Changes in health sciences students' perception of obstetric violence after an educational intervention.

BACKGROUND
Obstetric violence is a type of gender-based violence that is presented structurally. This type of violence has physical and psychological consequences for both the women who experience it and health professionals. The World Health Organization adds that health professionals need training to ensure that pregnant women are treated with compassion and dignity.


OBJECTIVES
The objective of the study was to evaluate health sciences students' perception of obstetric violence and to identify possible changes after an educational intervention.


DESIGN
A pre-post quasi-experimental study was carried out between January and June 2019.


SETTINGS AND PARTICIPANTS
Students of medicine and nursing from Jaume I University (Universitat Jaume I) (Spain).


METHODS
An ad hoc scale comprising 33 items was designed to measure the students' perceptions. In addition, sociodemographic and control variables were collected. Descriptive analyses of the sample and the scale were carried out, and a bivariate analysis was performed.


RESULTS
Of the students surveyed, 89.7% were women, and the majority was nursing students. Of the 33 items, 28 (84.84%) showed statistically significant changes in the pre-post-intervention measurement. Twenty-five of the 33 items (75.75%) showed a relationship with the sociodemographic variables of gender, field, course and ever having been pregnant.


CONCLUSION
This study shows the change in health sciences students' perceptions of obstetric violence after an educational intervention. In addition, the normalization of this type of violence was observed with the progression of training and with personal obstetric experience.


Introduction
The World Health Organization (WHO) states that "all women have the right to receive the highest level of health care, which includes the right to dignified and respectful care in pregnancy and childbirth, and the right not to suffer violence or discrimination" (WHO, 2014). In 1985, the European regional office of the WHO, the Pan American Health Organization and the regional office of the WHO for the Americas, at a conference on appropriate technology for childbirth, created a series of consensus recommendations among obstetricians, paediatricians, midwives, psychologists, epidemiologists, mothers and other professionals. The result was the "Declaration of Fortaleza" (World Health Organization, 1985), which the WHO considers applicable to all perinatal services throughout the world.  (Diaz-Tello, 2016;Pérez D'gregorio, 2010).
The WHO warns that an increasing number of studies on the experiences of women during pregnancy and, in particular, during childbirth present an alarming scenario, indicating that many women around the world experience disrespectful, offensive or negligent treatment during labour (WHO, 2014). In addition, it describes the practices that make OV visible: disrespectful and offensive treatment during childbirth, physical abuse, profound humiliation and verbal abuse, medical procedures performed without consent or under coercion (including sterilization), lack of confidentiality, failure to obtain the complete informed consent, refusal to administer analgesics, flagrant violations of privacy, refusal of admission to a health centre, negligence towards women during childbirth and the retention of women and new-borns in health centres due to their inability to pay (WHO, 2014), among others.
Concerning this type of violence, several conjectures have been raised regarding possible variables that favour its social stratification (Castro and Frías, 2019): lower socioeconomic level (Brandão et al., 2018;Santiago et al., 2018); youth, race, poor economic status and women's ignorance of their rights (Perera et al., 2018); or having dark skin (Grilo Diniz et al., 2018). However, an analysis of the main reasons and places of occurrence of this type of violence can reveal that OV is a type of structural violence. Structural violence is one J o u r n a l P r e -p r o o f The main reason for OV is gender bias, in which women's right to choose is nullified and replaced (Jardim and Modena, 2018). Regarding sites of occurrence, OV occurs throughout the world. Evidence shows that it exists in countries such as Mexico (Castro and Savage, 2019;Castro and Frías, 2019;Santiago et al., 2018) Ecuador (Brandão et al., 2018), Italy (Castro and Frías, 2019;Ravaldi et al., 2018;Scambia et al., 2018), the United States (Perera et al., 2018), andNigeria (Ishola et al., 2017), among other places. The structural nature of OV makes the health professional who exercises it unaware of it and even normalizes this practice (Borges, 2018).
Practices characterized by OV have physical and psychological consequences for both the women who experience them and the health professionals who practise or witness them. Women have shown how their physical, sexual and psychological health has been negatively affected (Chattopadhyay et al., 2018), and a very meaningful experience in their lives has been transformed into a violent and negative one (Borges, 2018;McGarry et al., 2017). On the other hand, the literature suggests that personnel who witness this type of violence during childbirth may suffer from compassion fatigue or secondary traumatic stress in response to observing the traumas the woman experiences first-hand (Sadler et al., 2016). The WHO states that it is necessary to generate data related to respectful and disrespectful care practices, responsibility systems and valuable professional support, adding that health professionals need support Journal Pre-proof J o u r n a l P r e -p r o o f and training to ensure that pregnant women are treated with compassion and dignity (WHO, 2014).
Is important to noted that no similar studies have been found in the literature; more specifically, there are no studies analysing health sciences students' perspectives regarding OV. Therefore, the main objective of this study was to evaluate the perceptions of health sciences students at Jaume I University (Universitat Jaume I) (Spain) have regarding OV and to detect possible changes in these perceptions after an educational intervention.

Design and sample
A pre-post quasi-experimental study was conducted among health sciences students at Jaume I University (Universitat Jaume I) (Spain) between January and June 2019.
Returned questionnaires with up to 10% of items incomplete were excluded from the study analysis. A sample size calculation was performed using the GRANMO programme, which determined that a sample of 99 subjects was sufficient. The values considered for the calculation of the sample size included a confidence interval of 95%, highlighting an initial proportion of events of 0.1 percentage points and a loss to follow-up of 20%.

Variables and instruments
The sociodemographic variables that were considered were age, gender, field (medicine, nursing), course, health experience in gynaecology and obstetrics services (yes, no), duration of experience (less than 1 year, between 1 and 4 J o u r n a l P r e -p r o o f years, more than 4 years), having been present at a birth (yes, no), duration of experience being present at births (less than 1 year, between 1 and 4 years, more than 4 years), personal experience with pregnancies and births (yes, no), time since pregnancy and birth (less than 1 year, between 1 and 4 years, more than 4 years).
The perception of OV was measured with an ad hoc questionnaire composed of 33 items that referred to OV practices and were divided into 4 key moments (before delivery, during delivery, in case of caesarean section and after delivery). These items were measured using a 5-point Likert scale (1 strongly disagree -5 strongly agree).

Analysis
A descriptive analysis of the data was performed considering means, standard deviations and 95% confidence intervals for the quantitative variables and the distribution of frequencies and percentages were taken into account for the qualitative variables. For the bivariate analysis, applicability was determined using parametric tests, the Kolmogorov-Smirnov normality test and Levene test for the homogeneity of variances. After these conditions were confirmed as acceptable, Student's T test was applied for paired data, with the intention of detecting the effect of the change in the different measurements, and the Mann Whitney U and Kruskal Wallis tests were used to determine the relationships of the sociodemographic variables and perceptions of OV with the responses on the pre-intervention measurement. The analysis was carried out with the statistical package Statistical Package for Social Sciences (SPSS), version 21.
A statistical significance level of p < 0.05 was established.

J o u r n a l P r e -p r o o f
The study was approved by the management of the Jaume I University (Universitat Jaume I) Nursing Research Group. The intervention was approved by the directorate of the Nursing Department and the dean of the School of Health Sciences of Jaume I University (Universitat Jaume I). Before data collection, the students received information about the objectives of the study as well as its methodology and the voluntary and anonymous nature of participation. The data collection tool did not include any personal data that could compromise the identity of the participants. The project was designed in accordance with the December 5 Protection of Personal Data and Guarantee of Digital Rights Organic Law 03/2018. In addition, the principles of the Declaration of Helsinki (charity, nonmaleficence, autonomy and justice) were respected. To respect the anonymity of the data and to match the first and second measurements, an ID was created consisting of the last two digits of the student's cell phone number, the last two digits of his or her National ID and his or her initials.

Results
A total of 107 questionnaires were collected. The mean age of the students was 22.5 years (± 5.87). Women represented 89.7% of the sample (n = 96). A total of 86.9% (n = 93) of the students belonged to the degree programme in nursing, and 28% (n = 30) of the students had completed clinical practice in the gynaecology and obstetrics departments. Of the sample, 20.6% (n = 22) had been present at a birth; only 4.7% (n = 5) had been pregnant, and 2.8% (n = 3) had given birth (Table 1).

J o u r n a l P r e -p r o o f
The Kolmogorov-Smirnov test (p <0.01) and the Levene test (p <0.05) verified the appropriateness of the Student's t-test for paired data. Table 2 shows the descriptive analysis of the variables for the pre-and post-intervention measurements and the bivariate analysis. All analyses of the OV perception scale by paired data showed statistically significant differences between the pre-and post-intervention measures, except on the items related to performing a pelvic examination without consent (p = 0.368); not preserving the privacy of the woman (p = 0.389); not considering the woman's decision (p = 0.086); taking pictures without permission (p = 0.379); saying "Stop complaining, it is not that bad" (p = 0.181); Separating the mother and new-born (p = 1.00); and giving formula to the baby without the mother's consent (p = 0.320).  Table 4).

Discussion
First, is important to emphasize that students' participation in the proposed teaching activity was much greater than initially expected. The students seemed very motivated by the central theme, and their involvement was notable; therefore, the capture of individuals from the sample exceeded the calculated sample size. The high percentage of women in the sample (89.7%) should be noted; it may have occurred because women were especially motivated by the issue or because female representation is increasing in the health sciences (Bernalte-Martí, 2015). It is noteworthy that students in different years of their programme were equitably represented in the seminar, although more secondand fourth-year students than students of other years were in attendance. The representation of medical students was low; some possible reasons for this low attendance may be low dissemination of the activity among these students or the possibility that nursing students feel more linked to this type of practice (Olza-Fernández and Ruiz-Berdún, 2015). Because our sample was young in relation to the mean age (32.58 years) for maternity in Spain according to data J o u r n a l P r e -p r o o f from the National Institute of Statistics (Instituto Nacional de Estadística -INE)(«Edad Media a la Maternidad por orden del nacimiento según nacionalidad (española/extranjera) de la madre(1579)», s. f.), very few participants had a personal medical history of pregnancy or birth, although 30% of the sample had experience in gynaecology and obstetrics.
Regarding the comparison of the pre-and post-intervention measures by paired data, it is noteworthy that with the exception of a few items, all measures presented statistically significant differences. Even so, it is worth highlighting the low results found in the pre-intervention measurement. The Fortaleza document (World Health Organization, 1985) states that a family member chosen by the mother may accompany her during childbirth and throughout the postnatal period to promote her well-being; the new-born should stay with his/her mother whenever possible; immediate breastfeeding should be promoted; the dorsal position of the woman in lithotomy during dilation is not recommended; and the shaving of pubic hair, the administration of enemas, the systematic use of episiotomy and the early artificial rupture of membranes should not be performed as routine procedures. All these procedures were included in the questionnaire, and of them, the only one that was identified as violence preintervention was "giving formula without the mother's consent".
It should be noted that health science students should be trained in the latest available evidence (Aglen, 2016). Apparently, this does not occur in the field of obstetrics in Spain, highlighting the need for all women of child-bearing age to Another feature of this OV is that it is rooted in a system that stands in the way of optimal health outcomes (Castro and Savage, 2019); thus, it also has a structural nature (Bhattacharya and Sundari Ravindran, 2018). In this way, the researchers assumed two facts that were confirmed through this study: a) the normalization of this type of violence according to the student's year of study, i.e., a lower perception of OV among more advanced students and a relationship between perceptions of OV and having participated in obstetric practices during study; and b) the normalization of this type of obstetric J o u r n a l P r e -p r o o f practices in relation to the participant's personal experience with pregnancies and births (a decreased perception of OV after having been pregnant or given birth). A larger study is necessary to determine the degree of normalization and the normalization process; however, given these preliminary data, it is essential to change the training of health personnel, who should have a solid foundation in ethics and gender and human rights because emotional factors or burnout may be among the reasons for practising OV (Olza Fernández, 2013). The strain on health personnel is so high that many professionals have to abandon their job and even their profession (Beck and Gable, 2012). Therefore, public policies must direct attention towards humane and respectful treatment that is based on and supported by the latest available evidence. However, in order for this to happen, health personnel who work with pregnant women must abandon the traditional hierarchy and structure in which medical supervision implies a subordination of women's bodies and sexuality. This fact is further aggravated when all attention is paid to techniques, and the value of how people are treated is lost (Grilo Diniz et al., 2018;Mselle et al., 2018). Education that promotes respect and informs and raises awareness among future professionals, along with policies, guides, protocols and education, will eradicate OV (Brandão et al., 2018;Diaz-Tello, 2016;Grilo Diniz et al., 2018;Mselle et al., 2018;Sen et al., 2018). Education is a fundamental aspect for ending the normalization of OV in society; it approaches the problem from the root and will evolve until the rights of women are respected. The results should be taken with caution because there are some limitations. This is a quasi-experimental study without a control group, carried out in a single institution and, also, the post-test data were collected immediately after the intervention. In spite of these limitations, the J o u r n a l P r e -p r o o f results obtained are interesting because they highlight the consideration of OV by health students and add value because they show that it is possible to change this perception with an educational intervention directed specifically at OV. Among the future lines of research, the authors propose designs that overcome these limitations, for example, considering a control group and longterm monitoring of students.

Conclusions
According to the results of this study, health sciences students integrate the normalizations of OV during their studies. A formative activity aimed at making this type of violence visible and reflecting on OV helps to create awareness among students, making it possible for them to notice this type of violence and be able to identify it. It is noteworthy that from the beginning, the women in the study have perceived all the points raised on the OV scale as having higher OV; additionally, OV becomes normalized as a result of being present at a delivery, the progression of training (depending on the course) and obstetric experience itself, including pregnancies and births.   J o u r n a l P r e -p r o o f