Job conditions, job satisfaction, somatic complaints and burnout among East African nurses

Authors

  • Margot van der Doef,

    1. Authors:Margot van der Doef, PhD, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, Leiden, The Netherlands; Femke Bannink Mbazzi, MSc, Program Coordinator, AVSI Uganda, Kampala, Uganda; Chris Verhoeven, PhD, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, Leiden, The Netherlands
    Search for more papers by this author
  • Femke Bannink Mbazzi,

    1. Authors:Margot van der Doef, PhD, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, Leiden, The Netherlands; Femke Bannink Mbazzi, MSc, Program Coordinator, AVSI Uganda, Kampala, Uganda; Chris Verhoeven, PhD, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, Leiden, The Netherlands
    Search for more papers by this author
  • Chris Verhoeven

    1. Authors:Margot van der Doef, PhD, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, Leiden, The Netherlands; Femke Bannink Mbazzi, MSc, Program Coordinator, AVSI Uganda, Kampala, Uganda; Chris Verhoeven, PhD, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, Leiden, The Netherlands
    Search for more papers by this author

Margot van der Doef, Assistant Professor, Institute of Psychology, Clinical, Health and Neuropsychology, Leiden University, PO Box 9555, 2300 RB Leiden, The Netherlands. Telephone: +31 (0)71 527 3987.
E-mail:doef@fsw.leidenuniv.nl

Abstract

Aims.  To describe job conditions, job satisfaction, somatic complaints and burnout of female East African nurses working in public and private hospitals and to determine how these well-being outcomes are associated with job conditions.

Background.  Insight into job conditions, health and well-being status and their interrelation is virtually lacking for East African nurses.

Design.  Cross-sectional survey of 309 female nurses in private and public hospitals in Kenya, Tanzania and Uganda.

Methods.  Nurses completed a survey assessing job conditions and job satisfaction (the Leiden Quality of Work Life Questionnaire – nurses version), somatic complaints (subscale of the Symptom CheckList) and burnout (Maslach Burnout Inventory).

Results.  The East African nurses show high levels of somatic complaints, and nearly one-third of the sample would be labelled as burned out. In comparison with a Western European nurses reference group, the nurses score unfavourably on job conditions that require financial investment (e.g. workload, staffing, equipment and materials). On aspects related to the social climate (e.g. decision latitude, cooperation), however, they score more favourably. In comparison with private hospital nurses, public hospital nurses score similarly on aspects related to the social climate, but worse on the other job conditions. Public hospital nurses have a lower job satisfaction than private hospital nurses, but show comparable levels of somatic complaints and burnout. Strongest correlates of low job satisfaction are low supervisor support and low financial reward. Burnout is mainly associated with high workload and inadequate information provision, whereas somatic complaints are associated with demanding physical working conditions.

Conclusions.  Improvement in job conditions may reduce the high levels of burnout and somatic complaints and enhance job satisfaction in East African nurses.

Relevance to clinical practice.  Efforts and investments should be made to improve the job conditions in East African nurses as they are key persons in the delivery of health care.

Introduction

So far, the majority of the research on nurses’ job conditions and health and well-being status has been conducted in the USA and Western European countries. These studies generally indicate that working in the nursing profession is quite stressful (for reviews, see Irvine & Evans 1995, McVicar 2003). Nurses experience higher rates of stress-related disease, burnout symptoms, psychiatric admissions and general physical illness than the general population (Hillhouse & Adler 1997). Job conditions that have been pinpointed as sources of stress and burnout include workload, interpersonal relationships and professional conflict and emotional cost of caring (Duquette et al. 1994, McVicar 2003, Ruotsalainen et al. 2008). Recent reviews on nurses’ job satisfaction identified interpersonal relationships at work and providing good patient care (Utriainen & Kyngäs 2009), physical working conditions, pay, promotion, job security, responsibility and hours of work (Lu et al.2005) as predictors of job satisfaction.

Due to the small number of studies conducted in Eastern Europe, Latin America, Africa and Asia, however, knowledge regarding nurses’ job conditions and health and well-being status in non-Western contexts is limited. This study focussing on job conditions and health status of nurses in public and private hospitals in three East African countries (Tanzania, Kenya and Uganda) aims to fill part of this gap. On the one hand, this study is conducted to provide information regarding job conditions and health and well-being of East African nurses. Furthermore, the relationships between job conditions and health and well-being outcomes are examined to identify those job conditions that would be most worthwhile to target in the East African hospitals. Improving job conditions would not only be of importance for nurses’ well-being, but could also be expected to positively affect quality of patient care (Currie et al. 2005, Poghosyan et al. 2010).

Recent cross-national research on nurses demonstrates that both job conditions and health outcomes vary significantly across countries, even in Europe (see e.g. Schaufeli & Janczur 1994, Van der Schoot et al. 2003, Pilsjar et al. 2011, Pisanti et al. 2011). For instance, Pilsjar et al. (2011) found worse job conditions and health status among East European hospital employees in comparison with West European hospital employees. Bambra and colleagues indicate that a country’s social, economic and cultural situation may underlie these cross-national differences (Bambra et al. 2005). For example, differences in the organisation of the healthcare system, in the public expenditure on health care and in the labour market, have been mentioned as important factors (Pilsjar et al. 2011, Pisanti et al. 2011).

As such, the findings of this study must be interpreted against the background of the East African healthcare situation. Health care in these countries is characterised by low availability of healthcare facilities and low budgets. The total expenditure on health per capita per year (in US dollars) is $65 in Kenya, $75 in Uganda and $29 in Tanzania, whereas it is $6714 in the USA (WHO 2006). Similar striking differences are seen in the number of nurses per capita: 1·14 per 1000 in Kenya, 0·37 per 1000 in Tanzania and 0·61 in Uganda, vs. 4·0 per 1000 in the USA.

Another issue determining working life of East African nurses is the fact that they face different health problems at work than nurses in countries outside East Africa. In East Africa, malaria, tuberculosis and AIDS are the leading causes of death. Especially AIDS patients occupy hospital beds, and this has consequences for physicians, nurses and other health workers (Raviola et al. 2002). Not only the impact of the disease as such is affecting them (no cure, limited availability of treatment, also affecting young children, risk of contamination), but also the high number of patients is an important stressor at work.

The few studies examining nurses’ well-being in a sub-Saharan healthcare context show high burnout and low job satisfaction levels (Engelbrecht et al. 2008, Hagopian et al. 2009). Among nurses in primary health care in the Free State, South Africa, the majority of the nurses reported high levels of emotional exhaustion (69%) and depersonalisation (85%) and moderate levels of personal accomplishment (91%) (Engelbrecht et al. 2008). In a study among Ugandan health workers, fewer than half of them were satisfied with their jobs (Hagopian et al. 2009). Satisfaction with salary was particularly low, followed by low satisfaction with working conditions.

In this research, on the basis of this context, we expect East African nurses to experience more unfavourable job conditions and a less favourable health and well-being status than a European comparison group of nurses.

The health services in East Africa can be divided into public (governmental) and private services. The public health services are organised and subsidised by the Ministry of Health. The services offered include curative and preventive programmes, as well as other specialised services such as dental and mental health care. All personnel are employed by the government and are managed by the rules and regulations that govern civil service (Raviola et al. 2002).

The private system is characterised by a variety of hospital types, both profit and non-for-profit. Significant in the private system are the church-controlled health facilities, initiated by the missionaries in East Africa. In the private system, funding is coming from contributions from either individuals or religious organisations. In the African context, private hospitals are financially better off and are often more able to invest in employees, equipment and materials than public hospitals (Israr et al. 2000). In the research of Hagopian et al. (2009), Ugandan health workers in the private sector were more satisfied with their working conditions than those in the public sector. However, compensation and job security were rated lower in the private sector than in the public sector (Hagopian et al. 2009). Similar results were found in a study among South African nurses (Pillay 2009): overall, private sector nurses had higher satisfaction rates than public sector nurses. Private sector nurses were only dissatisfied with their pay and career development opportunities. In contrast, public sector nurses were most dissatisfied with their pay, the workload and the resources available to them.

In this research, East African nurses working in private and public hospital settings will be compared on job conditions and health and well-being indicators. We expect the nurses working in private hospitals to have more favourable job conditions, as well as a more favourable health and well-being status, than the nurses working in public hospitals.

Also with regard to associations between job conditions and nurses’ health and well-being, research has shown important cross-national differences (Schaufeli & Janczur 1994, Glazer & Gyurak 2008, Pilsjar et al. 2011, Pisanti et al. 2011). For example, in the research of Pilsjar et al. (2011), job autonomy was positively associated with health status among West European hospital employees, whereas this job condition was unrelated to health status in the East European subsample. Also in a qualitative study among nurses from Hungary, Israel, Italy, the UK and the USA, the job factors the nurses identified as underlying their experiences of occupational stress varied between countries (Glazer & Gyurak 2008). For example, for Hungarian nurses dealing with death and dying was particularly salient, whereas in the other samples, lack of staff was mentioned as the predominant stressor.

Focussing on the research performed in sub-Saharan Africa, a review of the literature indicates that only a very limited number of studies have examined job conditions in relation to nurses’ well-being. In Malawi, Peltzer (1997) studied stressors among nurses in an urban hospital and found external factors (e.g. low salaries and lack of equipment) and interpersonal relations among staff (especially nurses and matrons) to be important stressors. Bassett et al. (1997) concluded that overwork and patient ingratitude were the main work stressors among nurses in Zimbabwe. A study among nurses residing in Kampala, Uganda, investigated stress levels, factors influencing stress and ways of coping with stress (Baguma 2001). Results showed that nurses experienced stress caused by organisational factors, including lack of participation in decision making and lack of job autonomy. Personal factors, except professional ranking, were not found to influence stress.

According to our review of the literature, only three studies in sub-Saharan Africa have examined the relationship between job conditions and burnout in healthcare workers. In a study on South African military nurses, four stressors were identified as significant predictors of burnout: lack of support from supervisors, high responsibility, long working hours and task overload (Van Wijk 1997). In a more recent study among nurses in primary health care in the Free State (Engelbrecht et al. 2008), high workload, interpersonal conflict at work and low resources (e.g. faulty equipment, insufficient information) were the strongest predictors of the burnout dimensions. Furthermore, examining burnout among Kenyan medical residents, Raviola et al. (2002) conclude that the large number of HIV/AIDS patients, lack of resources and poor communication between hospital staff affect the job, the morale and daily activities of these healthcare workers. They indicate that these factors could be the crucial adverse working circumstances leading to symptoms of burnout.

Examining job satisfaction, Ndiwane (2000) found that hospital nurses in Cameroon were more satisfied with their job in general when they perceived a high level of community support. Among South African nurses, career orientation and self-concept (Bester et al. 1997) and working as part of a team and recognition (Erasmus 1998) were identified as determinants of job satisfaction.

As this overview indicates knowledge on the relationship between job conditions and well-being in African nurses is limited and virtually non-existing with regard to East African nurses. The few studies conducted in sub-Saharan Africa suggest that both psychosocial job characteristics (workload, autonomy and social support) and environmental/organisational factors (e.g. pay, lack of resources) play a role in nurses’ well-being. Furthermore, it is evident that one of the most examined indicators of nurses’ well-being, that is burnout, is understudied in the African context.

To increase knowledge on these issues and identify those job conditions that would be most worthwhile to target in the East African hospitals, we will examine the relationships between job conditions and health and well-being outcomes, that is, job satisfaction, burnout and somatic complaints, in a sample of East African nurses.

Research questions

The present research addresses the following three research questions:

  •  What is the current situation of East African nurses with respect to job conditions and health and well-being status? We look into environmental/organisational job conditions (e.g. equipment and materials, financial rewards), as well as psychosocial job characteristics (e.g. workload, decision latitude and social support). A range of outcomes indicative of health and well-being is examined in the study, including job satisfaction, burnout and somatic complaints. Regarding job conditions and job satisfaction, we relate these findings to a European comparison group consisting of hospital nurses. On somatic complaints and burnout, the East African sample will be compared with normative samples of the questionnaires assessing these outcomes.
  •   Do East African nurses in public and private hospitals differ with regard to their job conditions and health and well-being status?
  •   Which specific job conditions are associated with an unfavourable health and well-being status in this sample of East African nurses?

Method

Sample and procedure

One of the researchers personally approached hospital staff in public and private hospitals in Kenya, Tanzania and Uganda. After permission from the hospital director, an appointment with the matron or nurse-in-charge was made. Together with the matron, the researchers distributed the questionnaires in various wards of the hospital. Another appointment was made to collect the completed questionnaires after a week. In 11 hospitals, a total number of 465 questionnaires were distributed to nurses. Four of these hospitals (two in Kenya and two in Tanzania) were public hospitals, and the other seven hospitals were private hospitals (two in Kenya, two in Tanzania and three in Uganda). In Uganda, it was impossible to gain access to a public hospital for this research, as affiliation to the Makerere Medical School or other research institutions linked to the public hospitals proved necessary. Participation of nurses was voluntary and anonymous. No incentives, except pencils to fill out the questionnaire, were offered to the participants. Ethical approval for this study was given by the Institutional Review Board of the Institute of Psychology of the University the authors are affiliated with.

Of the 465 nurses, 344 returned completed questionnaires, reflecting a response rate of 74% (Tanzania: 74·9%; Uganda: 70·7%; and Kenya: 76·4%). Considering the low percentage of male nurses in the sample and their unequal distribution across countries and type of hospital, it was decided to include only the data from the 309 female nurses in this study.

The analysed sample includes 309 female nurses working in 11 different hospitals in Tanzania (= 119), Uganda (= 105) and Kenya (= 185). Of this sample, 106 nurses (34·3%) worked in public hospitals and 203 nurses (65·7%) in private hospitals. The age of the subjects ranged from 19 to 62 years, with a mean age of 33·7 (SD 8·7). The majority of the sample is married (55·8%), only a few per cent of the nurses is divorced (1·0%) or widowed (3·6%). With regard to their living situation, almost half of the sample is living with their husband, and 20% is living with other relatives or with their children. The other nurses are living in the hospital (14·8%) or somewhere else alone (15·5%).

Over 64% of the nurses have children, on average 2·7 (SD 1·5) children, ranging from one child up to nine children. In many cases, the respondents take care of more children than just their own. Whereas 35·6% of the nurses are childless, only 24·3% has no children to take care of. On average five persons depend on the income of one nurse. In 45·2% of the cases, more than five (up to 15) persons depend on the nurses income.

The nurses work a mean of 5·4 days a week (SD 0·67), ranging from 3–7 days. The average number of working hours on a day is 8·1 (SD 1·44), ranging from 5–15 hours. The number of years of education in nursing varies between 2–7 years, with an average of 3·5 years (SD 0·9).

Instruments

A questionnaire was compiled containing various socio-demographic questions and several (translated) questionnaires measuring job conditions and well-being. The questions were posed in English. To ensure correct interpretation, a Swahili translation was added in Tanzania and Kenya.

The Leiden Quality of Work Life Questionnaire for Nurses (LQWLQ – nurses version) (Maes et al. 1999) was used to assess job conditions. This questionnaire is based on the general Leiden Quality of Work Questionnaire (Maes et al. 1993, Van der Doef & Maes 1999), which aims to measure the key concepts of the Job Demand-Control-Support model and relevant work aspects from other theoretical models, such as the Michigan model (Caplan et al. 1975). In 1999, an occupation-specific version of the LQWQ for nurses was developed. The original job conditions were measured with items rephrased to be specific for the nursing profession. Furthermore, items measuring organisational and environmental conditions based on the Tripod model (Wagenaar et al. 1990) and items measuring job stressors that are specific for the nursing profession were added. The questionnaire consists of 129 statements that are rated on a four-point Likert scale, ranging from 1–4 (1 = entirely disagree, 2 = disagree, 3 = agree and 4 = entirely agree). The questionnaire measures the following job conditions: staffing (six items, e.g., In my department, there are enough nurses to provide good care), workload (six items, e.g., I must care for too many patients at once), decision latitude (six items, e.g., I have a say in decisions that concern my work), social support of supervisor (five items, e.g., I can count on the support of my direct supervisor when I face a problem at work), social support of colleagues (six items, e.g., My colleagues give me emotional support when I need it), information (three items, e.g., I regularly receive incomplete and/or incorrect patient information from doctors), interdepartmental cooperation (four items, e.g., In matters of patient care, my department works well with other departments), equipment and materials (five items, e.g., I must work with materials, equipment and/or instruments that are of insufficient quality), physical working conditions (five items, e.g., I must sometimes work in unsafe surroundings) and financial reward (six items, e.g., I am paid well for the work I do). For all subscales, higher scales scores indicate more favourable job conditions. Job satisfaction was assessed with the LQWLQ-nurses version (seven items, e.g., If given the choice, I would take this job again).

The Maslach Burnout Inventory (MBI) (Maslach & Jackson 1986) was used to measure job-related feelings of burnout. This questionnaire consists of 22 items and assesses emotional exhaustion, depersonalisation and personal accomplishment. Each statement is rated on a seven-point Likert scale ranging from 0–6 (0 = never, 6 = every day). Principal component analysis confirmed the factor structure of the MBI in this sample. High scores on emotional exhaustion and depersonalisation and a low score on personal accomplishment reflect high burnout levels.

The somatisation scale of the Symptom Checklist (SCL-90) (Derogatis 1977) was used to measure somatic complaints, such as nausea, headache and chest pain. The somatisation scale consists of 12 items. Each statement is rated on a five-point scale, ranging from 1–5 (1 = not at all, 5 = very much).

The intercorrelations and estimates of internal consistency (Cronbach’s α) of the measures included in this study are presented in Table 1. The results indicate that in this sample, the reliabilities of the quality of work subscales are generally satisfactory to good (α = 0·69–0·86), with the exception of the subscales ‘information’ (α = 0·57) and ‘financial reward’ (α = 0·57). The reliabilities for the outcomes range from 0·66 to 0·82, with the lowest alpha for the MBI-scale depersonalisation. This latter finding is in line with earlier reports on the reliability of this subscale (Garden 1987).

Table 1.   Intercorrelations and internal reliabilities of job conditions and outcomes (n varies between 290 and 305 respondents, due to missing values)
Variable (number of items)123456789101112131415
  1. *< 0·05, **< 0·01, ***< 0·001.

1. Workload (6)α = 0·69              
2. Decision Latitude (6)−0·06 α = 0·75             
3. Social support of supervisor (5)−0·000·30*** α = 0·86            
4. Social support of colleagues (6)0·010·32***0·35*** α = 0·79           
5. Staffing (6)0·41***0·24***0·27***0·15** α = 0·82          
6. Information (3)0·100·16**0·070·29***0·11 α = 0·57         
7. Interdepartmental cooperation (4)0·24***0·27***0·26***0·43***0·40***0·20*** α = 0·76        
8. Equipment and materials (5)0·39***0·19***0·18**0·20***0·44***0·24***0·28*** α = 0·70       
9. Physical working conditions (5)0·40***0·050·100·17**0·29***0·14*0·15*0·38*** α = 0·72      
10. Financial reward (6)0·33***0·110·110·050·35***0·050·19***0·32***0·36*** α = 0·57     
11. Job satisfaction (7)0·25***0·30***0·38***0·32***0·41***0·18**0·39***0·32***0·24***0·34*** α = 0·82    
12. Emotional exhaustion (8)−0·24***−0·12*−0·16**−0·24***−0·16***−0·27***−0·16**−0·15**−0·22***−0·17**−0·33***α = 0·80   
13. Depersonalisation (5)0·03−0·12*−0·13*−0·12*−0·03−0·100·01−0·00−0·06−0·02−0·110·46***α = 0·66  
14. Personal accomplishment (7)−0·020·18***0·08**0·050·040·100·16**0·13*0·04−0·030·050·13*0·09α = 0·71 
15. Somatic complaints (12)−0·14*−0·07−0·11−0·08−0·16**−0·12*−0·08−0·12*−0·24***−0·15*−0·17**0·36***0·16**−0·04α = 0·82

Data analysis

Descriptive procedures, correlations, Student’s t-tests and multiple regression analyses were used to answer the research questions. By means of t-tests, differences in job conditions and outcomes between the East African sample and Dutch reference samples and between private and public hospital nurses were examined. Multiple regression analyses were used to examine the relationships between job conditions and each of the outcome measures (job satisfaction, burnout and somatic complaints). Analyses controlling for socio-demographic variables (e.g. age, number of children, number of working days and tenure) and analyses without these control variables yielded the same results. The results of the analyses without the socio-demographic variables are presented.

Data screening indicated that the distributions of the outcomes depersonalisation, personal accomplishment and somatic complaints were skewed. To meet the assumption of a normal distribution for regression analysis, a square root transformation was applied for the scale depersonalisation and a reflected square root transformation for the scale personal accomplishment, and the inverse was calculated for the somatic complaints scale.

Results

Well-being

In comparison with norm scores for nurses (Schaufeli & van Dierendonck 2000), the East African nurses score unfavourably on emotional exhaustion (= 13·2, < 0·001) and depersonalisation (= 3·60, < 0·001); 33·9% of the nurses can be labelled as ‘very highly’ (95th percentile) and an additional 31·6% as ‘highly’ (75th percentile) emotionally exhausted and 13·5% as ‘very highly’ (95th percentile) and an additional 17·7% as ‘highly’ (75th percentile) depersonalised. However, on personal accomplishment, the East African sample scores more favourable in comparison with the norms (= 11·34, < 0·001). 6·7% of the nurses score below the 95th percentile and an additional 6·1% below the 75th percentile score on personal accomplishment. Applying the cut-off points for burnout in nurses as provided by Schaufeli and van Dierendonck (2000), 32·1% of the East African nurses sample would be labelled as burned out.

In comparison with norm scores for a female general population (Arrindell & Ettema 1986), the East African nurses have more somatic complaints (= 5·37, < 0·001). In line with this, 9·9% of the sample can be labelled as experiencing ‘very high’ (>95th percentile) and an additional 27·7% as experiencing ‘high’ (>80th percentile) levels of somatic complaints. With regard to job satisfaction, the East African nurses are compared with a large sample of Dutch nurses (Gelsema et al. 2005). The job satisfaction score of the East African sample does not differ significantly from the score in this reference group (= 1·28, > 0·05).

Job conditions

The scores on the job conditions are compared with reference scores for female nurses derived from a large Dutch sample (Gelsema et al. 2005). The East African nurses experience a higher lack of staff (= −3·67, < 0·001) and a higher workload (= −10·91, < 0·001) than Dutch nurses. Furthermore, the East African nurses are less satisfied with the availability and quality of equipment and materials (= −2·37, < 0·05) and work in less favourable physical working conditions (= −2·54, < 0·05). The East African nurses, however, experience higher decision latitude (= 3·95, < 0·001) and higher supervisor support (= 2·42, < 0·05). In addition, the East African nurses score also more positively on information provision (= 14·27, < 0·001) and interdepartmental cooperation (= 6·74, < 0·001) than the Dutch nurses. On social support from colleagues (= 0·12, > 0·05) and financial reward (= 1·39, > 0·05), the sample does not differ significantly from the reference group.

Job conditions and well-being: comparing private and public hospital nurses

Table 2 reports the means and SD on the job conditions and outcomes for the private and public hospital nurses. Student’s t-tests indicate that the public hospital nurses have more unfavourable job conditions than the private hospital nurses with regard to staffing, workload, interdepartmental cooperation, equipment and materials, physical working conditions and financial reward. Public hospital nurses and private hospital nurses, however, score comparably on decision latitude, social support from supervisor and colleagues and information. With regard to well-being, the public hospital nurses experience a lower job satisfaction than the private hospital nurses. However, on the other well-being outcomes (burnout dimensions, somatic complaints), the public hospital and private hospital nurses do not differ.

Table 2.   Means (SD) on job conditions and outcomes in the East African nurses sample (= 309) and comparison of public hospital nurses (= 106) and private hospital nurses (= 203) by means of Students t-test. A higher score on a job condition indicates a more favourable situation
 Total sample
M (SD)
Public hospital nurses
M (SD)
Private hospital nurses
M (SD)
t-value
  1. **< 0·01, ***< 0·001.

Job conditions
 Workload2·18 (0·49)1·96 (0·49)2·30 (0·45)−6·04 ***
 Decision Latitude2·81 (0·51)2·86 (0·48)2·78 (0·52)1·21
 Social support supervisor2·91 (0·59)2·90 (0·60)2·91 (0·59)−0·14
 Social support colleagues3·03 (0·44)2·97 (0·43)3·06 (0·44)−1·73
 Staffing2·35 (0·58)2·18 (0·55)2·43 (0·55)−3·71 ***
 Information2·84 (0·55)2·80 (0·57)2·87 (0·53)−0·97
 Interdepartmental cooperation2·78 (0·56)2·66 (0·60)2·85 (0·54)−2·78 **
 Equipment and materials2·50 (0·58)2·21 (0·51)2·64 (0·55)−6·61 ***
 Physical working conditions2·27 (0·56)2·10 (0·62)2·35 (0·52)−3·57 ***
 Financial reward1·93 (0·49)1·81 (0·52)2·00 (0·47)−3·21 **
Outcomes
 Job satisfaction2·65 (0·58)2·50 (0·60)2·72 (0·56)−3·16 **
 Emotional exhaustion2·60 (1·30)2·72 (1·45)2·53 (1·21)1·17
 Depersonalisation1·36 (1·02)1·35 (1·09)1·37 (0·99)−0·15
 Personal accomplishment4·79 (1·15)4·81 (1·11)4·78 (1·17)0·15
 Somatic complaints21·0 (7·96)21·9 (8·05)20·8 (8·08)1·13

Associations between job conditions and well-being in East African nurses

Pearson correlation coefficients were computed to explore the relationships between job conditions, job satisfaction, burnout and somatic complaints (Table 1). All job conditions are significantly correlated with job satisfaction and, although less strongly, with emotional exhaustion. For depersonalisation, personal accomplishment and somatic complaints, only significant correlations with some of the job conditions are found, the correlation pattern suggesting differential effects. The outcome measures are not very strongly correlated (max. = 0·46, < 0·001 for emotional exhaustion and depersonalisation), indicating that these outcomes represent different aspects of well-being.

To examine the relationship between job conditions and outcomes, multiple regression analyses were conducted (Table 3). As expected on the basis of the correlation table, the job conditions explain the highest amount of variance, nearly 36%, in job satisfaction. A higher job satisfaction is related to better staffing, higher supervisor support, better interdepartmental cooperation and adequacy of financial reward.

Table 3.   Job satisfaction, emotional exhaustion, depersonalisation, personal accomplishment and somatic complaints regressed on job conditions (= 309)
Predictors: job conditionsJob satisfaction
β
Emotional exhaustion
β
De-personalisation
β
Personal accomplishment
β
Somatic complaints§
β
  1. *< 0·05, **< 0·01, ***< 0·001.

  2. Square root transformation.

  3. Reflected square root transformation.

  4. §Inverse transformation.

Workload0·08−0·21**0·000·080·09
Decision Latitude0·10−0·04−0·08−0·15*0·03
Social support supervisor0·21***−0·09−0·12−0·030·10
Social support colleagues0·10−0·15*−0·070·08−0·03
Staffing0·14*0·010·080·060·05
Information0·05−0·20***−0·08−0·040·07
Interdepartmental cooperation0·13*0·030·09−0·17*0·00
Equipment and materials0·040·070·05−0·11−0·01
Physical working conditions0·02−0·09−0·08−0·040·16*
Financial reward0·18**−0·07−0·030·100·02
Model R²0·358***0·172***0·0530·074*0·082**

The job conditions explain 17% of the variance in emotional exhaustion. A higher workload, lower social support from colleagues and problems concerning information provision are related to higher levels of emotional exhaustion. The job conditions fail to explain a significant proportion of the variance in depersonalisation. For personal accomplishment, 7·4% of the variance is explained by the job conditions. Higher decision latitude and better interdepartmental cooperation are associated with higher personal accomplishment. Eight per cent of the variance in somatic complaints is explained, with physical working conditions being the only significant predictor. Worse physical working conditions are associated with more somatic complaints.

Discussion

This research aims to provide more insight into job conditions and health and well-being outcomes of nurses in public and private hospitals in East Africa. The first research question focuses on whether East African nurses are in an (un)favourable position with regard to job conditions, job satisfaction, somatic complaints and burnout. In comparison with reference groups, East African nurses experience more understaffing, heavier workload, less favourable physical working conditions and less availability and poorer quality of equipment and materials. Other studies also highlight the problems regarding these job conditions in their research among African healthcare workers (Raviola et al. 2002, Fournier et al. 2007, Leshabari et al. 2008, Hagopian et al. 2009). The experiences on-site of one of the authors confirm these findings. The various wards visited generally showed lack of equipment and heavy physical job circumstances. During the rounds, the matrons indicated that a considerable number of the patients were terminally ill. These patients needed intensive care with only few nurses available. In terms of job conditions, it might be concluded that East African nurses have more difficult job conditions in these respects. However, East African nurses show more favourable conditions with regard to decision latitude, social support from supervisor, information and interdepartmental cooperation. These job conditions have in common that they are relatively independent on the financial investments a hospital can make; they are reflecting the social climate in the hospital setting. A basic difference in hospital organisational structure could underlie this different social climate. In Western Europe, hospitals are generally characterised by a complex organisational structure with a strict hierarchy. Such an organisational structure might hamper communication and limit the decision latitude of individual employees. The East African hospitals in this sample were mostly quite small-sized hospitals where communication lines are more direct.

Besides these differences, on some job conditions, East African nurses are comparable to the Dutch comparison group. Social support from colleagues does not differ between the two samples; in both samples, it is on average quite high. Furthermore, the appreciation of financial reward is equally rated. This remarkable finding is explicable if one takes into account that appreciation of financial reward is likely to be the result of a social comparison with the financial status of employees in comparable jobs (Major & Forcey 1985) and with the financial status of the general population.

Despite these differences in job conditions, job satisfaction in the East African sample is comparable to a Western Europe nurse reference group. Possibly, the more unfavourable job conditions (understaffing, high workload and high physically demanding circumstances) are to a certain extent counterbalanced by the more favourable social climate. According to various theories [e.g. the Job Demand-Control-Support model (Johnson & Hall 1988, Karasek & Theorell 1990), the social support buffer hypothesis (Cohen & Wills 1985)], high job control and a good social support system may act as buffering factors.

Somatic complaints, in contrast, are more prevalent among East African nurses. This seems partly attributable to the less favourable physical working conditions in the East African hospitals. However, differences in culturally acceptable ways of expressing psychological and physical distress can also play a role in the higher report of somatic complaints in the East African sample. In Europe, African migrants also show more psychosomatic complaints than the European population (De Jong & van der Berg 1996).

The burnout scores indicate high levels of emotional exhaustion and depersonalisation for the East African nurses. Strikingly, however, East African nurses have a relatively high sense of personal accomplishment. Similar high levels of emotional exhaustion and depersonalisation, in combination with moderate levels of personal accomplishment, were found among nurses in primary health care in the Free State (Engelbrecht et al. 2008). For our sample, this relatively positive finding regarding personal accomplishment might be explained by the favourable organisational climate (high support/cooperation, high decision latitude), which enhances nurses’ possibilities to adequately function in the demanding situation. Nevertheless, it should be noted that nearly one-third of the East African nurses in our sample would be labelled as burned out on the basis of their high emotional exhaustion and depersonalisation scores (Schaufeli & van Dierendonck 2000).

Given the working circumstances and the impact of working with high numbers of HIV/AIDS patients (Raviola et al. 2002), the high levels of emotional exhaustion and depersonalisation in this sample are not surprising. Coping with the suffering and high rates of death and loss of many young people may make it hard to maintain a personal sense of empathy for patients and not becoming ‘emotionally numb’ to patients’ subjective experience of suffering. A certain level of depersonalisation might even be an adaptive and necessary coping strategy to function as a nurse in the current situation in East African hospitals.

Furthermore, one should realise that whereas in many countries with an adequate social security system burnout is a cause for prolonged absenteeism or for quitting the nursing profession, in the East African context, one will encounter burned out nurses still at work. However, one should also keep in mind that the burnout definition is written from a western perspective. People from African cultures might present (the consequences of) their workload and burden in a more expressive way (De Jong & van der Berg 1996). They can also have a different perception of emotional involvement with patients. In the East African context, characterised by a low life expectancy and the acceptance of death as a normal part of life, nurses may have a different perception of death and loss of patients than nurses in Western Europe.

The second research question centres on whether there are differences in job conditions, job satisfaction, somatic complaints and burnout between nurses in public and private hospitals. Generally, in line with expectations, nurses in public hospitals report more unfavourable job conditions. This is in line with a study among Ugandan health workers, which consistently found private sector workers to rate their working conditions more positively than those working in the public sector (Hagopian et al. 2009). Regarding decision latitude, social support and information provision, the public hospital nurses seem comparable to the public hospital nurses. As mentioned, these job conditions have in common that they are relatively independent on the financial resources of a hospital.

Contrary to expectations, however, the public hospital nurses only score more unfavourably on a single well-being outcome, namely job satisfaction. This lower job satisfaction among public hospital nurses is understandable given their worse working conditions. We also expected to find higher rates of somatic complaints and burnout in the public hospital nurses. As already indicated, the relatively favourable situation on job resources (i.e. decision latitude and social support) might be functioning as a protective factor (Karasek & Theorell 1990).

The third research question focuses on the relationship between job conditions and well-being outcomes among East African nurses. Job satisfaction is best predicted by the job conditions, with 36% of variance explained. The explained variance of the other outcomes varies from 5–17%. This is in line with earlier research (see e.g. Van der Doef et al. 2000). While job satisfaction is greatly influenced by job conditions, burnout and somatic complaints are more likely to be influenced by non-work factors as well.

The strongest predictors for job satisfaction and burnout in this African sample are workload, information provision, supervisor support and financial reward. These results are in line with findings on South African nurses (Van Wijk 1997, Engelbrecht et al. 2008) and on Kenyan medical residents (Raviola et al. 2002). Furthermore, physical working conditions emerge as the strongest correlate of somatic complaints. Our study suggests that improvements in especially these job conditions may lead to better health and well-being in East African nurses.

Our study seems to be the first research examining job conditions and job-related outcomes, such as burnout, in East African nurses. Focusing on its strengths, it should be noted that this study is unique in its sample, consisting of 309 East African nurses from 11 hospitals in East Africa. Within 10 weeks, three East African countries were visited to collect the data, and a high response rate was obtained. Furthermore, job conditions were assessed with a comprehensive, occupation-specific questionnaire. Of course, this study also has some methodological limitations. Although the questionnaires to measure job conditions and outcomes have been used in various studies, they have not been used in this cultural context before. To ensure correct understanding of the questionnaires in our sample, a Swahili translation of the questionnaires was made for the Kenyan and Tanzanian nurses. Furthermore, principal component analyses and reliability analyses support the factor structure of the questionnaires. As for practical reasons only self-report questionnaires were used in a cross-sectional design, the study is subject to common method variance and no conclusions regarding the causality of the relationships between job conditions and outcomes can be drawn.

Despite these limitations, it seems valid to conclude that with regard to job conditions and health and well-being outcomes, there are differences between East African and Western European nurses and between public and private hospital employed East African nurses. It would be interesting to further examine the relationships between job conditions and health and well-being in other African countries, as well as in Latin American and Asian countries. This could clarify which job conditions to target to improve health and well-being of nurses in the different healthcare contexts. Furthermore, gaining more insight into whether and to which extent job stressors like high workload can be counterbalanced by favourable job resources, such as high job control or social support, is valuable, because reducing certain stressors through financial investments may not be possible in all contexts.

Future studies examining health and well-being outcomes in African settings may further examine the validity of the concept of burnout. Given the cultural and religious background and the current situation in East African countries, depersonalisation may represent a necessary coping style, instead of signalling a negative well-being state. Besides the increase in patients who need intensive care and the risk of infection, coping with higher rates of death and loss of many young people has become an important part of the nurses’ jobs (Raviola et al. 2002). In this context of limited resources as well as increasing numbers of AIDS patients in relation to staff levels, it is hard to deliver comprehensive nursing care, maintain interest and enthusiasm in nursing work, keep a personal sense of empathy for patients and retain a sense of hope and idealism (Raviola et al. 2002, Walusimbi et al. 2002, Hagopian et al. 2009).

Conclusions

This study indicates that East African nurses experience unfavourable job conditions with regard to staffing, workload, the availability and quality of equipment and materials and physical working conditions. Efforts should be made to improve these conditions, especially in the public hospitals, as nurses in these settings are worse off. This could be attained by increasing the number of staff and investing in the purchase and maintenance of equipment and materials. It should also be noted that East African nurses show more favourable conditions with regard to decision latitude, social support from supervisor, information and interdepartmental cooperation. Maintaining these job resources at a high level should also be considered important, given their potential to help nurses cope with the demands of their job.

Furthermore, this study indicates that the most important correlates of health and well-being of East African nurses are workload, supervisor support, information, physical working conditions and financial reward. Improvement in these job conditions seems essential in striving for a better health and well-being status of East African nurses.

Relevance to clinical practice

It is clear that many of these interventions to ameliorate job conditions require financial investment in both public and private hospital settings. Furthermore, to attain more adequate staffing levels, it seems necessary to address the current shortage of trained nurses in East Africa also on a national level. This requires investments in nursing education and efforts to counteract the migration of trained nurses to Western countries (Nguyen et al. 2008, Hagopian et al. 2009).

Given these recommendations and the lack of financial resources in most (public) hospitals in East Africa, it seems that action on different levels, locally in the hospitals, nationally and internationally, is necessary. In hospitals, also those lacking financial resources, the social work climate could be a focus of attention to ensure good communication, social support and decision latitude. Governments could allocate donor funds and own funds to invest more in their public health services, to improve job conditions, such as staffing levels and material resources. Likewise, donors (e.g. international organisations) could also be stimulated to increase their investments in the well-being and health of nurses, besides their focus on the patients, as nurses are key persons in the delivery of care to the patients. Besides having a positive direct effect on well-being and health of nurses, creating better job conditions might also be an avenue to counteract the migration of trained nurses to Western countries (Nguyen et al. 2008) and enhance patient quality of care (Currie et al. 2005, Poghosyan et al. 2010).

Acknowledgements

We would like to thank the nurses, nursing officers and matrons of the following hospitals for their participation in this study: in Tanzania, Ocean Road Cancer Institute, Mwazi Mmoja Hospital, Aga Khan Dar Es Salaam, and KMMC; in Uganda, Mengo Hospital, International Hospital, and St. Francis Hospital Nsambya; and in Kenya, Kenyatta National Hospital, Mombasa Hospital, Aga Khan Mombasa, and Coast Hospital. Furthermore, we thank Ingmar Pfähler for her assistance in the data collection, and Anita Kisyeri-Hettema for assistance in the Swahili translation of the questionnaire.

Contributions

Study design: MvdD, FBM, CV; data collection and analysis: MvdD, FBM, CV; manuscript preparation: MvdD, FBM, CV.

Conflict of interest

The authors declare no conflict of interest.

The Journal of Clinical Nursing (JCN) is an international, peer reviewed journal that aims to promote a high standard of clinically related scholarship which supports the practice and discipline of nursing.

For further information and full author guidelines, please visit JCN on the Wiley Online Library website: http://wileyonlinelibrary.com/journal/jocn

Reasons to submit your paper to JCN:

High-impact forum: one of the world’s most cited nursing journals and with an impact factor of 1·228 – ranked 23 of 85 within Thomson Reuters Journal Citation Report (Social Science – Nursing) in 2009.

One of the most read nursing journals in the world: over 1 million articles downloaded online per year and accessible in over 7000 libraries worldwide (including over 4000 in developing countries with free or low cost access).

Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jcnur.

Early View: rapid online publication (with doi for referencing) for accepted articles in final form, and fully citable.

Positive publishing experience: rapid double-blind peer review with constructive feedback.

Online Open: the option to make your article freely and openly accessible to non-subscribers upon publication in Wiley Online Library, as well as the option to deposit the article in your preferred archive.

Ancillary