Spatial patterning of occupational stress and its related factors in Iranian critical care nurses using a hierarchical Bayesian technique

Abstract

BACKGROUND:

In each community, health problems’ patterns and geographical changes are of prime importance to determine high and low-risk areas.

OBJECTIVES:

This study aimed to investigate the Spatial patterning of occupational stress and its related factors in Iranian critical care nurses using a hierarchical Bayesian technique

METHODS:

The current research was a cross-sectional descriptive-analytical study. The data includes the number of critical care unit nurses who show a high stress level based on a questionnaire. We used variables such as age, gender, collaboration status, working time, marital status, clinical experience, education, supervisor support, stress score, and working on holiday days for this study. The survey participants had to be at least 18 years old, a registered nurse, and working in the intensive care unit (ICU). OpenBUGS version 3.2.3 was used to implement the Bayesian hierarchical Poisson model and find partial patterning of occupational stress and its related factors.

RESULTS:

The final sample size was 17414 nurses. The overall prevalence of occupational stress in ICU nurses was estimated at 70%. The lowest and highest prevalence was 65.8% in the North Khorasan province and 75.2% in Golestan province. Occupational stress had a statistically significant association with collaboration status, but with demographic variables, shift work, supportive supervisor, and working on holidays had no statistically significant association.

CONCLUSIONS:

According to the findings, it is necessary to eliminate or reduce job stress and increase efficiency in Iranian nurses, encourage teamwork and collaboration as an essential element of a healthy workplace environment.

Keywords

Community health critical care nurses occupational stress spatial analysis

1 Introduction

In today’s modern and developed societies, human resources are the most valuable asset of any organization. However, this asset faces many problems and issues that can negatively affect their quantitative and qualitative efficiency and, ultimately, the organization’s overall performance. In recent decades, stress is one of the problems that has plagued human societies with the modernization of life [1]. Human reaction to the set of internal and external incompatible and unpredictable factors is called stress. So that whenever the mental and emotional harmony is lost, stress is created [2]. Stress is one of the most common diseases of the 21^st century that affects humans in different situations [3]. Stressors are factors or situations that affect people’s performance and health [4]. Job status is one of these stressors. Occupational stress is commonly defined as the harmful physical and emotional responses that occur when the demands of the job exceed the worker’s capabilities, needs, or resources. Job pressures that a person faces to adapt to their job requirements are inevitable and may be tolerable quickly. However, in a long time, a person’s physical and mental resilience deteriorates and eventually leads to job exhaustion [5].

For this reason, the occupational stress issue and its outcomes in employees are currently one of the most common problems in the service sector [6]. Among health care workers, nursing is recognized as one of the high-risk occupations for fatigue and illness. Nursing care is an essential component of a country’s healthcare system, and nurses play a vital role in the health care system [7, 8]. Problems such as work pressure, shift time, nursing service type, nursing care, communication with critically ill patients, and intra-organizational communication are well-known sources of occupational stress in nurses [9]. The US National Institutes of Health reported that nurse’s rank in referral to a psychologist or psychiatrist is 27th out of 130 [10, 11]. Poor nursing support, conflict with colleagues and patients, high job demand, and overtime are some of the stressors that can cause stress in nurses’ work environments [12, 13]. Among nurses in different units, nurses working in the intensive care unit face more stress than nurses working in other units [14]. Stressful situations in the intensive care unit are working with nurses and healthcare staff. Communicating with the patient and his companions, the need for a high level of knowledge and skills, high workload, the need for prompt and immediate response to immediate situations, and the heavy responsibility of patient care. These factors cause stress and ultimately job exhaustion [14, 15].

Since one of the most important areas of sustainable health development in human societies is the health sector and nurses, they are very active as efficient and effective human resources to care for patients and increase their efficiency in hospitals. Therefore, it is needed to ensure their health. However, due to the shortage of nurses in developing countries compared to developed countries, proper and appropriate planning should be done to eliminate and reduce stress-causing factors by policymakers to maintain health and especially mental health [16].

We believe that the most critical topic in mental health science research is applying applied research to recognize the prevalence patterns of this phenomenon to prevent most mental disorders in Iran. Drawing geographic maps to identify the root causes of stress and related information can be the first step. In actuality, disease mapping is a set of statistical approaches for obtaining precise mortality or disease incidence estimates and compiling them into geographic maps [17]. Epidemiology, which can be defined as the estimation and presentation of summary measures of health outcomes, has a long history with disease mapping. Disease mapping is a useful method for identifying areas at higher risk for health problems and in need of therapeutic or intervention programs. Disease distribution has a spatial form in disease mapping issues. The mapping goals are to find the departure from the expected value of disease in society and identify the areas with higher risk than expected. The Standardized Mortality Ratio (SMR) has traditionally been employed for disease mapping presentation, although it has some disadvantages. First, the observed-to-expected ratio estimate leads to significant changes in estimate while small changes in expected value.

Furthermore, the SMR will be pretty high when a minimum expectation is obtained, indicating that this SMR represents a saturated relative risk estimate [18, 19]. As a result, crude risks are not reliable values for mapping diseases since they are imprecise in minimal areas, and rates with considerable chance variation tend to highlight the map, making it difficult to read [20]. In epidemiology, disease mapping has a long history as part of the basic triangle of person/place/time. In the 19th century, John Snow utilized the point map approach to analyze plague epidemics in London [21]. The scattering of data and the possibility of observed data correlation in nearby locations is a significant hurdle for which a Bayesian hierarchical model has been developed. Bayesian model-based disease mapping predicts the local area risks ensemble in the best possible way, separates systematic variability from random noise and produces clean maps from random noise and any population variation outcomes [22]. When there is a spatial correlation between observations in neighboring regions, Clayton and Kaldor introduced hierarchical models and related empirical Bayesian inference for standardized mortalities [23]. Various geographical models for hidden hierarchical levels have been proposed in recent years. Because adjacent areas have the similar disease or mortality rates, the spatial pattern must be taken into account for estimating map parameters. The Bayesian method has been presented to analyze such situations, combines information about cases seen in each region and information about the relative risks of the entire region in terms of the prior distribution [17]. To put it another way, the goal of Bayesian spatial models is to provide smooth estimates of odds ratios (OR). Such maps are often created using Bayesian hierarchical models, in which the spatial pattern in disease risk is represented by a series of random effects. A conditional autoregressive (CAR) prior is frequently used to assign these random effects [24, 25].

Therefore due to the spatial correlation in the geographical distribution of nurses’ stress, models that consider the hierarchical structure should be used to obtain reliable estimates [23, 26].

Conventional regression models (such as multiple linear regression, etc.) cannot consider this spatial correlation, and the estimates obtained from these models are not very accurate. Therefore, in such studies, to achieve higher performance, the generalized or Bayesian models are used, which consider the correlation between data (such as mixed generalized linear models, multilevel models, etc.). There have been many studies on nurses’ stress in Iran, but the geographical and epidemiological mapping of nurses’ stress levels and Bayesian modeling has not been done. This study is the first to model this issue and prepare a map for the data. This study investigates the factors affecting intensive care unit nurses’ stress and geographic distribution in Iran [17 , 28].

Epidemiological studies on occupational issues, including work-related accidents, are superior to other types of studies. Because they can identify knowledge about the factors that increase the likelihood of such accidents occurring, determine the effectiveness of preventive studies, and increase social awareness and inform policymakers about the severity and importance of the problems [29 –31].

2 Materials and methods

The present study is a cross-sectional descriptive-analytical study. This study aimed to determine the geographical distribution of stress in intensive care unit nurses and its factors, using the hierarchical Bayes model in Iranian public hospitals. The study population is all intensive care unit nurses of Iran. Surveys were collected in 2015. The geographical units of the study were 31 neighborhoods (province). Data were collected using multistage cluster sampling. The ethics committee of Baqiyatallah University of Medical Sciences, Tehran, Iran approved the study (ethical code: IR.BMSU.REC.1399.192).

In the first step, to decrease selection bias, ten hospitals were selected from each province using cluster random sampling. In the second step, five hospitals with more than 100 nurses were selected from each province. In 5 provinces, all hospitals had fewer than 100 nurses; therefore, all ten hospitals were selected. Age 18 years, registered nurse, works in the intensive care unit (ICU), and willing and able to complete the survey were inclusion criteria.

Based on the results of the pilot study, the sample size was calculated. According to the odds ratio of 0.90 (the association between years of experience and stress levels), with a proportion of stress of 0.50, 95% confidence level, and 90% power, the sample size was calculated as 21767 nurses. Thus, the final sample size reached 17414 nurses by considering a 20% dropout rate [32].

The researchers delivered sealed envelopes containing project summaries and questionnaires to accredited intensive care nurses in the clinical setting. If the nurse is satisfied with the participation, the responsible nurse performs the tasks related to clinical tasks coverage during the survey. After completion, the form was returned to the researcher with a sealed envelope provided in the initial packet. Three investigators evaluated each de-identified questionnaire to determine suitability for inclusion (based on the completeness of the data and the nurse approval status).

The questionnaire consisted of 2 parts: [1] socioeconomic, professional, institutional variables, and [2] work stressors. The utilized questionnaire was previously derived and validated using a three classic round Delphi technique. Socioeconomic (age, gender, marital status, number of children, body mass index [BMI]), professional (education level and years of critical care nursing experience), and institutional (shift schedule, frequency of working holidays, city, hospital, ICU bed number, nurse-patient ratio, degree of nursing collaboration type of CCU, ICU system, and having a supportive supervisor) are the variables in the study. The collaboration variable had a 5-domain Likert-type scale (very low, low, moderate, large, and very high. The supportive supervisor’s presence had two options (yes or no). The questionnaire’s part 2 had a 22-item inventory of work stressors. This questionnaire has been reported in the study by Vahedian-Azimi et al. [32].

The questionnaire had a 5-point Likert-type scale ranging from “causes me no stress” to “causes me extreme. The minimum and maximum scores of the questionnaire were 22 and 110, respectively. The cutoff point was obtained 67 for stress based on a pilot study. Values higher than this cutoff point indicate significant stress [33]. This cutoff point was extracted using both quantitative and qualitative assessments, and based on that, a score of 67 was selected. Quantitative assessment was obtained using the receiver operating characteristic curve analysis. The qualitative assessment was performed by the expert group consisting of psychiatric nurses (n = 5), psychologist (n = 1), psychiatrist (n = 1), ICU nurses (n = 5), CCU nurses (n = 5), dialysis unit nurses (n = 5), intensivists (n = 3), cardiologists (n = 3), nephrologists (n = 3), and ICU administrators (n = 5). The k agreement coefficient test was performed after the two sessions. 89. Data for this study was taken from a previous study conducted by Vahedian-Azimi et al. [32].

This study’s data consist of the intensive care unit nurses’ number that has a high-stress score. We used age, gender, collaboration status, working shift, marital status, clinical experience, education, supervisor support, stress score, and working in holiday days’ variables for this study.

In epidemiology and environmental health research, the Bayesian strategy has become standard. It has named “hierarchical” because it iteratively employs numerous levels of analysis. Unlike classic statistical inference, which produces parameter estimates for each analysis unit by borrowing information from all analysis units, hierarchical Bayesian modeling produces parameter estimates for each analysis unit by borrowing information from all analysis units, the Bayesian “borrow of strength” effect. The variance is equal to the mean in a standard Poisson model. However, many Poisson models have larger variances, and they are called over-dispersed Poisson models. These “extra variances” are identified utilizing hierarchical Bayesian models. In the non-Bayesian method to spatial statistics, a regression that only explains a small variance is achieved if the regression model is highly uncertain. The “unexplained variance” in a hierarchical Bayesian model, on the other hand, is commonly described as either spatially-correlated effects or heterogeneity effects. Bayesian inference is a powerful forecasting technique in various fields due to its ability to incorporate prior knowledge without being constrained by classical distributional assumptions [34].

The number of nurses that have high-stress score follows a Poisson distribution, $O_{i} \sim Poisson (E_{i} μ_{i})$ where O_i, E_i and μ _i are the observed number of nurses with a high-stress score, the expected number of cases, and the relative risk for province i, respectively.

The number of expected events is calculated as follows: $E_{i} = n_{i} (\frac{\sum_{i} 0_{i}}{\sum_{i} n_{i}}), i = 1, 2, \dots . I$ where n_i is the number of intensive care unit nurses in province i [35].

The Poisson model is applied to the count data. An overdispersion variance is common in these models. The reason for overdispersion can be the presence of spatial autocorrelation. If we assumed the local estimates of disease risk are correlated for neighboring areas, it implies spatial autocorrelation. Hierarchical models such as the Besag, York, and Mollie (BYM) model have been presented to neutralize these challenges. In this model, spatial factors that not measured were controlled by using suitable random effects, as shown below: $log (μ_{i}) = α + x_{i} β_{i} + u_{i} + v_{i}$ (1)

X_i’s are region-specific covariates, and β _i ’s are fix effects. Also, α is the log-relative risk baseline, and v_i and u_i indicates random random-effects components regarding spatial and non-spatial factors.

Conditional autoregressive (CAR) models are used to explicate the spatial autocorrelation across neighborhoods. For example, the CAR model within the BYM model is as follows: $[v_{i} | v_{j}, i \neq j, τ_{v}^{2}] \sim N ({\bar{v}}_{i}, τ_{i}^{2})$

where ${\bar{v}}_{i} = \frac{1}{\sum_{i} w_{ij}} \sum_{i} v_{i} w_{ij} and τ_{i}^{2} = \frac{τ_{v}^{2}}{\sum_{i} w_{ij}}$

If areas i and j are adjacent, w_ij is equal to 1; otherwise is 0. Risk factors with non-spatial structures are represented by u_i random effect. In the BYM model, this random effect follows the below distribution: $u_{i} \sim N (0, τ_{u}^{2})$

The parameters $τ_{v} = \frac{1}{σ_{v}^{2}}$ and $τ_{u} = \frac{1}{σ_{u}^{2}}$ are precision parameters.

For estimating parameters in the model, we implemented a Markov chain Monte Carlo (MCMC) simulation. We ran the MCMC model with 220,000 iterations, ignoring the first 20,000 iterations as burn-in. The thin value was 20. we specified the same gamma prior distribution for τ _v and τ _u . OpenBUGS version 3.2.3(http://www.openbugs.net/w/Downloads) was used to implement the BYM model [17, 24].

3 Results

In total, 17414 nurses were enrolled. About 70% of them had high-stress scores. Table 1 shows the distribution of nurses according to study variables. Overall, 31.2% of the nurses were male. The majority of nurses (60.7%) had a fixed shift working. 11.5% of nurses were single. A high level of collaboration was reported by 39.7%, a moderate level of collaboration by 42.2 percent, and a low level of collaboration by 18.1 percent. The supervisor was supportive of 36.8% of the nurses. Approximately 96 percent of them went to work during the holidays.

Table 1
Sociodemographic characteristics of the nurses

Variable Number Percentage

Collaboration Status Low 3144 18.1

Moderate 7348 42.2

High 6922 39.7

Working Shift Fixed 10573 60.7

Rotation 6841 39.3

Marital Status Single 2001 11.5

Married 15413 88.5

Gender Male 5433 31.2

Female 11981 68.8

Education Bachelor and Associate 15421 88.6

Masters and PhD 1993 11.4

Supportive Supervisor Yes 6400 36.8

No 11014 63.2

Working Holiday Days Yes 16589 95.3

No 825 4.7

Variable		Number	Percentage
Collaboration Status	Low	3144	18.1
	Moderate	7348	42.2
	High	6922	39.7
Working Shift	Fixed	10573	60.7
	Rotation	6841	39.3
Marital Status	Single	2001	11.5
	Married	15413	88.5
Gender	Male	5433	31.2
	Female	11981	68.8
Education	Bachelor and Associate	15421	88.6
	Masters and PhD	1993	11.4
Supportive Supervisor	Yes	6400	36.8
	No	11014	63.2
Working Holiday Days	Yes	16589	95.3
	No	825	4.7

Table 2 shows the prevalence of stress by province. The lowest prevalence was in the North Khorasan province with 65.8%, and the highest prevalence was related to the Golestan province with 75.2%. Both of these provinces are located in northeast Iran. The prevalence of stress was close to 70% in four provinces (East Azerbaijan, Tehran, Qom, and Markazi). The results in Table 2 show that the prevalence of stress is high among intensive care nurses.

Table 2

Prevalence of stress by province

Province	Number	Percentage
Ardabil	369	73.8
Esfahan	525	70.9
Alborz	296	74.0
Illam	268	67.0
East Azerbaijan	498	70.0
West Azerbaijan	339	67.8
Bushehr	367	73.4
Tehran	671	70.2
Chaharmahal and Bakhtiari	294	73.5
South Khorasan	345	69.0
Razavi Khorasan	616	67.0
North Khorasan	329	65.8
Khuzestan	408	68.0
Zanjan	292	73.0
Semnan	346	69.2
Sistan and Baluchestan	445	74.2
Fars	663	73.5
Ghazvin	296	74.0
Qom	209	69.7
Kurdistan	368	68.9
Kerman	497	71.9
Kermanshah	368	73.6
Kohgiluyeh and Boyer-Ahmad	267	66.8
Golestan	376	75.2
Gilan	345	69.0
Lorestan	356	71.2
Mazandaran	330	66.0
Markazi	481	70.3
Hormozgan	372	74.4
Hamadan	331	66.2
Yazd	637	72.7

Table 3 reports the results of the BYM Model for the 10,000 samples after the burn-in period. Posterior mean, standard deviation, Monte Carlo error, median, 2.5th percentile, 97.5th percentile (with the last two items being simply the 95% credible interval), and relative risk, computed as the exponential of posterior mean presented in this table. The sign (positive or negative) and the size of the parameters indicate the direction and magnitude of the fixed effects for the regression coefficients. In this study, of the variables, only collaboration status indicates an effect on stress relative risk. That value was about 0.98. The means that collaboration status shows a 2% decrease in relative risk for each increase in the size of its standard deviation. Other variables in the model (demographic variables, shift work, supportive supervisor, and working on holidays) did not significantly affect the relative risk of stress.

Table 3

Posterior summaries for regression coefficients

Parameter		Mean	SD¹	MCE²	CI³ 2.5%	Median	CI 97.5%	Relative Risk
Intercept		–1.94	0.97	0.09	–4.02	–1.85	–0.26	–
Collaboration Status	Low	–0.0118	0.0054	0.00052	–0.0219	–0.0118	–0.0013	0.9882
	Moderate	–0.0155	0.0052	0.00051	–0.0256	–0.0149	–0.0066	0.9846
	High	–0.0124	0.0052	0.00051	–0.0216	–0.0122	–0.0021	0.9876
Working Shift	Fixed	–0.0024	0.0085	0.00085	–0.0124	–0.0057	0.0123	0.9976
	Rotation	–0.0039	0.0086	0.00086	–0.0142	–0.0070	0.0115	0.9961
Marital Status	Single	0.0138	0.0125	0.00123	–0.0075	0.0140	0.0329	1.0138
	Married	0.0077	0.0122	0.00122	–0.0114	0.0073	0.0244	1.0077
Clinical Experience	–0.0001	0.0468	0.00453	–0.0939	–0.0028	0.0881	0.9999
Gender	Male	0.0078	0.0125	0.00125	–0.0145	0.0109	0.0278	1.0078
	Female	0.0067	0.0125	0.00125	–0.0155	0.0104	0.0264	1.0067
Education	Bachelor and Associate Degree	–0.0022	0.0104	0.00104	–0.0171	–0.0044	0.0189	0.9978
	Masters and PhD	0.0073	0.0115	0.00110	–0.0126	0.0059	0.0325	1.0073
Age	0.0583	0.0476	0.00474	–0.0353	0.0504	0.1572	1.0600
Supportive Supervisor	Yes	–0.0038	0.0037	0.00037	–0.0097	–0.0042	0.0056	0.9962
	No	–0.0036	0.0036	0.00036	–0.0091	–0.0040	0.0052	0.9964
Working in Holiday Days	Yes	0.0064	0.0040	0.00039	–0.0017	0.0070	0.0138	1.0064
	No	0.0046	0.0040	0.00039	–0.0036	0.0052	0.0120	1.0046

¹Standard Deviation; ²Monte Carlo error; ³Confidence Interval.

Figure 1 plots the map of the standardized incidence ratio (SIR) of stress in the study area. The standard incidence ratio is the crude ratio of observed high-stress counts to expected counts. The expected counts are considered fixed and proportional to the known population, and no variable or random effect is considered in their calculation. The SIR = 1 means that observed and expected counts are equal. In provinces with SIR > 1, observed counts are more than expected counts. Fourteen provinces had observed counts more than expected counts. The means that stress incidence is high in 45% of the provinces. High-stress relative risk was less than or equal to 1 in 55% of the provinces. In most provinces with borders with other countries, there is less risk of stress (SIR < 1).

Fig. 1

Standardized Incidence Ratio of High-stress in Iran.

Figure 2 presents the map of the fitted incidence ratio of stress in the study provinces. The fitted incidence ratio is SIR multiplied by the exponential of μ _i , which is modeled in [1]. In this index, the fixed effects, spatial and non-spatial effects, have been accounted. As can be seen in Fig. 2, crude SIR was modified in some provinces. In other words, Fig. 2 shows the smoothed SIRs from the BYM model. Kermanshah, Golestan, and Sistan, and Baluchestan provinces have high-stress relative risk in the BYM model (Relative Risk > 1). While in the raw model (crude SIR), these provinces are among the low-risk provinces (SIR < 1).

Fig. 2

Fitted Incidence Ratio of High-stress in Iran.

4 Discussion

This research aimed to map occupational stress’s spatial distribution and its related factors in Iranian critical care nurses. Spatial analysis can present useful details at different geographic layers. A hierarchical Bayesian analysis indicates a level of inequality in the spatial distribution of occupational stress.

As mentioned, the overall prevalence of occupational stress was 70%, which indicates a high prevalence. In Northern Ireland and Thailand hospitals, the prevalence of occupational stress was 57.4% and 26.2%, respectively. In the United Kingdom, stress prevalence was very high, and nurses are highly exposed to occupational stress [36 –38]. In the present study, the relative risk of stress is low at 45% of the provinces (e.g., Zanjan, Gilan, and Hamadan). In comparison, 55% percent of the provinces have a high relative risk (e.g., Yazd, Fars, and Tehran); the prevalence of stress was 74.4% in Hormozgan province. Also, the findings of the study by Tajvar et al. showed that 83.9% of ICU nurses suffered from high stress in Bandar Abbas hospitals [39]. The results of these two studies are almost consistent. The BYM model shows that the relative risk of stress was high in the Hormozgan province; the study by Masoumy et al. showed that 67.8 % of nurses have moderate to high stress in Bushehr Hospitals ICU [40], which was consistent with our study. In a study conducted by Mohebbifar et al., 95.6% of nurses in Ghazvin had stress [41], whereas in our study the stress prevalence for Qazvin province was 74%. This difference could be due to changing working conditions, such as increasing the number of nurses in the ICU ward. In other studies, the prevalence of stress was 76.3% among nurses working in intensive care units of hospitals in East Azerbaijan, which is more than the present study [42]. The results of the study by Moallemi and Adroom were consistent with our study, and the prevalence of stress among nurses was about 75% [43]. However, the questionnaire for the evaluation of stress in these studies was not the same as ours.

It was recognized that there was an association between heavy workloads and stress levels. The workload burden caused high stress in nurses and may lead to severe psychological disorders. This difference in prevalence rates can be due to different workloads in different hospitals in different countries. In countries where the nurse ratio to the patient is appropriate, the prevalence of stress is lower. Due to the high prevalence of stress, it is necessary to recognize the causes of stress and take influential health policymakers to eliminate and reduce these factors to decrease stress prevalence in nurses and increase work efficiency.

In the present study, there was no significant relationship between age and stress relative risk. This result was consistent with Yim et al., Kakemam et al., Faraji et al., and Mohebbifar et al. [41 , 44–46]. However, some studies have reported a burnout incidence of 28% to 33% among ICU nurses18,32 and are associated with younger age,18,41 [47, 48], and Wu et al.’s findings have a significant relationship between age and occupational stress [49]. Therefore, we believe that the working environment of the ICU and the condition of patients hospitalized in this unit can impact stress at any age.

There was no statistically significant association between the occupational stress and gender variable. Our results are in line with the findings of other studies [41 , 46]. Nevertheless, they are not consistent with the studied by Royani et al. and Kakemam et al. [11, 44]. The study by Vahedian-Azimi [32] found that the male gender has lower levels of stress than peer collaboration that can be explained by the fact that women generally have more stress at work. However, from our point of view, occupational stress can impact any person regardless of gender.

There was not any statistically significant relationship between the nurses’ occupational stress and their education. Our results are in line with the studies of Yim et al. and Royani et al., but they are not consistent with other studies findings [11 , 46]. It is expected that nurses with a high level of education have a better performance than nurses with lower education levels. The expectation of others from the nurse and their expectations can effectively create and increase nurses’ occupational stress.

We did not find any significant association between the occupational stress and clinical experience variable. Faraji et al., Kakemam et al., Sharma et al., Sahraian et al., and Mohebbifar et al. also came to the same conclusions [41 , 51]. However, Miandoab et al. showed a significant statistical relationship between nurses’ occupational stress and work experience, so that with increasing work experience, the occupational stress Increased [52]. In our opinion, increasing the work experience leads to a reduced level of occupational stress of nurses with high work. The experience is the clinical competence that over time increases through obtaining different occupational experiences in the field of patient care.

Similar to the studied by Mortaghy et al. and Ali et al. [53, 54], we did not find a statistically significant relationship between the nurses’ occupational stress and marital status. However, in the study by Parveen et al., unmarried nurses had less job stress. The difference between their and our study might be due to the fact that Parveen et al. only included female participants [55]. Women are more stressed than men due to the stress of working at home and maternal and marital responsibilities, and this has caused a difference between the results of the two studies, but our study included both sexes. In contrast with this finding, Rostamani et al. reported that single employees compared to married employees had higher occupational stress due to couples sharing their problems and sympathizing with each other [56]. According to Darvishiet al., this difference is because single persons are mostly younger and less experienced [57].

In our results, shift time, working on holidays, and supportive supervisor variables were not significantly associated with occupational stress. In addition, Shojaei et al. and Faraji et al. showed no statistically significant relationship between the nurses’ occupational stress and their shift time [58, 59].

5 Conclusion

According to the findings, there is a statistically significant association between occupational stress and the status of cooperation, so it is necessary to eliminate or reduce job stress and increase efficiency in Iranian nurses, encourage teamwork and collaboration as an essential element of a healthy workplace environment.

Limitations

The Iranian healthcare system has certain features that limit the generalizability of the observations. For instance, in the United States, academic and tertiary ICUs are mosrly closed ICUs, but they are typically semi-closed or open in Iran. Also, we did not obtain data on the psychological symptoms of individual nurses.

Footnotes

Acknowledgments

The authors express their gratitude to the participants of the study and staff of the Military Health Research Center of the Baqiyatallah University of Medical Sciences.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declare no potential conflicts of interest.

Availability of data and materials

The raw data and materials used to support the findings of this study are available from the corresponding author (Mehdi Raei) upon reasonable request.

References

Hazavehei

SMM

, Kharghani Moghadam

, Bagheri Kholenjani

, Ebrahimi

. The influence of educational interventions to reduce occupational stress: A systematic review. Health Safety at Work. 2017;7(4):363–74.

Ghasemzadeh

, Abbaszadeh

, Hassani

, Hashemi

. Structural modeling of personality traits on stress and job performance considering the mediating effects of individual accountability. Iran Occupational Health. 2013;10(2):54–64.

Evans

, Kelly

. Pre-registration diploma student nurse stress and coping measures. Nurse Education Today. 2004;24(6):10.

Nejad

, Nikpeyma

. Occupational stressors in nursing. Iran Occupational Health. 2009;6(1):22–7.

Mohajan

. The occupational stress and risk of it among the employees. International Journal of Mainstream Social Science. 2012;2(2):17–34.

Ghaleei

, Mohajeran

. The relationship among spiritual intelligence, mental health and job stress in nurses in imam khomeini hospital of mahabad. Avicenna Journal of Nursing Midwifery Care. 2015;23(4):14–21.

Josephson

, Lagerström

, Hagberg

, Hjelm

. Musculoskeletal symptoms and job strain among nursing personnel: a study over a three year period. Occupational Environmental Medicine. 1997;54(9):681–5.

Shareinia

, Khuniki

, Bloochi Beydokhti

. Comparison between job stress among emergency department nurses with nurses of other departments. Quarterly Journal of Nursing Management. 2018;6(3):48–56.

McCoy

. Sources of stress and coping among emergency department nurses: Gonzaga University; 1997.

10.

Al-Aameri

. Source of job stress for nurses in public hospitals. Saudi Medical Journal. 2003;24(11):5.

11.

Royani

, Sabzi

, Mancheri

, Aryaie

, Aghaeinejad

, Goleij

. The Relationship Between Job Stressors and Coping Strategies From Critical Nurses’ Perspective. Iranian Journal of Psychiatric Nursing. 2016;4(3):25–32.

12.

Garrosa

, Rainho

, Moreno-Jimenez

, Monteiro

. The relationship between job stressors, hardy personality, coping resources and burnout in a sample of nurses: A correlational study at two time points. International Journal of Nursing Studies. 2010;47(2):205–15.

13.

Biganeh

, Abolghasemi

, Alimohammadi

, Ebrahimi

, Torabi

, Ashtarinezhad

. Survey of Occupational Stress Effects on Burnout among Nurses. Journal of Knowledge & Health. 2018;13(1):9.

14.

Amini

, FarahBakhsh

, NikoozadehKordmirza

. Comparative analysis of life satisfaction, resilience and burnout among intensive and other units nurses Quarterly Journal of Nursing Management. 2013;1(4):9.

15.

McGrath

, Reid

, Boore

. Occupational stress in nursing. International Journal of Nursing Studies. 2003;40(5):555–65.

16.

Nayeri

, Nazari

, Salsali

, Ahmadi

. Iranian staff nurses’ views of their productivity and human resource factors improving and impeding it: a qualitative study. Human Resources for Health. 2005;3(1):9.

17.

Lawson

, Browne

, Rodeiro

CLV

. Disease mapping with WinBUGS and MLwiN: John Wiley & Sons; 2003.

18.

Rao

, Molina

. Small area estimation: John Wiley & Sons; 2015.

19.

Maryam

, Abbas

, Mohsen

. Mapping the obesity in iran by bayesian spatial model. Iranian Journal of Public Health. 1970;42(6).

20.

Venkatesan

, Srinivasan

, Dharuman

. Bayesian Conditional Auto regressive model for mapping Tuberculosis prevalence in India. IJPSR. 2012;3(1):1–3.

21.

Gordis

. Epidemiology. Forth edition. United States of America: Elsevier; 2009.

22.

MacNab

. Hierarchical Bayesian spatial modelling of small-area rates of non-rare disease. Stat Med. 2003;22(10):1761–73.

23.

Clayton

, Kaldor

. Empirical Bayes estimates of age standardized relative risks for use in disease mapping. Biometrics. 1987:671–81.

24.

Besag

, York

, Mollié

. Bayesian image restoration, with two applications in spatial statistics. Annals of the Institute of Statistical Mathematics. 1991;43(1):1–20.

25.

Lee

. A comparison of conditional autoregressive models used in Bayesian disease mapping. Spatial and Spatio-temporal Epidemiology. 2011;2(2):79–89.

26.

Devesa

. Atlas of cancer mortality in the United States, 1950-94: National Institutes of Health, National Cancer Institute; 1999.

27.

Waller

, Gotway

. Applied spatial statistics for public health data: John Wiley & Sons; 2004.

28.

Goovaerts

. Geostatistical analysis of disease data: accounting for spatial support and population density in the isopleth mapping of cancer mortality risk using area-to-point Poisson kriging. International Journal of Health Geographics. 2006;5(1).

29.

Nurminen

, Karjalainen

. Epidemiologic estimate of the proportion of fatalities related to occupational factors in Finland. Scandinavian Journal of Work, Environment & Health. 2001;27(3):161–213.

30.

Villanueva

, Garcia

. Individual and occupational factors related to fatal occupational injuries: A case-control study. Accident Analysis & Prevention. 2011;43(1):123–7.

31.

Thacker

, MacKenzie

. Preface: the role of the epidemiologist in injury prevention and control—an unmet challenge. Epidemiologic Reviews. 2003;25(1):1–2.

32.

Vahedian-Azimi

, Hajiesmaeili

, Kangasniemi

, Fornés-Vives

, Hunsucker

, Rahimibashar

, et al. Effects of stress on critical care nurses: a national cross-sectional study. Journal of Intensive Care Medicine. 2019;34(4):311–22.

33.

Hashemian

SMR

, Farzanegan

, Fathi

, Ardehali

, Vahedian-Azimi

, Asghari-Jafarabadi

, et al. Stress Among Iranian Nurses in Critical Wards. Iranian Red Crescent Medical Journal. 2015;17(6):e22612–e.

34.

Zhu

, Gorman

, Horel

. Hierarchical Bayesian spatial models for alcohol availability, drug” hot spots” and violent crime. International Journal of Health Geographics. 2006;5(1):54.

35.

Ayubi

, Barati

, Dabbagh Moghaddam

, Reza Khoshdel

. Spatial modeling of cutaneous leishmaniasis in Iranian army units during -using a hierarchical Bayesian method and the spatial scan statistic. Epidemiology and Health. 2018;40:e2018032–e.

36.

Aoki

, Keiwkarnka

, Chompikul

. Job stress among nurses in public hospitals in Ratchaburi province, Thailand. Journal of Public Health Development. 2011;9(1):9.

37.

Smith

. The scale of perceived occupational stress. Occup Med (Lond). 2000;50(5):294–8.

38.

Nicholl

, Timmins

. Programme-related stressors among part-time undergraduate nursing students. Journal of Advanced Nursing. 2005;50(1):93–100.

39.

Tajvar

, Saraji

, Ghanbarnejad

, Omidi

, Hosseini

SSS

, Abadi

ASS

. Occupational stress and mental health among nurses in a medical intensive care unit of a general hospital in Bandar Abbas in 2013. Electronic Physician. 2015;7(3):1108–13.

40.

Masoumi

, Tahmasebi

, Jalali

, Jafari

. The Study of the relationship between Job stress and spiritual health of nurses working in Intensive care ward at Bushehr Hospitals. Nursing of the Vulnerables. 2016;3(8):37.

41.

Mohebbifar

, Khosravizadeh

, Sadeghi

, Ahansazan

. Job stress and its related factors in nurses of Qazvin University of Medical Sciences. The Journal of Medical Education Development. 2015;7(1):55–63.

42.

Farnaz

, Mojgan

, Vahid

, Farzad

. Relationship between general health, occupational stress and burnout in critical care nurses of Tabriz teaching hospitals. Iran Journal of Nursing. 2010;23(66):54–63.

43.

Moallemi

, Adroom

. Comparison of Job Stress and Job Satisfaction Amongst Nurses of Different Units. Military Caring Sciences. 2016;3(3 (9)):165–73.

44.

Kakemam

, Raeissi

, Raoofi

, Soltani

, Sokhanvar

, Visentin

, et al. Occupational stress and associated risk factors among nurses: a cross-sectional study. Contemporary Nurse. 2019;55(2-3):237–49.

45.

Faraji

, Karimi

, Azizi

, Janatolmakan

, Khatony

. Occupational stress and its related demographic factors among Iranian CCU nurses: a cross-sectional study. BMC Res Notes. 2019;12(1):634.

46.

Yim

H-Y

, Seo

H-J

, Cho

, Kim

. Mediating Role of Psychological Capital in Relationship between Occupational Stress and Turnover Intention among Nurses at Veterans Administration Hospitals in Korea. Asian Nursing Research. 2017;11(1):6–12.

47.

Embriaco

, Papazian

, Kentish-Barnes

, Pochard

, Azoulay

. Burnout syndrome among critical care healthcare workers. Current Opinion in Critical Care. 2007;13(5):482–8.

48.

Erlen

, Sereika

. Critical care nurses, ethical decision-making and stress. Journal of Advanced Nursing. 1997;26(5):953–61.

49.

, Chi

, Chen

, Wang

, Jin

. Occupational stress among hospital nurses: cross-sectional survey. Journal of Advanced Nursing. 2010;66(3):627–34.

50.

Sahraian

, Davidi

, Bazrafshan

, Javadpour

. Occupational Stress among Hospital Nurses: Comparison of Internal, Surgical, and Psychiatric Wards. International Journal of Community Based Nursing & Midwifery. 2013;1(4):182–90.

51.

Sharma

, Takkar

, Purkayastha

, Jaiswal

, Taneja

, Lohia

, et al. Occupational Stress in the Indian Army Oncology Nursing Workforce: A Cross-sectional Study. Asia-Pacific Journal of Oncology Nursing. 2018;5(2):237–43.

52.

yoosefian miandoab

, charkhat Gorgich Eoh , Rezvani Amin

, Shahrakipoor

. Occupational stress in operating room and anesthesia technicians of ZahedanAli ebneAbitalebHospital. Beyhagh. 2015;20(2):33–41.

53.

Mortaghy Ghasemy

, Ghahremani

, Vahediane Azimi

, Ghorbani

. Nurses Job Stress in Therapeutic Educational Centers in Zanjan. Journal of Research Development in Nursing and Midwifery. 2011;8(1):42–51.

54.

Dalirooyfard

, Masoumi

, Forouzan

. Occupational Stress among Emergency Department Nurses of Golestan and Emam Khomeini Hospitals in Ahvaz. Jundishapur Scientific Medical Journal. 2015;14(3):6.

55.

Parveen

. Investigating occupational stress among married and unmarried working women in Hyderabad city. Bahria Journal of Professional Psychology. 2009;5:21–37.

56.

Rostamani

, Rostamani

, Fallah

, Mousavi

, Jenabian

. Occupational Stress among Female Employees of the Golestan University of Medical Sciences in 2015. Journal of Clinical and Basic Research. 2017;1(4):36–40.

57.

Darvishi

, Sadeghi

, Saed Panah

. Evaluation of effective factors on occupational stress in firefighting personnel. Health System Research. 2015;11(1):184–92.

58.

Faraji

, Valiee

, Mazidi

, Ramazanh

, Rezaee farahani

. Relationship between job characteristic and job stress in nurses of Kurdistan University of Medical Sciences educational hospitals. IJNR. 2012;7(25):54–63.

59.

Fatemeh

, Azizallah

, Roya

, Mostafa

, Maryam

, Samira

. Occupational Stress and its Determinants among Nursing Staff of Hospitals Affiliated to Zanjan University of Medical Sciences in 2011. Health & Development Journal. 2013;2(1):74–0.