Lived experience of burnout and fatigue in perioperative healthcare professionals in Rwanda: a qualitative study

Abstract

Background

There is a lack of qualitative data on the negative effects of workplace stressors on the well-being of healthcare professionals in hospitals in Africa. It is unclear how well research methods developed for high-income country contexts apply to different cultural, social, and economic contexts in the global south.

Methods

We conducted a qualitative interview-based study including 64 perioperative healthcare professionals across all provinces of Rwanda. We used an iterative thematic analysis and aimed to explore the lived experience of Rwandan healthcare professionals and to consider to what extent the Maslach model aligns with these experiences.

Results

We found mixed responses of the effects on individuals, including the denial of burnout and fatigue to the points of physical exhaustion. Responses aligned with Maslach's three-factor model of emotional exhaustion, decreased personal accomplishment, and depersonalisation, with downstream effects on the healthcare system. Other factors included strongly patriotic culture, goals framed by narratives of Rwanda's recovery after the genocide, and personal and collective investment in developing the Rwandan healthcare system.

Conclusions

The Rwandan healthcare system presents many challenges which can become profoundly stressful for the workforce. Consideration of reduced personal and collective accomplishment, of moral injury, and its diverse downstream effects on the whole healthcare system may better represent the costs of burnout Rwanda. It is likely that improving the causes of work-based stress will require a significant investment in improving staffing and working conditions.

Editor's key points.

• •In high-income countries, preventing burnout among healthcare professionals is increasingly understood as a key factor in ensuring healthcare systems function effectively.
• •Very little research has been undertaken in the global south where hospitals are often resource poor, and the impact of burnout may well be significantly greater.
• •This study provides evidence from an African country that burnout among perioperative healthcare professionals has significant impact on both individual staff and the healthcare system as a whole.

Healthcare professionals have been described as one of six fundamental ‘building blocks’ for strengthening healthcare systems.¹ However, human resources for health should be considered to have a special place within this model, because of lead time, potential impact, and their unique vulnerability.² Compared with other factors such as financing, information technology, and the provision of essential drugs and medications, change in human resources is slow, requiring many years of training.³ This is especially the case for physicians and specialist care. When human resources in healthcare are well developed, they have a disproportionate potential to affect care, impacting other building blocks, including through leadership and agency outside of clinical care delivery. As the human component of the healthcare system, healthcare professionals are at risk of work-related stresses, care-fatigue and burnout. A contextual vulnerability is that where the development of healthcare workers is most needed, they are fewest, resulting in excessive workload and increased risk of physical, psychological, and social harm. These harms increase the likelihood that healthcare professionals leave the profession,4, 5, 6, 7 risking a vicious cycle of healthcare workforce scarcity.⁸

The World Health Organization (WHO) estimates that by 2035 the current shortage of 7.2 million healthcare workers worldwide will increase to 12.9 million.⁹ Among the seven recommendations identified by the WHO to bridge the provider gap is the importance of ‘respecting the rights of the [healthcare] worker, who in turn must embrace the right to health’.¹⁰ The human resource shortage is particularly acute amongst the surgical and anaesthesia workforce,^11,12 contributing to lack of access to safe, timely surgical care for over 70% of the world's population.⁹ The global burden of surgical disease accounts for more deaths than HIV, tuberculosis, and malaria combined.^13,14

There is an extensive literature on burnout in healthcare professionals, primarily from high-income countries. A 2016 meta-analysis reported that physician burnout has reached ‘epidemic levels’ in the United States. Consequences include irritability, impaired alertness, frequent errors, substance abuse, and suicidal ideation; these symptoms lead to lower staff retention.^15,16 There is a much smaller, but increasing body of literature on burnout from East Africa,¹⁷ and also from other low- and middle-income countries.^18,19 However, this latter body of literature often assumes that conceptual models of burnout, such as Maslach's three-factor model (work-related emotional exhaustion, depersonalisation, and reduced personal accomplishment),²⁰ can be applied unreflexively across very different social, cultural, and economic contexts. There is also a huge body of literature on fatigue in healthcare professionals, relating fatigue to workload, performance, and patient outcomes,21, 22, 23 but a paucity of literature on the causes and effects of fatigue on healthcare professionals from outside of high-income countries.²⁴

Rwanda is a small, land-locked and mountainous low-income country in East Africa with unique social, historical, and cultural context. Originally the Kingdom of Rwanda, it has a complex colonial and post-colonial history, including colonial rule by Germany and Belgium, civil war, and the devastating genocide in 1994. Recent history has been defined by political stability and economic growth in the recovery from conflict.²⁵ A focus on collective good and community building is exemplified by the culture of Umuganda (‘coming together for a common purpose’ in Kinyarwanda), when all Rwandans report for community work on a monthly basis.²⁶ Rwanda has a rapidly developing but vulnerable healthcare system characterised by low densities of healthcare professionals, but with significant investment in human resources for health both by the Rwandan Government and its partners, including the US$150 million 7-yr Human Resources for Health programme 2012–2019.³ Research suggests that burnout, as described by the Maslach model, is common,¹⁶ and that the health system struggles with retention of healthcare professionals. However, there is a lack of in-depth qualitative data on the negative effects of workplace stressors on healthcare professionals' physical, psychological and emotional well-being. To bridge this gap, we planned to interview perioperative healthcare professionals from across Rwanda to identify how they experience the effects of workplace-based stressors. We aimed to identify the ‘lived experience’ as reported by participants, considering a broad bio-psycho-social perspective on the consequences of workplace stress and reported effects on patient care. As the Maslach model of burnout has been applied widely in disparate contexts, a secondary goal was to consider to what extent this framework aligned with the experiences of our participants.

Methods

We used a qualitative interview-based study design to understand lived experiences of healthcare work of perioperative healthcare professionals at different career stages and across a variety of healthcare institutions within the Rwandan Healthcare System. The study design was guided by the pre-existing theoretical framework of the Maslach three-factor model, where burnout is characterised as a combination of emotional exhaustion, reduced personal accomplishment, and depersonalisation¹⁶ as well as a bio-psycho-social model of fatigue.¹⁹

Research ethics approval was provided by the University of Rwanda (No. 227/CMHS IRB 2019) and the University of Ottawa (ID: 20190180-01H). Permission was also provided by the leadership of each hospital to interview their staff. Reporting of this study is aligned with both the consolidated criteria for reporting qualitative research (COREQ)²⁷ and the standards for reporting qualitative research: a synthesis of recommendations (SPQR).²⁸ The COREQ checklist can be found in the supplementary material online.

We sought to make our research team diverse in terms of skills (including clinicians, clinician-researchers with qualitative expertise, and academics with backgrounds in sociology, psychology, and medical anthropology), gender, and ‘insider’/‘outsider’ perspectives (including Rwandans, investigators from high-income countries and other low- or middle-income countries outside of Rwanda). We used purposive sampling to ensure we included participants from (1) all five provinces in Rwanda; (2) each level of hospital within the Rwandan healthcare system (district hospital, referral hospital, tertiary hospital); and (3) all relevant perioperative professional groups: midwives, nurses, general practitioners who perform surgical procedures, specialist surgeons and obstetricians, physician anaesthetists and non-physician anaesthetists, and trainee specialist surgeons, obstetricians, and anaesthetists.

To some extent, sampling was influenced by access: practicalities of road travel and accommodation for the data collection team and permissions granted by the leadership of individual hospitals.

Data collection was in-person, from December 2, 2019 to February 11, 2020. Interviewers used a semi-structured interview guide (questions with prompts and probes for further depth), with participants free to elaborate (see the interview guide in the Supplementary Appendix). This interview guide was developed using data from a previous study by one of the principle investigators (PIs) - MDB,²⁹ other relevant literature and discussions amongst co-investigators who live and work (or have lived and worked) in Rwanda (ET, CN, MC, MG, MDB). We started with a pilot data collection (December 2–6, 2019; MDB), then an initial phase of analysis and further development of the interview guide. The main phase of data collection (January 13, 2020 to February 11, 2020; CC) was concurrent with analysis.³⁰ Some participants spoke in Kinyarwanda, which was translated contemporaneously into English during the interview by an interpreter. Audio-recordings were made of all interviews and field notes were taken. There was generally no pre-existing relationship between the interviewer and the participant, but some of the participants from the pilot phase had previously worked with the interviewer.

All interviews were transcribed verbatim by a professional transcription company. Participants and any named individuals were de-identified in the transcripts, with codes kept in a separate key. Transcripts were imported into NVivo 12 (QSR International, Doncaster, Australia) for analysis. Analysis began with detailed phrase-by-phrase coding. During this phase there were frequent meetings of the coding team (LY, MDB, CR, ET, LW), keeping an audit trail of the team's analysis decisions. Following standard procedures of thematic analysis,³¹ our iterative process consisted of both inductive and deductive analysis. Deductively we followed the existing theoretical model (Maslach's three-factor model) and developed a coding ‘tree’ that included the themes raised by participants and considered themes that were both consistent and in tension with the Maslach model. The team also engaged in open coding that captured patterns of data not included in the dimensions contained in the Maslach's model. Thematic analysis continued with an iterative process, with repeated revision of codes focusing on identifying themes that were most representative (consensus, where present) or felt to have most impact, even if mentioned by a minority of participants (important dissenting opinions). There was no further data collection after the main phase, but all analysts felt that the data were adequate for thematic sufficiency.^32,33 Transcripts were not returned to participants.

Results

Eight interviews were conducted in Kigali in December 2019 (by MDB) and a further 56 interviews in January–February 2020 (by CC). Participant characteristics are detailed in Table 1, and included 18 women (29%) and 45 men (71%). Data were missing for age for 37 participants (59%), but the median [range] age was 34 years [25–54]. Data were missing for years in practice for 26 participants (25%), but the median [range] duration was 3 yr [0–17] for those who had completed training. No-one refused participation in this study, with 100% response rate, but one hospital declined to have any of their staff approached for recruitment in the study. Some interviews were cut short because of clinical workload. Interviews were generally conducted close to the workplace, although eight interviews were conducted at a conference venue. Interviews lasted between 12 and 78 min. The institutions that participants worked in are detailed in Figure 1 (37% teaching hospital, 6% referral hospital, 52% district hospital, 5% unknown).

Table 1.

Participant information.

	Female	Male	Total
Anaesthesia resident	3	3	6
Anaesthesiologist		5	5
Anaesthetist (non-physician)	3	7	10
General surgery resident		3	3
General practitioner	1	7	8
General surgeon		1	1
Midwife	6		6
OB/GYN		5	5
OB/GYN resident		3	3
Orthopaedic surgeon		1	1
Registered nurse	5	10	15
Total	18	45	63

Fig 1.

Location of interviews. Provinces are numbered Eastern Province (1), Northern Province (2), Kigali City (unnumbered), Eastern Province (4), and Southern Province (5). The number in the circle refers to the number of participants included; the blue circles are tertiary/teaching hospitals, the green circle is a referral hospital, and the red circles are district hospitals. Map adapted with permission from Wikimedia Commons (https://commons.wikimedia.org/wiki/File:Rwanda,_administrative_divisions_-_Nmbrs_-_colored.svg).

In this paper, we will briefly report on workplace stressors to provide context; however, we plan to report more detail on the structural conditions of labour elsewhere. We will report findings relating to the negative effects of workplace stress on healthcare professionals according to four main themes: fatigue and physical response, emotional response, social response, and downstream effects on participants and their patients. Although these findings often aligned with the Maslach three-factor model for burnout, additional themes were identified. Finally, we report sources of resilience and positivity by participants, to provide a balanced account of the experiences reported to us.

Workplace stressors

Core goals for the Rwandan healthcare system, such as reduction of maternal mortality in a context of low availability of trained clinical workers poses pressures of a magnitude that is difficult to understand in resource-rich countries where most burnout research has taken place. As one midcareer obstetrician recounted:

‘When I came, we had a problem with maternal deaths. They were very high. So one of the strategy to reduce them was that I should be available whenever I am needed.’ Participant ID number (#) 37, obstetrician

The reported workload could be extremely high, and consequently was the most frequently mentioned as the primary source of stress. Study participants across all specialties described high numbers of patients, long hours of continuous work and up to years of continuous on-call commitment as a sole practitioner. Participants were frequently interrupted during the interview for clinical work, and often had to leave the interview early.

‘Sometimes you need do like parallel cases. So maybe someone is having an emergency c-section and another one is coming also for another emergency c-section. That’s … it’s not a situation where you may even call someone to help you. So you do them’ #24, non-physician anaesthesia provider (NPAP)

In this context, poor outcomes were often identified by participants across all specialties as being associated with deficiencies across the whole healthcare system. These deficiencies were further sources of stress: lack of essential drugs, equipment, and sterilisation, especially when this leads to an inability to provide appropriate care resulting in poor outcomes. Other factors cited as contributing to poor outcomes included poor resource organisation leading to inefficiencies and chaotic patient flow leading to long unpredictable work hours with protracted stretches of wasted time between cases; lack of availability, training, and seniority of other members of the perioperative team, and practice in a remote and unsupported location; lack of the skills to provide effective care; social determinants of health including poverty; and late presentation of disease (Table 2). Other reported stressors included debt and low salary, such that doctors struggled to afford basics such as housing, transportation, school fees for their children, and even food.

Table 2.

Commonly cited causes of workplace stress.

Workload •‘For us in residency, working 24 hours or even 36 hours, it is not strange’ #35, OB/GYN resident •‘A very bad day is a day they call you at 4:00 am in the morning. So you come maybe for a PPH or someone is having a uterine rupture and they are also doing another c-section and after finishing it, maybe you finish just the case at 7:00 and you start also the scheduled cases for the day. So there comes a case that are hard to manage, that is stressful to manage and cases of general surgery, maybe open laparotomy. So you are having maybe a c-section there and they came with a septic patient. A septic patient who needs, maybe, a ventilator or who needs infusion of dopamine. So that's where it becomes really stressed because you need someone to concentrate on and the other who really also needs care. So that's really when I feel like, when you have those like a few laparotomies and have like five c-sections a day, that's the day you go home without even … you didn't even have time to drink even water. Yes you didn't have time even for a break. Everything is really …. Frustrated. Frustrated, just even like you're empty, stressed, you're having headaches, hypoglycaemia’ #23, Non-physician anaesthesia provider (NPAP) •‘It was terrible. So I had to work 24 hours a day, 7 days a week because if you are alone, even if you are supposed to be resting, there is a phone calling …. For two years and a half’ #35, OB/GYN resident •‘I can do 15 C-sections, Caesarean sections without even help of another anaesthetist’ #46, NPAP •‘I have to travel from … Kigali to Bushenge … (by) public transport, it's almost six hours and a half. So when I came here, it's almost I'm tired, I feel tired’ #27, OB/GYN •‘I blame the administration … they should have provided some tea or some food just to drink so that people might feel comfortable to work.’ #24, NPAP •‘We work too much … and we are not safe.’ #24, NPAP

Resources •‘Imagine if I don't have oxygen. Fortunately, this hospital has the plant, they produce the oxygen …. But there are some hospitals they don't have plant. They go with the tank, they buy, they bring the … oxygen and you want the oxygen right now. And the hospital, they say, “We are signing some papers.” Paper cannot save patients.’ #20 anaesthesiologist •‘So sometimes you may lack essential medication and then the patient can die. Not for your skills, not for the team’ #7, anaesthesiologist •‘I have to find a place where I can work with materials. Because you can't still work, it's like we're using like 10% of the knowledge I have gained from the school’ #52, surgeon •‘there was no blood. Then I referred the patient to the hospital where there is blood. Then she died on my way …. Due to a lot of bleeding with, there was a big distance. She got two times cardiac arrest. I resuscitate. Then we could go, we continue. Second time, then she passed away.’ #24, NPAP

Lack of pay •‘Why not having my own car …. You have to be a doctor who is … there is rain, there is a patient who is critically ill but I'm stopped by the rain. It's no sense.’ #54, anaesthesiologist •‘Only teaching hospital they pay double than ours …. Those who work there, they have studied the same, we are in the same class. And they were best performer than then, but we are paid less than them … they get twice than my salary.’ #18, non-physician anaesthetist •‘You find that you don't have the money or are struggling food’ #36, general surgical resident

Medical culture •‘Sometime they really treat you in a very harsh way. Yeah, we all know that in medicine it's like military so you know, the superior has … he has all the rights.’ #49, GP •‘We have protocols and everything hang on the walls but they are not really being used. You know, for example …. If you have a checklist with from the WHO, for example, they are saying every teamwork should be introduced their selves but sometimes you can be inducing. No surgeon around. No nurse around which is really embarrassing.’ #28, anaesthesia resident •‘He just insulted me like used some words after five days without sleeping …. I called him for a patient to operate during the night at like three am. And when he came … I think he was maybe sleeping when I called. He came angry and said like, “Are you mad” …. Like he was not happy with maybe the way I called him during the night, I don't know. It was an extreme emergency.’ #40, surgeon •‘If we would decide to delay the case then to discuss with the senior, sometimes [the obstetricians] get angry.’ #3, NPAP

Lack of training •‘As we are GP, we are not able because of not enough knowledge, to help them …. Sometime people died, pregnant woman died during Caesarean section and so another reason why we said no, we have to learn to develop our knowledge.’ #30, anaesthesia resident •‘That was my frustration too. Because in my perception I had to do what could be done but the frustration was coming I have no training. No guidance. So what am I going to do during the night’ #28, anaesthesia resident

Poor outcomes •‘Because the woman that comes for delivery, actually she is … we, for us, we don't say she's sick. She come here and at home they are preparing for party to receive the newborn and suddenly you may have the fetus dead or other complications can come and yet they have been preparing for dancing and enjoying everything So that makes obstetrician very scared, stressed and even … so the number of patients, the expectations' #41, obstetrician •‘So when I was called by that doctor from another hospital, because we have another hospital nearby there. But it is some … like two hours from our hospital …. When the doctor called me, he called me saying that she has a patient with postpartum haemorrhage but it was post Caesarean section … the patient reached the hospital four hours later. What … what I saw, the patient was already dead, almost dead … There was no … okay, a heartbeat but there's no pulse, there's no blood pressure …. She was intubated and they tried to do what I have to do. But so almost dead. The patient passed away in like 20 minutes.’ #37, obstetrician •‘That mother would have been saved. But then the mother reaches the referral unit, dies on arrival. They start CPR but without success. So that literally for me I sit down and cry.’ # 51, obstetrician

Contextual issues/post-conflict society •Interviewer: ‘Do you still think about that?’ Participant: ‘You can't stop to think about that.’ #23, GP •‘I say ah, okay. I am a survivor. Survivor. Have to do something …. Take strength from that.’ #20, anaesthesiologist •‘I lost my mom when I was 7, which is difficult …. And also my father was abroad and as with my brother and my sister, three. It's a long story but it have affected me in different areas. I can remember I did primary school and I was with my sister only in the house because my brother was in secondary school. He has to stay in school. You can imagine two young children living alone, study alone, it's terrible. I can say I have lost probably lots of opportunities because of this tragedy and also because of lack of parents' advice. But always I can say thanks God because I am still living. Maybe I have got also some inspiration because if you face difficult condition, maybe you can think more and more.’ #31, anaesthesia resident

The negative effects of workplace-based stress on the healthcare professional

Without detailing formal criteria for burnout, or referencing Maslach's three-factor model, most participants noted a phenomenon that was work related, resulting in negative emotional, physical, and social consequences and ultimately affecting performance, work, and patient outcomes (Table 3). This was noted to be either a chronic phenomenon, or an acute crisis, or perhaps most often an ‘acute-on-chronic’ phenomenon with crises occurring on the background of chronic exhaustion. Some participants noted the interaction of this phenomenon with other separate mental health issues in colleagues, including depression, bipolar disorder, and post-traumatic stress disorder. Often, participants noted that they had not expected the poor quality of life or work-life balance when they chose a career in healthcare and had anticipated a better salary and more time to spend with family. They generally noted they would dissuade their own family members from following in their footsteps.

Table 3.

Physical and emotional responses to workplace stress.

Physical exhaustion •‘5pm [the doctor goes to] another hospital and they're getting fatigue and they fall down’ #46, NPAP •‘You can sometimes be dizzy and sometimes can faint. That's why, because of the stress.’ #50, midwife •‘You don't even have time to drink water’ #24, NPAP •‘I have a colleague who fainted in theatre. Yes. Fainted in the theatre. Woke up some minutes later. Continued operations.’ #35, obstetrics resident •‘I was so much exhausted but I felt like I almost lost a patient and well, in the morning report, it was …. So other people have their brain working, functioning normally but mine is compromised by the busy night, so and they are requesting for more information that I am not able to generate. I could see they asking me good questions but I was failing to generate the answers to them’ #34, obstetrics resident

Substance abuse •‘[A sole practitioner] anaesthetist … unfortunately now died. But he used it even up to some point of opioid abuse so that he can sleep’ #25, registered nurse •‘Maybe addicted to alcohol and drugs.’ #5, NPAP •‘I said “My friend, I look you all the time. When you come on duty night, you like somebody who took alcohol. What's wrong?” And he [a GP] told me, “I can't do night or do operation, surgical operation in night if I didn't take alcohol.”’ #30, anaesthesia resident

Negative emotions •‘You can care for emergency case during the work and then you do all you can and then finally you transfer the patient to tertiary care and then when you are at home you still keep thinking if the patient reached, they are healthy, still alive, or the patient died in the midway or not. So you keep thinking about it at home.’ #47, midwife •‘It's really scary and stressful and tiresome’ #24, NPAP •‘Stressed … nervous … not able to control himself’ #7 anaesthesiologist •‘Discouragement or depression or some anxiety related mostly to the work’ #9 anaesthesiologist

Hopelessness •‘Most people come and tell me “Doctor, I don't know what else to do. I am … I feel I am done. I don't know even why I entered into medicine”’ #51, obstetrician

Lack of empathy •‘the family said they don't have money. I was frustrated, our patient died … you may become less sensitive to people who are sick … you feel you are not sympathetic’ #54, anaesthesiologist

Supressing emotions •‘because in our culture, we have to hide it up. Like everybody has to be hard to burnout … burnout sounds like more a weakness’ #49, GP

Many participants described ‘hidden’ burnout, noting that in Rwandan culture this may be seen as a weakness and that hiding negative responses to workplace stress is normalised. Participants also described the assumption that stress is just part of being a doctor.

‘I think it’s also a problem in that people who are burnt out, they cannot … it’s very hard for them to express it out, like to say it out because in our culture, we have to hide it up. Like everybody has to be hard to burnout … Very difficult to deal with. Because if they can’t say it out, how can you deal with it.’

#49, general practitioner (GP)

A small proportion of participants stated either that they had never heard of burnout, or that it was not possible to experience this phenomenon in Rwanda, or even that it was not possible that it could be a significant problem in Rwanda as it had not been identified as such by the Rwandan government. Some participants noted that burnout as a phenomenon was not well defined in the Rwandan context.

‘Yes it's a problem in Rwanda. It’s a problem in Rwanda even if it’s not … it’s not talked about … there's no research done about burnout. They … people know about burnout but it’s not debated.’ #27, obstetrician

Fatigue and physical response

Participants reported fatigue and physical strain in response to these stressors, to the point of exhaustion and literally falling down, fainting, or appearing ‘drunk’ with tiredness (#52, surgeon). A resident reported always being tired and never awaking rested. Other physical sequelae reported include hunger, headaches, vertigo, insomnia, hypertension, chronic backache, and the perception of hypoglycaemia and chronic gastritis. These physical symptoms were often described as getting better if the healthcare worker was allowed sufficient time away to rest.

‘When we start this job it’s an opportunity of opening your eyes and opening your mind …. When you take this long time in this job, like 10 years …. Your back start to bend … to come in deformity … your mind starts to be … I say to be depleted … your reasoning starts to be destroyed’ #21, registered nurse

Emotional response

Participants reported a wide range of negative emotional responses to these stressors, including anger, frustration, becoming easily irritated, disappointment, anxiety, shock, low mood, boredom, emptiness, and guilt. Some participants only described positive emotions relating to work, and a registered nurse, rural southern province (#17) described feeling ‘courageous and happy to come to serve people’. A general surgical resident (#33) in Kigali, said that when starting a 48-h shift he felt like he was ‘preparing [him]self for a fight’. An obstetrician who had tried to care for two patients simultaneously, and one had died while the caring for the other, said:

‘You feel like you are worthless. So you think maybe I did not, maybe I should have done this. Maybe I should have delayed that one. Maybe I should have left this area then go to the other one.’ #29, obstetrician

There were often descriptions of moral injury, for instance from an obstetrician at a referral centre who received a woman who was experiencing peri-arrest from bleeding, after the referring hospital could not manage the patient and sent them 4 h on the road. A NPAP described the road being so bad on a transfer to another hospital that the patient was extubated because of potholes in the road while the NPAP was manually ventilating the patient. Experiencing these adverse outcomes and knowing what the correct management should have been, if resources were available, appeared to violate their own values and ideas about how a healthcare system should function.

‘We should be at least two. Having someone on call …. If someone is going to take 10/20 minutes to arrive, if you are called in, you are asleep, so you get up, you put on something and then you come straight, it will take you like 10/15 minutes. So someone who is not breathing, someone who failed intubation, so it’s really stressful … when it happens, that’s death to the patient.’ #24, NPAP

For some participants, the negative responses to stress were the norm in the workplace:

Interviewer: ‘In one week, how many good days do you have and how many bad days to do you have?’ Participant: ‘It depends on the week maybe; it depends on the week but maybe sometimes we have like bad days … bad days are the commonest.’ #24, NPAP

Social response

Some participants noted that workplace stress affected relationships family, friends, and the wider community outside of work.

‘You can bring it into the family’ #50, midwife

‘The social consequences, you can reach the point where you can’t … you can’t be able to produce … for in the community, for your family’ #27, obstetrician

Doctors often reported long travel time between the family home and the hospital, and participants would often describe having to live away from family to work.

Downstream effects on participants and their patients (Table 4)

Table 4.

Downstream effects of workplace stress.

Leaving profession •‘I [am] tired and I don't have no one to replace me. I feel like it is very difficult and I would want to quit.’ #46, non-physician anaesthesia provider (NPAP) •‘Because we see people leaving their jobs. They leave a medical career and go for other jobs and the jobs that they seek are not necessarily paying much compared to the work they have been doing before. But they prefer leaving because of stress, overwork and the stress in the medical career. So they prefer leaving for other jobs, not necessarily because there are other jobs they are seeking that are paying much money, but because they feel overwhelmed by it.’ #41, obstetrician •‘And people are leaving the job mostly because of the burnout. What I have seen. You see you are working but no one is taking care of your health …. Resident, they leave. Intern they leave. GP they leave. Consultant leaves. Even the other non-physicians. Like nurses, anaesthetists, laboratory. Everyone. They leave because of just this kind of overworking’ #36, general surgery resident

Home life •‘For sure as a parent it was not good … they think I am not giving them enough time to care for them.’ #37, obstetrician •‘Because even in the weekend when I am at home, I feel very tired. Even to go for social events, it's not for me, yes. I don't … I don't participate’ #27, obstetrician

Financial issues •‘A house, it's … we may do a lot of mathematics to find one. It's … it's not easy.’ #54, anaesthesiologist •‘You have to rent a house, transport. You know, for other, oh, I have children. Sometimes they don't have … they are not able to pay school fees for their children’ #36, general surgery resident

Motivation •‘A nurse, like you see him not caring for the patient. You see him like may … he or she or he may leave in your room and go like at home without telling you where he is going.’ #29, anaesthesia resident •‘Their performance decreases. They are no longer working at best to do before. And they even give up their job’ #29, obstetrician

Patient care •‘When they call me, sometimes I can't give, I can't give help for them because I am alone’ #46, NPAP •‘She can behave cynical or harsh, harsh behaviours towards patient’ #5, NPAP •‘And you see he's not much interested in the patient. He's like he's not worried now anymore, he's not … the critical condition of the patient is not … sorry, he's not his business anymore’ #1, registered nurse •‘And the patient is experiencing pain andyou see, the doctor, he doesn't care, he doesn't care about that patient who is in pain. It's not because they don't want to understand, but his mind is absent. He doesn't feel what the patient feels.’ #11, registered nurse

Participants described the opportunity cost of many hours of unpaid work—of time they could be spending resting with their family or earning income by other means. Stresses, and especially financial stresses, resulted in doctors seeking opportunities for work outside of Rwanda, for example, working for international non-governmental organisations (NGOs). An anaesthetist said:

‘One day I got a very low salary and I said, What can I do with this? This cannot pay the house. I have to pay the school fees for my wife. AndI had to live, at that hospital, of course I had to eat …. The opportunities that was there was to go to the mission, outside mission.’ #54, anaesthetist

Financial stresses also resulted in healthcare professionals seeking work in private clinics outside of their government practice, further worsening work-life balance, a different source of stress.

‘In Rwanda we have a proverb that says when you have not … when you can’t have milk from a calf with the right hand … use the left hand (laughs). So for certain … I am thinking … how can I extract the milk with the left hand … maybe find another contract … and have maybe an additional income.’ #20, anaesthetist

Participants also worked outside of their medical practice, in small businesses, to raise sufficient income to support their family, either as a side-business or leaving healthcare altogether.

Other consequences of workplace stressors included being rude, aggressive, even abusive, with both patients and colleagues and a lack of motivation for work. Participants described negative effects to patient care, including lack of concentration, impaired judgment and clinical decision-making, and errors such as giving the wrong medication to patients. Participants described providing lower quality care resulting in poor patient outcomes. Some described simply being unable to work altogether.

Sources of resilience

Although most of the reported causes of work-related stress cited by participants were systemic, when asked about potential solutions, most described solutions at the level of the individual—learning to cope with burnout, talking through things with colleagues, or having short breaks away. Many participants noted self-care or social activities to relieve stress, including spending time with family, sport, music, watching films/television, or simply rest. Many participants mentioned the value of their religious faith to cope with stress.

Almost universally, participants described resilience as coming from the good days at work, when they were able to take care of patients to the best of their abilities, without complications or negative outcomes. An obstetrician described hearing the first cry of a healthy baby as the best part of medicine, what they were expecting from their skills and knowledge. However, many participants described having more bad days than good. Participants often cited finding strength due to a sense of duty and necessity.

‘So with that I know if I can’t do it, there is no other one to do it. I have to be strong.’ #33, surgical resident

‘The only way to cope with that is to feel like it is my responsibility. It’s my duty to be there.’ #9, anaesthetist

For further context, some participants described lifelong hardship, including growing up in poverty. Some participants noted the context of living in a post-conflict society. This was both listed as a source of resilience and of an example of ongoing personal stress. Some participants also spoke of growing up as refugees in neighbouring countries before the genocide, where they were treated poorly, and only being able to return to Rwanda after the end of the genocide.

‘On my father’s side … my mother’s side. Many people were killed. So and I was left with widows, orphans … and you have to support this one, this one, and now you …. I don’t have no money actually. No one believes you. No one believes you.’ #20, anaesthetist

Participants often noted their backgrounds of poverty and that having been successful from such a background provided motivation to continue.

‘I have resiliency from my background … from … hardships of life. Those hardships actually, they do not affect me negatively. I hope that they affect me and give me energy to endure. Not to give up.’ #35, obstetric resident

‘I was not even dreaming of becoming a doctor because we suffered. I and my young brother, because we didn’t even think of having sugar at home or salt because it was really difficult to find work to eat. And then really we could not dream about that … you need to fight for your future’ #28, anaesthesia resident

Participants noted getting resilience from trying to improve their healthcare system, and also informal discussions with colleagues that were reassuring.

Although participants often described their lived experience as very different from how they expected their work would be, many were resigned to ongoing workplace stresses, feeling that there is little than can be done to improve the situation. Some participants noted that the job is simply difficult, and not for everyone. Generally, participants reported feeling conflicted about how they felt about their job: participants appreciated having a challenging job that was often rewarding, but felt that the demands on them were just too high and most felt that it was not worth it.

Discussion

Our data included diverse lived experiences of workplace-based stresses that is fundamentally different to more well-resourced contexts and not comprehensively described by the Maslach three-factor model. Although the workplace stressors listed may seem to be similar to those described in the literature from high-income countries such as overwork, lack of financial and other forms of recognition, and lack of support, the degree of overwork was such that it is essentially a different phenomenon. For example, it is difficult to imagine a specialist hospital doctor in a high-income country not to be able to afford adequate housing or nutrition. Other stressors were different to those from the high-income country literature, including lack of access to essential drugs and equipment and working as a sole physician in a remote area, being unable to take leave in some cases for years on end, and being on call 24 h a day for years at a time. Participants commonly reported extreme fatigue and varied psychological responses, including resilience, denying the existence of burnout in Rwanda, moral injury, and emotional exhaustion. Downstream effects included a negative impact on home life, disengagement with work, and being pushed to work outside of medicine, or Rwanda, or the public sector.

Our findings are consistent with other qualitative data from sub-Saharan Africa. In particular, low staffing and an overwhelming number of patients appears central to occupational stress.34, 35, 36 Data from Malawi suggest this may have become worse during the COVID-19 pandemic.³⁷ Other factors such as culture, training, political factors, and workplace conditions seem to be very contextually specific. We did not find comparable literature on the lived experience of work-related stress in the region, with other studies tending to focus on causative factors.

Of the three factors in the Maslach model of burnout, descriptions aligned with ‘emotional exhaustion’ were common, although participants more often described physical fatigue or exhaustion. The factor of ‘lack of personal accomplishment’ is intrinsically linked to expectations for personal goals and this is inseparable from contextual and cultural factors. Our participants often described perspective on accomplishment that was more collectivist that individualist, and this should be considered when applying Maslach's three-factor model to societies often referred to as ‘non-Western’. In Rwandan culture, social cohesion and patriotism are considered very important because as they have provided stability and have often supported economic growth and prosperity. In addition to descriptions of decreased personal accomplishment, it was common for participants to detail both progress and ongoing frequent failures across the healthcare system. Participant's ambitions were often framed by narratives of Rwanda's recovery after the genocide and in particular investment in developing the Rwandan healthcare system.38, 39, 40, 41 The overlap of personal goals and national strategy was cited as a source for hope, patience, and resilience but also as a cause of frustration when policies seemed ineffective or even counterproductive or wasteful. The third factor, variously characterised as ‘depersonalisation’ or ‘cynicism’, although cited by some participants, was less commonly described than the other two factors. Relationships with patients and their emotional responses were diverse, from describing experiencing joy in making a difference in patient's lives to becoming frustrated, impatient, and cynical towards patients. Further responses ranged from moral injury in response to poor patient outcomes, to an expectation that patients tolerate the reality of the Rwandan healthcare system, and even personal anxiety about the legal repercussions from negative outcomes. Depersonalisation can be considered a facet of the complex ways that healthcare professionals view their patients and form relationships with them. It seems plausible that these relationships are context sensitive and future research may be required to further explore this complexity in the Rwandan context. Alternatively, infrequent reporting of depersonalisation may be related to the context of an interview led by an interviewer from outside the Rwandan culture, and simply a limitation of our study.

Although burnout as a theoretical model seems useful for a Rwandan healthcare context, the three factors are clearly not the only issues to consider when analysing the effect of workplace-based stressors in this context. Neither do our data support a premise that these three factors should be considered with equal weighting, nor that they are experienced the same way in Rwanda as in a North American or European setting. We suggest a re-framing of these factors for the Rwandan context. Firstly, that physical exhaustion seems to commonly co-exist with emotional exhaustion, was often reported as a more prominent concern, and should be considered when burnout is examined in Rwanda. Physical exhaustion in this case was the result of extreme under-staffing conditions that were unparalleled by those experienced in most high-income settings. Secondly, that lack of accomplishment of teams and their collective goals across the healthcare system, with resulting moral injury, may be as or more important as individual goals, and measurement of burnout may need to consider new multi-level models including ‘collective burnout’.^42,43 Thirdly, ‘downstream’ effects go beyond cynicism or depersonalisation and include problematic relationships with other staff, often caused by understaffing. Finally, financial stresses contribute to a lack of motivation to work and feed the intention to leave the profession or the public sector, which may not be captured by the Mashlach Burnout Inventory - Human Services Survey (MBI-HSS). These findings may also be relevant for other contexts, but further research is required to examine them, as well as to evaluate instruments that measure ‘burnout’ with a collective/multi-level model, and that consider the broader downstream impact. The extent that this is generalisable to other highly collectivist cultures should be a research priority.

The strengths of our study include broad representation of participants across professions and geography, a diverse collaborative research team, and adding to the dominant theoretical model for burnout. Our study also has limitations. We had limited numbers of participants from the mid-level referral hospitals. Although English is a national language, it is not the first language for most Rwandans and some interviews, or parts of interviews were in Kinyarwanda. This was more of an issue for nurses and midwives than for physicians. Some participants may not have felt free to speak openly in a relatively short interview with a stranger about some topics that are either controversial or relate to personal trauma. Some nuanced perspectives to burnout or other issues may have been lost in translation. We note that many participants told us that Rwandan culture generally prefers not to discuss negative responses to workplace stressors that may be seen as a weakness. As a qualitative study, these findings are not necessarily generalisable to other jurisdictions, or even other groups of healthcare professionals in Rwanda. Our data were collected over a relatively short period of time, and many important factors may change quickly due to either development of the healthcare system, or new challenges including the COVID-19 pandemic. We only collected data from public hospitals, and this may not be transferrable to private healthcare within Rwanda. Finally, this study was not a formal or psychometric evaluation of the MBI in the Rwandan context and is intended to be hypothesis generating in terms of future approaches to evaluating fatigue and burnout.

In conclusion, the Rwandan perioperative healthcare system presents many challenges which can become profoundly stressful for a chronically understaffed health workforce. Physical fatigue and exhaustion in an East African context are under-represented in the literature. A consideration of the effect of reduced collective as well as personal accomplishment, of moral injury, and of diverse downstream effects on the whole healthcare system may better represent the true status and costs of burnout in Rwandan healthcare professionals. It seems likely that supporting the health of the workforce would require a significant investment in staffing and working conditions.

Authors’ contributions

Principal investigators: MDB, ET

Protocol development: MDB, ET, CK, MC, MG, CR

Data collection: MDB, CC

Data analysis: CR, MDB, ET, LW, LY

Writing the manuscript: MDB,

Critical review of the manuscript: MDB, ET, CK, MC, MG, CC, CR

All authors read and approved the final manuscript.

Acknowledgements

Thanks to Heather O'Reilly for input on the protocol and early stages of analysis.

Declaration of interest

The authors declare that they have no conflicts of interest.

Funding

University of Ottawa Department of Anesthesiology and Pain Medicine Research Operating Grant.

Handling Editor: Rupert Pearse

Appendix A. Supplementary data

The following are the Supplementary data to this article: