Evaluating Caregiver-Clinician Communication for Tracheostomy Placement in the Neonatal Intensive Care Unit: A Qualitative Inquiry

Objective Identify stakeholders’ tracheostomy decision-making information priorities in the Neonatal Intensive Care Unit (NICU). Study Design English-speaking caregivers and clinicians who participated in NICU tracheostomy discussions between January 2017 and December 2021 were eligible. They reviewed a pediatric tracheostomy communication guide prior to meeting. Interviews focused on tracheostomy decision-making experiences, communication preferences, and guide perceptions. Interviews were recorded, transcribed, and analyzed using iterative inductive/deductive coding to inform thematic analysis. Results Ten caregivers and nine clinicians were interviewed. Caregivers were surprised by the severity of their child’s diagnosis and the intensive home care required, but proceeded with tracheostomy because it was the only chance for survival. All recommended that tracheostomy information be introduced early and in phases. Inadequate communication limited caregivers’ understanding of post-surgical care and discharge requirements. All felt a guide could standardize communication. Conclusions Caregivers seek detailed information regarding expectations after tracheostomy placement in the NICU and at home.


Introduction 47
In the Neonatal Intensive Care Unit (NICU), about half of premature infants born before 48 Two cohorts of participants were eligible for the study. The first included English-93 speaking caregivers with primary custody of infants admitted to the NICU who discussed 94 tracheostomy placement for their infants between January 2017 and December 2021. Both those 95 who received and those who declined tracheostomy were eligible. The second cohort included 96 English-speaking clinicians caring for NICU infants who regularly participate in tracheostomy 97 discussions with caregivers. Eligible clinicians included attendings, fellows, and nurse 98 practitioners from the pediatric divisions of neonatology, allergy and pulmonology, 99 otolaryngology, and palliative medicine, and NICU bedside nurses. 100 Potential caregiver participants were identified by referral from research staff and via 101 electronic medical record search for "tracheostomy" (ICD-10 code Z93.0) in charts of NICU 102 patients admitted between the study inclusion dates. We completed an individual chart review of 103 each patient from the query, and patients were excluded if tracheostomy was discussed after 104 NICU discharge, if the family required an interpreter, or if the caregiver no longer had primary 105 custody. Clinician participants were identified via referral by research staff. 106 107

Data Sources 108
If participants expressed interest, a teleconference interview was scheduled, and research 109 staff emailed the example guide along with study information and consent letter. Verbal consent 110 was provided at the beginning of each interview. The remainder of the interview was audio-111 recorded. The IRB-approved interview guide was based on the Consolidated Framework for 112 Implementation Research (CFIR) 19 and the Ottawa Decision Support Framework. 20 The 113 interview explored participant informational priorities in the decision-making process, 114 perceptions of the example guide, and views on implementing it in the NICU. Refer to 115 Appendix 1 and 2 for the caregiver and clinician interview guides, respectively. Once the 116 recording ended, participants completed a brief demographic survey. With caregiver consent, 117 research staff collected infant clinical data from electronic medical records including date of 118 birth, gestational age at birth, length of stay in the NICU, and co-morbidities. 119 120

Data Analyses 121
Audio recordings were professionally transcribed verbatim, and the interviewer reviewed 122 the transcripts prior to analysis for accuracy. Using an inductive/deductive approach, 123 investigators KB and AH, a research team member with expertise in qualitative research 124 methods, developed a codebook based on the interview guide and the Ottawa Decision Support 125 Framework. Standard qualitative analysis procedures were then followed. A minimum of two 126 investigators (KB, AH, KC) independently coded all interviews. The team met regularly to 127 resolve coding discrepancies through consensus, discuss updates to the codebook, and improve 128 understanding of the codebook to achieve consistent application across transcripts. Additional 129 codes identified in the transcripts were added to the codebook iteratively and updates were 130 applied to all transcripts. 131 Next, the research team met to explore the relationships and concepts within and across 132 transcripts. 21,22 The research team outlined patterns to identify conceptual linkages and themes. 133 Per standard protocol, recruitment ended when the interviews reached thematic saturation, and 134 no additional themes were identified during analysis. 23 135 136

137
Between November 2021 and October 2022, the research team called 35 potential 138 caregiver participants, spoke with 24, and consented 10 (42%) for participation. Concurrently, 139 we emailed 14 potential clinician participants and consented 9 (64%) for participation. Those 140 who declined either were not interested or had scheduling conflicts that prohibited participation. 141 Participant characteristics are described in Table 1. Of the caregivers, eight identified as 142 female; six were White, three were Black, and one was Asian. All completed at least some Of the clinician participants, five were physicians, two were nurse practitioners, and two 153 were bedside nurses. They worked in the pediatric divisions of neonatology, allergy and 154 pulmonology, otolaryngology, or palliative medicine. Six identified as female; all were White. 155 Analysis of the interview transcripts identified four phases in the tracheostomy process: 156 1) preparation, 2) deliberation, 3) action, and 4) home management and reflection. Definitions of 157 phases and considerations in each are described in Table 2. 158

159
The Preparation Phase 160 In the preparation phase, providers introduce tracheostomy as an option to caregivers. 161 During this time, clinicians assess caregivers' emotional readiness for such a discussion. Clinicians also recognized that it can be a shift in caregivers' thinking and planning: 177 "It's a very fundamental change in the way that they will have to care for their child, and 178 it is a very obvious difference from the baby that they probably imagined that they would 179 have and so it can be a hugely emotional conversation." [Clinician, P126] 180 181 The intensive home care that is required appeared to induce fear amongst caregivers 182 including worries of what living with a child with a tracheostomy would entail, and whether they 183 would be capable of providing the care needed. Caregivers thought that being introduced to the 184 tracheostomy early in the hospital course and then revisiting it multiple times with small amounts 185 of information would help alleviate this stress: 186 "I don't know if they're taught to do it that way or if that's just kind of how it happens. 187 But… it would kind of come up, but then it would be like, 'But we don't need to talk 188 about that now.' I think that helped that we didn't try to know the ins and outs of it before 189 we actually had to. But we were hearing it." [Caregiver, P105] 190 191 The clinicians thought similarly, many thinking the conversations should be started earlier: 192 diagnosis of BPD or other congenital anomalies would allow tailoring the material to suit the 314 differences between the two, including reasons for tracheostomy placement, hospital course, and 315 the path towards decannulation. Infants with BPD have lung disease that may improve after 316 years of time and growth while infants with congenital anomalies or structural airway 317 abnormalities may have more complicated courses including multiple airway surgeries without a 318 clear guarantee of decannulation. 319 "I think that's another thing, too, is there should be more education for the different… 320 times a trach is needed. You know, is it needed just for BPD? Because they need time for 321 their lungs to grow? Because that has a different outcome than if the baby has a 322 chromosomal abnormality and can't initiate breaths on their own. And when you're… 323 looking for research and for outcomes and things like that, it's all lumped together. You 324 … can't find, you know, things just on the babies that needed more time to grow." 325 [Caregiver, P102] 326 327 Another consideration is that clinicians may be under time constraints and leave the guide 328 for families to read independently. However, multiple caregivers and clinicians reiterated that the 329 discussion that accompanies a guide is more important than the guide itself: Lastly, caregivers overwhelmingly expressed that the benefits from tracheostomy 335 placement outweighed their fears. They highlighted benefits such as facilitating discharge home, 336 aiding in comfort and development, participating in therapies, reaching developmental 337 milestones, and improving oral feeding. No caregiver expressed regret over the decision for 338 tracheostomy: 339 "I think one of the things that you can't grasp in a guide is how much your child is going 340 to be able to, like, thrive, and you can't grasp that in a guide. You just have to almost do 341 it and then see that they start rolling and they start sitting up and they start walking and 342 they start doing the normal things." [Caregiver, P101] 343 344 "But for us, I will say, we feel choosing the trach was the best thing we ever did." revisiting the topic over time. Caregivers mentioned that they would prefer to receive 364 information earlier in the process, even if it is uncertain whether their child will definitively need 365 a tracheostomy, rather than being surprised and unprepared at time of tracheostomy placement. 366 Discussions which include clinicians whom the family trusts may be better received than 367 discussions led by a medical team member without an existing rapport. Identifying trusted 368 individuals can be done by asking families whom they would like to participate in future 369

conversations. 370
The themes identified during the deliberation phase echo previous literature. Clinicians During the action phase, all caregiver participants ultimately agreed to tracheostomy for 383 their infants, but for many, it took time to accept this course of treatment. Clinicians can prepare 384 for caregivers to cycle between deliberation and action phases multiple times and be available to 385 answer questions, provide more information, and aid caregivers in working through their fears. 386 None of our caregiver participants mentioned that they considered redirection of care. In contrast 387 to other diagnoses requiring tracheostomies for a stable airway, the infants of families in our 388 study almost all had BPD, a disease with the potential to improve overtime. This may have 389  Lastly, when reflecting on the process, clinicians noted that every physician approaches 395 the topic of infant tracheostomy differently. Both clinician and caregiver participants felt that a 396 guide could standardize the process and provide a more structured presentation of information 397 across families. The guide used could be tailored to the NICU population, potentially by creating 398 one for infants with BPD and one for infants with other diagnoses. In both cases, caregiver and 399 clinician participants thought it would be useful for the document to include pictures of nurseries There are some notable limitations to this study. We interviewed a small sample of 406 participants. Despite recruitment efforts, caregivers who declined a tracheostomy and those 407 whose infants passed away were not represented in our study. These are important viewpoints 408 that should be explored in future work. In addition, clinician participants represented a subset of 409 clinicians at one institution. Despite the sample limitations, these qualitative data provide rich 410 insight of participants' experiences navigating this process. Second, the chosen guide was shared 411 with participants prior to their interviews so that they could review and prepare their thoughts. 412 However, this early introduction could have impacted their thoughts on their experiences and 413 preferences. Third, the interview guide did not specifically probe for thoughts on alternative 414 options to tracheostomy placement but should be included in future work.