Exploring the space for task shifting to support nursing on neonatal wards in Kenyan public hospitals

Sustainable Development Goal 3.2 calls for a reduction in neonatal mortality to at least 12 per 1 000 live births [1]. High impact, low cost interventions could avert more than 71% of neonatal deaths, but depend to a large extent on effective coverage with facility-based care [2]. However, the performance of public sector hospitals in low- and middle-income countries (LMICs) is often poor [3, 4], and this is particularly the case for newborn care [5, 6]. Some of the contributing factors are inadequate material resources and equipment, poor adherence to evidence-based guidelines, inadequate human resources for health (HRH) and poor management of material and human resources [6–8].

Nurses are key to the provision of quality care and have a particular influence on newborn health outcomes in hospitals as primary care givers to this highly dependent group [9]. However, according to the Global Health Workforce Alliance (GHWA) 2014 report ‘A Universal Truth’, the global shortage of health workers is estimated to be over seven million [10]. Kenya has an acute shortage of nurses in the public sector, with densities ranging between 1.2 and 0.008 per 1 000 population across counties compared to a new suggested minimum health workforce threshold of 4.45/1000 population for doctors, nurses and midwives combined [11]. Nairobi county, the specific focus of our study, suffers from both major shortages of nurses providing frontline neonatal care in the public sector and very poor neonatal outcomes [12]. Indeed, in Nairobi’s public hospitals, recent work suggests nurse: baby ratios of 1:15 [13]. In countries such as the United Kingdom, even for babies who do not require intensive care, guidelines suggest one nurse for every two to four sick babies [14, 15] with evidence suggesting a relationship between lower nurse ratios and higher mortality [14]. Kenya’s public-sector nursing workforce challenge is however complex. Recent data reveals the country has more than 50 000 nurses registered to practice, but fewer than 17 000 offering care in the public sector, which is particularly relied upon by the poor for inpatient care because of inadequacies in public finance [16]. Workforce solutions must therefore carefully consider budget impacts.

The World Health Organization (WHO) suggests task shifting and sharing as a means to lessen the problems of HRH shortages, while potentially improving access and maintaining or improving quality [17]. ‘Task shifting’ is a phrase used to cover a variety of interventions, but WHO uses it to group activities in which: ‘…trained cadres who do not normally have competencies for specific tasks deliver them and thereby increase levels of health care access’ [18]. The related term ‘task sharing’ is described as ‘the rational distribution of tasks among trained and supervised health professionals and health workers’. In this study, we use the term task shifting although in the context under study tasks that are shifted might still be supervised by a nurse.

In most cases, task shifting and task sharing aim to provide services at reasonable cost through a new cadre of worker when an absolute shortage of staff puts unreasonable demands on existing cadres who are unable to meet requirements [19]. A series of studies and systematic reviews intended to inform WHO’s ‘Recommendations for Optimizing Health Worker Roles to Improve Access to key Maternal and Newborn Health Interventions through Task Shifting’ (OPTIMIZEMNH) (WHO [18]) have reported ‘barriers’ such as training and supervision challenges, problems with professional hierarchies and poor integration of new cadres into formal health systems [20, 21]. Reported facilitators of task shifting included public recognition, creation of visible ties to the formal system where formal health workers were involved in training new cadres [22, 23]. However, Mijovic et al. [24] highlight that in all the African case studies they reviewed, the cadre taking up additional shifted tasks exceeds the formal mandate, taking on responsibilities additional to those ‘shifted’. Hence, while the financial case for task shifting is clear [25], implementation is difficult and the intervention ought not be thought of as a ‘cure-all’.

Task sharing and shifting have been used extensively in LMIC contexts in response to HIV/AIDs and in anaesthetic and surgical care where tasks that are traditionally the preserve of trained physicians are performed by non-physician clinicians or nurses [26]. Task shifting and task sharing is also now common in high-income countries’ hospitals where health care assistants undertake many ‘basic’ nursing tasks under the supervision of nurses [27]. Historically, Kenya had a cadre of non-professional assistants referred to as ‘nurse aides’. Our respondents confirmed that this position was phased out in the early 1990s having been linked to reports nurse aides were over-stepping role boundaries (typically when operating without supervision) and thus putting patients at risk. This might partly explain reluctance to formally re-introduce such a cadre [28]. Anecdotal evidence indicates, however, that shifting of nursing tasks to non-qualified personnel is happening although not officially allowed in the Kenyan public sector.

Our work aimed to explore the potential for task shifting to support the provision of basic nursing care in hospitals’ neonatal units. In 2017, the Kenyan Ministry of Health launched the Task Sharing Policy 2017–2030 and the Task Sharing Policy Guidelines [29]. These policies focused on legitimating task-shifting between existing health professionals by redefining scopes of practice to reflect the realities of routine work and its evolution over more than 20 years (e.g. nurses run primary care, prescribe and put up intravenous fluids in clinical settings). The policy, however, makes no mention of new hospital-based cadres to support basic inpatient nursing care.

Prior task shifting and sharing approaches in LMICs highlight the importance of well-designed interventions linking newly established cadres to existing professionals at the micro level, and their integration into formal structures of health systems [20–22, 30, 31]. Relatedly, De Sardan [32] describes the disconnect resulting from implementing standardized interventions without consideration of everyday contexts. In particular, he describes ‘practical norms’ as ‘the various informal, de facto, tacit or latent norms that underlie the practices of actors, which diverge from the official norms (or social norms)’ ([33]: 26). It is important then to investigate the socio-cultural dimensions or ‘software’ of health systems, including practical norms, as part of any potential task-shifting design efforts [34, 35].

This study aimed to explore the current operation of Nairobi’s New Born Units (NBUs) using an ethnographic approach. It is part of a body of work exploring major gaps in quality of neonatal care and potential solutions, including task shifting. The study reveals nursing routines, highlights areas of working practice that might easily be shared with a lower cadre, describes nursing stakeholders and nurses’ perceptions of task-shifting for this context and explores the overall potential for task shifting in Nairobi’s public hospital NBUs.

This work first started with an exploration of the views of influential nursing leaders on the idea of task shifting as a possible intervention to improve in-patient care for sick newborns. The results of the first phase were complemented by a formal stakeholder analysis [28] and informed empirical work in the second phase of the research. Phase two explored how nursing tasks are performed within the busy, highly normative culture of Kenyan newborn nursing.

Our work sought to gain an understanding of how neonatal care is organized and structured [40] on a daily basis and what opportunities exists for task shifting. We believe that the professional cultures, occupational jurisdictions and de facto templates of care of frontline workers will directly influence the success or failure of task-shifting interventions. We have attempted to reveal these contextual factors so that the design of any task-shifting intervention is more considered and hence more likely to succeed.

As there were no explicit job descriptions and standard work guidelines for routine newborn work, the working model of nursing was normatively formed over many years in response to the practical realities of the environment. Hospitals are replete with organizational timetables and schedules [41], but to the nurses, this structuring offered limited guidance on how to organize and delegate certain tasks. Therefore, as a way of coping with task ambiguity, nurses developed de facto routines that provided direction on what they should be doing at any given time (see Fig. 1). However, because of the unpredictable nature of events in these settings [42], the routines would often be disrupted, and nurses would find themselves quickly having to make difficult decisions. In such situations, nurses prioritized nursing tasks based on patient needs using ‘subconscious triage’, which often included delegating and shifting tasks to others.

Delegation of tasks was, however, ad hoc as support staff, students and mothers frequently worked without supervision, and none seemed concerned about undertaking tasks that they were not trained for. This contrasted with findings of an expert meeting conducted as part of our broader programme of work [39] where attendees firmly expressed consensus on tasks that should be formally conducted by nurses and on tasks where delegation might be allowable.

From a policy perspective, our insights suggest that there is potential space for formal task shifting within the everyday routines of neonatal nursing in Kenya. We suggest that careful study of context can help inform task-shifting design. When asked formally about the prospects for task-shifting, the nurses we spoke to had doubts and concerns, but their near continual reliance on organic forms of task shifting suggest that a carefully designed programme could be accepted. Kessler, Heron and Dopson [27] reveal in the United Kingdom health context that post-implementation, health care assistant roles are nearly always viewed very positively by nurses, patients and the new cadres themselves. They do however warn that the specific usefulness of these new human resources is rarely well understood, and hence, they are not strategically deployed despite their obvious potential in addressing HRH needs. Our findings also indicate that considerable work may be required at the policy level, dominated by professional institutions, experts and senior managers, to help bridge the apparent gap in acceptance of task shifting in this clinical arena.

Despite delegating many ‘nursing tasks’ to others, nurses maintained their distinctive role and power by authorizing to whom and how delegation of tasks was orchestrated in their wards. Their authority was mainly drawn from physically spending most of their time in the NBU and being constantly engaged in direct patient care, in contrast with the other more medical cadres. Nurses experiential knowledge of the wards, and what works and does not work, consequently authored the patterns of social organization and culture of work within the ward [43]. These are reinforced in the NBUs by the routines nurses keep. Routines and rituals exist to serve different needs [44, 45], and in our case, a template of routines helped nurses mentally manage tasks within shifts. They offer nurses a sense of control and accomplishment within chaotic work environments characterized by limited resources and staff shortages that make completion of all tasks, as the experts and seniors would have them conducted, an impossibility.

The description of organic practices and local norms established in response to work pressures provides useful guidance for the design of task-shifting initiatives [22]. For instance, we see that nurses prioritize their more technical roles. This inadvertently moves them further away from their traditional roles of providing bedside care and consequently helps delineate potentially acceptable nursing and task-shifting roles [46, 47]. To achieve multi-disciplinary holistic care, it is, however, important that professional role boundaries are negotiable and occupational jurisdictions remain flexible [43, 48], but the starting point for such endeavours is ensuring that roles are defined, that scopes of practice are understood, and that task allocations match capabilities.

Past studies in LMICs have described how lay health workers’ (LHWs) credibility is enhanced through visible ties to the health system through, among other things, visible contact with health professionals through referrals, supervisory visits and involving health professionals in training LHWs [20, 21]. In short, preliminary evidence suggests lay health workers are well-liked by patients. Health systems are however complex, dynamic and political systems, and this is particularly true in Kenya after the recent devolution, which has exacerbated human resources management challenges including disruptions, delays, and discrepancies in health workers’ salaries; resulted in political interference and discrimination in HRH management; and prompted frequent industrial actions by health workers [49]. The working environment of health workers across health facilities in the country has been characterized by fear, anxiety, mass resignations and low health worker morale. Many of these issues would affect a new cadre of health workers as they do the existing cadres. Challenges of health worker shortages and human HRH management in Kenya are chronic and largely remain unaddressed, and it is important that stakeholders do not perceive task shifting as a ‘fix all’ solution.

In summary, our work described a routine template of neonatal nursing work which nurses used to enhance predictability when working in chaotic, resource-limited work environments. We detailed how this model of care deprioritizes less technical or less clinical tasks, which are then left to students, parents or other staff through ‘organic’ task shifting. Such insights are critical when considering more formal task-shifting projects as they help to delineate the tasks most easily shifted, and how best to support and work with busy nurses. While the profound effects of existing health system challenges (e.g. devolution, frequent strikes, delayed salaries and complexities around pay on hospital operations) on human resources management must also be addressed, our findings suggest there is space for a well-designed task-shifting programme that could have a positive and supportive effect on nursing care and on nurses themselves.