Nursing professionalism and associated factors in Ethiopia: a systematic review and meta-analysis

Systematic review and meta-analysis.

The included primary studies were conducted among nurses working in health institutions of Ethiopia. The Ethiopian nursing association, which is a legally recognized professional nursing association, represents more than 68,961 nurses working in the country [35]. The number of nurses and midwives was 0.8 for 1000 people, based on the World Bank report of December 2024.

This systematic review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA, 2020) guidelines [36] (S1 File). Before database searching, we have assessed the prospective register of systematic reviews (PROSPERO) to check for duplicated work, and there was no registered review with our title. Then, the PubMed, Science Direct, HINARI, African Journals Online (AJOL), online universities repository in Ethiopia, and Google Scholar databases were accessed from 15/10/2024–30/10/2024. The references of the articles retrieved were also searched to avoid missing any article from different sources. Boolean operators AND/OR combine the keywords nursing, professionalism, associated factors, and Ethiopia. Synonymous and medical subject heading (MeSH) terms were also used to search the databases (S2 File).

The research questions were constructed using condition, outcome, context, population (CoCOPop) strategies.

The prevalence of higher level of nursing professionalism and associated factors.

Ethiopia.

The primary outcome was the pooled prevalence of a higher level of nursing professionalism, and the secondary outcomes were factors associated with a higher level of nursing professionalism. A higher level of nursing professionalism was determined on the basis of those nurses who scored above the mean score of the professionalism questions for binary outcome studies, and those nurses who scored high among the tertile ranked studies categorized as low, medium and high.

Nurses.

We included cross-sectional studies conducted on the prevalence of nursing professionalism or associated factors in Ethiopia. The articles included were either published or unpublished without time bounds, but were restricted to the English language.

We excluded studies conducted with nursing students about professionalism because nursing students are not licensed in the nursing profession. We also excluded articles that were not fully accessible and published in language other than English. Additionally, studies that lacked our outcome measures on the prevalence of nursing professionalism and associated factors in Ethiopia were excluded.

After database searching, we exported the data to the EndNote 7 citation manager and removed duplicates. Titles and abstracts were independently screened by two reviewers (MT and WC). Discrepancies among the two reviewers were subsequently resolved by consulting a third reviewer (AM) and a thorough revision and discussion among the team members to achieve a consensus. We used the Newcastle–Ottawa scale (NOS) quality assessment checklist for cross-sectional studies to assess the quality of the included articles [37]. We included articles scoring 7 or more out of the 10 assessment criteria. Reviewers MT and YA critically appraised each article independently, and any discrepancy noted was resolved with another third reviewer (WC) (S3 File).

We used a Microsoft Excel 2010 spreadsheet to extract the data (S4 files). The first author’s last name, study year, sample size, study region, odds ratio (OR) with 95% confidence interval (CI), frequency of highest level of nursing professionalism, response rate, and NOS scale were extracted (Table 1). Two reviewers (MT and WC) extracted the data independently, crosschecked for any discrepancy, and then, resolved the data with discussion [38].

We used STATA 17 statistical software for data analysis. The heterogeneity of the included articles was assessed via the I-squared (I²) statistic. We used the fixed effects model when the I² values were less than 50%, and the random effects model was used when the I² values were greater than 50%. Additionally, meta-regression and the Galbraith plot were also executed. Furthermore, Egger’s-based test and funnel plots regression were assessed to determine publication bias among the included studies. We performed a sensitivity analysis to examine the effect of a single study on the overall pooled result. We also conducted subgroup analysis by sample size and study period to identify their effects on heterogeneity [39].

Overall, 3008 articles were retrieved from online databases. After being exported to the EndNote 7 citation manager, 1342 duplicates were removed, and 1198 were excluded after title and abstract screening. A total of 455 full articles were assessed for eligibility; 456 were excluded because of outcomes of interest. Finally, 12 articles were included in this systematic review and meta-analysis [23‐34] (Fig. 1).

Twelve cross-sectional studies with a total of 3710 nurses were involved in this review. Among them, four were unpublished studies [27, 31, 32, 34]. Four studies were from the Amhara region [23, 25, 27, 28], two studies were from Addis Ababa [29, 34], three studies were from the Oromia region [30, 32, 33], one study was from the Tigray region [26], one study was from the Sidama region [24], and one study was from the central Ethiopian region [31]. The smallest and largest sample sizes were 103 [28] and 407 [23], respectively (Table 1).

The pooled prevalence of higher levels of nursing professionalism according to the random effects model was 43% (CI: 32%, 53%), I²=97.92%, p<0.001 (Fig. 2).

The funnel plot results revealed an asymmetric distribution, but Egger’s based regression analysis revealed a p-value of 0.245, which indicated lack of publication bias among studies. Furthermore, the Galbraith plot indicated that all studies had CIs of 95%, which confirmed the absence of any outliers (Fig. 3). We also computed a meta-regression analysis considering sample size and study year as covariates which could identify heterogeneity sources. However, there was no significant influence of heterogeneity among these covariates (Table 2).

We also conducted subgroup analysis on the basis of the study period before and after 2019, by a sample size of less than 350 and greater than or equal to 350 and by study region. Hence, the pooled prevalence of higher levels of nursing professionalism did not significantly differ by study period and sample size. By region, it was greater than the pooled result in Addis Ababa, Amhara and Oromia regions, but it was lower than the pooled result in Tigray, Sidama and Central Ethiopian regions (Table 3).

We also conducted a sensitivity analysis by removing one study at a time to examine the influence of each study on the pooled result, but there was no single study influence on the overall effect size (Fig. 4).

The factors associated with higher levels of nursing professionalism were education, payments, work experience, self-image, satisfaction and training (Table 4).

Nurses with BSc and above educational status had a higher level of nursing professionalism than diploma nurses did (POR: 1.80, CI: 1.38, 2.33), I²=0%, p=0.499. Eight primary studies with 2487 nurses were included in this category of factor analysis.

Additionally, nurses who learned from government colleges had a higher level of nursing professionalism than those who learned from private college did (POR: 2.14, CI: 1.34, 3.42), I²=0%, p=0.343. Two primary studies with 454 nurses were included in this category of analysis.

Nurses with long year of work experience had a higher level of nursing professionalism than did nurses with short periods of work experience (POR: 2.15, CI: 1.73, 2.68), I²=25%, p=0.259. Four primary studies with 1373 nurses were included in this category of analysis.

Nurses who received better payments had a higher level of professionalism than those who received smaller payments did (POR: 1.85, CI: 1.16, 2.98), I²=13.2%, p=0.33. Six primary studies with 1732 nurses were included in this factor analysis.

Nurses who had positive self-images also had higher levels of professionalism than those who had negative self-images did (POR: 3.85, CI: 2.17, 6.84), I²=20.1%, p=0.263. Two primary studies with 421 nurses were included in this category of analysis.

Satisfied nurses had a higher level of nursing professionalism than unsatisfied nurses did (POR: 2.42, CI: 1.49, 3.95), I²=64.9%, p=0.058. Three primary studies with 644 nurses were included in this category of analysis.

Finally, nurses who received training had a higher level of nursing professionalism than those who did not receive training did (POR: 2.88, CI: 1.14, 7.32), I²=82.4%, p=0.017. Two primary studies with 327 nurses were included in this category of analysis.

The included studies did not represent all regions of the country, which may result in a lack of country representativeness. Although we used the random effects model, there was high heterogeneity across the studies included. Only English-language literature was included, which could be another limitation of this review. Furthermore, none of the included articles demonstrated a cause-effect association because they were all cross-sectional in design.

This meta-analysis can enhance nursing care provisions, improve patient outcomes, inform policies, and foster evidence-based nursing practices through enhancing nursing professionalism in Ethiopia. In this review, health institutions can develop and implement strategies to improve the level of nursing professionalism. It can also contribute to the global body of knowledge and practices related to the level of nursing professionalism and associated factors. This review highlights the gaps that need to be filled by nurse managers to enhance their employees’ sense of professionalism by resolving the factors that reduce the level of nursing professionalism. It can also indicate the area of research, especially the need to explore different factors in a qualitative study design. This meta-analysis also highlights the need to evaluate and update the nursing curriculum and books to include nursing professionalism concepts. Hence, it can contribute to educational reform by indicating the need for including broader concepts of nursing professionalism in nursing education and curriculum. Additionally, policymakers may also benefit from the findings in designing different nursing programs around the country.

The pooled prevalence of a higher level of nursing professionalism was low, since less than 50% of nurses had a higher level of nursing professionalism. The educational status and the college they attend, payment, work experience, self-image, satisfaction, and training are factors that determine the level of professionalism. These factors can be modified to increase the level of nursing professionalism in Ethiopia.

Nursing managers should facilitate better payment and satisfaction mechanisms and retain more experienced nurses in their organization for a higher level of nursing professionalism. They should also facilitate educational and training opportunities for nurses to achieve a higher level of professionalism. The ministry of health should monitor private colleges regularly for their focus on the concept of nursing professionalism for the students. The Ethiopian nursing association should work to enhance the level of nursing professionalism by addressing the factors that affect it. Researchers need to focus on the qualitative inquiry of nursing professionalism to investigate additional factors that affect it.