Workplace Violence in the Health Sector in Turkey: A National Study

Abstract

Violence in the workplace is an increasing occupational health concern worldwide. Health care workers are at high risk of assault. To develop, monitor, and manage prevention policies, baseline data should be available. This cross-sectional study was designed to determine the current extent of workplace violence nationwide in Turkey. The study population of 12.944 health care workers was a stratified sample of all health care workers (612,639) in the country. A probabilistic sampling was made on the basis of the “multistage stratified random cluster sampling method.” This study was conducted by a structured questionnaire in a face-to-face interview. The questionnaire items were adapted and translated into Turkish based on questionnaires of International Labor Organization, International Council of Nurses, World Health Organization, and Public Services International. The percentage of health care workers who experienced workplace violence in Turkey in the previous 12 months was 44.7%. The types of violence included physical 6.8%, verbal 43.2%, mobbing (bullying) 2.4%, and sexual harassment 1%. Multivariate analysis showed that level of health care system, type of institution, gender, occupation, age, working hours, and shift work were independent risk factors for experiencing workplace violence (p < .05). Our study indicates that the workplace violence among health care workers is a significant problem. The results of the study can serve as the basis for future analytical studies and for development of appropriate prevention efforts.

Keywords

workplace violence health care workers Turkey

Introduction

Violence in the workplace is an increasing global concern (Beech & Leather, 2006; Cooper & Swanson, 2002). Health care workers are at high risk in terms of workplace violence. More than half of health care workers are exposed to some type of violence, such as verbal abuse, during their careers (International Labor Organization [ILO], International Council of Nurses [ICN], World Health Organization [WHO], & Public Services International [PSI], 2005). Health care providers need to maintain a close and supportive relationship with patients and their families, particularly under unfavorable and stressful conditions. Patients may be aggressive due to their medical conditions or medications that they are on (Beech & Leather, 2006; WorkSafeBC, 2005). Moreover, tight security measures are not welcome in health care or social service settings (Calnan, Kelloway, & Dupre, 2012). A visible security presence to protect health care workers creates a barrier between health care workers, patients, and patients’ families. Absence of security increases the risk for health care workers.

Violence in health care institutions poses an ongoing risk for both employees and the mission of the health care workplace (Baron & Neuman, 1998; LeBlanc & Kelloway, 2002). Any workplace violence has negative effects on victims’ physiological and/or psychological well-being (Chappell & Di Martino, 2000; Flannery, Hanson, & Penk, 1995; Leather, Beale, Lawrence, & Dickson, 1997; Leather, Lawrence, Beale, Cox, & Dickson, 1998; Wykes & Whittington, 1994). Moreover, workplace violence may affect victims’ careers in a negative way (Flannery et al., 1995). Violence in the workplace causes considerable financial and productivity losses (Chappell & Di Martino, 2000). Therefore, workplace violence is an important primary public health as well as a health services management problem (WHO, 1995).

Estimates of the frequency of workplace violence incidents vary due to the lack of both consistent and compatible definitions and standard measurement methods (Bowie, 2000; Brady & Dickson, 1999). However, the number of incidents officially recorded is known to be considerably lower than the real number (Beech & Leather, 2006).

In 2002, the ILO, ICN, WHO, and PSI conducted several studies in numerous countries involving health care providers. Among the studies, the most frequent types of violence in the reporting countries were physical violence in South Africa (17%), verbal abuse in Australia (67%), mobbing (bullying) in Bulgaria (30.9%), and sexual harassment in Brazil (5.7%; Di Martino, 2002).

There is no comprehensive study searching workplace violence in the health sector in Turkey. This study was an effort to determine the real extent of workplace violence occurring in different “health regions” nationwide in Turkey; the study gathered information about the type, level, and structure of the violent incidents to collect data to serve as the baseline in formulating policies and regulations for violence prevention.

Materials and Method

The study was cross-sectional in design and involved all health care workers in Turkey as the health care worker universe. The sampling frame therefore consisted of 612,639 health care workers who were working in primary, secondary, and tertiary health care institutions of the Republic of Turkey.

Health care workers were classified into nine occupational groups using International Standard Classification of Occupations (ISCO-08) with few modifications (International Standard Classification of Occupations: International Labor Organization [ILO], 2012). These groups were called (a) physician, dentist; (b) nurse, midwife; (c) therapist, pharmacist, psychologist, dietitian, audiologist, speech therapist, social worker; (d) health officer; (e) ambulance and emergency medical services worker; (f) laboratory, radiology, sterilization technicians; (g) security staff; (h) administrative staff; and (i) support services, janitor, kitchen staff, maintenance staff, receptionist.

Sampling Design

The sample size was calculated to provide reliable estimates for the whole country and each of the 22 health administrative regions that was determined by the Ministry of Health of the Republic of Turkey. A probabilistic sampling was made on the basis of the “multistage stratified cluster sampling method.” The sample selection was undertaken in two stages. The first stage of selection included random selection of health care institutions as a “primary sampling unit.” A complete list of health care institution, based on Ministry of Health of Turkey data, created a total of 508 primary care and 144 secondary and tertiary care institutions that were selected randomly. In the second stage, the health care workers, who participated in the study, were selected from these institutions by simple random sampling method. MS Excel was used to generate random numbers in the sampling. The final sample consisted of 14,451 staff.

Inclusion and Exclusion Criteria

Health care staff who had been working at least 12 months at the institution were eligible for inclusion in the study. Health care workers who worked in private physician offices and private pharmacies were excluded to avoid duplications, and military health care institutions were excluded as well. The participants who were not present at their workplace for interview on any of the three visits, for any reason, were considered to be “missing” and excluded from the study.

Questionnaire

The survey was prepared based on “Workplace Violence in the Health Sector Country Case Studies Research Instruments-Survey Questionnaire, Geneva 2003,” which was developed by ILO/ICN/WHO/PSI; ILO, ICN, WHO, & PSI, 2003).

The questionnaire included 130 items to identify socio-demographic characteristics of participants, features of workplaces, details of violent types (psychological violence [verbal abuse, mobbing/bullying, and sexual harassment] and physical violence) and experience. The participants were asked about the last violent events in their previous 12 months, and they were asked to elaborate on the details of the violence. The details of violence were characterized into four sections.

Bullying/mobbing is described as “repeated and over time offensive behaviour through vindictive, cruel, or malicious attempts to humiliate or undermine an individual or groups of employees” by ILO/ICN/WHO/PSI. WHO generally uses of the mobbing term; thus, we have preferred to use the “Mobbing” instead of “Bullying.” Because mobbing is a complex concept, detailed questions were added that ask about frequency (at least once a week) and duration (at least 6 months) of hostile behavior to identify whether the incident is real mobbing or something else, according to principles set by Leymann (Leymann, 1996).

All participants were informed that participation to the survey was voluntary. To avoid coercion, the survey was conducted in locations in which privacy was established. The survey was conducted by face-to-face interviews.

The research was conducted in compliance with the Decision No. 11, which was issued by Kirikkale University Clinical Research Ethics Committee, dated June 29, 2012 and in Session 12-08.

Statistical Analysis

The statistical analyses were performed by SPSS for Windows version 21.0 by using complex samples modules. Weights were calculated by using p1 (probability of selecting institute in the stratum), p2 (probability of selecting health care staff in institute), and non-response rate of stratum (NRR). Frequencies were calculated according to multistage samplings. To calculate sample errors, “Taylor Linearization method” was used. Frequencies, percentages, and 95% confidence intervals were calculated. Chi-square test was applied to evaluate the relationship between demographic variables and exposure of any kind of workplace violence and Cramer’s V was calculated as an effect size measure (Cohen, 1988). Multiple logistic regression analysis was used to determine the risk factors of exposing any kind of violence in workplace in the previous 12 months. The variables whose significance level was less than .20 (occupation, level of health care system, type of institution, gender, occupation, age, length of professional experience, education, shift working, working between 18:00 and 07:00 hr, working alone) were included as possible risk factors in multiple logistic regression model (Daniel & Cross, 2013; Field, 2009).

The study was conducted between September 2012 and March 2013. A total of 12,944 survey interviews were conducted nationwide, and the response rate was 89.6%; 1,507 individuals did not respond for various reasons, 54% in primary health care (815 individuals) and 46% in secondary and tertiary health care (692 individuals). The rates of participants from the primary, secondary, and tertiary health care systems were 28.9%, 44.1%, and 27%, respectively.

Results

The mean age of the participants was 36.3 ± 9.2 years (min = 18; max = 80); and 40.4% of the participants were male and 59.6% were female. Medical doctors and nursing/midwifery professionals constituted the 59.6% of the participants and the other health care workers constituted the rest (40.4%). Characteristics of the participants are summarized in Table 1.

Table 1.

Demographic Data of the Participants and Their Distribution by Each Category.

Variables	Number	Percent
Level of health care system
Primary	3,734	28.9
Secondary	5,710	44.1
Tertiary	3,500	27
Gender
Male	5,221	40.4
Female	7,713	59.6
Age
29 and younger	3,386	26.3
30-39 years	5,045	39.2
40-49 years	3,261	25.3
50 and older	1,181	9.2
Education
Primary school	695	5.4
High school	2,617	20.3
University	6,958	53.9
Master’s/doctorate/specialist in medicine or higher	2,613	20.3
Occupation
Physician, dentist	3,377	26.1
Nurse, midwife	4,343	33.5
Therapist, pharmacist, psychologist, dietitian, audiologist, speech therapist, social worker	190	1.5
Health officers	765	5.9
Ambulance, emergency medical services workers	309	2.4
Laboratory/radiology/sterilization technicians	1,306	10.1
Security staff	58	0.4
Administrative staff	1,371	10.6
Support services, janitor, kitchen staff, maintenance staff, receptionist	1,224	9.5
Total	12,944	100

Nearly half of the health care workers (44.7%) had experienced violent behavior in the previous 12 months. Violence types were physical 6.8% (95% CI [6.1,7.5]), verbal 43.2% (95% CI [41.8,44.7]), mobbing 2.4% (95% CI [1.9,2.9]), sexual harassment 1% (95% CI [0.7,1.3]), and any type of violence 44.7% (95% CI [43.2,46.2). See Table 2.

Table 2.

Univariate Chi-Square Test Results of Workplace Violence in the Previous 12 Months.

	Physical Violence		Verbal Abuse		Mobbing/Bullying		Sexual Harassment		Any Type of Violence
Variables	No	Percent	No	Percent	No	Percent	No	Percent	No	Percent
Violence frequencies	875	6.8	5,595	43.2	249	2.4	129	1	5,786	44.7
Level of health care system
Primary	63	5.1	424	34.6	23	2.5	11	0.9	444	36.3
Secondary/tertiary	813	6.9	5,172	44.1	226	2.3	119	1	5,342	45.6
χ²	5.773		40.907		0.049		0.284		38.558
p	<.001		<.001		.768		.484		<.001
Effect size (Cramer’s V)	0.021		0.056		0.002		0.005		0.055
Gender
Male	419	7.9	1,190	37.7	84	2	28	0.5	2,085	39.5
Female	455	5.9	3,597	47	166	2.6	102	1.3	3,691	48.2
χ²	19.607		110.785		4		21.008		96.903
p	.008		<.001		.191		.008		<.001
Effect size (Cramer’s V)	0.039		0.093		0.019		0.040		0.087
Occupation
Physician, dentist	315	10.5	1,701	56.9	97	3.6	19	0.6	1,738	58.2
Nurse, midwife	278	6.2	2,197	49.3	91	2.4	71	1.6	2,273	51
Therapist, pharmacist, psychologist, dietitian, audiologist, speech therapist, social worker	4	1.8	70	31.3	4	1.9	1	0.5	70	31.6
Health officers	47	7	278	40.8	11	2.1	2	0.3	289	42.5
Ambulance, emergency medical services workers	35	11.2	137	43.5	3	1.4	16	5.1	141	44.8
Laboratory/radiology/sterilization technicians	69	4.7	494	33.5	17	1.4	16	1.1	503	34.1
Security staff	15	21.5	24	33.7	0	0	0	0	29	41.8
Administrative staff	43	3	445	30.8	21	1.9	4	0.2	465	32.2
Support services, janitor, kitchen staff, maintenance staff, receptionist	68	5.3	250	19.3	5	0.6	1	0	277	21.4
χ²	159.991		761.957		36.619		97.383		750.955
p	<.001		<.001		.023		<.001		<.001
Effect size (Cramer’s V)	0.111		0.243		0.059		0.087		0.241
Age
≤29 years	327	9.5	1,788	52.1	71	2.5	69	2	1,848	53.8
30-39 years	342	6.6	2,371	45.4	105	2.4	48	0.9	2,444	46.8
40-49 years	156	5	1,119	35.8	56	2.2	11	0.3	1,163	37.2
≥50	47	4.3	287	26.2	17	2	1	0.1	300	27.4
χ²	68.003		319.332		0.914		58.094		328.475
p	<.001		<.001		.952		<.001		<.001
Effect size (Cramer’s V)	0.073		0.157		0.009		0.067		0.160
Length of experience
1-4 years	266	8.9	1,411	47.3	53	2.2	39	1.3	1,462	49
5-9 years	221	8.1	1,277	46.8	79	3.5	43	1.6	1,321	48.4
10-14 years	133	6.3	990	46.9	29	1.6	21	1	1,017	48.1
15-19 years	121	6	879	43.2	43	2.5	19	0.9	913	44.9
≥20 years	121	4.2	963	33	44	1.9	7	0.3	992	34
χ²	34.490		169.771		19.308		28.379		183.010
p	<.001		<.001		.082		.029		<.001
Effect size (Cramer’s V)	0.052		0.115		0.043		0.047		0.120
Education
Primary school	32	4.4	108	14.9	3	0.6	0	0	119	16.4
High school	180	6.5	1,061	38.5	31	1.5	26	0.9	1,111	40.4
University	400	5.8	3,027	43.8	119	2.1	76	1.1	3,121	45.2
Master’s/doctorate/specialist in medicine or higher	261	10.4	1,376	54.8	96	4.2	27	1.1	1,410	56.2
χ²	69.358		401.090		46.493		8.396		392.076
p	<.001		<.001		<.001		.421		<.001
Effect size (Cramer’s V)	0.073		0.176		0.066		0.026		0.174
Shift working
Yes	426	8.6	2,465	49.6	78	2	79	1.6	2,565	51.6
No	446	5.6	3,105	39.2	171	2.6	50	0.6	3,195	40.4
χ²	41.76		133.372		4.535		28.166		155.773
p	<.001		<.001		.188		<.001		<.001
Effect size (Cramer’s V)	0.057		0.102		0.021		0.047		0.110
Working between 18:00 and 07:00 hr
Yes	589	9.4	3.231	51.5	118	2.3	92	1.5	3,333	53.1
No	281	4.3	2.270	35.1	131	2.5	37	0.6	2,356	36.4
χ²	127.674		352.180		0.570		25.062		362.335
p	<.001		<.001		.634		<.001		<.001
Effect size (Cramer’s V)	0.100		0.166		0.007		0.044		0.169
Working alone (>50% of work time)
Yes	41	5.1	282	35.3	15	2.4	11	1.4	294	36.9
No	816	6.8	5,288	43.8	234	2.4	119	1.0	504	63.1
χ²	3.374		21.695		0		1.096	20.715
p	.308		.009		.998		.579	.010
Effect size (Cramer’s V)	0.016		0.041		0.001		0.009	0.040
Type of institute
Private sector	49	2.5	533	27	15	1	19	1	546	27.6
Public sector	825	7.5	5,062	46.2	234	2.6	110	1	5,240	47.8
χ²	67.033		250.529		14.492		0.021		274.326
p	<.001		<.001		.021		.941		<.001
Effect size (Cramer’s V)	0.072		0.139		0.037		0.001		0.146
Numbers of accompanies to patients
Themselves	10	8.2	26	22	0	0	3	2.2	33	27.8
1 person	100	5	875	44	33	1.9	9	0.5	897	45.1
2-3 persons	479	7.9	3,211	52.9	115	2.3	66	1.1	3,297	54.3
4-5 persons	155	14.6	693	65.4	37	4.5	33	3.1	713	67.3
≥6 persons	46	16.9	143	53.1	10	5.6	9	3.2	147	54.5
χ²	108.417		167.906		26.476		49.249		166.574
p	<.001		<.001		.058		.003		<.001
Effect size (Cramer’s V)	0.107		0.133		0.058		0.072		0.132

We observed that 39.5% of male and 48.2% of female health care workers reported any type of workplace violence in the previous 12 months, so female health care workers reported more violence than males (p = .000). Violence in the public sector was much higher (47.8%) than in the private sector (27.6%, p = .000). The details of univariate analysis to identify the factors for risk of any type of violence in the previous 12 months is summarized in Table 2.

Among the factors, level of health care system, type of institution, gender, occupation, age, working between 18:00 and 07:00 hr, and shift working were independent risk factors for experiencing workplace violence (Table 3). Physicians and dentists were exposed to violence 3.2 times (95% CI [2.16,4.67]) more frequently than support service workers. Similarly, nurses/midwives and health officers were exposed to violence 1.7 times (95% CI [1.19,2.45]) more frequently than support service workers.

Table 3.

Results of Multiple Logistic Regression to Determine Risk Factors of Any Type of Workplace Violence in the Previous 12 Months.

		95% CI
Variables	OR	Lower	Upper	p
(Intercept)	1.00			.000
Gender (female)	1.24	1.06	1.47	.010
Level of health care system (2-3 level)	1.39	1.23	1.58	.000
Shift working	1.26	1.06	1.50	.009
Working between 18:00 and 07:00 hr	1.43	1.21	1.68	.000
Age				.000
≤29 year	2.36	1.57	3.54
30-39 year	1.56	1.11	2.20
40-49 year	1.36	1.02	1.81
50+ year	1.00
Length of experience				.058
1-4 years	1.04	0.74	1.45
5-9 years	1.25	0.93	1.67
10-14 years	1.38	1.04	1.82
15-19 years	1.28	1.00	1.64
20+ years	1.00
Education level				.154
Primary	1.00
High school	1.64	1.07	2.52
University	1.54	0.99	2.40
Master/doctorate/specialist in medicine	1.50	0.92	2.46
Occupation				0
Physician, dentist	3.18	2.16	4.67
Nurse, midwife	1.71	1.19	2.45
Therapist, pharmacist, psychologist, dietitian, audiologist, speech therapist, social worker	1.07	0.58	1.96
Health officer	1.69	1.11	2.56
Ambulance/emergency medical services worker	1.65	0.96	2.82
Laboratory/radiology/sterilization technicians	1.11	0.76	1.62
Security staff	1.54	0.62	3.79
Administrative staff	1.45	1	2.20
Support services, orderlies/janitor, kitchen staff, maintenance staff, receptionist	1
Type of institution	2.23	1.78	2.80	0
Working alone	1.06	0.80	1.40	.696

Note. CI = confidence interval.

The risk of violence was 2.4 times higher among younger people (≤29 years). Education levels, length of experience, and working alone were not significant in multivariate analysis.

Physical violence was mostly perpetrated by the relatives of patients (45.7%), followed by patients themselves acting together with their relatives (30%), and by patients acting alone (23.6%). Other perpetrators who performed physical attacks included colleagues (4.9%), administrators (2.4%), and the general public (2%).

Verbal violence was mostly done by relatives of patients (71.1%) and patients themselves (34.3%). Other groups who abused verbally were colleagues (8.6%), administrators (4.9%), and the general public (1.9%). Victims reported that 65.3% of the victims were mobbed by their superiors, 31.9% by colleagues, and 14.4% by subordinates. 38.2% of sexual harassment was performed by patients.

The study concluded that 38.2% of the sexual harassment was performed by patients, 36% by patients’ relatives, 30.8% by colleagues, 1.6% by administrators, and 5.9% by the general public.

Victims of physical violence reported that 77.5% of the aggressors were male, 13.3% were female, and 9.2% were of both male and female genders involved in attacks. Frequency of verbal abuse performed by male (72.6%), female (18%), and both genders (9.4%) were reported. Among the aggressors, 49% were male, 41.3% were female, and 9.7% were both genders; sexual harassment was performed by male (78.5%) and female (21.5%) aggressors.

We tried to analyze how many relatives accompanied the patients on average (Table 2). The more relatives that come with patients, the more frequently violence incidents occur (p < .001). However, responses to this question were often incomplete: 9,508 participants (73.5%) responded and 3436 participants (26.5%) did not. Among the respondents, 3299 (25.5%) participants declared that they did not have any direct contact with the patients and among them nobody responded to this question as expected.

Participants were also asked if they have ever been exposed to any kind of violence in their workplace throughout their entire career except the previous 12 months, without requesting detailed information. Among those who responded, 52.3% had experienced at least one type of workplace violence over their entire careers. Violence types were physical 10.4%, verbal 48.2%, mobbing 15%, and sexual harassment 1.6%. There was also significant difference among the occupations according to frequency of violence (p < .001). It was observed that 70.2% of the physicians and dentists, 57.4% of the nursing and midwifery professionals reported that they had experienced at least one type of workplace violence during their career. Females (54.3%) experienced more violence than males (49.4%; p = .002).

Discussion

This national study is the largest and the most comprehensive study on workplace violence in Turkey and provides reliable estimates for the country as a whole and its regions. The dimensions of violent incidents experienced by health care workers as well as violence types, incident frequencies, and their distribution by professions and socio-demographic characteristics were determined. In our study, the incident of workplace violence was 44.7% in the previous 12 months. ILO/ICN/WHO/PSI conducted similar studies in 7 countries (Brazil, Bulgaria, Lebanon, Portugal, Thailand, South Africa, and Australia) involving all health care providers, using similar instruments. A majority of health care workers were found to have experienced at least one incident of physical or psychological violence in the previous year: 75.8% in Bulgaria, 67.2% in Australia, 61% in South Africa, 60% in a health center and 37% in a hospital in Portugal, 54% in Thailand, and 46.7% in Brazil (Di Martino, 2002). Frequency of violence looks quite high in all studied countries and our results from Turkey look particularly similar to Brazil.

The most prevalent type of violence is verbal (43.2%) in our study. Several other studies also reported that health care workers most frequently were exposed to verbal abuse (Di Martino, 2002; Fujita et al., 2012; Hills, Joyce, & Humphreys, 2012; Winstanley & Whittington, 2004). Verbal abuse was reported to be 70.6% in Australia among physicians (Hills et al., 2012), 68% in England (Winstanley & Whittington, 2004), and 29.8% in Japan (Fujita et al., 2012) among all health care workers. In ILO/ICN/WHO/PSI studies, in all countries, the most frequent violence type was verbal. Frequency of verbal abuse was between 67% and 37.2% in these countries within the previous year (Di Martino, 2002).

Our study revealed 6.8% physical violence is in our study cohort. The physical violence was 27% in England (Winstanley & Whittington, 2004), 32.3% in Australia (Hills et al., 2012), and 15.9% in Japan (Fujita et al., 2012). ILO/ICN/WHO/PSI studies reported the incidents of physical violence between 2.6% (Portugal) and 17% (South Africa) in 7 countries. Physical violence in Turkey is similar to Brazil (6.4%), Lebanon (5.8%), and Bulgaria (7.5%; Di Martino, 2002).

This study yielded 2.4% of participants that were complained of mobbing in the previous 12 months. ILO/ICN/WHO/PSI studies reported the frequency of mobbing between 10.5% and 30.9% in studied countries. Our result looks to be significantly lower than the previous studies, but we are not confident that this result represents a real difference. The ILO/ICN/WHO/PSI studies gave a short definition of mobbing at the beginning of the interview. We think that the practice of providing a brief definition of mobbing to guide questionnaire completion may lead to misunderstandings when used alone, without confirmation that the subject understood the meaning. To avoid such misunderstandings, we included detailed definition of mobbing and some additional questions such as frequency and duration of maltreatment in our questionnaire. We believe that our survey captured the extent of mobbing more accurately.

Sexual harassment was the least commonly seen type of violence (1%) in our study and compares with 9.9% in Japan. Comparative ILO/ICN/WHO/PSI studies revealed sexual harassment between 1.9% and 5.7% over the previous 12 months. The differences among the studies might be due to differences in perception of sexual harassment among communities. There may be more obstacles to reporting sexual harassment in Eastern cultures, particularly by women. This may lead to underestimating sexual harassment. However, to capture as accurate information as possible, a detailed definition and types of sexual harassment were provided in the beginning of the relevant section in our study survey.

Direct comparison of studies may lead misinterpretations due to different study methods, differences in the populations studied, sampling design, and definitions of violence. In the literature, some studies focus solely on nurses (Estryn-Behar et al., 2008; Gerberich et al., 2004) or physicians (Hills et al., 2012; Perrone, 1999). Furthermore, some studies refer only to patients’ and visitors’ aggressive behavior and fail to consider worker-on-worker violence (Hahn et al., 2010). ILO/ICN/WHO/PSI studies are more comparable with our study because we used the same definitions of violence, similar questionnaire and study design. However, these studies were not comprehensive in representing all health care workers, as ours did (Di Martino, 2002).

Further analysis of our data showed that the level of health care system, type of institution, gender, occupation, age, working hours, and shift working were independent risk factors for experiencing workplace violence in the multiple logistic regression model. Although there were statistical significances, we determined the weak relationship between independent variables and exposure to workplace violence because of the small effect size.

Physicians and dentists were the highest risk group for workplace violence. Most of the previous studies reported that nurses are the highest risk group for violence (Estryn-Behar et al., 2008; Gerberich et al., 2004; Winstanley & Whittington, 2004). In our study, nurses ranked the second highest risk group. One reason might be that the people perceive physicians and dentists as the front line, or primary representative, of all health services and the entire health care system in Turkey. People often ask them to handle difficulties in managing the health care system that are outside the scope of their duties.

Being female, short length of experience (1-4 years), younger age (≤29 years), and working between 18:00 and 07:00 hr were risk factors for exposure to violence in workplace. Being female was reported to be a risk factor for violence in health care settings (Arnetz, 1998). Inexperienced personnel are reported to be exposed to more workplace violence as it is the case in our study. This might be as a result of the staffs’ poor communication skills that increase patients’ dissatisfaction and level of disappointment (ILO, 1998). Coping with difficult patients is a very hard duty for health care workers, especially for inexperienced workers, so the management should provide special training against difficult patients and relatives. Those working late at night and early in the morning are at risk of violence (ILO, 1998; LeBlanc & Kelloway, 2002). Higher education level and working alone are important risk factors in univariate analysis in our study, but they lost their significance in multivariate analysis. However, lower education level has been reported in previous studies to be an important risk factor for violence, in contrary to our study (Whittington & Wykes, 1996). Working alone has been shown to increase the risk of workplace violence in previous studies (LeBlanc & Kelloway, 2002).

In our study, 52.3% of the participants experienced at least one type of workplace violence over their entire careers. All kinds of violence were relatively higher in the entire career. In all, 80.5% of victims exposed to violence in the previous 12 months have been exposed to violence, at least once, in their entire career. We think that recent exposure to violence may be biased by easer recall.

We also noticed that frequencies of mobbing events over an entire career were greater than the period prevalence of events as the previous 12 months might seem to suggest. This could be explained by several reasons. In the definition of mobbing by Leymann that has been used, hostile behavior should last more than 6 months, it is hard to identify the violence in previous 12 months and reasonable to detect it in entire career time. Another reason is recall bias. Although people tend to forget other violent events in the past, they recall the mobbing more (Flannery et al., 1995; Leymann, 1996). Moreover, people who were victims and also those identified as the aggressors may be more likely to have been relocated, promoted, or left their job.

In our cohort, perpetrators of violent events were mostly male and relatives of patients, as previous studies. However, patients themselves were the most common group performing the act of violence in these studies (Gerberich et al., 2004; WorkSafeBC, 2005). However, the number of accompanying relatives with the patient is directly proportional with violence incidents. In the Turkish culture, patients tend to accompany their relatives to health care institutions as a demonstration of support, respect, love, and concern. These companions usually interact on their own initiative with health care workers and often (it is probably fair to say usually) try to get information about the patient from health care workers, who do not have unlimited time to provide details. This can cause communication problems and conflicts. It is therefore not surprising that the more relatives accompany the patient, the more frequent violent incidents would occur.

The present study was conducted on the sample of professional groups that represent the whole country and nearly all types of health care workers. However, it is limited in certain sectors of health care because it did not include military hospital workers, private physicians’ offices, and pharmacies. Our study is a cross-sectional study, with the usual limitations of inaccurate recall of past events and of possible contamination by current events. It was also relied on self-reported measurements and was therefore limited to perceptions of abuse from a worker’s perspective.

Conclusion

Our study demonstrates that the workplace violence targeting health care workers is a significant public health problem in Turkey. The results of the study may serve as the basis for future analytical studies and the basis for development of appropriate prevention and control efforts. Although it may be impossible to prevent workplace violence completely, comprehensive and analytical preventive programs to tackle the issue via a multi-dimensional approach, such as social, cultural, environmental, political, institutional, organizational, and individual aspects, may decrease the number of violent incidents as well as their severity.

Footnotes

Authors’ Note

It was presented as a moderate poster at 6th European Public Health Conference in November 13 to 16, 2013, in Brussels. All authors have participated sufficiently in the study and they take responsibility for the contents. All authors have contributed to the conception, design, analysis, and interpretation of data; the drafting or revision of the manuscript; and they approved the final version of the manuscript. All steps of the research were carried out by the Kirikkale University Faculty of Medicine. The funding organization had no role in the design, and in the collection, analysis, and interpretation of data of the study. This is a country-based cross-sectional study performed on 12,944 health care workers by face-to-face interview. Although it is a country-based study, we believe that outcomes of the study will serve universally and make significant contributions to existing knowledge in the field.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the General Directorate of Health Research, Turkish Ministry of Health.

Author Biographies

Tevfik Pinar, MD, PhD, Associate Professor of Public Health and Occupational Medicine in Hacettepe University, Institute of Public Health, Turkey.

Cengizhan Acikel, MD, Professor of Public Health in Gulhane Military Medical Academy, Department of Biostatistics.

Gul Pinar, PhD, Associate Professor of Nursing in Yildirim Beyazit University, Health Science Faculty, Department of Gynecology and Obstetrics.

Erdem Karabulut, PhD, Associate Professor of Biostatistics in Hacettepe University, Faculty of Medicine, Department of Biostatistics.

Meral Saygun, MD, PhD, Professor of Public Health and Occupational Medicine in Kirikkale University, Department of Public Health.

Elif Bariskin, PhD, Professor of Psychology in Hacettepe University, Faculty of Medicine, Department of Psychiatry.

Tee L. Guidotti, MD, Professor, Occupational Medicine, Editor of Archives of Environmental and Occupational Health, Medical Advisory Services, Rockville, MD, USA.

Recep Akdur, MD, Professor of Public Health in Ankara University, Faculty of Medicine, Department of Public Health.

Hilmi Sabuncu, PhD (Deceased), Professor of Chemistry in Yeniyuzyil University, Faculty of Medicine, Department of Public Health.

Said Bodur, MD, Professor of Public Health, Balikesir University, Faculty of Medicine Department of Public Health.

Mucahit Egri, MD, Professor of Public Health, Gazi Osmanpasa University, Faculty of Medicine Department of Public Health.

Bilal Bakir, MD, Professor of Public Health, Gulhane Military Medical Academy, Department of Public Health.

Emrah M. Acikgoz, MD, Resident of Public Health, Kirikkale University, Faculty of Medicine Department of Public Health.

Ismail Atceken, MD, Resident of Public Health, Kirikkale University, Faculty of Medicine Department of Public Health.

Mustafa Cengiz, MD, Professor of Radiation Oncology in Hacettepe University, Faculty of Medicine Department of Radiation Oncology.

References

Arnetz

J. E.

(1998). The Violent Incident Form (VIF): A practical instrument for the registration of violent incidents in the health care workplace. Work & Stress, 12, 17-28.

Baron

R. A.

Neuman

J. H.

(1998). Workplace aggression—The iceberg beneath the tip of workplace violence: Evidence on its forms, frequency, and targets. Public Administration Quarterly, 21, 446-464.

Beech

Leather

(2006). Workplace violence in the health care sector: A review of staff training and integration of training evaluation models. Aggression and Violent Behavior, 11, 27-43.

Bowie

(2000). Current trends and emerging issues in workplace violence. Security Journal, 13(3), 7-23.

Brady

Dickson

(1999). Violence in health care settings. In Leather

Brady

Lawrence

Beale

Cox

(Eds.), Work-related violence: Assessment and intervention (pp. 9-10). London, England: Routledge.

Calnan

Kelloway

E. K.

Dupre

K. E.

(2012). SAV-T first: Managing workplace violence. In Hughes

Kinder

Cooper

L. C.

(Eds.), International handbook of workplace trauma support (pp. 105-120). Malden, MA: Wiley-Blackwell.

Chappell

Di Martino

(2000). Violence at work (2nd ed.). Geneva, Switzerland: International Labor Organization.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Cooper

C. L.

Swanson

(2002). Workplace violence in the health sector—State of the art. Geneva, Switzerland: International Labor Organization.

10.

Daniel

W. W.

Cross

C. L.

(2013). Biostatistics: A foundation for analysis in the health sciences (10th ed.). New Jersey : John Wiley.

11.

Di Martino

. (2002). Workplace Violence in the health sector: Country case studies (Brazil, Bulgaria, Lebanon, Portugal, South Africa, Thailand)—Synthesis report. WHO, Geneva, Switzerland.

12.

Estryn-Behar

van der Heijden

Camerino

Fry

Le Nezet

Conway

P. M.

, . . . NEXT Study Group. (2008). Violence risks in nursing—Results from the European “NEXT” Study. Occupational Medicine, 58, 107-114.

13.

Field

(2009). Discovering statistics using SPSS (3rd ed.). Los Angeles, CA: Sage.

14.

Flannery

R. B.

Hanson

M. A.

Penk

W. E.

(1995). Patients’ threats: Expanded definition of assault. General Hospital Psychiatry, 17, 451-453.

15.

Fujita

Ito

Seto

Kitazawa

Matsumoto

Hasegawa

(2012). Risk factors of workplace violence at hospitals in Japan. Journal of Hospital Medicine, 7, 79-84.

16.

Gerberich

S. G.

Church

T. R.

McGovern

P. M.

Hansen

H. E.

Nachreiner

N. M.

Geisser

M. S.

. . . Watt

G. D.

(2004). An epidemiological study of the magnitude and consequences of work related violence: The Minnesota nurses’ study. Occupational & Environmental Medicine, 61, 495-503.

17.

Hahn

Muller

Needham

Dassen

Kok

Halfens

R. J. G.

(2010). Factors associated with patient and visitor violence experienced by nurses in general hospitals in Switzerland: A cross-sectional survey. Journal of Clinical Nursing, 19, 3535-3546.

18.

Hills

D. J.

Joyce

C. M.

Humphreys

J. S.

(2012). A national study of workplace aggression in Australian clinical medical practice. Medical Journal of Australia, 197, 336-340.

19.

International Labor Organization. (1998). Encyclopedia of occupational health and safety ( Stellman

J. M.

Ed.) (4th ed., Vol. II, 51.4). Geneva, Switzerland: Author.

20.

International Labor Organization. (2012). International standard classification of occupations: ISCO-08. Geneva, Switzerland: Author.

21.

International Labor Organization, International Council of Nurses, World Health Organization, & Public Services International. (2003). Workplace violence in the health sector country case studies research instruments. Geneva, Switzerland: World Health Organization.

22.

International Labor Organization, International Council of Nurses, World Health Organization, & Public Services International. (2005). Framework guidelines for addressing workplace violence in the health sector: The training manual. Geneva, Switzerland: World Health Organization.

23.

Leather

Beale

Lawrence

Dickson

(1997). Effects of exposure to occupational violence and the mediating impact of fear. Work & Stress, 11, 329-340.

24.

Leather

Lawrence

Beale

Cox

Dickson

(1998). Exposure to occupational violence and the buffering effects of intra-organizational support. Work & Stress, 12, 161-178.

25.

LeBlanc

M. M.

Kelloway

E. K.

(2002). Predictors and outcomes of workplace violence and aggression. Journal of Applied Psychology, 87, 444-453.

26.

Leymann

(1996). The content and development of mobbing at work. European Journal of Work & Organizational Psychology, 5, 165-184.

27.

Perrone

(1999). Violence in the workplace. Australian Institute of Criminology Research and Public Policy Series No. 22. Australian Institute of Criminology, Canberra. Retrieved from http://tinyurl.com/qzphma8.

28.

Whittington

Wykes

(1996). Aversive stimulation by staff and violence by psychiatric patients. British Journal of Clinical Psychology, 35, 11-20.

29.

Winstanley

Whittington

(2004). Aggression towards health care staff in a UK general hospital: Variation among professions and departments. Journal of Clinical Nursing, 13, 3-10.

30.

WorkSafeBC. (2005). Preventing violence in health care: Five steps to an effective program. British Colombia, Canada, Vancouver: Author.

31.

World Health Organization. (1995). Violence: A public health priority. Geneva, Switzerland: Author.

32.

Wykes

Whittington

(1994). Reactions to assault. In Wykes

(Ed.), Violence and health care professionals (pp. 105-126). London, England: Chapman & Hall.