Health professional’s job satisfaction and its determinants in Ethiopia: a systematic review and meta-analysis

Background Health professional’s job satisfaction is directly related to patient satisfaction and quality of care. Without satisfied health professionals the health system is not functional, and the national and global health related plans are not achieved. However, little is known on the level of health professional’s job satisfaction in sub Saharan African countries including Ethiopia. In addition, in Ethiopia there is no summarized evidence helped us an input to design strategies. Therefore, we aimed to assess the pooled prevalence of health professional’s job satisfaction and its determinants in Ethiopia. Methods Articles were searched from PubMed, PsycINFO, Hinari, Science Direct, web of science and African journal of online (AJOL) databases, Google and Google scholar. A standardized Microsoft excel spread sheet and STATA software version 16 were used for data extraction and analysis respectively. We followed the Preferred Reporting Items for Systematic reviews and Meta-Analysis to write this report. A random effect meta-analysis model was used to determine the pooled prevalence of job satisfaction. I2 was done to check heterogeneity. Egger’s test and funnel plot were conducted to detect publication bias. Subgroup analysis was also conducted. Association was expressed through pooled odd ratio with a 95% CI. Result In this review and meta-analysis, a total of 29 studies were included. The pooled prevalence of health professional’s job satisfaction was 46.17% [95% CI (43.08, 49.26)]. The heterogeneity and publication bias test results were I2 = 87.3%, P <  0.001 and Eggers’, P = 0.16. Female sex; OR: 2.20 [95% CI (1.63, 2.97)], working environment; OR: 9.50 [95% CI (6.25, 14.44)], opportunity for professional growth and development; OR: 5.53 [95% CI (1.56, 19.56)], staff relationship; OR: 3.89 [95% CI (1.65, 9.17)] and supportive supervision; OR: 5.32 [95% CI (1.77, 15.92)] were associated with health professional’s job satisfaction. Conclusion More than half of professionals were dissatisfied with their jobs. Therefore, the ministry of health and stakeholders better to design strategies to increase the level of satisfaction. Furthermore, it is better to strengthen staff relationship and making the working environment more attractive and equipped. Supplementary Information The online version contains supplementary material available at 10.1186/s13690-021-00664-7.


Background
Job satisfaction is a positive, favourable attitude and feeling what people have about their job [1,2]. Health professional's job satisfaction is significantly related with quality of care and patient satisfaction [3]. To achieve the Sustainable Development Goals (SDGs) strategy was planned on human resources by World Health Organization (WHO) [4]. Furthermore, motivation of the health care providers has a pivotal role for achievement of this strategy and it depends on many factors. Job satisfaction is one of the most important factors [5].
Shortage of skilled health professionals is a global burden and it is high in most African countries including Ethiopia [20]. According to WHO report by the year 2035 the health workforce shortage will reach up to 12.9 million; 47 and 25% of the shortfall will be in South-East Asia and Africa [21]. To prevent this challenge the health profession's intention to stay should be increased by increasing their job satisfaction level [22]. Therefore, determining the level of job satisfaction has paramount significance. Studies conducted in Africa on job satisfaction showed a prevalence that ranges from 17.4 [23] to 82.6% [24].
In Ethiopia, there is no representative and summarized data on health professional's job satisfaction. The conducted single studies showed inconsistent prevalence that ranges from 31.7 [33] to 74.4% [34]. Besides, there are some contradicting findings on factors associated with job satisfaction. Therefore, this systematic review and meta-analysis was aimed to estimate the pooled prevalence of health professional's job satisfaction and its determinants in Ethiopia.

Searching strategy
We followed the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines to write this review and meta-analysis [35]. The presence of similar reviews was checked using this link https:// www2.le.ac.uk/library/find/databases/p/Prospero. Primary articles were searched from PubMed, PsycINFO, Hinari, Science direct, web of science and African journal of online (AJOL) databases. Moreover, grey literatures were retrieved from Google and Google scholar. The reference list of published articles was searched to recognize other relevant articles that didn't showed in databases. Searching was restricted to studies conducted from 2010 until 2020 on humans and full English version articles. Probing of primary articles was started on September 20, 2020 and ended on October 5, 2020. We used "prevalence OR magnitude AND health care professionals OR health care providers OR health workers AND Ethiopia" for objective one and "Determinants OR factors OR predictors AND health care professionals OR health care providers OR health workers AND Ethiopia for the second objective as keywords for searching. Published and unpublished articles were searched and included in this systematic review and meta-analysis. Endnote version X6 was used to manage citations and to check duplication of articles.

Eligibility criteria
BG and JN independently executed eligibility assessment in an unblinded identical manner based on the stated inclusion and exclusion criteria. We solved disagreements by consensus and invitation of the remaining two authors.

Inclusion criteria
Observational studies conducted between 2010 and 2020 in Ethiopia among health professionals and published in English language were included. Furthermore, articles reporting the level of health professional's job satisfaction in proportion were included. Published and unpublished full articles reported the prevalence and/ or associated factor were considered.

Exclusion criteria
Studies qualitative in design were excluded.

Outcome measures
In this systematic review and meta-analysis two objectives were assessed. The first was to estimate the pooled prevalence of job satisfaction among health professionals in Ethiopia and it was estimated by dividing the number of satisfied health care professionals to the total number of health care professionals included in this review and meta-analysis, and multiplied by 100. The second objective was to determine the pooled effects of factors on health professional's job satisfaction in Ethiopia. In this systematic review and meta-analysis, variables identified as a factor for job satisfaction in at least three studies were considered. We used odds ratio (OR) to express the pooled effect.

Quality assessment and data extraction
Newcastle Ottawa Scale was used to assess the quality of studies [36]. BG and JN have appraised the studies independently using the above tool. The following items were included to evaluate the methodological quality, comparability, and the ascertainment of the outcome. The tool encompassed 10 criteria's for rating different quality elements and studies scored 6 and above out of 10 were included in to this review and meta-analysis. During the quality assessment any disagreements were solved through discussion and before agreement reached weighted kappa index was done, which was 0.76 [37].
BG and JN extracted all the necessary data independently using a standardized Microsoft Excel. Two data extraction formats, one for each objective were used. The first data extraction format that was prepared for objective one (prevalence of job satisfaction) comprised author name, publication year, region the study conducted, study design, sample size, response rate, profession and prevalence of job satisfaction. To extract data for objective two (factors for job satisfaction) we used two by two tables. Any disagreements during the data extraction time between the two authors (BG and JN) were resolved through discussion, two fold checking the varying data together, and through invitation of either of the remaining two authors.

Publication bias and heterogeneity
To check publication bias funnel plots [38] and Egger's statistical test [39] were conducted. We used a p-value < 0.05 to declare the statistical significance of publication bias. After a comprehensive examination of the included studies, I 2 test was conducted to assess the heterogeneity and declared as low, moderate, and high heterogeneity if it is < 50, 50-75%, and > 75% respectively [40].

Statistical method and analysis
All extracted data from each study using a Microsoft excel spread sheet were exported to STATA version 16 software for analysis. Binomial distribution formula was used to calculate the standard error of prevalence for each original article. Due to high heterogeneity between the included studies we used a random-effect model for analysis [41]. Subgroup analysis based on profession type and region was conducted to check the source of heterogeneity [42][43][44]. Using separate groups of meta-analysis the effect of the selected associated factors on the outcome variable was examined. We used texts, tables, forest plots, and OR with 95% confidence intervals (CI) to describe the features of the included articles and to display the finding of this review and meta-analysis.

Study search and selection
By fixing the searching scope on full-text articles, human studies, published in English language between 2010 & 2020, we found a total of 2280 primary articles from PubMed HINARI, web of science, PsycInfo, AJOL and science direct databases. Furthermore, we also found from Google scholar and Google. Of which, a total of 1039 and 1204 articles were excluded due to duplication and unrelated to our study that was screened by title and abstract respectively. In this systematic review and meta-analysis, only 37 articles were selected for full reading and 8 were excluded; 4 due to the overall job satisfaction level was not reported [45][46][47][48], 1 due to poor quality, since very small sample size used [34], 2 articles due to the outcome variable was not reported with proportion [49,50] and 1 due to it was qualitative study [51]. Lastly, a total of 29 articles that fulfill all the inclusion criteria's were selected for meta-analysis [16,33, (Fig. 1).

Meta-analysis
As shown on Fig. 2, forest plot was conducted to show the result of the included studies. Twenty nine studies were included in this systematic review and metaanalysis to estimate the pooled prevalence of health care professional's job satisfaction. The heterogeneity of the included studies was I 2 = 87.3% with p < 0.01. Due to this heterogeneity, we used the random-effect model to estimate the pooled prevalence of health professional's job satisfaction and it was 46.17%; [95% CI (43.08, 49.26)]. Based on the subjective inspection of funnel plot, more than half of the studies were distributed within the plot (Fig. 3). In addition, Eggers test (P = 0.16) was also done, so we have no evidence for publication bias.
To detect the source of heterogeneity subgroup analysis was conducted based on profession and publication year. As shown in Table 2

Factors associated with health professional's job satisfaction
In the present review and meta-analysis, factors identified as significant for health professional's job satisfaction in at least three primary articles were included. As a result, salary [53,60,65,75], working experience [33,55,[67][68][69]76] and profession type [33,66,69] were identified as factors for job satisfaction in three and above studies, due to they were reported through various category methods that create a difficulty to extract data we exclude them from analysis after discussion and agreement by authors. Support from supervisor [ [59,69,76] were factors that fulfill the criteria and selected for meta-analysis. However, support from supervisor, working environment, staff relationship, sex and opportunity for professional growth were identified as significant determinants for health professional's job satisfaction.
In the current review and meta-analysis, health professionals who had good support from their supervisors were 5.3 times more likely to satisfy as compared to their contrary health care professionals (OR: 5.32 [95% CI (1.77, 15.92)]). Health professionals who were worked in an attractive or good environment had 9.5 times higher opportunity to satisfy in contrast to their counterparts (OR: 9.50 [95% CI (6.25, 14.44)]). Moreover, health professionals who were worked in an organization that provide opportunity for professional growth and development like training and education access had 5.5 times higher satisfaction level than those working in an organization that doesn't provide any opportunity for professional growth and development (OR: 5.53 [95% CI (1.56, 19.56)]) ( Table 3).
As shown below in Table 3, staff relationship and sex were another factor identified for health professional's job satisfaction. Health professionals who were worked in an organization that had good relationship between staffs and female in sex had 3

Discussion
Job dissatisfaction worsen patient-health professionals ratio and staff turnover [22]. Job satisfaction is strongly associated with workers depression and anxiety [6]. The aim of this review and meta-analysis was to assess the pooled prevalence of health professional's job satisfaction and its determinants in Ethiopia. The pooled prevalence of health professional's job satisfaction was 46.17% [95% CI (43.08, 49.26)]. We used random effect model for pooled analysis of included studies, since heterogeneity was 87.3% and the Egger test value was 0.06 before excluding one single study done in Harare region with very small sample size (43) [34] and after excluding this single study the value of Egger test was 0.16. While inspecting funnel plot more than half of the articles were within the plot. Support from supervisor, working environment, staff relationship, sex and opportunity for professional growth were identified as significant determinants for health professional's job satisfaction.
Health professionals who had good support from their supervisors were more likely satisfied. This finding was supported by study done in Pakistan [84] and studies conducted in three African countries [86,87]. This might be due to the fact that good support from supervisors increases professional's motivation that leads to satisfaction. In addition it might be due to the fact that supportive supervision increase staff job satisfaction level [88].
Working environment of the employees was significantly associated with job satisfaction. Health professionals who were working in an attractive and good area had 9.5 times more likely satisfied as compared to their counterparts. This finding was supported by qualitative study conducted in Pakistan [89]. This might be the fact that good and attractive working environment have positive psychological and physical influence on employees and motivate them.
In the current review and meta-analysis, health professionals who were working in an organization that had good staff relationship were more satisfied. This finding was supported by studies done in India [90] and Bangladesh [91]. This similarity might be due to good staff relationship create good communication between staffs that leads to good working condition and satisfaction.
Employees who were working in an organization which provide an opportunity for professional growth and development had 5.5 times more odds of satisfaction. This finding was supported by studies conducted in South Africa [92] and Nepal [26]. This might be due to the fact that such kind of opportunity increase professional's income, level of knowledge and motivation that lead to satisfaction with their job. Moreover, it might be due to the fact that professional development have a direct positive effect on job satisfaction [93].
The last determinant identified in this review and meta-analysis was sex. Female health professionals had more satisfied as compared to their counterparts. This finding was supported by studies done in China [3] and USA [94]. This might be because of the fact that female  doesn't express their dissatisfaction openly rather internalizing the feeling of misery [95].
The first limitation of this study might be difficulty of establishing temporal relationship between the outcome and independent variables since all the included articles were cross sectional studies. Second, the presence of heterogeneity might be the indicative for the existence of underling effect among primary studies. Third, searching is limited to Ethiopia as result the evidence might not be generalized to other countries in the world.

Conclusion
In this review and meta-analysis, more than half of health professionals were dissatisfied with their job. Good support from supervisor, working environment, staff relationship, female sex and availability of opportunities for professional growth and development were associated with health professional's job satisfaction. In order to increase the level of job satisfaction the federal ministry of health and stakeholders better to develop a strategy and strength staff relationship, support from supervisors, professional growth opportunities and make the working environment more attractive.