Self-reported turnover intention and associated factors among health professionals in Kafa Zone, Southwest Ethiopia

Abstract

Objective:

The aim of this study was to assess turnover intention and its associated factors among health professionals in Kafa Zone, Southwest, Ethiopia, 2018.

Methods:

Cross-sectional study design relied on institution that was employed from 1–30 May 2019. Questionnaires were distributed to 427 participants who were selected by simple random sampling technique from selected health facilities. Epi Data Statistical software version 3.1 for data entry and Statistical Package for Social Sciences software version 21.0 for data analysis were used. Binary logistic regression analysis was used to identify factors associated with turnover intention. Odds ratios with 95% confidence intervals were used to examine associations, and a p-value less than 0.05 was considered significant.

Result:

Overall, 427 questionnaires were distributed, and 389 were returned making 91.1% response rate. Majority of the respondents were between the age of 21 and 30 years and more than half were male. More than half, 219 (56.3%), of the participants reported that they intended to leave their institution. Being female (adjusted odds ratio 2.65, 95% confidence interval (1.62–4.33)), married (adjusted odds ratio 0.57, 95% confidence interval (0.34–0.96)), degree holders (adjusted odds ratio 0.55, 95% confidence interval (0.34–0.89)), autonomous (adjusted odds ratio 0.51, 95% confidence interval (0.31–0.84) and not satisfied with living place condition (adjusted odds ratio 1.68, 95% confidence interval (1.0–2.83)) were found to have statistically significant association with intention to leave among health professionals in Kafa Zone.

Conclusion:

Turnover intention of health professionals was high in study area. Sex, educational status, marital status, autonomous, and living place condition were the identified predictors of turnover intention among health professionals. The health managers and stakeholders at different levels should have a discussion session with staff to cut the intention to leave the organization.

Introduction

Turnover is a leaving of an employee from an organization voluntarily. It occurs as a result of working conditions, organizational and psychological factors that can affect workers’ attitudes in and toward the institution. ¹ Since health industries are dynamic, needs of the staffing become changing. By considering the changing environment, retaining the present workers and recruiting new staff must meet these changing needs. ² For employee, turnover is unpredictable; proper human resource strategies are important to reduce the attrition rate that can make the organization advantageous. ³

Turnover is a result of different issues that make the employee to leave his or her institution. The reasons why staff may depart from the organizations are the wishes for job change, better chances, high salary, and so on. Since high-staff turnover rates are a bad sign for any organization, it is time to identify the source of problem when there is more resignations noticed than usual. ⁴

High staff turnover rate is challenging for the management of the organization. Thus, it is important to identify turnover intents as early as possible to take action by HR planners. ⁵

Healthcare is an industry where professional service relies on the skills of each staff member. Searching and protecting this skill is a vital activity of an organization to become cost-effective company that provides special services. ⁶ Health institution is one of the most vital companies where there is high staff turnover and job stress. Dora Wang reported that according to a study by Nursing solution Inc. (NSI), the national turnover rate among health professionals in all hospitals is 16.5%, and the costs are high. ⁷ Although most of the health systems search to minimize staff turnover, many nosedive to aware its cause, consequences and have no appropriate strategies to improve it. About 81.8% of the institutions mention retention is a crucial strategy, but only 51.5% have proper plans to use it. Healthcare industries are put on third uppermost in turnover rates at 18.9% as compared to others. ⁶

Staff turnover most likely harms organization’s outcome, wasting time, salary and important resources. The Human Resource Management (HRM) Society found that an organization costs averagely 6–9 months of a worker’s salary to replace her or him. ⁸ The impact of turnover is not only affecting the satisfaction of a patient and outcomes of treatment but it is also a high cost. It costs the hospital in average US$300,000 annually by every percentage increment of nurse turnover. ⁹

Health professional’s turnover is a big challenge in all countries. Crisis of nursing profession is a big challenge in South Africa as they leave the country seeking for profitable career. ¹⁰ The studies also indicated that turnover intention of health professionals is high in Ethiopia: the study done in Gondar University Referral hospital showed 52.5%, in Jimma Zone health facility 59.4%, and the turnover intention of nurses in Ethiopia is 52.4% and in East Gojjam of Amhara Region is 59.4%.^11–14

To minimize turnover in healthcare, it is important to work with HR on talent acquisition process improvements as a first action to improve talent management processes. Recruiters and hiring leaders must evaluate the applicants for cultural fit by considering the required competencies for a specific role rather than looking only to fill the vacancy. Employing individuals who do not culturally fit and lack alignment with the company’s mission, vision and values negatively affects the employee engagement, morale and patient satisfaction. ⁹

Some factors of turnover intention among health professionals are type of profession, management style, salary, benefits, working environment, inadequate opportunitiesfor professional development and so on.^13,15 In addition to these, the reasons for departing from the company include incivility, lack of autonomy, unfair treatment, feeling psychologically unsafe at work, supervisory support and emotional exhaustion of job satisfaction. ¹⁶

Even though human resource management is an important issue in health institutions to deliver necessary services for the community, currently health professionals’ turnover is a critical problem. The status of turnover intention and its associated factors among health professionals is not known in the study area. Therefore, this study aimed turnover intention and its associated factors among health professionals that provide valuable information about it for local and national health institution managers and policymakers to design appropriate retention strategies.

Methods and materials

Study area and period

This study was done from 1–30 May 2019 in Kafa Zone. This Zone is one of the Ethiopian Southern Nations, Nationalities, and Peoples’ Region (SNNPR). It is bordered on the south by Debub Omo, on the southwest by Bench Maji, on the west by Sheka, on the north by the Oromia Region, and on the east by Konta Zone. The administrative center of the zone is Bonga town, which is 465 km far from the capital city, Addis Ababa. The zone has three hospitals and 45 health centers. In this zone, there are 861 health professionals of which 526 were female and 335 were male.

Study design

Institutional-based cross-sectional study design was employed.

Source population

All health professionals in Kafa Zone were the source population.

Study population

All sampled health professionals in the selected health facilities.

Inclusion and exclusion criteria

Health professionals in selected health institutions who were more than 6 months after employment were included and who were seriously ill, on maternal, and annual leave were excluded.

Sample size and sampling technique

Sample size determination

Single population proportion formula was used to determine the sample size based on the following assumptions

$$n=\frac{z\left(\frac{\alpha }{2}\right)2\times P\left(1-p\right)}{d^2}$$

where n is the desirable calculated sample size; Z (α/2) = 1.96 (95% confidence level for two side); P = is the proportion of expected prevalence 59.4%; ¹² d is the degree of accuracy desired setting at (5%).

Therefore, the value of n will be calculated as

$$n=\frac{{(\mathbf{1}.\mathbf{96})}^{\mathbf{2}}\times \mathbf{0}.\mathbf{594}\left(\mathbf{1}-\mathbf{0}.\mathbf{594}\right)}{{\left(\mathbf{0}.\mathbf{05}\right)}^{\mathbf{2}}}=\mathbf{371}$$

Considering 15% non-response rate, the final sample size becomes 427.

Sampling technique

The study was conducted in three hospitals and 20 other randomly selected health centers of Kafa zones. The number of participants was allocated to each facility proportionally. Finally, after taking the list of health professionals from each health facility and coded, respondents were selected by simple random sampling technique.

Study variables

Turnover intention was a dependent variable, whereas independent variables were as follows:

• Socio-demographic variables: age, sex, marital status, profession, educational status, family size, and family condition.

• Organizational factors: organizational commitment, salary, work environment, co-worker relationships, opportunity to develop, perceived organizational support, and pay satisfaction.

• Job factors: job satisfaction, workload, and autonomy

• Living place condition, workplace condition

Operational definition

• Turnover intention—The extent at which the health workers desire to leave the public health institution. It was measured by 5-point Likert-type scale: strongly disagree, disagree, neutral, agree and strongly agree and assessed using three items. It was measured as intended to leave and not intended to leave depending on the mean score.

• Organizational commitment—The Organizational Com-mitments Questionnaire (OCQ) consists of nine items. High organizational commitment if scored above mean value and low commitment if less than mean value

• Job satisfaction— The level of satisfaction employees feel about their work, which can affect performance. It was measured by six items. Employees are satisfied if they score greater than mean value and dissatisfied if they score less than mean value

• Opportunity to develop—The statement indicates that the employees expand their knowledge, skills and abilities and apply the competencies they have gained to new situations. The opportunity to gain new skills and experiences can increase employee’s motivation and job satisfaction and help workers more effectively to manage job stress. To measure these, three-item questions were used. Based on this, respondents were classified into good (have opportunity to develop) if greater than mean value and poor (no opportunity to develop) if less than the average mean value.

• Co-worker (staff) relationship—have good staff relation if the score is greater than average mean value and have poor staff relation if less than the average median value. It will be measured by seven items.

• Perceived organizational support—It is suggested that POS is related to employees’ subjective well-being. However, their organization values and cares about them that making the workplace more pleasant. It was measured by four items and classified as low and high based on the computed mean score

• Workload/pressure—workload present in the health facilities. It was measured by two items of work pressure. Have workload if less than the mean value and does not have workload if greater than the mean value.

• Autonomy—Job characteristics that enable health professionals to make individual decisions about daily practice and also the feeling of professionals about independence in the work. Autonomy was measured using four items with 5-point Likert-type scale in which 1 denotes very dissatisfied and 5 denotes very satisfied. Respondents considered autonomous and satisfied about it when they score above the computed mean for the subscale.

• Living place condition—It is measured by seven items and since all the questions were uniformly stated in negative form, the value was not reversed. It was categorized as satisfied if less than mean score and not satisfied if greater than mean value.

• Working Environment—It was measured by two items and both questions were uniformly stated in negative form. It was classified into satisfied and not satisfied depending on the mean value.

Data collection technique and tools

Data collection instruments

Pre-tested structured questionnaire was used to collect data. The self-reported questionnaire was adapted after reviewing different literatures.^{11,13,17–25} The tool consists of socio-demographic and factors of intention to leave questions, such as job factors, organizational factors, and living and working condition. Predictors of intention to leave were measured using 5-point Likert-type scale. The five alternatives for Likert-type scale type of questions were coded as strongly disagree = 1; disagree = 2; neutral = 3; agree = 4 and strongly agree = 5. The negative questions become reversed after entering the data except the questions uniformly stated negatively like living place condition and working environment variables. The Cronbach’s alpha was done to test the reliability of tools. Accordingly, scores were turnover intention (0.75), job satisfaction (0.805), pay satisfaction (0.83), burden of work (0.747), autonomy (0.786), organizational commitment (0.84), co-worker relationships (0.84), opportunity to develop (0.822), perceived organizational (0.86), living conditions (0.848) and working environment (0.7)

Data collection facilitators

In total, 10 clinical nurses who work in nearby health facility were recruited purposively to facilitate data collection and 5 supervisors who have MSc in health were recruited.

Data collection procedure

The list of health workers was taken from the HR department of each institution and coded; then, self-reported questionnaires were distributed to health professionals selected using simple random sampling technique.

Quality control measures

Pre-test was done on 5% of the total study subjects on non-selected health and based on findings, necessary amendments were made. Data collection facilitators were trained for 1 day on the study instruments and data collection procedure. The recruited supervisors closely supervised the activities of data collectors on a daily basis to make sure they are collecting data as per the procedure of data collection they have been trained before data collection. To insure the quality of data again, careful data entering and cleaning was done before the beginning of the analysis.

Statistical analysis

Data were entered by EPI data software version 3.1 and analyzed by Statistical Package for Social Sciences (SPSS) software version 21.0. Descriptive statistics were done for all variables that can answer the objectives of research. Variables with p-value of less than 0.25 in binary logistic regression analysis were entered into the multivariable logistic regression analysis to control confounds. Odds ratios with 95% confidence intervals were calculated to examine the associations and p-value < 0.05 was considered significant. Finally, the result was presented using tables, charts and in description form.

Ethical considerations

Letter of ethical clearance from the Institutional Review Board of Mizan–Tepi University and permission letter from concerned authorities were obtained. Written consent was taken before the study. Respondents were reassured about the confidentiality of their response. They were also ensured their voluntarily participation.

Result

Socio-demographic characteristics of the study participants

Overall, 427 questionnaires were distributed, and 389 were returned that makes 91.1% response rate. The reasons for non-response were refusal and heavy workloads. Majority of the respondents were between 21 and 30 years of age and more than half were male. Regarding the marital status of the respondents, around half, 193 (49.6%), of them were single and majority 274 (70.4%) were orthodox in religion. Almost half, 200 (51.4%), of the participants were diploma in educational status and 185 (47.6%) accounts for nurses followed by midwives 68 (17.5%) in their type of profession. Majority, 274 (70.4%), were from hospitals and live in urban 224 (57.6). More than half, 222 (57.1%), of the respondents live far away from their family and 321 (82.5%) had ⩽ 5 family size (Table 1).

Table 1.

Socio-demographic characteristics of the health professionals study in Kafa Zone, Southwest, Ethiopia, 2019.

Variable	Category	Frequency (n = 389)	%
Age (years)	18–20	10	2.6
	21–30	324	83.3
	31–40	45	11.6
	⩾41	10	2.6
Sex	Male	210	54.0
	Female	193	49.6
Marital status	Single	193	49.6
	Married	188	48.3
	Widowed	2	0.5
	Divorced	6	1.5
Residence	Urban	224	57.6
	Rural	165	42.4
Religion	Orthodox	274	70.4
	Muslim	32	8.2
	Protestant	78	20.1
	Other a	5	1.3
Educational status	Diploma	200	51.4
	Degree	181	46.5
	Master’s and above	8	2.1
Type of profession	MD	22	5.7
	Nurse	185	47.6
	Pharmacy	32	8.2
	HO	37	9.5
	Midwifery	68	17.5
	Medical laboratory	30	7.7
	Other b	15	3.9
Position	Staff	344	88.4
	Head/coordinator	45	11.6
Type of health facility	HC	115	29.6
	Hospital	274	70.4
Work experience (years)	⩽5	250	64.3
	5–10	110	28.3
	10–15	13	3.3
	>15	16	4.1
Family condition	Live with family	167	42.9
	Far from family	222	57.1
Family size category	⩽5	321	82.5
	>5	68	17.5

MD: medical doctor; HO: health officer; HC: health center.

^aOther = Catholic.

^bOther = Environmental health, Ophtha, and biomedical.

Turnover intention

Turnover intention of the respondents was assessed by three items and the mean score was computed. The three items used to measure turnover intention were as follows: I often think about leaving my present job, I will probably look for a new job in the next year, and as soon as possible, I will leave the organization. More than half, 219 (56.3%), 95% CI (51.2–61.2), of the respondents claimed that they intended to leave their current organization.

Factors related to job

About half, 199(51.2%), of the respondents claimed that they were satisfied with their job. Regarding the burden of work, only 143 (36.8%) of the participants were loaded with work and around half, 206(53%), of the participants were autonomous on their work. Moreover, 300 (77.1%) reported that there is work schedule in their institution and 274 (70.4%) satisfied with this work schedule (Table 2).

Table 2.

Factors related to the job of health professionals in Kafa Zone, 2019.

Variable	Category	Frequency (n = 389)	%
Job satisfaction	Not satisfied	190	48.8
	Satisfied	199	51.2
Workload	Loaded	143	36.8
	Not loaded	246	63.2
Autonomy	Not autonomous	183	47.0
	Autonomous	206	53.0
Working schedule	Yes	300	77.1
	No	88	22.6
Satisfaction with working schedule	Yes	274	70.4
	No	113	29.0

Organizational factors

About half, 202 (51.9%), of the participants had high organizational commitment and 227 (58.4%) reported that there is an opportunity to develop in their institution. Only half of the participants, 198(50.9%), satisfied with their payment and 182 (46.8%) perceived that there was organizational support. About half, 202(51.9%), of the participants reported that there was good staff relationship (Table 3).

Table 3.

Organizational factors of health professionals in Kafa Zone, 2019.

Variable	Category	Frequency (n = 389)	%
Organizational commitment	Low	189	48.6
	High	202	51.9
Opportunity to develop	Yes	227	58.4
	No	162	41.6
Perceived organizational support	Low	207	53.2
	High	182	46.8
Co-worker relationship	Poor	187	48.1
	Good	202	51.9
Pay satisfaction	Not satisfied	191	49.1
	Satisfied	198	50.9

Living and workplace condition

Only 175 (45%) of the participants satisfied with their living place condition and 129 (33.2%) workplace environment (Figure 1).

Figure 1.

Living and workplace condition of the health professionals in Kafa Zone, 2019.

Factors affecting intention to leave

In multivariable logistic regression analysis, factors contributing to intention to leave were identified; sex, marital status, educational status, autonomy and living place condition were found to have statistically significant association with intention to leave among health professionals in Kafa zone.

Female health professionals were three times more likely intend to leave their organization as compared to males (AOR 2.65, 95% CI (1.62–4.33)). Being married in marital status, health professionals was less likely intends to leave by 43% as compared to their counterpart (AOR 0.57, 95% CI (0.34–0.96)) and degree holders in educational status were less likely intend to leave by 45% as compared to diploma holders (AOR 0.55, 95% CI (0.34–0.89)).

Health professionals those were not autonomous were less likely intend to leave their organization by 49% as compared to their counterpart (AOR 0.51, 95% CI (0.31–0.84)). Health professionals those are not satisfied with their living place condition were two times more likely intend to leave as compared to their counterpart (AOR 1.68, 95% CI (1.0–2.83)) (Table 4).

Table 4.

Multivariable logistic regression analysis of intention to leave among health professionals in Kafa Zone, Southwestern Ethiopia, 2019.

Variable	Category	Intention to leave		COR (95% CI)	AOR (95% CI)
		No	Yes
Age (years)	<20	7	3	1	1
	21–30	140	184	3.1 (0.78–12.1)	2.43 (0.56–10.49)
	31–40	19	26	3.2 (0.73–13.97)	3.3 (0.63–17.26)
	⩾41	4	6	3.5 (0.55–22.3)	2.42 (0.23–25.55)
Sex	Male	106	104	1	1
	Female	64	115	1.83 (1.218–2.75)	2.65 (1.62–4.33)
Educational status	Diploma	73	127	1	1
	Degree	96	85	0.25 (0.03–2.06)	0.55 (0.34–0.89)
	Master’s and above	1	7	0.13 (0.01–1.05)	4.19 (0.48–36.6)
Marital status	Single	83	110	1	1
	Married	83	105	0.95 (0.64–1.43)	0.57 (0.34–0.96)
	Divorced and widowed	4	4	0.75 (0.18–3.11)	1.04 (0.22–4.96)
Work experience (years)	⩽5	115	135	1	1
	5–10	42	68	1.38 (0.87–2.18)	1.38 (0.79–2.42)
	10–15	7	6	0.73 (0.24–2.23)	0.75 (0.10–2.91)
	>15	6	10	1.42 (0.5–4.03)	2 (0.39–10.26)
Job satisfaction	Not satisfied	92	96	0.66 (0.44–1.0)	0.74 (0.45–1.21)
	Satisfied	78	123	1	1
Workload	Loaded	99	151	1.6 (1.05–2.42)	1.56 (0.96–2.55)
	Not loaded	71	68	1	1
Autonomy	Not autonomous	96	87	0.51 (0.34–0.76)	0.51 (0.31–0.84)
	Autonomous	74	132	1	1
Organizational commitment	Low	93	83	0.5 (0.34–0.76)	0.65 (0.38–1.13)
	High	77	136	1	1
Perceived organizational support	Low	96	103	0.68 (0.46–1.12)	1.15 (0.68–1.93)
	High	74	190	1	1
Co-worker relationship	Poor	96	91	0.55 (0.36–0.82)	0.9 (0.53–1.53)
	Good	74	128	1	1
Living place condition	Satisfied	90	85	1	1
	Not satisfied	80	134	1.774 (1.18–2.66)	1.68 (1.0–2.83)
Working place condition	Satisfied	54	75	1	1
	Not satisfied	116	144	1.16 (0.77–1.74)	0.8 (0.47–1.38)

COR: crude odds ratio; CI: confidence interval; AOR: adjusted odds ratio.

Discussion

This study was aimed to identify the magnitude of turnover intention and its associated factors among health professionals. The magnitude of turnover intention in this study was 219 (56.3%). This indicated that the significant number of health professionals intended to leave their organization makes health care industries difficult to give quality services for the community. This finding was somewhat similar with the studies conducted at the University of Gondar Referral Hospital (52.5%), ¹¹ Jimma Zone health institutions (59.4%), ¹² and East Gojjam of Amhara Region (59.4%). ¹⁴ However, it is lower than the study in Horro Guduru Wallaga Zone that was 65% ¹⁵ and in Jimma Zone 63%. ²² The difference might be because of the time. The finding of this study is greater than the study on nurses in Saudi Arabia that was 40% ²⁰ and Ethiopia, Sidama Zone which was 50%. ²³ The discrepancy is because of the difference in socioeconomy and developmental status of the town or access of the facilities in the area.

In this study, some associated factors were identified: among socio-demography, sex, educational status and marital status were associated to turnover intention. Females are more likely to leave their institution than males. This is inconsistent to the study done in Saudi Arabia. ²⁰ The Saudi female nurses have the opportunity to work in their living areas than male nurses that is why they not intend to leave their organization. The study done in Ethiopia, Sidama Zone indicated that female nurses are 47.4% less likely to be satisfied with their current job aspect. ²³ This study also shows that only 47.5% female health professionals satisfied with their work. Marital status is the other predictor of turnover intention. Married health professionals were less likely intended to leave their current institution by 43% than single individuals. This is consistent with the study done on nurses in Saudi Arabia and Sidama Zone of Ethiopia.^20,23 It is known that it is not easy for married health professionals to search for other opportunities of work like single individuals do. Single health professionals have less family responsibilities so they do not have to consider moving family members when transferring to another organization. ²⁰

Educational status is the other predictor of turnover intention in this study. Researchers show that as the educational status increases the intention to leave the current organization increases.^11,13,20 In contrary to this, the finding of this study indicated that the degree holders health professionals were less likely intend to leave the current institution by 45% than the diploma holders. The possible justification for this finding might be due to the fact that the degree holders are encouraged by their salary, position and training than the diploma holders.

The finding of this study indicates that non-autonomous health professionals less likely intended to leave their current organization. The possible explanation for this might be due to the fact that the autonomous individuals can easily make decisions in their life to search for other opportunities as they can do at their working institution.

Living place condition was other identified predictor of turnover intention in this study. Living condition (such as lack of transportation, poor access of water and electric at home, high cost, lack of housing, shortage of telephone services) was studied in different literatures in different ways.13–15 For instance, the study done in Amhara region ¹⁴ showed that lack of transportation is significantly associated to turnover intention. The study conducted in Horro Guduru ¹⁵ also showed working and living condition is a predictor for turnover intention among health professionals. This is not amazing to live in one area, it is important to get the access of some basic needs, such as water, electric, telephone, house, and so on.

As strength, this study was conducted on sufficient sample size and includes both hospitals and health centers. However, this study may have its own limitation like since it is self-report there may be bias; in addition to this, since it is not follow-up, we do not know if they leave their organization or not.

Conclusion

The finding of this study showed that high number of health professionals intended to leave their organization in the study area. Sex, educational status, marital status, autonomous and living place condition were the identified predictors of turnover intention among health professionals.