Implementing a change process to support respectful communication in newborn units in Kenya: A qualitative study

Abstract

Background

In contexts of high workloads, resource shortages, and environmental inadequacies, neonatal nurses in many low- and middle-income countries face high levels of stress and burnout, as well as significant communication challenges. Advancing care in these ‘on the edge’ emotionally fraught contexts requires a multi-layered systems approach, including context-sensitive courses that support respectful communication.

Methods

We share our learning from the implementation of a communication and emotional competence training process in two public hospital newborn units in Kenya. We adapted an existing course in collaboration with newborn unit nurse managers, with the aim of enhancing relationships among staff and with parents, as well as improving the quality of care. We drew on course data, 60 semi-structured interviews, and 70 h of post-course observations in newborn units. We developed a theory of change to inform our research. Data were analyzed thematically using NVivo version 12 software.

Findings

Even in these highly challenging contexts, we documented positive effects on interactions with parents and colleagues, staff well-being, and management processes. However, there were significant differences between the two hospitals, and some unintended consequences, linked to relational readiness and capacity to create and maintain safe spaces. Key gains across both hospitals were a better awareness of communication behaviors and emotions, greater self-efficacy, and a reignited sense of professional values. Positive signs for spreading and sustaining gains included role-modeling new norms and incorporating modules into wider initiatives.

Conclusion

There is potential value in incorporating participatory training processes into wider change initiatives. Nevertheless, given the massive workforce gaps and persistent space and resource challenges that undermine staff well-being and contribute to parental distress, tackling resource shortages remains essential.

What is already known

• •Neonatal nurses in low-resource settings face high levels of stress and burnout and significant communication challenges.
• •Advancing respectful communication in these settings requires multi-layered approaches, including context-sensitive training courses.
• •There is limited literature on developing and implementing nurse training in communication and management of emotions in low-resource settings.

What this paper adds

• •Implementing a participatory communication and emotional competence training process in Kenyan newborn units is feasible.
• •Relational readiness and capacity to create and maintain safe spaces in training processes are important.
• •Initiatives such as this should complement rather than detract from efforts to address massive underlying workforce and resource gaps.

1. Background

Neonatal mortality accounts for 45 % or more of all mortality under five years of age, making the reduction of neonatal mortality by 2030 a global priority (Sustainable Development Goal 3.2). Major expansion in the provision of high-quality, facility-based newborn unit care is required to achieve this in low and middle-income countries (Ashorn et al., 2023; Kruk et al., 2018). However, in these settings, newborn units often have low staffing levels, high workloads, resource shortages, environmental inadequacies (such as crowding and lack of privacy), and high mortality rates. In these high-stress contexts, staff are expected to provide medical interventions, careful monitoring, infection prevention, and support feeding, among many other tasks (Nzinga et al., 2019). Staff should also foster open, respectful communication with patients' parents to support the delivery of compassionate care, and improve parent satisfaction, understanding, and adherence to therapy (Bry et al., 2016; Disch, 2012). This crucial nurse-parent communication is determined by nurses' emotional competence and by the work environment, including their relationships with colleagues and supervisors, workloads, and access to resources (Horwood et al., 2019; Larson et al., 2017). In low and middle-income countries, newborn units face particular challenges linked to the rapidly evolving health conditions of sick newborns and the emotional distress of patients' parents, while mothers themselves are often exhausted and anxious, physically unwell and in pain, and separated from their normal social networks (Oluoch et al., 2023). Indeed, even in well-resourced settings, some mothers describe the experience of having their baby admitted into the newborn unit as a devastating, traumatic, and life-altering event (Horwood et al., 2019).

1.1. Staff well-being and communication in newborn units in low- and middle-income countries

Interventions in these settings to advance newborn care are overwhelmingly focused on scaling up access to technologies and biomedical interventions (NEST360, 2025). However, we and others have documented high levels of stress and burnout, and significant communication challenges in these units, often characterized by major staff shortages, with one nurse caring for 10–30 sick babies (English et al., 2020; Kwame and Petrucka, 2020). In these settings, handling infant deaths, communicating bad news with families, and being reprimanded by supervisors in front of parents or colleagues are particularly emotionally distressing for nurses (Dyuti, 2023). Many nurses describe ‘lashing out’ in anger with colleagues and parents when they feel overwhelmed and exhausted, or withdrawing and providing only the very minimum care and less communication while experiencing a wide range of emotions, from professional pride and satisfaction through to devastation, heartache, and indifference (Dyuti, 2023; McKnight et al., 2020). In such circumstances, nurses preserve their identity as capable professionals through subconscious rationing to focus on biomedical interventions and avoid emotional work (Maben, 2008; McKnight et al., 2020; Ramsey et al., 2024). For parents of babies admitted to newborn units, this can be experienced as disinterest or disrespect (Oluoch et al., 2023). Nurses' actions in these challenging environments can reinforce the above communication challenges, further undermining staff motivation and teamwork, in what becomes a vicious circle. Although we have observed many acts of kindness, care, and mutual support among staff and between staff and parents (Dyuti, 2023). The predominant culture is typically far from the ideal of respectful, compassionate, and nurturing care, with profound implications for maternal and neonatal outcomes.

1.2. Reframing improvement as an organizational, systemic concern

Improving neonatal care at scale in low and middle-income countries demands more than technologies, with research suggesting a significant role for communication skills and emotional competence training for nurses and nurse managers (Kwame and Petrucka, 2020). However, any such initiative is intervening in a system that is ‘on the edge’; organizational deviance from expected standards has been normalized, and mistreatment of parents can be institutionalized (Okondo et al., 2022; Ramsey, 2022). Such a situation is enabled by system conditions of normalized scarcity at the macro-level, by meso-level actors struggling to achieve organizational goals with limited autonomy and resources, and by a regulatory environment that cannot adequately monitor, prevent, or address professional misconduct (Ndwiga et al., 2022; Ramsey, 2022; Ramsey et al., 2024).

While we would not necessarily characterize interactions between providers and parents of newborns as ‘institutionalized mistreatment,’ Ramsey's emergent organizational theory resonates with the potential organizational drivers of high reported burnout rates among staff in public sector newborn units and allied units in Kenya. Any intervention must therefore acknowledge the effects of resource scarcity and imbalanced power relations that could undermine change processes to strengthen the emotional well-being of staff. Staff in these environments may be suffering from moral distress or ‘moral injury’, where they know what care patients need but are unable to provide it due to constraints beyond their control (Dean et al., 2019; Ramsey et al., 2024). This framing, rather than ‘burnout’, is valuable in focusing the problem and potential solutions away from the individual (who might be seen to lack the skills, resources, or resilience to cope with the work environment) to the drivers of system challenges. It also emphasizes ‘creating a health care environment that …[values]… developing the trust, understanding and compassion essential to providing quality care’ (Dean et al., 2019).

1.3. Communication skills and emotional competence training

Given such complex layering of the drivers of emotional distress and communication challenges, any initiative aimed at advancing nurturing care by strengthening staff communication and emotional competence skills should be implemented as part of a multi-pronged individual and organizational process. Paradoxically, there is potential that ‘reawakening’ emotional awareness may undermine emotional coping strategies and increase distress (Taylor et al., 2024). Therefore, interventions must consider the potential for unintended negative consequences. These realities also suggest that initiatives should seek to build upon and reinforce leadership practices and coping strategies that strengthen self-awareness and efficacy of individuals, team relationships, interactions with parents, and quality of care (Nzinga et al., 2021; Witter et al., 2022).

In this paper, we share our learning from the implementation of a communication and emotional competence training process in two public hospitals' newborn units (and allied units – maternity units, postnatal wards) as part of work examining interventions targeting improvements in neonatal care in Kenya.

2. Study setting and methods

2.1. A multi-level strategy incorporating communication training

The ideas informing a wider, multi-level, large-scale change strategy targeting improvements in care across Kenyan newborn units have been described elsewhere (English et al., 2020). In brief, the change efforts build on an already established Clinical Information Network, now extending to 24 hospitals in Kenya. The Clinical Information Network aims to leverage collaboration between researchers, policymakers, pediatric professional association members, and hospital pediatric teams. The network generates and uses routine clinical data, audits, feedback, and peer interaction to promote the adoption of best practices in the form of improved health records, guidelines, and the use of recommended technologies in newborn units and pediatric hospital wards (English et al., 2020). Communication and emotional competence training was an additional specific change process introduced to improve trust within teams and relationships between staff and families. The training and evaluation were based on our emerging theory of change, developed through team meetings between PM, DO, SM, and the lead trainer (MB), with inputs from ME (Fig. 1). The focus was on the immediate outcomes, assumptions, and mechanisms to test the approach's potential for success concerning longer-term outcomes and impacts.

Fig. 1.

Theory of change: how the training process impacts outcomes.

The participatory training process was adapted by MB with Kenyan nurse managers from an existing course (the iCARE-Haaland Model) aimed at strengthening participants' awareness of their own, their colleagues, and their patients' emotional triggers and responses. (The course materials are available at TGHN, 2022.)

The training process began with a basic course for eighteen senior newborn unit nurse managers from an initial 14 Clinical Information Network hospitals. The nine-month training process was primarily on-the-job, directed self-observation and reflection work, with two skills-building workshops: the first halfway through and the second at the end of the process (see Box 1). The course was tailored to participants' workplace stresses and realities, drawing on their shared experiences and case studies. An early evaluation suggested that the process strengthened nurse managers' awareness of their own, their colleagues', and their patients' emotional triggers and responses, as well as improved relationships among staff and between staff and parents (Musitia et al., 2022). There was interest in spreading the learning to colleagues.

Box 1. iCARE-Haaland Model course for NBU nurse managers: March–November 2019.

Main components: a)an initial set of reflective tasks to become conscious of own communication behaviors to support focused periods of self-observation in managers' normal workplaces (March–May 2019). b)a three-day workshop, using managers' observations to illustrate communication theories; using interactive participatory methods to teach skills (June 2019). c)another set of reflective tasks in normal workplaces around putting skills learnt from the workshop into practice (June to Oct 2019); and d)a final three-day follow-up workshop to share realities and success stories (Nov 2019).

Alt-text: Box 1

2.2. Cascading the training to hospitals through a training of trainers process

To support sustainability and wider dissemination, six of the initial 18 participants were selected for a trainers course as part of the process of cascading out the training across their respective facilities. The trainers' course ran concurrently with implementing the iCARE courses in two newborn units, enabling mentee trainers to immediately apply and practice their newly acquired skills. The mentee trainers led the courses in the two facilities, following the basic process outlined in Box 1, with support from the training leads (MB and HD) as shown in Table 1, Table 2. Both newborn units had infrastructural challenges typical of the Clinical Information Network (Tuti et al., 2016). However, one was located in a more densely populated urban metropolitan area and was a generally busier environment, with an average 1:11 nurse-to-baby ratio (Hospital A) compared to a 1:8 (Hospital B).

Table 1.

Trainers involved in the training of trainers and the course rollout.

	Hospital A	Hospital B
Trainers involved	Newborn unit nurse manager: Registered Nurse, Higher Diploma in Paediatric Nursing, 13 years working in the Newborn unit, Certified Newborn Emergency Triage and Treatment Trainer, and Neonatal Essential Solution Technologies (NEST 360) mentor. Other mentee trainers: 1.Senior Registered Nurse with Higher Diploma in neonatal nursing, 14 years' experience in newborn unit; Newborn Emergency Triage and Treatment Trainer and Simulation trainer. 2.Senior Registered Nurse with 12 years of experience in the newborn unit. Support in planning and course delivery from lead trainers: MB and HD	Newborn nurse manager: Registered Nurse, Higher Diploma in Paediatric Nursing, 12 years working in the Newborn unit, Newborn Emergency Triage and Treatment Trainer, and Neonatal Essential Solution Technologies (NEST 360) mentor. Other mentee trainers: 1.Newborn unit nurse manager - Registered Nurse with a Higher Diploma in Paediatric Nursing; 17 years working in the Newborn Unit. 2.Newborn unit nurse manager; Registered Nurse; 13 years working in Newborn Unit; Newborn Emergency Triage and Treatment Trainer, and Neonatal Essential Solution Technologies (NEST 360) mentor. Support in planning and course delivery from the lead trainers: MB & HD

Table 2.

Pilot course rollout process and data collected.

Phases	Activities/content covered	Data collected
Phase 1: June-Sept 2021	a)One-day planning meeting to understand challenges and needs. b)Pre-training individual interviews. c)Self-observation and reflection tasks, on the job -Participants received weekly guided tasks on how to listen, ask questions, and how emotions influence their communication. Aim: To discover one's communication behavior patterns and become aware of the effects on others. These discoveries can trigger the inner motivation to take responsibility for changing how they communicate.	•Notes from the planning meeting. •Filled baseline questionnaires (36 participants) •Submitted significant stories of change by participants (34) •Interview transcripts (20)
3-day skills workshop (Sept 2021)	Skills training workshop tailored to participant needs and linking their self-observation and reflection to theories, using interactive participatory methods. Modules covered: 1.Introduction of the course, participants, and main concepts 2.Communication and learning (How do adults learn, Basic communication skills) 3.Understanding and handling emotions (dealing with anger, conflict, and stress, attitude and behavior change). 4.Using communication skills and emotional competence to educate and empower patients.	•Filled post-workshop open-ended self-administered questionnaires. •Workshop observation notes by PM (‘fieldnotes’).
Phase 3 September–December 2021	Another set of guided self-observation and reflective assignments to deepen and confirm learning. -To strengthen communication with colleagues and supervisors.	•Submitted feedback on reflective tasks (14) •Endline evaluation questionnaires* (30)
Phase 4: 3-day skills workshop (January 2022)	A three-day follow-up workshop to share learning and success stories in using the new skills, strengthen confidence and empowerment, and build new skills, particularly in handling difficult emotions: Modules covered 1.Introduction and feedback from the learning process: The Big Changes. Gold standard strategies to communicate with emotional competence. 2.Understanding and dealing competently with strong emotions (anger, conflict, bullying, death and dying, and burnout) 3.Building and practicing strategies to communicate with emotional competence at work	•Filled pre- and post-course evaluation open-ended. •Workshop observation notes by PM (‘fieldnotes’).
March–April 2022	Post-training individual interviews (both trained and non-trained participants). Observations of trained participants in practice, done in the two Newborn Units.	•Interview transcripts (40) •Fieldnotes made by PM (70 h of non-participant observation).

The training of trainers began before the activities outlined in Table 2, with a one-day basic online skills workshop on experience-based learning approaches. It continued with online or in-person workshops before each of the four course phases in Table 2, where course materials were adapted together, and facilitation skills were built through rehearsals with constructive feedback to the mentee trainers. The agreed selection criteria for the two pilot facility courses were staff working in the newborn units, and closely allied units, maternal and child health clinic, post-natal ward, maternity, and voluntary participation. In practice, the nurse managers adapted the participant selection process. In Hospital B, the course was open to all staff, with most attendees being experienced staff. In contrast, the manager in hospital A deliberately targeted young nurses seen to be problematic and with poor communication skills. Each facility had 20 course participants. Of the total 40 participants, 38 were nurses and 2 were nutritionists. Only two participants were male. Participants were aged between 25 and 58 years, with work experience ranging from 2 to 38 years.

2.3. Data collection

To assess the course value to participants and its potential for positive impact and scalability, the evaluation team (PM, DO, ME, and SM) combined data gathered from the roll-out course process (Table 2, right column starred) with additional observations of the workshops by PM (fieldnotes), and specifically organized semi-structured pre- and post-course individual interviews. The evaluation team comprised three social scientists and one clinician (ME), all with significant experience in the Kenyan public health system. PM and DO live and work in Kenya, with DO leading research on the experiences of mothers hospitalized with sick newborns (Maluni et al., 2025). ME and SM have lived and worked in Kenya for several decades; ME is leading extensive research on quality improvement initiatives for hospital care in Kenya (English et al., 2020). None were involved in the daily running of the training process. They developed interviews based on the theory of change with open-ended sections for any additional insights from participants. A total of 20 pre-course and 40 post-course interviews were conducted by PM one month before and two months after the end of the training process, including trainers, trained participants, and others (managers, colleagues of trained participants). Table 3. These data were supplemented by 70 h of non-participant observations conducted by PM, documenting nurses' daily work and interactions with patients and colleagues in the newborn unit. These observations focused on participants applying their skills in their usual workspaces within six months of completing the training. Additional sources included course materials, baseline and endline semi-structured questionnaires, and reflective tasks.

Table 3.

List of participants interviewed.

Cadre	Pre-course interviews (n = 20)	Category
Nurses	15	Course participants
Ward in charge	2	Course participants
Nutritionist	1	Course participant
Nursing service manager	1	Non-course participant
Hospital administrator	1	Non-course participant

Cadre	Post-course interviews (n = 40)	Category
Nurses	16	Trained
Nurses	14	Non-trained
Ward in-charges	4	Non trained
Ward in charges	2	Trained
Nursing service manager	1	Non-trained
Trainers	3

2.4. Data analysis

All interviews were recorded, transcribed, and managed in NVivo 12 software, and field notes were typed up and saved in a password-protected study folder. Our analysis process involved several steps. First, PM conducted a thematic analysis of the pre- and post-course interview data, with an initial coding framework agreed upon by PM, SM, and DO based on the course goals, our evolving theory of change, and anticipated unintended consequences. Second, PM developed two reports of findings organized by key themes (one report per site), incorporating insights from her observation notes and based on regular discussions with DO and SM. Third, the reports and overall findings were discussed in several longer in-person meetings between PM, DO, SM, and MB, to allow group reflection on similarities and differences across the two facilities. Our emerging analyses were shared through presentations with the remaining authors for input. This process allowed us to triangulate and incorporate different data and insights while maintaining some distance between the training team and evaluators. However, we acknowledge that those most centrally involved in the analysis have all been involved to differing extents in the design and evolution of the course itself.

2.5. Ethics approval and consent to participate

The Kenya Medical Research Institute, Science and Ethics Review Unit, approved this study. KEMRI/SERU/CGMR-C/161/3852. Informed consent was obtained from all participants for in-depth interviews and observations. Throughout the data collection period, the researchers adhered to ethical principles such as confidentiality of information, the right to withdraw, and voluntary participation.

3. Results

We share our learning around our evolving theory of change (Fig. 1). We focus on immediate outcomes, first about individual change and then in relation to participants' communication with patients/parents and colleagues. We then describe some of the key mechanisms of change, followed by some of the similarities and differences observed between the two facilities. The latter highlights the importance of pre-existing relationships, leadership style, and course implementation processes, both as assumptions and as influencing the ability to create safe spaces (the overarching mechanism).

3.1. Immediate outcomes - individual change

3.1.1. Strengthened self-awareness and personal change linked to new knowledge, skills, and strategies

Overall, the course was highly valued by the participants from both hospitals, with many highlighting the course as quite different from previous courses in feeling relevant and necessary for themselves and in inspiring self-awareness and personal change.

“This course is not like the other courses, this is a lifestyle…, even when we were finishing, we evaluated ourselves and we said, this is not a course that you went to learn for somebody else, it is a course that I'm learning for me.”

(IDI 016 Hospital B – Trainer)

Participants reported increased knowledge about the effects of communication behaviors on others, including colleagues and mothers of newborns in newborn units, and striving to be more intentional in their communication to enhance relationships. Most also reported an increased awareness of their emotional reactions when interacting with parents and colleagues and being better able to manage these through the enactment of newly learned skills and strategies.

Two related sets of skills or strategies were identified as most useful: learning to ‘put themselves in others’ shoes' and to ‘step back from automatic emotional reactions’ when upset or overwhelmed.

“It taught me a lot about how to communicate with others, especially my colleagues, and my patients, and the keyword I learnt is ‘awareness.’ When communicating, you should be aware and put yourself in somebody's shoes because most of the time we talk carelessly, especially when you are not happy or maybe you have stress, or are overwhelmed, so ‘awareness’ is key”.

(IDI 013 Hospital B – Trainee)

“That stepping back is something I find myself practicing, and when I find that I am projecting…if the environment is not conducive to acting at that time, you step back.”

(IDI 08 Hospital A – Trainee)

With regard to parents, participants emphasized that they gained greater knowledge and understanding of parents as individuals with their own different social and emotional needs. They discussed having learned how to put themselves into the parents' shoes, show respect through actively listening, and try not to rush into conclusions.

Drawing on their learning about the influence of emotions on behavior, many participants described being less judgmental of others, of the potential value in seeking to better understand the reasons behind behaviors.

“You see, it has changed me so much; I feel like someone should be given a listening ear. Just listening before I judge.”

(IDI 003 Hospital A – Trainee)

Most participants described having previously labeled mothers as ‘uncooperative’ when they did not follow instructions, without finding out the reasons for the lack of cooperation. They described now being less critical and recognizing that sometimes it was their explanations that were inadequate:

“When a baby comes with dehydration, before you could tell the mother, ‘Why don't you want to express milk, you're the one who is denying the child milk.’ But now when she comes… There is not that condemning, you understand her, you encourage her, and show her what she is supposed to do”.

(IDI 006 Hospital B - Trainee)

Participants from both hospitals emphasized the direct relevance and applicability of the knowledge and skills learnt not only in their daily work situations in the newborn units but also in many cases at home and in communities:

“This one is a lifestyle, it's something you would want to use daily, not only at work, in the community, in school, at home, and in church, you know it's 360-degree training.”

(IDI 016 Hospital B – Trainer)

3.2. Immediate outcomes – improved communication

3.2.1. Interactions with parents

Many participants described how their awareness, new knowledge, and skills were changing how they interacted with mothers. For example, one nurse explained a change in how she gives information to mothers about expressing breast milk:

“So, it taught me how to approach these mothers differently. And if you approach them differently, you even get better results. And being empathetic, trying to put yourself in their shoes. … So, you'll find that maybe if I get a mother who's struggling with expressing milk, I even go ahead and even teach her the technique to express the milk. If she gets 5mls at first, I will even appreciate her for that because it's not easy, and I encourage her.”

(IDI 009 Hospital A - Trainee)

These positive interactions between nurses and mothers were also observed by PM in the newborn unit in both hospitals, and by non-trained managers and colleagues beyond the newborn unit.

“Nurses are seen taking time to explain procedures to mothers, and even some of the patients seemed to be happy with the services in the units, from the way they are talking with each other. I informally join some mothers, and they are talking about how they are treated with respect compared to the neighboring private hospital X where they are shouted at. As one mother says, ‘Here is not like hospital X, where mothers are shouted at and beaten while giving birth; here nurses are nice.’”

(Non-participant observation hospital A)

Before the training, most nurses expressed fear in handling angry parents, or parents who had lost a baby. Deaths were especially challenging when staff felt blamed for the death by parents. However, after learning about how to approach death with sensitivity, by acknowledging parents' emotions, listening actively, and showing empathy, participants reported that they now allow parents to express their anger and grief. Additionally, they reported doing something many had not done before: allowing parents adequate time to hold their dead baby before taking them into the morgue.

“A mother lost her child, who was critically sick, but said there was negligence and wanted to carry her baby immediately. I have learnt to give them time to express their anger and console them with less talking or very minimal explanation unless need be”.

(Self-administered form_Hospital A - Trainee)

“Earlier, I could not allow them to hold the baby and allow time with the baby's body, but post-training, I give time for the mother to grieve and try to answer her concerns as much as I could.”

(Self-administered form Hospital A – Trainee)

“The nurse asks the mother about her baby's condition during her last visit. The mother responds that the baby was extremely sick at that time. Gently, the nurse informs her, ‘Mama, your baby was very sick. We did our best, but sadly, the baby passed away. We are so sorry. Please come, let me show you the baby’. The nurse walks the mother to the baby, carefully uncovering the body to allow her to confirm.”

(Non-participant observation - Hospital B)

Outside the newborn unit, trained staff in hospital A's children's ward described initially being mocked while applying their new skills as they were perceived to be too soft with the patients and unlikely to sustain the approach in the long run. However, over time, their colleagues, including their untrained line manager, could see that they were handling patients with respect, which led to their cooperation in the care process.

“Now the others are emulating what we were trained to do. I am not saying all of them, but there are some.”

(IDI 001_Hospital A - Trainee)

Using what they had learned about patient-centred care, several talked about going beyond routine nursing care to explore and respond to parents' worries.

“… before I had no patience, I could rush to conclusions. Nowadays, I am trying my best to put myself into patients' shoes and listen, take time to listen because what I have learnt when you take time to listen to a patient, they tend to open up and give you things … you show her you care, or you are also concerned about her problems. So, I have learnt to be patient, to be keen, to have an interest in the patient and put ourselves in their shoes.”

(IDI 013 Hospital B - Trainee)

In both institutions, non-trained managers and colleagues described observing how some trained colleagues had learned to stop acting automatically and were urging others to exercise sensitivity in front of patients.

“Before the training, colleague Y was very emotional; she could tell you off before patients, but now she tells us to observe what we say before patients. She encourages us to be confidential and not to talk carelessly.”

(IDI 020 Hospital A - Non-trained manager)

3.2.2. Interactions with colleagues

Participants reported many conflicts with colleagues before the training, including about the duty rota. Although the challenges remained post-training, they reported being more aware of when they became upset and angry and being better able to control their emotions rather than lashing out. Petty issues among colleagues were better managed:

“The challenges with the duty rota, they are getting worse, (laughs) they are getting worse, but apparently, we do not fight about the duty rota so much.”

(IDI 016 Hospital B – Trainer)

“Yeah, before, the relationship wasn't so friendly. Let me say so. Because before we were trying to work like, and everyone knows-it-all. … But now, at least we can step back, and if there is any conflict, we will sit down once he or she cools down. Now, can you call and say what the issue was? I'm sorry.”

(IDI 001 Hospital A - Trainee)

Non-trained colleagues had similar observations, commenting that there is now more open communication about duties, including difficult scenarios, and more understanding among colleagues, in some cases leading to line managers being empowered.

“She [colleague in a Paediatric ward] is one of those who was very emotional and who could tell you off, even before patients …One thing that she has also done differently is to handle the staff with difficulties; she can handle them, especially those who are ‘noise makers’ and would like to complain about duties.”

(IDI 013 Hospital A - Non trained)

Two areas that emerged as important were around the value of showing appreciation to colleagues and the value of role models. Before the training, nurses in both hospitals reported that there was a lack of a culture of appreciation among colleagues. Colleagues assumed that there is no reason to show gratitude for something that they are expected to do while working together. However, this was reportedly changed by the training process:

“Since I came back from the training, I learned that appreciation is important because sometimes, you stay with someone, and they do their job well, but you never even tell them. …I came to realize the value of appreciation. It looks simple, easy, and very small, but it's very important”.

(IDI 002 Hospital A - Trainee)

Regarding role-modeling, both hospitals reported that their relationships with patients had become more positive as a result of listening, showing empathy, and supporting parents better, and that this led them to be viewed as role models by their peers and to be asked for assistance in handling challenging cases. Their improved reputation among their colleagues inspired them to keep practicing their new skills. Similar reports were made by non-trained managers and colleagues in the postnatal ward in Hospital A, where they described their trained colleague as positively influencing others by modeling good communication behaviors with patients and colleagues in the department.

“So far, nurse Z has just greatly changed; the change is embracing the others, so now, because of that change, people want to attend the course so that they may look the same way she is.”

(IDI-13 Hospital A - Non-trained manager)

3.3. Mechanisms of change

The overarching mechanism that emerged through our analysis is the importance of creating and maintaining safe spaces to reflect, build trust, share, and test out new learning.

With regard to reflection, all the participants felt that the self-observation and reflection exercises before the taught courses were essential for their learning, allowing them to discover for themselves the effects of their behavior on patients and colleagues. Many reported that, by writing down their reflections, they could evaluate their weaknesses and identify areas for safe change without blaming others:

“… evaluating on the effects of what I did to the other person or what I said or how I behaved, it had not really occurred before… it gave me an opportunity to look at the situation that we are facing as a reality. Yeah, without blaming the workload, blaming the people you are working with.”

(IDI 016 Hospital B - Trainer)

“Having to sit back and write…at least you see you've changed from this person who would not even greet mothers in the morning, and you are now warm and welcoming, and you can understand these patients more. That was the best.”

(IDI 003 Hospital A - Trainee)

Participatory approaches, with practical scenarios from their work, helped them see themselves in those behaviors and understand the effects on patients, motivating them to change. Open sharing in workshops with no fear of judgment also made participants feel safe and that they were not alone in their struggles. Training over time in a step-by-step approach (reflection period, learning skills, practice, and follow-up) helped participants put new skills into practice and gain confidence. It also helped them see what works and how patients and colleagues responded to their new behaviors. The trainers' qualities of humility, deep local experience and insight, and calmness were described as supporting this process:

“What motivates me to practice my skills is so that I can be a good role model like the lead trainer. So, you see, like we admire her, personally, I really admired her, because in whatever situation she's always calm.”

(IDI 019 Hospital A - Trainee)

The majority said that by learning to step back from automatic responses when dealing with parents, they could relate to and better support parents in their concerns and the overall care of their babies. In turn, they received appreciation from mothers and colleagues for their good work. This positive feedback made them feel more respected and satisfied, and inspired them to continue practicing their new skills:

“You know, when you are appreciated, you get that urge to work better. So, if the patients are appreciative, you hear a patient say, ‘Hey, sister, you've helped me so much, thank you.’ That thing gives the motivation to go on using these skills.”

(IDI 006 Hospital B - Trainee)

3.4. Differences between facilities, influencing mechanisms, and testing assumptions

Despite the overall positive immediate outcomes in both hospitals, with positive potential for longer-term outcomes, several nurses noted that their difficult work environments, including heavy workloads and staffing shortages, made it difficult to practice and maintain the new skills, with the ‘old self’ popping back:

“It's too much workload. Sometimes the ward is too much. You are alone. At that time, you had very sick babies. You go from the nurses' station; another mother is calling you. Then you find a mother whom you instructed and has not done what you instructed. So, you're like, ‘Didn't I tell you how to do?’. So, the old self pops back.”

(IDI 03 Hospital A - Trainee)

Also, the positive impacts appeared to be far greater in Hospital B than in Hospital A. We observed less recognition of trained colleagues as role models in Hospital A, where trained nurses sometimes expressed frustration when sharing their new learning and skills with non-trained colleagues. These differences appeared to be related to several interrelated factors.

3.4.1. Pre-existing team dynamics

Before the training, nurses in Hospital B described a culture of mutual respect and support, reinforced by regular monthly meetings where they discussed clinical updates, fostering healthy professional relationships. This pre-existing culture was built upon through the training process, for example, through allocating specific time to discuss stresses and anything else affecting team relationships and interactions with patients. In this hospital, nurses also describe supporting each other in practicing their skills and teaching their untrained colleagues through additional training sessions that they organized themselves, including specifically organized learning sessions.

In Hospital A, the team was less cohesive. Before the training, two main groupings of nurses were in place: some junior nurses who were deployed to the unit as a disciplinary measure, and more senior nurses with several years of experience in the newborn unit. As the in-charge explained, the staffing officer told me, ‘I have given you new people-very difficult ones, but I know you will handle them’ - (IDI 018 Hospital A – Trainer). The tensions were that before the course, some nurses were not talking to each other. Although the course helped with some of these relationships, there were no physical meetings before or after the training, and all correspondence would take place via the team WhatsApp group. There are also indications that problematic relationships were reinforced at least for one senior nurse who was inspired by the course to stop being late to work and report her colleagues, which led to her being socially isolated:

“Like late coming. More often, I do afternoon shifts, so the night staff comes to release me late. While I changed and I come so early every day, I leave work very late…Now I went to report them to the in-charge… Therein, I have had enemies. Now I can say the people who do the night shift, they only talk to me when it is convenient to talk about something that is job-related, but they cannot ask me anything outside that job, so I feel the relationship has been affected”.

(IDI 003 Hospital A - Trainee)

There were also no reports of formal dissemination of the training to colleagues. Non-trained participants expected formal feedback about the training from their trained colleagues, but this did not happen, for reasons that are not entirely clear.

3.4.2. Leadership style

In Hospital B, nurses described their manager as approachable, motherly, respectful, and supportive, with no concern about sharing their reflective assignments through her as they considered her part of the trainer team. The course appeared to strengthen relationships across the team as staff saw her continuing to role model skills, develop new ways of handling challenges, and give them positive reinforcement and motivation:

“Our in charge is that person, you can even have a scenario with the patient, or something happens even with a colleague, and you'd sit, discuss with her, and the way she handles it, she'll not, even if you went individually, she'll not discuss with the other party, like you had discussed. She will handle it in a good way after listening on this side and the other sides. She is a good person. She is motherly. And in terms of professionalism, she still has it, so they correlate well.”

(IDI 001 Hospital B - Trainee)

“Whenever she supervised, let me say as part of her work, you could always hear, ‘Hey, I can see communication is working.’ Maybe you do something unconsciously, and she can point out something good that you've done that was there in the training. So, it reminds you and, in a way, it boosts you at work, and your communication skills continue.”

(IDI 001 Hospital B - Trainee)

In Hospital A, the in-charge had a relatively ‘command-and-control’ or top-down leadership style with colleagues describing her as being heavy-handed, favoring a select few, and intimidating. The latter is illustrated in an observation note by PM during a patient handover meeting with a new staff member:

I observed a bitter exchange between a nurse who had been posted into the unit against her will … In her previous ward, she was used to individual patient handovers between nurses, and here it was done as a group. During handover, she was resistant to the [new] handover procedure…. The in-charge kept telling her, ‘You will change … you will just have to change.’ The nurse did not like the whole handover procedure as she compared it to her former department. She even sneered at the nurse in charge when she left the desk.”

(Non-participant observation - Hospital A)

Nurses conveyed apprehension regarding the sharing of their reflective tasks via her, citing fear of reprisals. Post-course, the manager was described as less harsh and a little more approachable. However, she still reportedly refused requests for in-person meetings, suggesting that trust may not have significantly improved. The manager, on the other hand, believed that the nurses were taking advantage of her and inappropriately expecting her to respond to all their requests after the course.

3.4.3. Implementation of the course process

We noticed that the two institutions chose different participant selection procedures, potentially related to their past relationships and leadership styles. In Hospital B, the course was open to all staff, with most attendees being experienced staff. In Hospital A, the manager intentionally targeted young nurses perceived to be problematic and with poor communication skills; this was recognized by non-trained colleagues within the same hospital. This approach went against initial course agreements (see methods) and limited the participants' ability to role-model and share new knowledge and skills with older, more experienced colleagues. Additionally, in Hospital A, the manager was known for being effective at managing challenging staff and had such staff rotated into her unit for disciplinary reasons.

4. Discussion

Nurses working in newborn units in low- and middle-income countries face high workloads, resource shortages, environmental inadequacies (such as crowding and lack of privacy), and high mortality rates, with negative implications for interactions with parents, their well-being, and quality of care. Advancing care in these ‘on the edge’ contexts requires far more than technologies and biomedical interventions: multi-level change strategies that challenge normalized scarcity and associated realities for staff, parents, and mid-level managers are needed (Molinaro, 2025; Ramsey, 2022; Taylor et al., 2024; Witter et al., 2022). Specific initiatives that improve staff well-being and reduce parental distress can also be of value, especially where carefully tailored to the context with stakeholders and ideally implemented as part of wider strategies (Gray et al., 2019; Taylor et al., 2024).

Our research suggests that a participatory communication and emotional competence initiative can have positive effects on staff and parents, even in these very challenging contexts. In line with our earlier evaluation (Musitia et al., 2022) and as summarized in our theory of change (Fig. 1), the main immediate outcomes were that participants gained new knowledge and skills, greater self-awareness and self-efficacy resulting from better awareness of their own and others' communication behaviors, emotions, and their effects, and a stronger sense of professional identity and core values. Learning to ‘put ourselves in others' shoes and step back and refrain from automatic reactions when upset or overwhelmed’ was felt to be particularly valuable in contributing to improved interactions and relationships among teams and with families. Some staff were role-modeling good practice and sharing their skills with colleagues, and there were reports of changes to routine unit processes such as handling of deaths and staff rotas. Key mechanisms contributing to these effects included a strong motivation to change triggered by self-observation and reflection, supportive pre-existing team relationships, enabling leadership, creating and maintaining safe spaces to reflect, build trust, share experiences, and test out new learning. Positive reinforcement from colleagues and managers when role-modeling new behaviors also played a critical role. These indications of micro and even some meso-level changes appeared to feed back into nurses' satisfaction and well-being.

Despite generally positive outcomes, important differences emerged between the two hospitals, along with some unintended negative effects. For example, in Hospital A, one nurse faced isolation for speaking out against tardiness, and some trained colleagues were criticized by their peers for spending time with parents. The potential for such interventions to backfire in such ways was noted in Taylor et al.'s (2024) review. Two interrelated sets of factors appeared to modify the experiences and outcomes across our two hospitals, with implications for future similar initiatives: the ability to create and maintain safe spaces (an essential overarching mechanism in the theory of change), and organizational relational readiness before the course (which tested theory of change assumptions and challenged the ability to create safe spaces).

4.1. Creating safe spaces as an essential overarching mechanism

The overarching mechanism in our theory of change is creating and maintaining safe spaces to reflect, build trust, share, and test out new learning. Overall, in the less busy Hospital B, the process of creating and maintaining safe spaces was more effective than in Hospital A. Creation of safe spaces requires establishing psychological safety, which is defined as ‘the degree to which people view the environment as conducive to interpersonally risky behaviors like speaking up or asking for help’ (Edmondson et al., 2016). Psychological safety has been described as critical to health care improvement efforts and in nurturing ‘everyday resilience of health systems’ (i.e., the organizational capacity to absorb, adapt, and transform in the face of regular, routine, and recurring challenges) (Brown et al., 2021; Gilson and Barasa, 2024; Witter et al., 2022). It is recognized as essential to enabling learning and change in contexts characterized by high stakes, complexity, and essential human interactions, such as newborn units (Blacklock et al., 2022; Nembhard and Edmondson, 2006). In such settings, professional norms and hierarchical cultures within teams and organizations can create barriers to speaking up or asking for help, undermining psychological safety, and contributing to moral injury. However, psychological safety can also be built and nurtured by leadership capacities such as emotional and social intelligence that facilitate positive work environments (Edmondson et al., 2016; Nzinga et al., 2021).

In both of our study hospitals, ‘practical norms’ – forms of evolved informal socio-cultural rules – such as tardiness, task delegation, and emotional rationing, had evolved (McKnight et al., 2020). These norms both necessitated and challenged the building of interpersonal trust, which is essential to the development and maintenance of safe spaces and testing of new behaviors. Reflective tasks helped participants develop self-awareness and motivation for change, while facilitators' expertise in selecting examples and guiding discussions was crucial for fostering trust and finding relevant solutions (Van den Bossche et al., 2011). It was potentially embarrassing for participants to share experiences that exposed their vulnerabilities, and stigmatizing to test out new ideas and practices seen to take more effort or time (Taylor et al., 2024). However, in both hospitals, the training team used their leadership skills to role model communicating with emotional competence that helped reduce counterproductive norms. Positive feedback among colleagues and parents, and recognition by non-participants, added momentum and encouragement in both hospitals, and some course participants came to be seen as champions of a new way of interacting. Through their actions, they modeled fresh ways of understanding and addressing challenges, demonstrating the development of new cognitive and behavioral skills while gradually unlearning ineffective habitual practices (Alvesson and Sveningsson, 2003; Nzinga et al., 2019; Uhl-Bien and Marion, 2011). Overall, the more effective establishment of safe spaces in Hospital B was linked to greater relational readiness and the skills of the unit manager/facilitator.

4.2. Relational readiness tested assumptions and challenged the creation of safe spaces

Regarding relational readiness, the nurses in Hospital A were already divided into two main groups who interacted minimally. There was also a less nurturing and consultative approach to communication and decision-making than in Hospital B, and a broader organizational culture of shifting staff regularly between wards, including for disciplinary reasons. This contributed to some participants in Hospital A being targeted for the course because of their poor communication, contrary to the course assumption in the Theory of Change regarding voluntary participation, with negative implications for their positive engagement and trust in their hospital facilitator/manager (another course assumption). Making interventions mandatory has been observed elsewhere to lead to some staff feeling resentful, anxious, and exposed (Taylor et al., 2024). In our setting, this was potentially exacerbated by being applied to only some staff. In contrast, Hospital B benefited from stronger pre-existing relationships and a nurturing leadership ‘motherly’ style, which provided a stronger platform for training success. Drawing on Blacklock et al.'s literature review (Blacklock et al., 2022), breaking down workplace silos and fostering new social ties requires trust, mutual understanding, or a shared sense of urgency—factors that appeared more present in Hospital B than in Hospital A from the outset. The differential findings from our hospitals suggest the importance of initiatives that build relational readiness from the outset (Benzies, 2016), and that relational readiness could and should be centrally incorporated into an amended Theory of Change for the course process.

4.3. Positive potential for achieving longer-term outcomes

Achieving longer-term outcomes requires changing norms over time, which depends in large measure on constantly shifting social ties among trained nurses to peers, leaders, and other professions (Blacklock et al., 2022). For these reasons, sustaining improvements and changing norms is notoriously difficult (Mannion and Davies, 2018; Nzinga et al., 2021). Positive signs for longer-term outcomes and sustainability in our study include the hints of a positive impact of the course process on everyday leadership and the functioning of the units in ways that improve staff well-being, with the potential to reduce parental distress. Additionally, efforts have been made to integrate modules into continuous professional development, and funding from both government and non-governmental organizations has been secured for expanded training. There has been national nursing council accreditation, and elements of the course have been incorporated into broader initiatives like NEST360 (NEST360, 2025). Regular refresher sessions are being organized by the mentee trainers for their teams. Piloting of the online modules is also ongoing, with a view to their future use in undergraduate teaching and continuing professional development. Such initiatives have the potential, if implemented, to reach more people in different ways at different stages of their careers and ensure wider impact. Also ongoing is promising work to incorporate patient experiences into the course materials and process. Nevertheless, we reiterate that given the massive workforce gaps and persistent space and resource challenges that undermine staff well-being and contribute to parental distress, organizational support and change processes that tackle scarcity remain essential.

4.4. Strengths and limitations

This study's strengths include co-development of the training with nurse managers, use of diverse qualitative data sources, and grounding in a locally developed theory of change. Working across two hospitals allowed exploration of contextual influences on training outcomes. However, findings are limited by the small number of study sites and the lack of long-term follow-up. The study primarily used qualitative methods; while rich in depth, the need to identify quantitative approaches to measure impact in the future would also be valuable.

5. Conclusions

Our research suggests that a participatory communication and emotional competence initiative can have positive effects on staff and their interactions with parents, even in the very challenging contexts of public hospital newborn units. The literature and our findings suggest that organizational culture, leadership styles, and course implementation processes interplayed with individual drive and motivation to influence the nature and depth of course impact and the potential for sustainability and longer-term impacts in each hospital. The study underscores the significant impact of leadership style on team dynamics and participation in course activities. When leadership is perceived as unapproachable or unresponsive, it can foster an environment of mistrust and disengagement, ultimately eroding motivation, participation, and impact. To improve uptake and outcomes, future interventions should complement careful trainer selection processes by promoting inclusive leadership approaches that foster openness, equity, and support within teams. Although our evaluation was not designed to explore the links and overlaps with the other ongoing sets of interventions in these hospitals, our findings suggest the potential value of incorporating such training processes into wider change initiatives is urgently needed in such highly constrained and emotionally stressful environments.

CRediT authorship contribution statement

Mwanamvua Boga: Writing – review & editing, Writing – original draft, Formal analysis, Data curation, Conceptualization. Peris Musitia: Writing – review & editing, Formal analysis, Data curation. Dorothy Oluoch: Writing – review & editing, Formal analysis, Conceptualization. Dyuti Sen: Writing – review & editing. Hiza Dayo: Writing – review & editing. Ane Haaland: Writing – review & editing. Lisa Hinton: Writing – review & editing. Jacinta Nzinga: Writing – review & editing. Mike English: Writing – review & editing, Funding acquisition, Formal analysis, Conceptualization. Sassy Molyneux: Writing – review & editing, Writing – original draft, Supervision, Funding acquisition, Formal analysis, Conceptualization.

Funding

This work was supported by the Wellcome Trust, through a Senior Fellowship awarded to ME [207522/Z/17/Z]; and a Core award to the KEMRI-Wellcome Trust Research Programme [227396/Z/23/Z]; and by the NIHR (project reference: NIHR130812 and NIHR303168) using UK international development funding from the UK Government to support global health research and Bill and Melinda Gates Foundation Subaward [263771-5119872]. The views expressed in this publication are those of the author(s) and not necessarily those of the NIHR or the UK government. The funders had no role in the study design, data collection and analysis, the decision to publish, or the preparation of the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be available on request through dgc@kemri-wellcome.org. Once the request form has been shared, the corresponding author will be able to share the transcripts and any other related data.