Data for the prevalence of nurses׳ burnout in Iran (a meta-analysis dataset)

The present dataset was carried out using meta-analysis method towards investigation of the prevalence of nurses׳ burnout in Iran. To this end, the keywords were searched in the Iranian databases such as Medlib, SID, Iranmedex, Magiran or even some international databases such as Cochrane, Science-Direct, Scopus, PubMed, and Google Scholar. The data were analysed using the STATA Software Version 12. In ten articles with a sample size of 1758 subjects, an average age of 30.73 (54%) and the confidence interval of 43–64, the prevalence of burnout was reported. The obtained data indicated that Fars and Zanjan Provinces had the highest and lowest rates of burnout (72% and 26%, respectively). According to the acquired data, the total prevalence of burnout among men and women measured 46% and 65%, respectively. Given the high prevalence of burnout among the Iranian nurses in this dataset and the importance of nursing in public health which requires highly motivated and committed nurses with high job satisfaction, it is recommended that the intensity of burnout be reduced through supervising the nurses׳ professional performance, supporting, paying attention to their problems, following up and providing the necessary strategies to improve their environmental, economic, and personal conditions.


a b s t r a c t
The present dataset was carried out using meta-analysis method towards investigation of the prevalence of nurses' burnout in Iran. To this end, the keywords were searched in the Iranian databases such as Medlib, SID, Iranmedex, Magiran or even some international databases such as Cochrane, Science-Direct, Scopus, PubMed, and Google Scholar. The data were analysed using the STATA Software Version 12. In ten articles with a sample size of 1758 subjects, an average age of 30.73 (54%) and the confidence interval of 43-64, the prevalence of burnout was reported. The obtained data indicated that Fars and Zanjan Provinces had the highest and lowest rates of burnout (72% and 26%, respectively). According to the acquired data, the total prevalence of burnout among men and women measured 46% and 65%, respectively. Given the high prevalence of burnout among the Iranian nurses in this dataset and the importance of nursing in public health which requires highly motivated and committed nurses with high job

Value of the data
The obtained data of this dataset can be employed in further studies to investigate the prevalence of burnout among nurses in Iran.
The data of the present dataset was related to prevalence of nurses' burnout in throughout Iran.
Therefore, the data obtained are more valuable than studies that are done individually.
This dataset can be useful for provide various strategies to reduce the prevalence of nurses' burnout in Iran.
The method of data collection in this study is meta-analysis. Therefore, this methodology can be useful for future similar studies.

Data
Based on the inclusion criteria, out of 145 studies using the Maslach questionnaire to assess the prevalence of burnout, 10 cross-sectional studies were entered into the meta-analysis process [3][4][5][6][7][8][9][10][11][12] (Fig. 1). The sample size included 1758 subjects with an average of 175 subjects per study. In Table 1, the specifications of the selected studies are presented.
The prevalence of burnout based on the database and geographical regions are showed in Figs. 2  and 3, respectively. The findings demonstrated that the overall prevalence of burnout measured 54% (95% CI: 43-64). Based on subgroup analysis, the highest and lowest rates of burnout were reported in the area 2 (72%) and 3 (43%), respectively (Fig. 3).  Table 1 The Specifications of the articles in a systematic review and meta-analysis of prevalence of burnout among the Iranian nurses.

Author
Year   Based on the results of the meta-regression test, although the frequency of burnout increased in line with the years of conducting the studies, sample size and mean of age of subjects, this growing trend were not statistically significant ( Figs. 4 and 5). The obtained data of funnel plot indicated there was no publication bias in the present dataset.
The univariate meta-regression data for the prevalence of burnout, the years of conducting the studies, sample size and mean of age of nurses in Iran are showed in Table 2. Additionally, the funnel plot of the investigated studies and the overall prevalence of burnout in different geographical regions are presented in Figs. 6 and 7, respectively.

Search strategy
In this data article, the prevalence of burnout in Iranian nurses was reviewed based on the published studies without time limitations until December 2016. To this end, the keywords were searched in the Iranian databases such as Medlib, SID, Iranmedex, Magiran or even some international databases such as Cochrane, Science-Direct, Scopus, PubMed, and Google Scholar. The sources of related articles were also reviewed for access to other articles.

Inclusion and exclusion criteria
All articles addressing the prevalence of burnout in nursing staff using the Maslach Questionnaire were collected [2][3][4][5][6][7][8][9][10][11]. The studies were selected based on inclusion and exclusion criteria. The exclusion criteria included: non-relevant studies (reviews, editorials, non-research letters), studies with non-random sampling method, case reports, interventional studies, insufficient data, duplicate publications, not using the Maslach Questionnaire to assess the prevalence of burnout, the prevalence of burnout in other healthcare groups, and lack of access to the full text of studies.

Data extraction
To reduce the bias, the search of articles was independently done by two researchers, and in the event of disagreement, the study was judged by another expert in meta-analysis (DS). Then, the required information such as the title of the article, the first author, year of publication, prevalence of burnout, place of study, total sample size, sample size by gender, mean age of participants, geographical regions and province of studies were collected from the selected articles, and the prevalence of burnout was recorded in a form, too. The articles' screening and selection process was conducted according to the PRISMA Guidelines [12].

Data analysis
The point estimation and a confidence interval of 95% were calculated for the prevalence of burnout in each study using the Der Simonian and Laird's random effects model. Moreover, the Cochran Q test and I 2 index were used to investigate the heterogeneity between studies Table 2 The univariate meta-regression results for the prevalence of burnout, the years of conducting the studies, sample size and mean of age of nurses in Iran.

Variable
Coefficient SE t Confidence interval p  To evaluate the small effects of the study and potential population bias, a funnel plot was used based on the Egger Regression Test. In addition, to study the relationship between the prevalence of burnout and each of the years of conducting the studies and the sample size, a meta-regression analysis was used. Further, the subgroups analysis was applied to estimate the prevalence of burnout in each geographical region. As for data analysis, the STATA Software Version 12.0 was employed (Stata Corp, College Station, and TX).