The pediatric intensive care unit (PICU) is a high-stress environment where the majority of in-hospital pediatric deaths occur.1 Working in this context, critical care clinicians face well-documented risks of burn-out,2,3,4 moral distress,5,6 and compassion fatigue.7,8 External pressures, such as the COVID-19 pandemic, can excaberate these risks.9,10,11 Given current challenges with retention and burnout in critical care,10,11,12,13,14,15 developing a holistic understanding of clincians’ lived experiences at work can help policy-makers better address underlying challenges and structural contributors.

Prior to the COVID-19 pandemic, most Canadian PICUs had adopted values and practices consistent with family-centred care (FCC).16,17,18 All patients could access at least two primary caregivers 24/7, nonparent family members including siblings could be at the bedside with variable frequency, and family members were encouraged to participate in patient care rounds.16 For clinicians who understand and practice FCC values, policies that impede family members’ access to critically ill children pose risks such as moral distress.5,6

In response to COVID-19, inpatient units across the globe restricted family presence.19,20 In Canada, many adult intensive care units (ICU) prohibited all family presence.21 While PICUs allowed at least one caregiver,16 this was a significant reduction from usual practice. Researchers and clinicians have sought to understand the impact and efficacy of restricted family presence (RFP) policies in ICUs.19,20,22,23,24,25,26,27,28,29,30,31,32 Clear impacts have been shown for clinician mental health20,23,29 and unit morale29 within adult ICUs. While pediatrics did not experience the influx of COVID-19 cases seen in adult populations, novel situations associated with a lone caregiver and obtaining exceptions to restrictions for families introduced other sources of stress.16,23,33 To date, few studies have specifically addressed PICU clinician experiences and impacts associated with RFP.23,33,34,35 Therefore, we designed this study to understand how PICU clinicians were impacted by RFP policies.36

Methods

Study design

Following ethics approval from the University of Alberta Research Ethics Board (Edmonton, AB, Canada; REB #Pro00104410, September 2020), we conducted a qualitative study that followed an interpretive phenomenological design. We report the findings according to the Consolidated Criteria for Reporting Qualitative Studies reporting guidelines37 (Electronic Supplementary Material [ESM] eAppendix 1).

At its core, phenomenology argues that individuals’ lived experiences have value and are worthy of scientific study.36 Interpretive phenomenology involves interpreting potential meanings from these lived experiences that may or may not be known to the research participants themselves.36 Accordingly, the interpretive phenomenologist typically comes with relevant expertise and prior knowledge.

Reflexivity is a core tenet of qualitative research.38 The interviewers and analysis team members were early-to-mid career, cisgender White women living and working on unceded Indigenous land in Canada. The study leads were motivated to pursue this work based on first-hand experiences with RFP policies in their PICUs (pediatric intensivists: J. F., D. G.; PICU nurse practitioner: L. L.). This motivation could bring bias that RFP caused harm. One analysis team member had first-person experience as a parent of a hospitalized child and reflected that their experiences may have biased them against restrictions and to empathize with health care professionals having to implement the restrictions. These biases were mitigated by having the primary coders bring different perspectives (qualitative research, public health: M. R.; qualitative research, rehabilitation sciences: S. D.) to challenge assumptions that may be coming from their own experiences rather than the interviewees, as well as member-checking throughout the data collection and analysis. Participants knew the interviewer was a research coordinator with qualitative training and experience (M. R.), or a PICU nurse who had recently received interview training (J. B.). Participants had no prior relationship with either interviewer.

Participant sampling and recruitment

Participants were recruited from respondents to a survey of clinicians who worked in a Canadian PICU during periods of RFP who had agreed to be contacted for future studies (N = 27).33 We purposively sampled this group of 27 clinicians for maximum variation along the following characteristics: profession; region of practice; years of practice; gender; and self-identification as Black, Indigenous, and/or a person of colour. Between May and November 2021, J. B. and M. R. contacted potential interviewees by email and/or phone, according to the individual’s preference. For groups that remained underrepresented (e.g., by profession or geographic location), we snowball sampled from fellow members of the Canadian Critical Care Trials Group. Recruitment concluded once a diverse sample, based on maximum variation, had been achieved.

Data collection

We conducted one-time, 30–60-min, open-ended interviews over videoconferencing software or phone, depending on participant preference. The interview guide was designed to follow the flow of conversation and explore the aspects of RFP most meaningful to the interviewee. The initial guide was based on themes from earlier phases of the authors’ research program.33 This guide was reviewed every two to five interviews to explore emerging themes and to allow for member-checking (ESM eAppendix 2). Interviews were conducted at a prearranged time convenient for the participant. All interviews were audio-recorded and transcribed with participant consent. Participants received a copy of their transcript to review for accuracy once available. J. B. conducted interviews 1–5, with Dr. Marghalara Rashid’s supervision for interviews 1–2; M. R. conducted interviews 6–16. The team provided regular feedback to interviewers. The interviewers completed fieldnotes following each interview as part of an ongoing reflexive practice and to ensure that relevant nonverbal aspects were recorded. The fieldnotes were reviewed for context during the coding process but were not formally coded themselves.

Data analysis

We performed an interpretive, inductive thematic analysis following the steps outlined by Braun and Clark.39 All transcripts were first read for a sense of the whole and then independently, inductively coded using NVivo 12 qualitative analysis software (Lumivero, Denver, CO, USA). Coding was guided by the question: “What was meaningful about RFP for PICU clinicians?” To develop an initial codebook, the first three transcripts were independently coded line-by-line by M. R., S. D., L. L., J. F., and J. B. The remaining transcripts were coded by two primary coders (M. R., S. D.) who met regularly to ensure consistency, refine definitions, and define new codes. Coding occurred while data collection was ongoing to allow for member-checking in subsequent interviews and further interview guide refinement. Every five interviews, the analysis team (M. R., S. D., L. L., J. F.) met to discuss emerging themes and to adjust the interview guide as needed. Once all transcripts were coded and the team reached consensus on the final codebook (ESM eAppendix 3), the team independently identified themes through inductive thematic analysis.39 The team met to refine these themes and reach consensus. To see if this interpretive analysis resonated with interviewees, we sent the proposed themes with example quotes to interviewees in May 2022; 11/16 responded. We integrated this feedback into the final themes presented below.

Results

Sixteen PICU clinicians participated in an interview. Interviewees included registered nurses (n = 8), one of whom had a management role; social workers (n = 2); child life specialists (n = 2); physicians (n = 2); and respiratory therapists (n = 2) (see Table 1 for demographic details). We identified four themes that illustrated clinicians’ experiences with RFP policies; exemplary quotes (“Q#”) are listed in Table 2.

Table 1 Participant demographics
Full size table
Table 2 Exemplary quotes by theme
Full size table

Theme 1. “[Everybody] understood why it had to happen, but felt really sad for the patient”: Balancing infection control and family presence

For study participants, RFP policies put infection control and family presence in opposition to one another, introducing a tension to balance these competing priorities. Early pandemic rules were understood and largely appreciated by clinicians, who expressed that decision-makers were doing the best they could to keep everyone safe given what was known at the time (Q1, Q2). Although interviewees encountered family member suffering (Q1), the rules seemed necessary and in balance given limited available knowledge and infection control concerns (Q2). These early restrictions also helped put later restrictions into perspective; for example, a social worker expressed how much easier living with a two-caregiver rule was compared with the trauma felt when having to limit family presence to one caregiver (Q3). As the pandemic evolved, several clinicians did not feel that the rules evolved appropriately to maintain balance. Participants wanted restrictions to be responsive to the pandemic context and proportional to risks, and expressed increased recognition of the mental health risks posed by RFP (Q4, Q5). When an imbalance developed between infection control priorities and family presence, such as when bars could be open but two primary caregivers could not be in the unit with their critically ill child, participants experienced distress, sometimes described as moral distress, and feelings of injustice (Q4, Q5).

Theme 2. “Speaking into the wind”: Feeling disempowered by hospital and policy-making hierarchies

Clinicians described top-down, “one-size-fits-all” policies that were made without bedside clinician input (Q6–Q8). The policies and their implementation removed autonomy from bedside clinicians, making it difficult to act in the perceived best interests of families (Q9). Further, frontline clinicians with little institutional power compared with policy-makers had the most responsibility for implementing and enforcing policies day to day. This power dynamic showed a lack of recognition for interviewees’ FCC expertise when developing and implementing RFP policy (Q9, Q10). Bedside clinicians also felt unheard when they advocated for their patients and voiced concerns about RFP policies (Q7, Q11, Q12). As the professionals who had to manage the policy consequences daily (Q9), interviewees emphasized the need for frontline voices in policy design (Q10, Q13).

Interviewees also found that policy communication to frontline clinicians was top-down and overwhelming (Q14). This led to inconsistent policy application, frustrating both clinicians and the families with whom they worked (Q15). Hearing information directly from trusted managers and supervisors helped, and several clinicians described support as good communication via clear policies, clear rationale, and leadership presence in unit to help communicate rules (Q14, Q16). This hierarchy and its related power dynamics also had some benefits. For example, clinicians responsible for enforcing the rules appreciated being able to defer blame to leadership and would make this lack of control clear to families; this helped preserve clinicians’ therapeutic relationship with families (Q16, Q17).

Theme 3. “Exhausted with feeling for them”: Empathizing with family trauma

Throughout the interviews, clinicians shared their perceptions of how RFP policies and practices impacted PICU families, including many stories of bearing witness to situations in which they perceived that family members experienced trauma (Q18–Q21). Being present for these experiences impacted clinicians themselves; some interviewees described being present for these moments as “heartbreaking” (Q20). The act of having to deny family members’ access to a critically ill child weighed on clinicians (Q19). Child life specialists felt particularly distressed by separating siblings, both for the time lost in these important developmental relationships and for the potential harm to the patients’ sibling(s) (Q21, Q22). When discussing these situations and their emotional reactions, clinicians felt moral distress since they were unable to do for the patients and families what they felt was in their best interest (Q23). One clinician summed up the sentiment of many by saying: “I was just exhausted with feeling for them because they were suffering so much” (Q18).

The empathy for what families were experiencing helped clinicians identify fault lines in the policies, such as limiting available supports for families (Q23, Q24), and invited critical consideration as the pandemic evolved as to whether the infection control benefits still outweighed the costs to families (Q4, Q5). Although clinicians tried to provide support as best they could, they indicated that this could only go so far: “We don’t replace family […] it’s just not the same” (Q24). Driven by their empathy and commitment to FCC, study participants nonetheless found creative ways to keep families connected, provided additional support, and advocated for change (Q25–Q28). Given these insights, RFP may put clinicians at increased risk of compassion fatigue and moral distress; a risk that may be compounded by perceived threats to FCC and the therapeutic relationship.

Theme 4. “We became the bad guy”: Navigating threats to the therapeutic relationship

Although “caring for the medical disease” stayed the same, as expressed by a physician, “caring for the whole patient” was negatively impacted by RFP policies and practices (Q29). This limited clinicians’ ability to deliver FCC (Q30). Having to enforce policies or deliver bad news related to family presence limitations with insufficient support to do so was detrimental to the therapeutic relationship (Q31–Q33). For example, a nurse shared that she began to distance herself from families to make those enforcement conversations easier “which isn’t great for relationships in this role”; this contrasted with her pre-pandemic experience of having “really good relationships” with families (Q32). Clinicians shared several factors that helped protect the therapeutic relationship in face of threats. Respiratory therapists described being removed from policy enforcement and able to leave the bedside, which protected their own therapeutic relationship and minimized the perceived impact of the policies on themselves compared with other interviewees (Q34). Other protective factors included receiving support from leadership (Q16, Q35), providing families with social and physical support (e.g., facilitating innovative ways to keep families in touch with those who could not be physically present) (Q25–Q27), and providing and advocating for exceptions (Q28).

Discussion

Understanding clinician experiences and impacts from policies enacted during unprecedented contexts can inform both emergency-preparedness planning and everyday policies. In this study, we identified four themes reflecting PICU clinicians’ lived experiences with RFP policies during the COVID-19 pandemic: 1) balancing infection control and family presence; 2) feeling disempowered by hospital and policy-making hierarchies; 3) empathizing with family trauma; and 4) navigating threats to the therapeutic relationship. These findings largely align with the emerging literature in pediatrics and critical care more broadly, while providing important nuances specific to the PICU context.

Our findings differ from studies with adult ICU clinicians in that feelings toward the policies among participants in our study were more mixed,30 and our study did not highlight increased communication difficulties.40,41,42,43 This is likely related to the allowance of at least one family member, which seemed to mitigate some of the negative impacts experienced in adult ICUs, such as witnessing patients dying alone.16

Pediatric ICU clinicians play an integral role in maintaining FCC, especially during times of crisis.31,44 Being responsible for implementing policies that conflicted with such a foundational principle as FCC was difficult for clinicians in our study, especially since their presence at the bedside necessitated bearing witness to the impacts of restrictions on patients and family members. Conflict with PICU values, such as FCC, can be minimized through reframing family presence policies to focus on “how” to enable and optimize family presence rather than “if” or “when” parents can visit.45 Taking action to minimize value conflicts could protect against moral distress.5,6

The empathy interviewees experienced for patients and families deeply weighed on them, putting them at risk for compassion fatigue.7,9 Empathy is a cornerstone of strong therapeutic relationships between patients, families, and clinicians.46,47 A correlational study found that the more empathy critical care nurses displayed, the more accurately they assessed family needs.47 Similarly, our findings showed that empathy helped maintain the therapeutic relationship in the face of challenges by identifying and addressing family needs. It is worth recognizing these efforts, while also appreciating the burden they place on clinicians. It is also important to create proactive, sustainable systems that support delivery of high-quality FCC and that do not rely on reactive, individual-level actions alone.

When institutional policies pose potential risks for clinicians, the stakeholder perspective must be sought to understand impacts and improve assessment of policy proportionality. Based on our findings, frontline clinicians wanted meaningful opportunities to provide their input and expertise with initial and ongoing development of RFP policies. Further, pediatric frontline clinicians often face moral distress when they are “tasked with implementing policies that they did not develop,”45 characterizing a morally hazardous environment.48 This environment was exacerbated by the power dynamics illustrated above, whereby 1) those implementing restrictions were removed from decision-making and 2) those with the least institutional power bore the weight of unintended consequences resulting from RFP policy implementation. Similar tensions were described by Wall et al. in their study of organizational influences on moral distress for PICU clinicians: “[participants observed that] managers tend to ‘address the power angle’ by reminding nurses that the decision-making power belongs to management. As another nurse pointed out, ‘After a while, you realize that your voice is not wanted.’”49 Involving frontline clinician perspectives in RFP policy development may help mitigate this capacity for moral hazard and is supported by emerging recommendations from acute care,20 adult critical care,26,50,51 and pediatric palliative care.45

Strengths and limitations

This study has several strengths. The multidisciplinary study team, including PICU family partners (C. S., M. W.), supported a holistic presentation of findings, ensured different perspectives were considered, and minimized bias. Purposive maximum variation sampling ensured experiences of multiple professions were heard, reflecting the PICU’s interprofessional work environment. This study also employed several techniques to ensure validity, such as maintaining a detailed record of data collection activities, member-checking through an iterative interview guide design, interviewee verification of own transcripts, and interviewee feedback on preliminary themes.

Study limitations include the decision to recruit clinicians who completed a survey in an earlier phase of the same research program. There may be a selection bias as it is possible that people who complete a survey and volunteer for an additional interview have particularly strong feelings on a topic and their views may have differed from those of other respondents. There may also be recall bias as most PICUs had moved away from the strictest policies (e.g., one caregiver only) by the time of interview. Since the study design prioritized what was most meaningful to interviewees, it is likely that the most impactful and meaningful experiences were still easily accessible. Finally, this study only included clinicians who were comfortable being interviewed in English, consequently limiting input from non-English-speaking clinicians.

Conclusion

Canadian PICU clinicians were impacted by RFP policies during the COVID-19 pandemic. Although interviewees felt that these policies were necessary during early pandemic stages, many felt the policies did not evolve proportionally to infection risk and placed undue stress on families and their critically ill children. The impacts on families were hard to witness and weighed heavily on interviewees. RFP policies contributed to feelings of disempowerment and challenged clinicians’ perceived ability to provide the best FCC possible. Looking forward, frontline expertise should be reflected in future policy design and implementation to best support FCC in any context and minimize risks of moral distress. Policy-makers can better empower PICU clinicians and honour their expertise by directly involving them in policy development and by acting upon frontline feedback as the circumstances evolve.