SELF-REPORTED SLEEP QUALITY AND ITS ASSOCIATION WITH DEMOGRAPHIC AND HEALTH-RELATED FACTORS AMONG EMERGENCY DEPARTMENT NURSES

Continual sleep loss and disorder among Emergency Department (ED) nurses may negatively impact the quality of care. ED nurses are expected to act with great pace and accuracy in a crowded and high-risk environment. Detail information regarding sleep quality (SQ) and its components among ED nurses is still limited in Indonesia. The study aimed to explore SQ of ED nurses and its relation to demographic and health-related factors. This study utilized a cross-sectional, correlational descriptive design with 65 samples recruited using total sampling technique from 3 EDs in Central Sulawesi. Descriptive statistics and bivariate analysis were performed to examine the association between variables. Global PSQI score was 5.5±2.0 with the prevalence of poor sleepers was 43.1%. Only body mass index (BMI) had correlation with SQ (r=0.26; p<0.05). ED nurses are recommended to reach the ideal body weight to get a bigger chance of having a better sleep. Further studies are required to evaluate interventions that might enhance SQ of ED nurses.


Introduction
Epidemiological studies show that sleep disturbance is a common problem among adults. About 30-35% of the world's population complain about sleep problems and 10% of cases were caused by chronic insomnia (World Health Organization, 2018).
Nurses are the largest group of health workers (about 50-60%) in the hospital, cover shifts 24-hour working hours a day, and have the longest duration of contact with patients (Ministry of Health Republic of Indonesia, 2013). Like the general population, one of the problems that occur in nursing practice is inadequate sleep. The prevalence of sleep problems tends to increase every year. When nurses experience disturbed sleep, it may lead to work negligence (Thayeb, Kembuan, & Khosama, 2015). Specifically, Weaver, Stutzman, Supnet, and Olson (2016) found that sleep quality created a bigger influence on nurses' performance than sleep quantity. A better sleep quality may reduce the chance of doing errors.
Emergency Department (ED) nurses play an important role in providing nursing care and must perform various tasks with high pressure and demands. The unpredictable severity and number of patients causing the workload seems very heavy. Therefore, sleep disorders are common among those who are working in the ED (Stathopoulou, Karanikola, Panagiotopoulou, & Papathanassoglou, 2011;Zeng et al., 2020). The prevalence of poor sleep quality among ED nurses population ranged from 36.5% to 92.1% (Kim, Kim, Park, Kang, & Choe, 2013;Suleiman, Hijazi, Kalaldeh, & Sharour, 2020;Thayeb et al., 2015). Like in other wards, shift work is a significant cause of poor sleep quality for nurses. This could happen due to disruption of the circadian rhythm of someone who works on night shifts. (Kim et al., 2013).
The understanding of sleep quality is quite complex and is related to various factors such as fatigue, work stress, or other psychological factors (Chien et al., 2013). A study by Tareluan, Bawotong, and Hamel (2016) showed that an increasing number of patients in ED lead to fatigue that affects the nurse's sleep patterns. In congruent with Saftarina and Hasanah (2014) that stated heavy workload can cause stress, fatigue, and ultimately affect both the quality and quantity of nurses' sleep. Scott, Arslanian-Engoren, and Engoren (2014) mentioned that in critical care units, poor sleep quality among nurses may determine the clinical decision, as the patients in these units are more vulnerable compare to other units.
During the night shift, sometimes nurses in the ED are found asleep and woken up by patients and their families. This may have impact on response time delays and often lead to dissatisfaction with the performance of nurses. Overall, sleep quality of nurses were worse than general populations. However, detail information for better understanding of sleep quality among ED nurses in Indonesia is still limited. Therefore, the aim of this study was to assess the sleep quality of ED nurses, including its components by self-reported questionnaire. We also tried to identify potential demographic data and health-related factors for sleep quality among ED nurses.

Methods
This study used a descriptive correlation with crosssectional approach. The participants in this study were 65 ED nurses, recruited using total sampling technique from EDs of 3 general hospitals in Central Sulawesi, Indonesia. Data were collected through 1month-long survey in July 2019. The inclusion criteria were ED nurses who has been working at ED with minimum of one year experience, able to communicate in Bahasa Indonesia, and willing to join the study. The head nurse and those who were on leave or illness condition were excluded from the study.
Demographic data included age, gender, tribe, religion, education, marital status, working experience in the ED, and hospital status. Healthrelated factors were smoking, exercise, and body mass index (BMI). ED nurses' sleep quality was measured using Pittsburgh Sleep Quality Index or PSQI (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). This questionnaire consists of 24 questions, divided into 19 questions for participants and 5 for roommates (not included in the data analysis). There are 7 components for sleep quality assessment i.e. duration, latency, subjective sleep quality, duration, habitual sleep efficiency, use of sleep medication, and daytime dysfunction. Each components weighted in the range of 0-3. Therefore, global PSQI score ranged from 0-21. Scores more than 5 distinguished good and poor sleepers, where a higher score denotes worse sleep quality. The cronbach's alpha of PSQI in this current study was 0.77.
Permissions to conduct this study were obtained from the ethical committee of each hospital. Eligible nurses were listed from the head nurses. The researcher elucidated the purpose of study and procedure to potential participants. Anonymity and confidentiality were protected to all participants. As an agreement to join this study, each participant signed the informed consent before data collection. Participants were free to withdraw at any time. Data were collected through face-to-face interview and questionnaires.
Data collected in this study were analyzed using statistical software. Data was checked for any errors that might appear after retrieval. Data analyses approaches included: 1) Descriptive statistics (mean, standard deviation, frequency, percentage, minimum, maximum) was employed to describe all variables; 2) Cronbach's alpha was used to determine the reliability of PSQI; 3) Mann-Whitney, Kruskal-Wallis, and Spearman Rank Correlation was utilized to identify associations between variables. The level of significance was set at 0.05. Table 1 portrayed demographic data of 65 ED nurses. The age of participants ranging from 21 to 57 years with a mean of 33.4 (SD=7.5). Majority of participants were female (55.4%) and married (81.5%). The top 3 tribes were Pamonanese (24.6%), Togeanese (18.5%), and Bugisnese (13.8%). Those who have faith in Islam (61.5%) were more dominant than any other believers. The highest level of education was Diploma III or DIII (75.4%). The mean of working experiences was 4.9 (SD=4.7), ranging from 1-33 years. More than three-fourth of ED nurses were working in public hospitals (83.1%). Non-smoking participants were 90.8% and those who routinely doing exercise only 29.2%. The mean of BMI was 23.3 kg/m 2 (SD=3.0), ranging from 16.8 to 34.7 kg/m 2 .   The prevalence of sleep disturbances condition was determined by adding score categories of 2 and 3. As depicted in Table 3, "have to get up to use the bathroom" was the most frequent condition. There were 41 participants (63.1%) reporting of having this event. From those participants, 34 of them had nocturia one or two nights in a week.

Self-Reported Sleep Quality Components and Global PSQI
The second highest prevalence of sleep disturbances was "wake up in the middle of the night or early morning", which experienced by 58.4% participants. More than half of the participants (52.3%) had sleep problem due to midnight arousal or in the early morning one or two nights a week. The other most common condition was "cannot get to sleep within 30 minutes" which reported by 27 (41.5%) participants, where all of them had difficulties in initiating sleep within 30 minutes one or two nights in a week.

Factors Associated with Sleep Quality
The Mann-Whitney and Kruskal-Wallis test were utilized to measure the association between categorical data of demographics and health characteristics with global sleep score (see Table 4). The findings showed there were no relationships between variables among participants (p>0.05).

Discussion
The average age of respondents was 33.42 with intervals of 21-57 years. Thayeb et al. (2015) found that the age of emergency room nurses ranged from 20-40 years. Some of the nurses aged over 40 years rarely experiencing rotation to another wards of the hospital. Most of the respondents were women. Similarly, the study of Tareluan et al. (2016) found that in the ED, the nurses were predominantly women. There are two major of participants' tribes in this study, Pamonanese and Togeanese. These two tribes are originally form two regencies that the study was done.
Islam and Christianity are the religions most professed by the study participants, consistent with the distribution of religion in the province. Majority of ED nurses hold DIII degree. In contrast to the study by Tareluan et al. (2016), where most of their participants had bachelor degree. DIII program is a type of nursing education that produces graduates who have the competence to implement nursing care, predominantly refers to practice, so it tends to provide more action than analysis (Ekawati & Ardani, 2018).
Regarding marital status, the majority ED nurses were married. The findings were consistent with a study in the same population by Thayeb et al. (2015), where more than half of the participants were married. Married nurses, especially women, often face multiple role conflicts in family and work. The worst effect in this situation is that the quality of their services are poor (Wulandari & Dwiyanti, 2014).
The average of working experience among ED nurses was almost 5 years. Studies by Kasmarani (2012) and Suleiman et al. (2020) also found that the average years of experience of nurses in the ED was lower than 6 years. Job rotation was the main reason behind it. Periodic and proper job rotation can be used as a strategy to help nurses achieve better performance, expand skill and knowledge, and enhancing quality of care to patient. Moreover, it may increase job satisfaction, reduce conflict, and motivate nurses to keep focus on nursing career and prevent turnover rate (Pinhatti, Vannuchi, Sardinha, & Haddad, 2017).
Most of participants were working in public hospitals. This is quite plausible, where 2 out of 3 settings in this study took place in the ED of public hospitals. Majority of ED nurses in this study were non-smokers. This findings is similar with a study by Rahajeng and Tuminah (2011) who found only about one-third of their participants were smokers, while in contrast to Anggara and Prayitno (2013) were slightly more than half of participants were active smokers.
Participants who routinely do regular exercise were only about 29.2%. Workload, shift rotation schedule, and family are the reasons for the lack of time to do routine exercise. The mean BMI of the participants in this study was still considered in the normal category. In contrary to a study by Saftarina and Hasanah (2014) who reported the average BMI of hospitalized nurses was included in pre-obese group.
The main problem of sleep quality in this study participants was daytime dysfunction. In contrast, the result was inconsistent with Suleiman et al. The study revealed that the older the participants, the worse subjective sleep quality they reported. Increasing age compensates for both biological and psychological needs, resulting in stress levels that worsen the sleep quality (Susanti, Kusuma, & Rosdiana, 2017 (Vgontzas et al., 2008).
The use of sleeping medication had a correlation with BMI in this study participants. Limited information available regarding the relationship between these two variables. In a study among Australian nurses and midwifes, about one third of participants used prescribed sleeping medication as their sleep aid (Dorrian et al., 2011). Shao, Chou, Yeh, and Tzeng (2010) reported that female nurses in Taiwan who took prescribed medication for sleeping had poorer physical health.
Overall, only BMI had correlation with sleep quality in ED nurses. This is consistent with studies by Beebe, Chang, Kress, and Mattfeldt-Beman (2017) and Zeng et al. (2020) who found that those with higher BMIs were poor sleepers. Nurses tend to do extra eating or snacking during the night shift or when awake at night and this may lead to a higher BMI. This was done as a way to help their sleep problems and was thought to increase happiness during work (Arslan & Aydemir, 2019).
Some limitations were found in the current study. The use of cross-sectional design is limited to generalizing the findings, since it can only highlights associations but not a causal relationship. The data collection were only taken from three EDs in Central Sulawesi, thus the study findings may not presented the whole status of sleep quality among ED nurses in other areas of the province. Future research should expand the scope in larger area to have better understand the sleep quality among this specific population. Response bias may happen, due to the use of self-rated questionnaire.

Conclusion
Among the components of sleep quality, daytime dysfunction was the main problem of study participants, but most of ED nurses were good sleepers. Older ED nurses experienced poor sleep quality and had more sleep disturbances. Senior ED nurses were faster to fall asleep, compared to those who had fewer years of work experience. In addition, those who were obese or belong to a high BMI group, had more disturbed sleep, tend to use sleeping medication, and generally poor of sleep quality. The findings in this study may stimulate