Screening and selection process of included studies
During bibliographic data base search, a total of 23,589 studies were identified of which 21,930 were duplicates and removed. Of 1659 remaining studies, 1510 were found to be unrelated and excluded based on title and abstract screening. 97 articles were not eligible based on eligibility criteria and 52 full text articles were downloaded for further evaluation against eligibility criteria. 4 full text articles excluded as they missed prevalence of overall job satisfaction data. Two full text articles found unrelated for this systematic review. One full text article was out of Ethiopia and one full text article was out of recommended date and excluded. Finally we ended up with 40 full text articles[41-80] for evidence synthesis (Fig 1).
Characteristics of included studies
Among 40 included studies published from 2013-2022 G.C in Ethiopia, the smallest sample size was 60 on a study done in Tigray regional state and the largest one was 575 on a study done in Amhara regional state. The study settings included were health posts, health centers, hospitals, community pharmacies, private hospitals and clinics. These studies involved a total of 10,749 participants from health extension workers, nurses, midwives, pharmacy professionals, anesthetists and doctors (Table 2).
Methodological quality of included studies
Methodological quality assessment was done by two independent assessors, GDA and YKG using the JBI-Prevalence Critical Appraisal Checklist. Discrepancies between assessors were resolved through consensus and mutual understanding. The assessment results revealed that quality of included studies range from 77.8% to 100% (Annex 2).
Findings of the review
Based on 40 included studies with 10,749 participants, the pooled satisfaction level of health workers at Ethiopian health care setting was found to be 49.41% (95%, CI: 48.47- 50.36) for fixed effect model and 50.31%(95%, CI: 45.59- 55.01) for Random effect model at Q= Q=957.8331 ( I2 (inconsistency)= 95.93%, P < 0.0001), see Table 3. Only the result from random effect model was used for further discussion as there was significant heterogeneity between studies (Fig 2).
Publication bias analysis of included studies
The Egger’s test and symmetric funnel plot showed that studies included in this meta-analysis have no publication bias (Table 4 and Fig 3).
Sub group analysis
To investigate source of heterogeneity, sub group analysis has been done on sample size, study area, participant’s profession, year of publication and health facility type. All the above listed variables were significant source of statistical heterogeneity of which health worker category was the highest source of heterogeneity (I2=99.48%, Table 5).
Factors Associated with Health Worker’s Job Satisfaction at Health Care Settings in Ethiopia
Among 40 included studies used for pooled satisfaction level analysis, fifteen were assessed for factors associated with job satisfaction. Factors assessed were ; having additional responsibility, job autonomy, job description, coworker relationship, further educational opportunity, management support, nature of work, on job training, organizational policy, performance appraisal, reward/recognition, satisfied with monthly salary, supervision, work achievement, work environment, working equipment availability and workload. Of which autonomy (POR = 5.79, 95% CI: 1.99 - 16.90), on job training (POR=3.09, 95% CI: 1.69 - 5.67), organizational policy (POR=4.71, 95% CI: 2.09 - 10.61), reward/recognition (POR=4.58, 95% CI: 1.51 - 13.84), satisfied with monthly salary (POR=3.89, 95% CI: 1.77 - 8.54), adequate supervision (OR=5.34, 95% CI: 3.72 - 7.67) and work environment (POR=5.44, 95% CI: 2.80 - 10.58) were found to be significantly associated factors for overall staff satisfaction of health workers at health care setting in Ethiopia (Table 6).