectional online survey was administered to validate the survey instrument.

Survey development
Following experts' opinion the third MOR section "When I had my baby I was treated poorly […] because of" was considered incomplete because a number of common reasons for discrimination were missing. Therefore, it was extended with additional items naming other social categories as listed in the "General Act on Equal Treatment"(AGG) (27), the German anti-discrimination law: age, disability or chronic disease and socio-economic status. Furthermore, being HIV positive and being overweight was included because of the strong stigmatising effect of these attributes. In total, this resulted in nine items addressing discrimination (table 4).
In view of reports of childbearing women in Germany (19,28), three additional items were added to the mistreatment section, one asking about disparaging remarks against the childbearing woman, one about non-consented interventions (including episiotomy, caesarean section, labour augmentation, amniotomy, drug injection and venous access) and one about the use of fundal pressure during the second stage of labour (table 5).
Inclusion of a PTSD screening tool to assess convergent validity Convergent validity is established when two or more scales that measure related constructs show a high degree of association. Evidence on the relationship between inappropriate intrapartum care and trauma (2,3,29) suggested the use of a self-report tool that screens for posttraumatic stress disorder (PTSD) for convergent validation. Out of the available tools validated in the German language the "PTSD Symptom Scale -Self Report" (PSS-SR) (30) was chosen for the present study because of its reasonable number of items, its appropriate psychometric properties and its previous use in other post partum studies (29,(31)(32)(33). The PSS-SR is highly speci c for PTSD, with its 17 items corresponding directly with the symptoms for PTSD as presented in the DSM IV (34).

Pilot testing
The questionnaire was pilot tested by 14 women who had given birth in the last 3 to 6 months. They reported on the time needed to ll in the questionnaire and if they had any di culties answering the questions or problems with comprehensibility or clarity of wording. They also had the opportunity to provide general comments on the questionnaire. Participants endorsed all of the items, thus providing evidence of content validity from the perspectives of service users. Participants suggested to add a question about the lack of time or presence of caregivers. Abandonment of care is one of the Bowser and Hill (6) categories and part of Bohren et al.'s (16) typology, therefore the item: "The health care providers had no time for you when you needed help " was added in the mistreatment section of the survey.

Sampling procedure
The Facebook website of Roses Revolution Germany (www.facebook.com/RosesRevolutionDeutschland) with more than 10,000 followers was chosen for the distribution of the online survey. Through further distribution via social media, a snowball sampling process started. After four weeks, with 3,336 responses including 2,045 fully completed questionnaires, the survey was closed.

Inclusion criteria
Participation was restricted to women who had given birth in Germany between 2009 and 2018, to ensure that results re ect the current maternity care situation in Germany.

Ethical considerations
Approval to conduct the study was given by the ethics committee of Hannover Medical School and was approved on September 10th, 2018 (Ref No: 8075_BO_K_2018). Participation was anonymous and con dentiality was assured. Respondents could stop participating at any stage. We informed participants that recalling a previously traumatic event may be distressing, and provided contact details of the research team as well as free national counselling services for support.

Measures
The MOR-7 is the German translation of the original "Mothers on Respect" seven item scale developed in Canada. Scale scores ranged from 7 to 42. The long version of the MOR scale included an additional 6 items and was named "Mothers on Respect -German" (MOR-G). The possible range of scores was 13 to 78. Higher scores on the two scales indicated a higher degree of experienced respect. Possible MADM scores ranged from 7 to 42, higher scores indicating that the care provider facilitated autonomy in decision making to a higher degree.
In order to be able to investigate women's discrimination experiences as a separate concept instead of subsuming it under the concept of respect, the nine discrimination items were separated from the MOR scale. The Likert six-point response options were recoded into dichotomous variables with the values that indicate disagreement (1 -3) being grouped together and the values that indicate agreement (4 -6) constituting the group that experienced discrimination. Multiple forms of discrimination were assessed by summing up the recoded item values. The scores ranged from 0-9.
A mistreatment-index (MIST-I) was de ned according to the number of marked mistreatment items and calculated by summing up the dichotomous item values coded as 1 = "answered" and 0 = "not answered". The scores ranged from 0 -12.

Data analysis
Statistical analysis was performed via IBM SPSS, version 25. Descriptive statistics were conducted to assess sample characteristics and to describe scale scores. Cronbach's alpha was computed for the scales to assess internal consistency reliability. Corrected item-to-total-correlations were calculated to assess whether individual items correlate highly with the sum of the items. Items with high correlation coe cients indicate high conceptual overlap whereas items with low coe cients indicate that the item might not measure the same construct. Information about how much each item impacted the reliability of the respective scale was given by the comparison of the Cronbach's alpha if the item would be deleted with the overall Cronbach's alpha.
Factor analysis was performed for the included scales, i.e MOR-7, MOR-G and MADM to determine how many dimensions the scales had. For these very short scales it was desirable that they were not only reliable but also uni-dimensional. The suitability of the data set for factor analysis was assessed by the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and the Bartlett's test of sphericity. Factor extraction was performed by unweighted least squares; the scree test was used to con rm the number of factors to be retained.
Non-parametric correlation analysis was used to assess convergent validity of the translated MOR and MADM scales and the mistreatment index (MIST-I) against the PSS-SR scale. Spearman's rho values above 0.5 were interpreted as a large or strong positive correlation, values below -0.5 as a large or strong negative correlation.

Sample characteristics
Participating women (n = 2045) gave birth in all German federal states. The participants' mean age at the date of childbirth was 29.8 years (SD 4.5; min/max: 16 -44). About half of women had obtained higher education at the university level (51.4%), accordingly, the distribution of income in the sample displayed a strong tendency towards medium and high income, with only around 10% reporting low-incomes (less than EUR 1400 net household income). Women with a migration background represented 8.9% of the sample.
Most participants reported on the birth of their rst child (77.4%); 22.6% were multiparous. 59.5% had low-risk pregnancies without any health problems, the others marked one or more pregnancy risks. Two percent reported on the birth of twins and four participants had triplets or more. A stillbirth or a disabled child was reported by 0.8% and 1.2%, respectively.
A hospital was the place of birth for 86.5% of the participating women (planned 74.2%). Six percent of women had a hospital birth with a caseload midwife (planned 7.5%). Out of 401 planned community births (19.6%), 77 women (3.8%) gave birth in a birth centre and 77 (3.8%) at home. The birth mode was spontaneous in 56.0% of the cases, 11.1% had a vacuum extraction, 1.3% a forceps and 31.2% a caesarean section, including 7.1% emergency caesareans.

Experience of respect
Around half of the participants reported that they felt comfortable asking questions of their intrapartum care providers while making decisions, felt comfortable accepting or declining care that was offered, or that their personal preferences had been respected (table  1). In contrast, around half of the participants reported that they held back from asking questions because their care providers seemed rushed, because they wanted maternity care that differed from what their care providers wanted, or because their care providers might view them as di cult or would not value their opinion. Nearly 60% (n=1219) felt coerced into accepting care options their care provider suggested. A lower percentage felt discriminated against (18.2%, n=372) or had di culties understanding the language the care provider used (8.2%, n=168). The median MOR-7 score was 24 (5th/95th percentile: 7 -42, range: 35). The median MOR-G score was 47 (5th/95th percentile: 25 -77, range 65).

Experience of autonomy in decision making
Less than one third of the participants (28.5%, to 30% (items 3 to 6) and around 30% of women felt neglected or ignored by health care providers (items 7 and 8). Breaches of con dentiality were reported by less than ve percent of the participating women.
Internal consistency reliability and construct validity of the scales Table 6 provides a summary of scale characteristics and reliability statistics. The Cronbach's alphas were 0.95 for both MOR scales and the PSS-SR scale and 0.96 for the MADM scale, indicating very high internal consistency reliability of all included scales.
Corrected item-to-total correlations for MOR-7 exceeded 0.73 for all items (table 2). These strong correlations indicate that the scale is homogeneous, measuring one single underlying construct. The internal consistency reliability of the scale could not be improved by item deletion. For the 13-item version MOR-G the corrected item-to total correlations ranged from 0.37 to 0.87 (table 1). As for the seven-item version, the internal consistency reliability of the scale could not be improved by item deletion. Similarly to MOR-7, the MADM scale showed very high item-to-total correlations (0.81 to 0.90) and no improvement of internal consistency reliability by item deletion (  Non-consented care was the most commonly reported form of mistreatment: More than 40% of the participants reported that intrapartum interventions were carried out without their consent. These numbers con rm the ndings of  (37), who found non-consented care to be the most often occurring form of disrespect and abuse observed by doulas, nurses and midwives. Compared to the participants in the Canada-based "Changing Childbirth in British Columbia" (21) study women in the current study experienced far more disrespect and coerced decision making during childbirth. For example, in British Columbia around 90% of the survey participants agreed to have felt comfortable asking questions while making decisions, to have felt comfortable accepting or declining care that was offered and that their personal preferences had been respected. In the present study only around half of the participants agreed to the respective items. Furthermore, almost 60% felt coerced into accepting care options their care provider suggested, compared to 13% in the British Columbia sample and around 20% in the US Giving Voice to Mothers study (18). With respect to autonomy in decision making, similar differences were observed: The median "Mothers' Autonomy in Decision Making" (MADM) score was 18 in the present study, less than one third of the total scale range, and 39 in the British Columbia sample (38), which is very close to the highest possible scale score of 42. These discrepancies con rm the known selection bias resulting from recruitment via a mothers consumer organization representing women affected by obstetric violence in the present study, whereas the British Columbia study (21) embedded MOR and MADM into a large survey on maternity care experiences in a geographically and socioeconomically diverse population. In addition, the North American studies had a disproportionately high number of midwifery clients, and the US study also included many people who gave birth at home or in birthing centres (50% of sample). Previous research has shown that midwifery clients and those planning a community birth have much lower rates of mistreatment (18) and higher respect and autonomy scores (21,38).
Being coerced into accepting interventions instead of being engaged in a process of informed decision-making during childbirth undermines women's autonomy and constitutes a disrespectful and abusive practice. Coercion has been reported previously in the "Listening to Mothers III" survey, a national survey of 2,400 US women who gave birth in US hospitals from mid-2011 to mid-2012 (24): Women experienced pressure from health professionals to accept labour induction (15%), epidural analgesia (15%), or caesarean section (13%). Similarly, in the "Giving Voice to Mothers" study, 13.0% felt pressured to have labour induction, 7.3% to have epidural analgesia, and 10.6% to have a caesarean Sect. (18).
In the current study, one third of the participating women reported physical abuse. This constitutes a large proportion compared to the few "Giving Voice to Mothers" survey participants who agreed to the respective item (1.3%)(18). Besides the selection bias towards women affected by abusive intrapartum care in the present study, a broader de nition of physical abuse may explain this difference. As suggested by the German experts, painful vaginal examinations and insu cient anesthesia for the suture of an episiotomy were included as additional examples in the translated item.
The high number of women reporting the use of fundal pressure by maternity care providers in the second stage of labor is another important nding of the current study. Twenty-seven percent of the survey participants have experienced this intervention for which there is insu cient evidence (39). Fundal pressure is commonly used with the indication of maternal exhaustion or suspected fetal distress in order to avoid instrumental birth but often it is applied without formal indication (40). According to the guidelines of the International Childbirth Initiative (41), fundal pressure is among the harmful procedures to be avoided; in addition, its use is not recommended by the WHO (1). For women, this intervention is frequently experienced as physical violence and can be traumatising (28, 42).
Violation of physical privacy also was a frequently reported experience in the present sample, which is congruent with ndings from Vedam et al. (18). Being uncovered and having unknown people, e.g. medical students, watching the birth without the woman's consent can cause distress and loss of dignity as qualitative research showed (43)(44)(45).
A large number of women in the present study felt ignored or did not get help when needed. Both quantitative and qualitative studies found neglect and abandonment to be one of the most frequently cited mistreatment experiences and is linked to women's perceptions of traumatic childbirth (4,18,46). Neglect and abandonment of childbearing women, next to indicating disrespectful attitudes of care providers or a disrespectful facility culture, are likely to also be a consequence of structural constrains leading to sta ng shortage, which is described by Bohren et al. (16) under the theme "health system conditions and constraints". The shortage of midwives in German hospitals -with one midwife caring for up to four labouring women (47) -may play a key role for failure of professional standards and meeting women's needs. Given the fact that a delayed response to clinical warning signs has been found to be one of the most common types of contributors to maternal deaths (48), these ndings are alarming, especially in a high resource setting. Overall, the survey instrument in the current study showed very good psychometric properties. Feedback from the pilot testing phase and the expert review provided support for the content validity of all included measures. The internal consistency reliability of included scales exceeded 0.95 and was high compared to other published tools in the area of research (13)(14)(15) and comparable to the ndings of the scale developers (21,22). Similarly to the original scales, very high item-to-total correlations above 0.7 were found for MOR-7 and for MADM, i.e. each single item strongly correlated with the sum of all other items of the respective scale, thus providing strong evidence for the homogeneity of these scales (49).
Because of their homogeneity, uni-dimensionality of the scales was assumed. Factor analysis con rmed this assumption for all scales: With only one eigenvalue larger than one, scree plots showing a clear elbow curve and high loadings on one factor, MOR7, MORG and MADM -similarly to the original scales -form uni-dimensional scales with good construct validity. Future users of the MORG scale might discuss eliminating item f ("During my birth I held back from asking questions or discussing my concerns because my doctor or midwife didn't use language that I could understand.") because of its far lower factor loading in comparison to the other items of this scale.
The almost perfect correlation between MORG and MOR7 allows to consider the use of MOR7 as su cient for the assessment of respectful care. It would be a short and effective measure and, as it has not been altered in the validation process, it may be more useful for international comparisons than the adapted version MORG. On the other hand, MORG integrates more aspects of respectful care drawing a broader picture of the construct respect. In consequence, for German prevalence or intervention studies MORG would be the preferable option.
Finally, strong and signi cant negative correlations of MOR7, MORG and MADM scale scores with PSSSR scale scores assessed convergent validity of these scales, thus further con rming construct validity. Based on scienti c evidence on the relationship between inappropriate intrapartum care and trauma (2,3,5,29,43), it has been hypothesized that low perceived respect or autonomy during childbirth would be associated with increased posttraumatic stress symptoms. Hollander et al. (50), for example, found lack of autonomy in decision making to be attributed to childbirth trauma by 30% of the participants of their cross-sectional survey conducted in the Netherlands among 2,192 women with a self-reported traumatic birth experience. However, measures of respectful care, autonomy in decision making or mistreatment during childbirth have never been correlated with validated measures of postpartum PTSD before. A strong association of indicators of obstetric violence with postpartum depression has recently been assessed in a Brazilian cross sectional study with 10,468 women (51): Women who experienced neglect, verbal violence, or physical violence were found to have an up to seven times higher risk of developing postpartum depression than women without these experiences. It has to be noted, however, that the authors did not use a validated instrument to measure obstetric violence.

Strengths and limitations of the study
The present study is the rst quantitative study on disrespect and abuse during childbirth (D&A) in the German maternity care context. Furthermore, the current study assessed associations between indicators of D&A and symptoms of postpartum posttraumatic stress disorder for the rst time. Convergent validation against a measure of PTSD is a new and signi cant addition to the testing of validity and utility of the MOR and MADM scales.
Certainly, the high sample size is another strength of this study. As delineated above, the sample is not representative of the target population, and therefore generalisability of the results is limited. The selection bias towards women who experienced disrespect & abuse in the sample, however, facilitated tool validation in particular with regard to the relevance of the items. The recall bias is assumed to be minimal in this study because women's recall of their childbirth experiences has been shown to be very accurate when compared to medical records, even 10 to 15 years after the event (52).
Due to a transcription error from paper to the survey software, the PSS-SR item "Did you have trouble concentrating?" was missing in the survey. One possibility to overcome this problem could have been to replace the missing values by the respective means of the other 16 values to get an estimate of the PSS-SR scores. Although this option is an established statistical procedure to deal with a moderate number of missing data, it is not recommended when all values are missing because it can severely distort results of analyses (35). All calculations therefore were made with the remaining 16 items. The Cronbach's alpha was higher than has been reported for the original scale (30) and also higher than in later studies using the PSS-SR postpartum (33,53,54). Therefore the 16item scale was considered to be valid and useful for the purpose of this study to assess convergent validity between the constructs of respect, autonomy in decision making, and mistreatment and the construct of postpartum posttraumatic stress. The missing item, which refers to di culties concentrating after the event, moreover, can be considered as one of the less meaningful symptoms when used to assess posttraumatic stress in the puerperium. Di culties concentrating may be normal in this phase of reorientation, often accompanied by lack of sleep, instead of indicating trauma.
Despite the strong associations observed between indicators of D&A with postpartum PTSD symptoms, causality cannot be assumed because of the cross-sectional nature of the study measuring outcome and exposure simultaneously. Furthermore, the survey participants were not screened for prior trauma, one of the main predictors of postpartum PTSD (2).

Conclusions
This study presents a valid and reliable instrument derived from international evidence for the quantitative assessment of disrespect and abuse during childbirth in Germany. It addresses the four domains respect, autonomy in decision making, discrimination and mistreatment, which are relevant aspects for high-income countries and align with WHO de nitions of disrespect and abuse. The included translated and adapted scales showed very good psychometric properties when administered to a German sample of childbearing women. The study results demonstrate that childbearing women's experiences of disrespect and abuse are a relevant phenomenon in German maternity care.
Disrespect and abuse during childbirth may be associated with severe mental health problems postpartum. By providing respectful, digni ed and supportive care to women during childbirth, midwives and other maternity care providers play a key role in preventing postpartum trauma with its possible negative impact on women's, family and child health. Participation (online) was anonymous and con dentiality was assured. Respondents could stop participating at any stage.

Declarations
Consent for publication: The manuscript contains no individual person's data.
Availability of data and materials: The datasets generated and analysed during the current study are not publicly available because they are still used in further research but are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests.
Funding: The reported research was not funded.
Authors' contributions: MMG and JL made substantial contributions to the conception and design of the work. KS made substantial contributions to the data analysis and SV substantially revised the manuscript. All authors read and approved the nal manuscript and have agreed to be personally accountable for their contributions and have appropriately investigated and resolved all questions related to the accuracy and integrity of any part of the work.