This is a prospective cross-sectional study with descriptive intent. It took place at the University Teaching Hospital (CHU) of Bouaké in the Pediatric and Obstetrics & Gynecology departments from 26 January 2016 to 24 February 2016. The Pediatrics and Obstetrics & Gynecology departments, are the only tertiary-level referral services in the health pyramid in the Gbêkê region. They are located at about 347 km from Abidjan, the economic capital in the south of the country. The annual admissions in the department of Pediatrics, all units combined (Neonatology and General Pediatrics) is nearly 19,000 patients and has 99 workers all socio-professional category. Similarly, The Obstetrics & Gynecology Department has 138 active staff. These two services represent in terms of activities, nearly two-thirds of the total volume of activities of the University Teaching Hospital of Bouaké and therefore constitute the main gateway to this reference health facility in the Gbêkê administrative region. The study population is represented by the staff of both departments. Included in the study were all personnel working in these services, recognized by the Human Resources Department of the University Teaching Hospital (CHU) of Bouaké, present at the time of the survey and gave a written informed consent. All personnel absent at the time of the investigation due to illness, leave or annual leave were not included in the study.

We conducted a guided interview with all staff meeting the inclusion criteria. The information collected was recorded on two structured survey sheets, anonymized and validated by one independent expert. In this study, the level of education of staff was defined as follows: primary (≤6 years of education), secondary (7 and 13 years of education), higher (≥14 years of education). The data was captured and analyzed on the statistical software SPSS version 20. The analysis was descriptive and consisted of calculating the numbers and determining averages and proportions. Quantitative variables were analyzed as mean with standard deviation. The qualitative variables were expressed as a proportion. The comparison of the quantitative variables was done with the chi-2 test. The threshold of significance was set for a value of p ≤ 0.05.

They were materialized by the informed consent of the participants, the anonymity of the data collection forms and by the research authorization from the Scientific Medical Directorate of the University Teaching Hospital with copying to the heads of the services concerned.

From the total of 237 health professionals in both services, 129 participants were enrolled representing a participation rate of 54.4%. Seventy-one (71) were from the Pediatrics department and fifty eight (58) from the Obstetrics & Gynecology department. There were 72 women and 57 men, a sex ratio of 0.8. The average age was 36 ± 6 years [extreme 26 and 55 years]. The participants’ had a higher level of education in 65.8%, secondary education in 33.3% and primary education in 0.9%. Professional seniority ranged from 3 months to 26 years with an average of 6 years. 38.3% of the interviewed personnel were working in emergency services and 19.8% in the wards. The other socio-demographic characteristics of the participants are presented in Table 1.

One hundred of the 129 staffs interviewed said they had been victims of violence at least once i.e. a frequency of 77.5%. Forty-five were from the Pediatrics Department and 55 from Obstetrics & Gynecology. There were 49 men and 51 women i.e. a sex ratio of 0.96.

The violence was perpetrated in the emergency room in 36.8%, in reception room in 17.2% and the ward in 12.3% of cases. Aggression took place between 8 am and 6 pm in 54% of cases and between 6 pm and 8 am in 46% of cases. The characteristics of the violence are presented in Table 2. The assume perpetrator in 55% of cases was a man and in 45% a woman. The assumed age of the perpetrator was between 30 and 40 years in 47.2% of cases. The perpetrator was a parent of the patient in 71%, a hospitalized patient in 21% and a team member in 8% of the cases. The aggression perpetrated by the hospitalized patient on the staff occurred in 13 times on Day 1 of hospitalization, 6 times in Day 2 and twice in Day 3. The average number of days of hospitalization was 1.5 days (extreme 1 and 3 days). The sex of the victim was the only factor significantly associated with the violence (Table 3).

The four main feelings of the victims of violence were frustration (26.7%), discouragement (21.3%), feelings of insecurity (18.3%) and humiliation (17%). The victims reported that violence affected their working activity in 44% of cases. It was the lack of self-investment (59.6%), the desire to change jobs (24.2%) and the lack of compassion (14.1%).

Nine of the 100 victims (9%) reported having received medical assistance after the attack. All the victims claimed that they had no legal recourse after the attack. To prevent violence on the staff, the victims proposed to raise awareness against violence (27.6%), to campaign for respect for the staff working in the hospital (25%) and the implementation of a policy of zero tolerance to violence (12.5%).

This cross-sectional and descriptive study aims to investigate the causes of violence against staff working in the pediatric and Obstetrics & Gynecology departments of the University Teaching Hospital (CHU) of Bouaké with a view to prevent it and improve the professional practice. It relies on staff declarations based on honor from memory without confirmation of their accuracy by incident reporting forms. As a result, memory bias is possible. In addition, the study is mono-centric and does not take into account cases of violence against staff perpetrated in other health centers. As a result, the results of the study cannot be generalized to the whole country. Nevertheless, despite the methodological

Others*: Refusal of the prescription with respect to the government free admission policy (3), announcement of death (2), sorting of patients (2), discharges against medical opinion (2), welcoming of patients (2), pathological status (2), threat (2), fatigue related to hospital procedures (1), violation of hospital instructions (1), racketeering (1).

*p-value significant.

limitation, this work provided for the first time in Côte d’Ivoire some information on violence against health workers in Pediatric and Obstetrics & Gynecology services and could be further by other subsequent studies. The study raises the following points of discussion.

The overall participation rate for staff in both departments is 54.4%. A lower rate, 45%, was reported in the United States in 2009 by Karen et al. [16] . On the other hand, higher participation rates have been reported in Morocco (84.5%) and Saudi Arabia (100%) [13] [17] . This variability, linked to the inherent difference in methodology in each study, can also be attributed to the reluctance of the staff. Indeed, this is the first survey addressing this problem in obstetrics & gynecology and pediatrics department in Côte d’Ivoire. The staff members who were interviewed were predominantly female (56%). Similarly, Hossein et al. [18] reported 52% of women in 2013 in Iran. Furthermore, Abodunrin et al. [14] in Nigeria in 2014 and Mensah et al. [19] in Ghana in 2016 also found that participants were mostly young and predominantly female. In the study, the perpetrator has an average age of 36 years. Hossein et al. [18] reported in their work a comparable average in age of 34.2 years. About two-thirds of the staff surveyed have a seniority in service less than or equal to 5 years. These socio-demographic characteristics are roughly equal to or similar to those of studies elsewhere in Europe, Asia and Africa [16] [18] [19] [20] [21] [22] .

The study reported a prevalence of 77.5% of violence on health professionals. This rate is in the range with 66.8% to 90.7% reported between 2007 and 2015 in the literature [17] [18] [21] [22] . Staff violence occurs during “on-call” hours (53.7%) especially in the emergency units (36.8%). In the study by Turki et al. [16] , nearly 58.9% of the assaults took place during the night. In our case, this can be explained by several factors among others, the heavy workload of staff; the inadequacy of the service offer and the demand for care of the patients, the low level of understanding of the parents concerning the procedures of children care in emergency situation. In our study, the aggression is often multiple. 68% of victims were assaulted at least 2 times in the quarter preceding the survey with an average of 3 (range 1 and 5). This violence is mostly verbal followed by physical violence. Hossein et al. [18] in Iran in 2013 reported an average number of incidents of 4.96 (extreme 1 and 40). In 2007, Meyssirel et al. [22] study in Montpellier (France) found 77.4% verbal aggression, 12.9% moral violence, 6.4% physical aggression and 3.2% thefts. The study by Turki et al. [17] reported 88.5% verbal abuse and 41.1% physical abuse. Motasem et al. [20] in Palestine in 2015 had reported 8.6% sexual harassment. Cases of rape and sexual harassment were not found in our study. This can be explained by an information retention bias. The causes of violence are multiple. In the study, this concerns problems related to management (32.1%), visiting time (26%) and low level of understanding of the subject (16.5%). For Abodunrin et al. [14] in 2014, the reasons were the waiting time (80.6%), the insults of the staff against the patient (61.8%) and the unavailability of staff (43.1%). In Egypt, Samir et al. [23] in 2009 reported negligence (40.5%) and misconduct (35.8%) as the main causes. In Palestine Motasem et al. [20] in 2015 reported that the main reasons for the violence were the waiting time, lack of measures to prevent violence and the dissatisfaction of parents or patients. The perpetrator is often a parent or a visitor of the hospitalized patient (71.1%), of male gender (55%) with an estimated age between 30 and 40 years (47.2%). Often, it maybe the patients themselves (21%) or a staff member (8%). However, there is no significant difference between the sex, the assumed age and the nature of the violence. These results are comparable to data from the literature [14] [22] [23] . When the perpetrator is a staff member, it is a doctor in 43%, a care giver in 43% and an administrative staff in 14%. The work shows that the victim is mostly male (P = 0.045) but with no relationship to the socio-professional category, the service, the seniority in the profession and the seniority in the service. In the study by Aden et al. [13] carried out in 2012 in Morocco, the proportion of nurses and doctors victims of violence in the line of duty was respectively 86% and 78%. In our work, although there is no significant difference in the socio-professional category of victims, we nevertheless notice that security guards (100%) and assistants nursing (81.5%) are more assaulted than the nurses & midwives (76.7%) and the doctors (72.9%). This may reflect a lack of consideration by the perpetrators vis-à-vis of staff of lower educational backgrounds.

In the study, the abused staffs are mostly frustrated, feel humiliated and insecure, with discouragement, abandon or lack of concentration as drawback. The feelings experienced by the staff have already been reported by other authors in the literature [14] [21] [24] . In the study, only 9% of victims reported having received a medical assistance. None of these victims received legal and psychological support. The absence of complaint with regards to the aggression by the victims to the hospital management or either by ignorance or by negligence explains the lack of the medical, psychological and legal support observed in the study. Indeed, at the University Teaching Hospital of Bouaké, the process of taking care of the abused person begins only when the victim declares the incident to the hospital management by filling out the declaration form. Unfortunately, in the majority of cases, this declaration form that grants the right to medical, legal and psychological assistance is not completed by the victims. Proposals have been made by the staff to address the violence occurring in the hospital. All suggestions are aimed at preventing violence at different levels viz. awareness against violence (27.6%), campaign for respect for the staff working in the hospital (25%) and the implementation of a policy of zero tolerance to violence (12.5%). These suggestions have also been reported by other authors in the literature [17] [23] [25] [26] .