Anxiety levels in employees and students in psychiatric nursing

Introduction: Several research finding indicate that nursing care professionals are often faced with situations which may lead to anxiety. The aim of the present research was to determine the prevalence and typical signs of anxiety among nursing employees and nursing students in psychiatric settings. Methods: The Burns Anxiety Inventory was used as an assessment tool to measure anxiety. The research sample consisted of 242 participants. The data collected were processed by the descriptive statistics, Leveneʹs test, the ANOVA statistical test, the Welchʹs t-test, and the post hoc analysis. Pearsonʹs correlation coefficient was used to measure the strength of the association between the variables. Results: The results of the current study show that nearly half of the participants experience anxiety, but the differences were noted as regards their anxiety thoughts (p = 0.039). Anxiety feelings are more prevalent in female students (p = 0.046). Habitual smokers (p = 0.030) and casual smokers (p = 0.020) are more likely to develop anxious feelings and physical signs of anxiety. The anxiety signs are also more pronounced in the respondents with self-assessed lower economic status (p = 0.001) and poor self-rated health (p = 0.0001). Discussion and conclusion: The professionals and students in psychiatric nursing often encounter situations conducive to the development of anxiety. Further studies on the current topic are therefore recommended to design adequate educational programmes to timely recognise anxiety symptoms and to implement mutual and self-help measure. IZVLEČEK Uvod: Raziskano je, da se zaposleni v zdravstveni negi pogosto srečujejo z dejavniki za razvoj anksioznosti. Namen raziskave je bil ugotoviti pojavnost znakov anksioznosti med zaposlenimi na področju zdravstvene nege v psihiatriji in med študenti zdravstvene nege. Metode: Za raziskavo je bil uporabljen Burnsov kontrolni seznam anksioznosti. Sodelovalo je 242 anketirancev. Podatki so bili obdelani z uporabo deskriptivne statistike, Levenovega testa, ANOVE, Welchovega testa, analizo Post Hoc in Pearsonovo korelacijo. Rezultati: Ugotovljeno je bilo, da se zaposleni in študenti srečujejo z znaki anksioznosti, med seboj se razlikujejo pri anksioznih mislih (p = 0,039). Med študenti ženskega spola se v večji meri pojavljajo anksiozna občutja (p = 0,046). Anketiranci, ki kadijo redno (p = 0,030) ali občasno (p = 0,020), so bolj dovzetni za anksiozna občutja in telesne znake anksioznosti. Anketiranci, ki svoj ekonomski položaj (p = 0,001) in zdravje (p = 0,0001) zaznavajo slabše, bolj občutijo znake anksioznosti. Diskusija in zaključek: Zaposleni in študenti se pri svojem delu ter kliničnem usposabljanju pogosto srečujejo z dejavniki, ki povzročajo pojavnost anksioznosti. Potrebne bi bile nadaljnje raziskave, predvsem pa bi bilo treba izobraziti zaposlene in študente, da prepoznajo znake ter si tako pomagajo pri premagovanju le-teh.


Introduction
Stress at home and work-related stress are known to cause several types of anxiety (Gomboc, 2010). Various studies have demonstrated that nursing is stressful and that the prevalence of occupational stress-related burnout in the profession is high (Bégat, et al., 2005;Hegney, et al., 2014). According to Karanikolaand Papathanassoglou (2013) up to 10 % of nurses manifest clinically significant anxiety and depressive symptoms leading to altered professional attitudes.
A cross-sectional survey conducted by Gao and colleagues (2012) among 1807 registered nurses in seven public city general hospitals in China identified the key factors causing anxiety symptoms in nurses, including not highly valued work, chronic diseases, worse nurse-patient relationships and relatively low social status of nurses. The results of the study suggest that proper counselling, promotion of healthy lifestyle behaviour and improvements to the social environment in the workplace may be helpful toward reducing or preventing the anxiety symptoms. Another important finding was that a large proportion of nurses had anxiety symptoms (43.4 %), which warrants immediate investigation and intervention from the hospital administrators. These results are consistent with those demonstrated by Taghinejad and colleagues (2014), who ascertained that nurses experience mental problems among which anxiety is the most common.
Different types of work schedule and working environment may contribute to the development of anxiety symptoms. Shift work, especially during the night, was proved to be significantly associated with anxiety in medical and nursing employees. The prevalence rates of anxiety symptoms among the nurses were higher than among the referent population (Ardekani, et al., 2008;Melo, et al., 2010). The intensive care unit is a stressful environment especially for nurses and physicians and these stresses may have negative effects on their mental health and performance (Nooryan, et al., 2012). Stathopoulou and colleagues (2011) provided evidence of the correlation between the extent of workplace stress and subsequent anxiety symptoms in emergency nursing personnel. Contrary to these findings, the results of quantitative surveys conducted by Peters and colleagues (2013) suggested that the level of anxiety of nurses working in hospice care or in community services was not high. The recent research investigating the influence of gender on the development of anxiety demonstrated that anxiety levels were significantly higher in females than males (Kayalha, et al., 2013).
In a study conducted in seven city hospitals in China, 75 % of 1807 employees were included in the survey. It was established that the prevalence of anxiety symptoms in nurses was 43 %. Several factors, such as education, chronic disease, regular meals and physical activity, hospital grade, job rank, monthly salary, nurse-patient relationships, job satisfaction and others were all statistically significantly related to the prevalence of anxiety symptoms (Gao, et al., 2012).
These results accord with the observations of Mark & Smith (2012), who also provided evidence that social support, rewards, and skill discretion are negatively associated with mental health problems. Huang and colleagues (2013) conducted a prospective, randomized control study which documented a positive impact of different light exposure during night shift on the nurses' symptoms of clinical insomnia and subsequent anxiety, and depression. In nursing, rotating shifts are common and due to the nature of work the necessary rest periods between shifts can be shorter than stipulated by the health and safety legislation (12 hours). However, no association between the frequency of work shifts separated by less than 11 hours and excessive anxiety or depression was established (Flo, et al., 2014).

Aim and objectives
As a sequel to the literature review and recent research (Selič, 2010;Gao, et al., 2012), the study aimed to determine the prevalence of anxiety and anxiety symptoms in psychiatric nurses and the thirdyear nursing students of the Faculty of Health Care Jesenice during their clinical practice in psychiatric settings. The research questions: -What is the prevalence of anxiety symptoms among nurses and nursing students and which are the most frequently experienced anxiety symptoms? -What is the correlation between different work environments in psychiatric nursing and the development of anxiety symptoms? -What is the impact of rotating night shifts on the development of anxiety symptoms? -What is the correlation between the development of anxiety symptoms and the employees' work experience, gender, self-rated health and economic status, and smoking habits?

Methods
The study was based on a non-experimental quantitative research method. A standardised structured questionnaire was deployed as a research instrument.

Description of the research instrument
The structured questionnaire was divided into three sections. The first section inquired about demographics and workplace of the respondents. In the second section the Slovene translation of the free online test for anxiety -the Burns Anxiety Inventory (BAI) (Burns, 2002) was used for data collection. This structured questionnaire (BAI) is divided into three parts, namely, anxious feelings, anxious thoughts, and physical symptoms. To evaluate the level of anxiety indicated on the checklist, each item is added up numerically: 0 to 4 = minimal or no anxiety, 5 to 10 = borderline anxiety, 11 to 20 = mild anxiety, 21 to 30 = moderate anxiety, 31 to 50 = severe anxiety, and 51 to 99 = extreme anxiety or panic (Burns 2002). The third section of the questionnaire included statements about the self-assessed economic and health status and physical fitness/activities of the respondents. A Likert-type scale was used to determine the intensity of experience on a continuum from strongly agree to strongly disagree with a choice of five pre-coded responses. The Likert scale was combined with a 5 point assessment scale (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent). Cronbachʹs Alpha was used to estimate the reliability of a psychometric test. Its value equals 0.831 and confirms the reliability of the instrument used (Cencič, 2009).

Description of a sample
The survey was conducted among graduate nurses and nursing technicians in three psychiatric hospitals in Slovenia and the third-year nursing students of the Faculty of Health Care Jesenice during their clinical practice in psychiatric settings in the study year 2013/2014. Of the 270 questionnaires, non-randomly distributed among a probability sample of psychiatric nursing personnel, a total of 192 were returned, yielding a return rate of 71 %. Of 80 questionnaires distributed among nursing students, a total of 50 were returned, yielding a return rate of 62 %. All the returned questionnaires were fully completed and included in the statistical analysis.
The research sample consisted of 192 (79.3 %) nursing professionals employed in three Slovenian psychiatric hospitals and 50 (20.7 %) nursing students performing their clinical practice in psychiatric settings. The percentage of female participants (186, 76.9 %) was significantly higher than the percentage of males (56, 23.1 %). The study included 73 (30.1 %) of nurses from the Psychiatric Department of the University Clinical Centre Maribor, 60 (24.8 %) nurses from the Psychiatric Hospital Idrija, and 59 (24.2 %) nurses from the Psychiatric Hospital Begunje. According to the results of the analysis, 62 respondents (32.3 %) of the respondents had between 20 to 29 years of working experience, 52 (27.1 %) of the respondents had between 10 to 19 years of working experience, 41 (21. 4 %) of the respondents had less than 10 years, and 37 (19.3 %) of the respondents had 30 to 39 years of working experience. The majority of the respondents (157, 64.9 %) have completed secondary school education, 8 (3.3 %) have a higher professional education, and 77 (31.8 %) hold BSc degree in nursing. Nearly half of the participants 109 (45 %) come from urban areas, 73 (30.2 %) of the participants are from rural areas and 60 (24.8 %) come from the suburbs. A good half 125 (65.1 %) of the respondents work in places requiring special surveillance for more than half of their working hours. The same number of the respondents (125, 65.1 %) work rotating night shifts more than five times a month, and 13 (6.8 %) of the respondents work more than seven nights per month. Less than five night shifts are performed by 59 (30.7 %) respondents.

Description of the research procedure and data analysis
The approval to conduct the survey was obtained from the relevant authorities in all participating hospitals as well as the Faculty of Health Care Jesenice. All the participants were informed with the aim of the study and the content of the questionnaire, verbally and in writing. Participation was voluntary and the participantsʹ anonymity was ensured. The study was carried out from June to August 2014. The data collected were statistically analysed using SPSS, version 22 (IBM, 2013). The descriptive statistics was used to establish the differences between the variables. The variables are described by frequency distribution, displayed by percentages and central tendency, described by mean or average values ( ). The data collected were processed by different statistical methods, namely, the t-test for independent samples, descriptive statistics, Leveneʹs test, the ANOVA statistical test, the Welchʹs t-test, the post hoc analysis and the Pearsonʹs correlation coefficient. The level of statistical significance was set at p < 0.05.

Results
Results of the study show that both mental health nurses as well as the students in psychiatric setting may develop anxiety symptoms. These symptoms are, however, less frequently experienced by the nursing personnel than the students ( = 0.30 and = 0.35, respectively). The employees are more likely to develop physical symptoms ( = 0.33) while the students more commonly experience anxiety feelings ( = 0.36) and anxiety thoughts ( = 0.36). The t-test for independent samples revealed that anxiety thoughts are statistically significantly more frequent in students than in nursing employees (t = -2.077, p = 0.039). No statistically significant differences were established in relation to other symptoms (Table 1).
The data on frequency distribution of specific symptoms show that the most pronounced symptoms among anxiety thoughts are: Difficulty concentrating ( = 0.69), Fears of criticism or disapproval ( = 0.84) and Racing thoughts or having your mind jump from one thing to next ( = 0.30).The most common physical symptoms of the respondents include: Feeling tired, weak or easily exhausted ( = 0.87), Skipping or racing or pounding of the heart ( = 0.60) and Sweating not brought on by heat ( = 0.48). The employees of the Psychiatric Hospital Indrija ( = 0.41) experience anxiety more frequently than those of the Psychiatric Hospital Begunje ( = 0.25) and the Psychiatric Department of the University Clinical Centre Maribor ( = 0.24).
The nursing professionals working in departments under special surveillance more frequently experience anxiety feelings ( = 0.33) and anxiety thoughts ( = 0.26) compared with those working in open psychiatric units. These differences are, however, not statistically significant (t = 0.553, p = 0.581). Similarly, no statistically significant difference was established between employees working overnight or having rotating shifts in comparison to those who, on average, less frequently work night and rotating shifts (t = -1.395, p = 0.165).
The study aimed to establish the correlation between anxiety symptoms and the years of working experience. From the data in Table 2, it is apparent that the greatest differences exist in the occurrence of physical symptoms, but these differences are not statistically significant (t = 0.499, p = 0.683). Genderrelated differences can be observed in the prevalence of anxiety thoughts which are more prevalent in female ( = 0.26) than male employees ( = 0.21). In the cohort of students, the greatest gender-related differences were noted in experiencing the anxiety feelings. The female/male ratio was = 0.42 and = 0.19, respectively. There are no statistically significant gender-related differences among employees, while in students statistically significant differences were identified in experiencing anxiety feelings which are more frequent in females than males (t = 2.041, p = 0.046) ( Table 2).
The study explored into the relation between anxiety symptoms and the respondents' nicotine dependence. The respondents were divided into the group of regular smokers, occasional smokers and non-smokers. It was established that regular smokers are more liable to develop anxiety symptoms than non-smokers. Statistical significance of this variable was calculated with the ANOVA statistical test and the Welch's t-test. In the group of employees, significant positive correlation was found between smoking habits in all the anxiety symptoms (p < 0.05), while no such differences were identified among student respondents (p > 0.05) ( Table 3).   The survey also explored the impact of self-rated economic ( = 3.22) and health status ( = 3.83) on the development of anxiety symptoms in nursing employees. The Pearson's correlation coefficient was used to measure the relation between the variables. It was established that subjective economic and health status is inversely associated with the prevalence and intensity of anxiety symptoms. The anxiety symptoms in the respondents were more pronounced when the self-rated economic status and health condition scores were low (p = 0.01 and p = 0.001 respectively). In student respondents, no statistically significant differences in this respect were identified (Table 4).

Discussion
The purpose of the current study was to determine the prevalence of anxiety symptoms in nursing employees and the students during their clinical practice in psychiatric settings. The choice of the research sample consisting of nursing employees and students was based on the assumption that similar anxiety symptoms are displayed in both groups of respondents. Melincavage (2011) argues that anxiety experienced by student nurses may interfere with the acquisition of knowledge and skills during clinical practice. Nursing professionals have to cope with stressful situations on numerous occasions. Gao and colleagues (2012) reported that nurses work in a stressful environment which may affect their mental well-being. The prevalence of anxiety symptoms in nurses was found to be much higher than that in teachers, medical students, doctors or even cancer patients.
Results of the current study show that nursing employees develop mainly the physical symptoms of anxiety, especially fatigue, weakness and exhaustion, which accords with earlier observations (Stathopoulou, 2011). Previous research indicates that working in a psychiatric setting is associated with increased work stress, high job strain and psychological distress. Mental health nurses are exposed to stressful events, including patient-perpetrated aggression and violence. Personal safety and health concerns and the concern for the physical and psychological well-being of patients dominate and have a profound impact on nurses (Chen, et al., 2005;Magnavita & Heponiem, 2012;Lee, et al., 2015). The prevalence of anxiety symptoms among psychiatric nurses was found to be higher than among other nursing professionals. The research conducted in all Slovenian psychiatric hospitals in 2009 produced similar results as the current study (Čuk & Klemen, 2010).
The current study shows that nursing students most frequently experience anxiety thoughts and anxiety feelings. Statistically significant differences were established between the nursing employees and the students in regard to anxiety thoughts which are more common in the latter group. Chernomas and Shapiro (2013) claim that stress, depression, and anxiety can interfere with learning, affect academic performance, and impair clinical practice performance. The same anxiety symptoms were identified in the student respondents, which validates the results of the current research. Similar results were obtained in the study conducted by Szpak and Kameg (2013) confirming that the greatest anxiety levels are observed in the students when attending psychiatric clinical training. Melo and colleagues (2010) also reported that anxiety is more prevalent in nursing students than in the general student population. The mean scores on the STAI for this sample of nursing students were higher than those previously reported for the general population of university students. The study conducted among American university students confirmed that higher social anxiety may result even in more alcohol problems (Kenney, et al., 2014).
The results of the study did not show any significant work-related increase in anxiety levels of the respondents. It was established, however, that nursing employees working in the units under special surveillance more frequently develop anxiety thoughts and anxiety feelings than those working in general psychiatric units. In contrast to some other research findings (Nooryan, et al., 2012), these differences are, however, not statistically significant.
As some authors offer findings about work-related anxiety symptoms which differ from the results of the current study, further research on this topic is recommended. A study carried out among Taiwanese psychiatric nurses investigated the impact of patients' aggressive behaviour on the increased anxiety symptoms in nursing employees (Chen, et al., 2005). The established prevalence of aggression in psychiatric inpatient units ranges from 6.1 % to 35 % per hospital bed, or 2.54 incidents of the patients' aggressive or violent behaviour per year in comparison to only 0.73 incidents in other hospital wards. Patients' verbal and physical aggression causes significantly higher levels of state anxiety among psychiatric nurses than among other nursing professionals. Gao and colleagues (2012) conducted a study the purpose of which was to investigate anxiety symptoms and the associated factors in nurses working in public city hospitals. Results of the study indicate that among all the factors, the workplace conditions and work-related stressors stood out as the strongest associated factor for anxiety symptoms in nurses. The above study also reveals that the stressful situations and events most commonly occur in psychiatric wards under special surveillance.
In regard to factors conducing to the development of anxiety symptoms, the findings of the current research differ from some published sources (Melo, et al., 2010;Nooryan, et al., 2012). Contrary to some other research findings, the current study, for instance, does not demonstrate any elevated anxiety levels in nurses working in wards under special surveillance. Although previous research tried to establish an association between increased number of anxiety symptoms and working rotating night shifts, no evidence of this association was confirmed in the current study. This finding is in agreement with the results obtained by Portela and colleagues (2004), Ulas and colleagues (2012) and Øyane (2013) who suggest that nurses working rotating night shifts experience even fewer anxiety symptoms. The current study also did not establish any positive correlation between increased number of anxiety symptoms and work experience or gender which accords with observations of some other studies (Boya, et al., 2008;Atindanbila, et al., 2012). Gender-related occurrence of anxiety symptoms was, however, confirmed by the Iranian study of mental disorders among shift work hospital nurses (Ardekani, et al., 2008) which reported that anxiety and somatic symptoms are more common among females. The same results were obtained in the study among Japanese nurses which indicated that female nurses have significantly higher stress levels than males related to stress reactions of fatigue and anxiety (Yada, et al., 2014). Distinct from other findings in the current research, statistically significant higher prevalence of anxiety feelings and thoughts was established among female student respondents.
Multiple studies have demonstrated an association between cigarette smoking and increased anxiety symptoms or disorders (Koprivnikar, 2012;Jamal, et al., 2012;Moylan, et al., 2013). Nicotine-dependence is very common also among nursing professionals. This observation is in agreement with that of O'Donovan (2009) who found that nurses working within psychiatric care have the highest smoking prevalence (47.4%) among the nurses who smoked (O'Donovan, 2009). The most common factors preventing smoking cessation included fear of withdrawal symptoms including stress and anxiety (Berkelmans, et al., 2011). The current study identified a positive correlation between nicotine dependence in nursing employees and the increased prevalence of anxiety symptoms. However, no statistically significant correlation between nicotine dependence and occurrence of anxiety symptoms was established in the student respondents. These results are consistent with the results of the Iranian study conducted on a sample of 1020 high school students (Khademalhosseini, et al., 2015).
The current study confirmed that poor self-reported economic condition and health status present greater risk for the development of anxiety symptoms. Several studies revealed that poor economic conditions substantially and significantly increase the prevalence of anxiety symptoms (Spence, et al., 2002;Hjorleifsdottir, 2007;Gili, et al., 2013).

Study limitations
A considerable limitation of the current study is the non-probability sample, composed of a considerably larger proportion of female nursing professionals who work rotating night shifts in the wards under special surveillance. The sample composition therefore limits our ability to make broader applicability, translation and generalisations of the obtained results.