Introduction

Caring for critically ill and dying patients often triggers both professional and personal growth for medical trainees. Trainees must master critical care medicine, communicate with families about goals and values, negotiate conflict, and manage emotions. For pediatric trainees, the neonatal intensive care unit (NICU) is often noted to be among the most distressing settings. This chaotic milieu often does not permit trainees sufficient opportunities to reflect upon, and learn more from, those ethically complex experiences.

Understanding more about how trainees make sense of the recurrent stressors in the NICU would help target educational interventions. Salih and Boyle (2009) suggest that reflective narrative exercises are one way to integrate bioethics education into neonatal-perinatal training programs. Reflective writing can enrich clinicians’ understanding of their patients’ experiences and enhance learning from ethically challenging situations (Charon 2001; Verkerk et al. 2004; Wald et al. 2012).

We present a longitudinal descriptive analysis of narrative writing by pediatric residents and neonatology fellows throughout their clinical NICU rotations. Our goal was to gain a better understanding of not only trainees’ reflections on the most poignant encounters of these rotations but also how trainees are both professionally and personally challenged by and make sense of repetitive, ongoing conflicts when caring for critically ill and dying infants.

Methods

Study Procedures

Our NICU is a 45-bed, university-based level IIIB NICU in an urban setting, with subspecialty capacity to care for infants of any viable gestational age. Patient care is delivered by a two teams of clinicians; one team is designated as the physician trainee service. This team, typically composed of four to five residents, a neonatology fellow, and an attending physician, shares in the medical care of 25 to 30 NICU infants. This team also attends all deliveries of high-risk infants, provides inpatient prenatal consults when neonatal complications are expected, and acts as nighttime coverage for the well-baby nursery. Pediatric residents on this team do four-week rotations at least once per year of training; neonatology fellows do two-week rotations at least five times per year of training.

Pediatric residents and neonatology fellows were recruited for study participation at the beginning of their NICU clinical rotation. Participants were asked to independently journal about their NICU experiences throughout their rotation, for at least 20 minutes a day, at least three times a week. Completed journals were reviewed and edited by study personnel to remove identifying data about patients, staff, or participants. Participants also completed an electronic survey at the beginning and at the end of their rotation that included the Jefferson Scale of Physician Empathy (JSPE) (Hojat et al. 2002). We hypothesized that the process of journaling would be associated with an increase in the pre- vs. post-JSPE scores among participants, and we wondered if the narrative themes would differ among participants with higher vs. lower JSPE scores. Study protocol was approved by the institutional review board; written informed consent was obtained. Participant incentives reflected the number of study activities completed and ranged from $20 to $245.

Data Analysis

Through careful reading of the longitudinal narratives we performed thematic narrative analysis, with particular attention to trainees’ expressions of conflict, ethical questioning, emotion, and attempts to place their NICU experiences into the context of their personal lives.

Narratives were first analyzed individually to identify overall themes. Two authors (RDB, PKD) independently coded the narratives from five participants. Those authors then met to review the assigned themes, discuss and resolve discrepancies, and refine the coding strategy. This process was repeated for the narratives of the next five participants and resulted in an exhaustive code book. One author (RDB) used the code book to code the remaining narratives. Key themes were identified by their frequency in the narratives. We present narrative excerpts to illustrate the trainees’ experiences.

To explore whether narrative themes differed between trainees with the lowest versus the highest JSPE scores, we compared the writings of the five trainees with the lowest JSPE scores and the five trainees with the highest JSPE scores.

Statistical analyses were performed using SPSS 16. Descriptive statistics included means, medians, frequencies, and proportions.

Results

From November 2009 to June 2010, 31 of 46 eligible pediatric residents and six of seven eligible neonatology fellows were enrolled, for a total of 37 participants. Participant characteristics are noted in Table 1. The average number of narratives per trainee was 12 (range 6–16). The total number of writings available for analysis here was 441.

Table 1 Participant characteristics and JSPE scores
Full size table

The possible scores on the JSPE range from 20 (low empathy) to 140 (high empathy). Of the 34 of 37 participants in our sample who completed the JSPE, the JSPE scores ranged from 94–133, with a mean of 115.8 (SD 8). There was an inverse relationship between the mean JSPE score and years of training for participants in our sample (Table 1). There was no significant change in the JSPE scores on the pre vs. post participant surveys.

Conflict was the pervasive overall theme in the narratives and seemed to be the predominant force shaping many trainees’ NICU experiences. Trainees experienced conflicts with families and conflicts with other clinicians. Resulting from some of these experiences and from others, trainees also described conflicts of identity as members of the neonatology team, members of the medical profession, members of their own families, and as members of society.

Conflicts with Families: Placing and Facing Blame

Blame

Many trainees described interactions with patients’ families that were confusing, disturbing, or frankly upsetting. Some trainees felt angry at parents whose actions were perceived by the trainee to harm an infant. This included parents who abused illicit substances, who did not visit their infant in the NICU, or who demanded medical interventions perceived to increase infant suffering without benefit. Trainees regretted that such parents retained the right to make medical decisions. Some were demoralized by a perceived lack of parental sacrifice for their infant, which trainees weighed against their own personal sacrifices in order to be present as doctors for those infants. Trainees struggled to make sense of how these conflicts did, or should, impact patient care.

I watched a mother cry as I told her that her son was still seizing. I wanted to feel bad for her … but I also felt angry … because she tried to deliver him at HOME. … She wanted to blame and question everyone else for what happened, but she cannot admit to herself that … laboring at home without progression may have had a little to do with it. I felt angry because I felt like she was trying to put the blame on us. … Here in front of me was a beautiful full-term boy who likely is significantly devastated from a neurological standpoint who could have (maybe) been completely normal if he was not laboring (actively) at home (304).

She was a victim of domestic abuse and had a suicide attempt when pregnant. The father was the abuser. It makes me sick and causes a knot in my stomach; it is quite possible that physical abuse caused her to go into preterm labor … it’s so hard to not blame the father (501).

I almost yelled at a mom today. She’s 17 or 16, I’m not sure. She has not come once in the month that her child has lived in the NICU. And now we’re ready to discharge this baby, yet we can’t even get her in for a night to teach (302).

It’s hard, when you are putting in such long hours and feel that you are not quite aligned with the parents with a common goal for the wellbeing of a child (504).

Frankly, I wish I didn’t have to take care of the baby because I find the whole situation so upsetting. I don’t think I’ve ever felt that way before about a patient. I’ve been trying to channel my emotions into empathy for the family. I think if I think of them as victims of malpractice and certain cultural mindsets then I will at least be able to handle interacting with them and be empathetic. But it is so hard not to want to blame them as well. I can honestly say though, that whatever emotion I have about the situation cannot possibly compare to the suffering they are feeling. And maybe that thought alone will be enough (405).

Trainees felt personally blamed by some families, as when families denied that their baby needed NICU care or when families blamed clinicians for an infant’s illness.

I really wanted to say … do you think we want to keep your child away from you and keep her in the hospital forever? We just want your child to be safe when she goes home … and that isn’t worked out yet (301).

It is very frustrating when a parent is accusing you of trying to hurt her child. … I try and accommodate the emotions and concerns of a parent, even to the point of ordering a test I don’t think the child needs just to please them. BUT, I hate when a parent, who has no medical background, wants to dictate their child’s every order (301).

Nothing I said was good enough for this dad. He asked why we were treating with antibiotics … he either didn’t believe me, or didn’t like my answer. Either way he looked at me like I had 3 heads. … People always talk about how medicine is an art and that is because there are no guarantees but that doesn’t sit right with parents. I think they think I’m hiding something when I do not make promises but that’s because I don’t want to break a promise (502).

I don’t think the parents understand. They told [the] nurse that their baby has proven my attending wrong so far and they think she will continue to the point when they will come back to the NICU in several years with her healthy and rub it in our attending’s face. That comment sounds very spiteful to me and I don’t know where it is coming from. Maybe it is because my attending told them what they didn’t want to hear but she was just doing the best she could do. We all are just doing the best we can do … (502).

Conflict Blunts Emotion

Repeated conflicts with parents sometimes became barriers to trainees attaching emotionally with the babies of those families.

I find that I really like the babies with nice parents, but when parents are mean or demanding, I am less emotionally attached to the baby. Does that make me mean? (606).

One of the things that is hard about the NICU is how many families are unable or uninterested in being at the bedside … that leaves few families/moms who are around as much as I think I would be around if my baby was in the NICU; which means few opportunities to experience the kind of empathy I thought I would have … (404).

Sometimes people forget that the little one-kilogram baby is actually someone’s son or daughter or brother. It’s easy to forget b/c we hardly see parents in the NICU (107).

The opposite was also true.

I could see her mom holding her and reading to her. Gowned and gloved, holding her daughter attached to all sorts of tubing and alarm bells, there were the two of them enjoying a book. … This child’s trach, CRM, alarm beeping, and tubes were muted in light of this simple act. This moment almost brought me to tears. … The next day I came to this baby’s room like I do every morning … for the first time she seemed first and foremost a little baby to me, and I smiled as I reflected on that moment with her mom (604).

Conflicts With Clinicians: Doubted and Defeated

They Question My Compassion

Trainees described conflicts with peers, supervisors, or other staff members. Many trainees perceived conflicts with nursing staff. A common and very distressing conflict arose when trainees felt accused by nurses of not caring about patients.

There have been very few times … when a nurse has said something to me that I find hurtful enough to make me flustered and get red and hot in the face. I … often defer to their judgment, but I knew this patient. And to be accused, basically, of not caring for this patient enough, that maybe if I cared more about him I would act differently, made me want to both scream at her and cry (302).

I sometimes feel as though their intention is to protect the babies from us, as though we don’t have their best interest at heart, or as though what we may advocate for might be opposite of what the baby truly needs. Do they understand that we also care, that we also want to see these little ones thrive and grow out of their illnesses? … I can’t possibly understand why someone would feel like they have to defend their patient from me. In fact, the mere suggestion of it hurts me deeply (603).

She said that she “was really worried about this kid,” implying that we didn’t care (104).

I Question Their Decisions

Conflicts with supervisors revolved around negotiating control and responsibility. Trainees resented being asked to handle most day-to-day patient care without being granted some autonomy to make important decisions. Trainees who felt excluded from decision-making expressed more discontent with supervisors’ decisions, more confusion about the rationale behind decisions, and more conflict about carrying out those decisions.

There is a certain amount of independence and ownership I crave in order to feel like I’m even a worthwhile member of the medical decision-making team (404).

Sometimes you feel like you can’t make any of the decisions, and sometimes you feel like you’re doing all the paperwork (almost like a glorified secretary). Making no decisions but doing all the work is horrible, the worst combination (502).

It’s frustrating to me too, yet I have to act like we are all unified and know what we are doing, especially when so often it seems like it’s just somebody’s whim (102).

There is an overwhelming sense of futility here. We may not agree with the management decisions, there may be no evidence behind the management decisions, the management may vary widely depending upon who the attending is and we get minimal input into making the decision, yet we are expected to carry them out (401).

Normally I feel very invested in my patients, and obligated to fight for what I believe in, but I just felt very defeated about the whole thing, and got the message that no amount of arguing would change [the attending’s] mind (104).

Conflicts of Identity: Shame, Guilt, and Inadequacy

Trainees frequently felt conflicted about their identity as a member of the neonatology team and as a physician in general. Trainees were often uncertain and uncomfortable with what they were expected to do, and how they were expected to feel, about their responsibilities to patients and families. In addition, trainees felt conflicted about how to make sense of these experiences within their own understanding of family, parenting, and personhood.

My Identity as a Member of the Neonatology Team

Nearly all trainees described an experience where they questioned their moral responsibility as a member of the neonatology team. The most common and pronounced moral challenges were related to the resuscitation and treatment of extremely premature infants.

Some days I wonder if we are playing “God,” if we are doing the right thing for these babies. Maybe they weren’t meant to live? Maybe the mother went into preterm labor at 24 wks because the baby was not doing well and not making it. … It scares me that we are obligated to save every infant >25 weeks because they have a 50/50 chance of “survival.” … You can make many people survive all sorts of things … but it doesn’t mean that we should. The NICU leaves me with such big ethical dilemmas, and makes me question my own morals and what I would want for a 25-week premie of my own (304).

I mean they are alive but to what end. Life is precious and it’s a scary thing when people get to judge who will live and die based on disability but I can see that it would be hard on me to work in a place where you see the beginnings of a devastated child (307).

I think one of the hardest parts is seeing all you do to these poor tiny children—the needles, the intubation, the lack of human contact, and then seeing their MRIs and HUS showing bad, bad brain. Why am I doing this? (403).

I think we all struggle to do something meaningful, and it’s hard to keep going back to the NICU every day, when the most constructive thing you do is increase some kid’s feeds from 2cc to 4cc an hour, and spend the rest of your day prolonging the inevitable death of this ex-24-weeker (104).

Hers is representative of some of the most frustrating and morally clouded cases for me. … The mother went into labor at 24 weeks and delivered this half-kilo child whose chances in the world are next to nil. We’ve taken over almost all her bodily functions, and we’re still behind the eight-ball, all the while pumping tens of thousands of dollars into her care. … I sincerely hope that the near future brings developments to clarify either the moral or medical scenarios complicating her and my days (601).

Trainees felt guilty about a variety of their responsibilities on the neonatology team. Some trainees felt guilty for wanting to avoid responsibility for resolving uncertainty, conflict, or suffering. Others felt guilty for “giving” an infant a serious diagnosis or for not being certain about the prognosis. Some felt guilty for their emotional responses to the NICU. Trainees also regretted when they found satisfaction in performing the procedures and managing the clinical challenges that afflicted their patients.

I had to do a terrible thing today. … I had to introduce myself to the parents for the first time … then tell them … they’ve confirmed their baby has Down’s. This is what it felt like. Mom started to cry, and it occurred to me that my face may forever be drilled into her memory as the person who told her that her daughter definitely had Down’s. … I felt like I had caused her pain and I somehow had to undo it by hugging her, telling her it would be ok, I had this bizarre thought that I even wanted to take it back. Like my words themselves had given her daughter Down’s (302).

How do you come to terms with a dying child who does not die as expected? How do you deal with the news that your newborn will die within minutes of your giving birth and then have to deal with the fact that she is still alive despite having hardly any brain mass? (105).

I can’t speak for everyone, but in the NICU I have displayed emotion to people that I did not know that well, not because I was necessarily comfortable with the thought, but because I had no other option. The emotion, exhaustion, anger, frustration often come to the surface, whether you want it to or not … it forces us to be honest to one another about our experiences, when you lose the ability to censor things (1105).

A nursing student … said she couldn’t work in the NICU long term, she would cry every day. I told her I thought the same thing, but look where I am working. I don’t cry every day, I don’t cry most days, I don’t cry sometimes when I want to, and I try my best not to cry at home. My husband is just waiting for me to crumble (1110).

For the first time I didn’t hope the baby was crying. I thought, I’m ready if she’s floppy or whatever. And the baby came out, no cry, blue. I calmly grabbed the blade and intubated and sucked out meconium. Awesome! I did it. I was psyched (107).

I actually like those kinds of deliveries. A little excitement, adrenaline pumping, not knowing what to expect. I am glad the babies did well, but selfishly, I do enjoy a good resuscitation from time to time (301).

It was a good experience for me, but at the cost of a baby’s health. In general that is the case for us, the interesting patients are the sick ones (1104).

My Identity as a Physician

For some trainees, the ethical dilemmas that they personally experienced in the NICU challenged their identity as a member of the medical profession. Trainees sometimes felt inadequate to meet their responsibilities, resistant to fulfilling their responsibilities, and doubtful that any physician should have such a responsibility. Some trainees held a guilty wish that a patient would die and release physicians from making decisions that seemed too big.

She weighs 540 grams … how could a baby who weighs less than two apples ever survive a major surgery? I want to believe in the power and possibilities of modern medicine. … But deep in my heart, the most human side of me often feels shamed by our heroic efforts. Why should we ask so much of her? She is supposed to be inside her mother’s womb, doing nothing more than being nurtured and loved! Is putting a 24-wk old infant through mechanical ventilation, anesthesia, and invasive surgery nurturing? But what kind of doctor would I be if I expressed this sentiment aloud to anyone? A fearful, non-believing, weak one? But maybe for once, I don’t want to play the doctor role. Perhaps I just want to be a just, logical, and deeply sensitive human being (603).

It was my first day and I actually wished that one of my patient’s families would decide to withdraw care. I don’t think I’ve ever wished that before. Somehow it felt wrong, almost anti-doctor, to wish for death. On the one hand I feel guilty for even thinking it, but this baby would not have lived with any quality of life. I mean, that’s what I tell myself but who am I to judge what is quality of life. Sometimes it feels like I do not have much of a quality of life outside this hospital (502).

I returned today to find out that the baby I was afraid would hang on for several months with a horrific outcome passed peacefully last night. I was really happy. This is a new concept for me to be happy with death, but that’s how it feels. Today probably won’t be the last time I feel like this but I don’t really like it (502).

Ultimately one of the babies died. I have never actually been present for a patient’s death here. I realize now that this is pretty unique. Last night what struck me, honestly, was how distant I felt from everything. This was probably because I was not physically involved in what was going on. I was standing there, passive, watching it all occur before me. But what was possibly more disturbing, was the numbness that came after. I watched the other resident tear as she described the event and her response. Now maybe this was because she was more connected to this patient, or newer to these experiences, but I couldn’t help feeling that I should be explaining my reaction more than hers. Distant and numb, great (1105).

There were many times when I felt like I would crack, completely overwhelmed by the fact that I simply didn’t have enough time to talk with my patients’ parents … to witness their grief and despair as they struggled with having a very sick child that might not live, and if by some miracle they did, would likely live with a life of disability. This confused and frustrated me to no end … after all, isn’t having the opportunity to be present and listen truly the most important aspect of my work as a physician? (603).

The demanding NICU milieu left many trainees worried about making mistakes and fearful of the serious responsibility that their role demanded. These experiences often caused them to question their abilities as physicians.

The baby was doing well and with all the other sick children, I didn’t check to see that labs had been ordered so I missed it. What if that child ended up with a horrible bilirubin or indication of an infection? (403).

It was a small error that had no impact on patient care, but I knew I would never have made that mistake on a 12-hour shift (101).

I cried for the first time in residency in the NICU. I don’t know what it was exactly that broke me in. Sleep deprivation and a sick child and miscommunication among the team. But I couldn’t shake the feeling that I was guilty and really messed up. Although most of the time I feel in control and capable of completing all the tasks assigned for my patients—this time I couldn’t remember what had been decided on and it was a major oversight. I couldn’t stand the thought that I carelessly endangered a patient’s life—so I broke down. … Turned out that I actually hadn’t messed up but it was too late. I already had the doubt—how could I have messed up so badly? Am I not cut out for this? Is anyone going to trust me again? (305).

My Role as Family, Parent, Person

The experiences that trainees had in the NICU often colored their sense of themselves as a person and their interactions with those in their personal lives. Trainees were unsure of how to integrate their NICU experiences within their own understanding of family, parenting, and personhood.

Sometimes trainees’ intense emotions about the NICU seemed to overwhelm their personal relationships; several trainees described how numbness prevented them from empathizing with family and friends. It was hard for many trainees to leave the NICU behind at the end of the day. The experiences also clearly impacted trainees’ thoughts about their own families and potential families.

I am still exhausted from yesterday. It is amazing how much traumatic patient experiences stick with you. After the adrenaline of the acute event goes away, you sit down and think about the family and the patient on a more personal level. It is hard to maintain a balance, allowing you to process the emotional side, without having these situations take over your personal life and personal time (605).

I’m officially an aunt, as of last Tuesday. … My brother’s worked up because the baby’s had some difficulty with breast feeding. … It’s very difficult to be concerned about such issues and sympathetic, especially in light of the babies in the NICU (103).

Some days, I felt like a dead weight in our newly married relationship. We usually have a pretty jovial banter, but lately I haven’t had much of the spark to keep it going. … I am a little resentful towards the NICU, feeling as though I give so much to each family and seem to have little left to my own (604).

I had a miscarriage earlier this year. It was one of the most difficult things I’ve ever had to go through. I cried a lot and thought “why me” plenty of times. It just doesn’t seem fair that drug addicts and unmarried teenagers can get pregnant but I can’t (107).

Although I often groan about the sometimes seemingly futile care we provide to the sickest babies … I found it hard to express those beliefs out loud when I thought about the possibility that it could be the child of a close friend, and would that change anything (404).

Empathy Scores and Narratives

We compared the narrative themes of the trainees with the lowest JSPE scores (n=5; scores range from 94–106) with those of the trainees with the highest JSPE scores (n=5; scores range from 125–133). There was an inverse relationship in our sample between JSPE scores and years of training. Only one of the five trainees with the lowest JSPE scores was a PGY 1 (pediatric intern), but four of the five trainees with the highest JSPE scores were PGY 1.

Those trainees with the lowest JSPE scores wrote more about disliking the NICU rotation because of the heavy workload and demanding schedule. These trainees more often expressed resentment of NICU staff whom they perceived as unwilling to help trainees more. Conversely, the narratives from those trainees with the highest JSPE scores were more likely to reflect internal conflicts related to caring for infants likely to have very poor outcomes, a greater sense of personal responsibility for patient outcomes, and more poignant interactions with individual families or infants.

Discussion

The emotional and moral growth that physician trainees experience when caring for critically ill and dying patients is an important but often neglected aspect of medical education and professional development (Novack et al. 1997). Although the Accreditation Council for Graduate Medical Education requires that trainees across specialties gain competence in a range of end-of-life care skills, including communication, experience with patient death and dying, and personal awareness (Weissman and Block 2002), there is variability in the implementation of these guidelines. Some skills are formally taught, many become part of the “informal curricula” for residents and fellows, where trainees learn not via intentional curriculum but via the lived experiences—positive and negative—that they and their colleagues endure (Karnieli-Miller et al. 2010).

The longitudinal narrative writings from our sample of physician trainees in a busy neonatal intensive care unit reveal the moral and emotional burdens of accompanying patients and their families through life-threatening illnesses. The writings reflected a process whereby trainees often internalized moral conflicts, so that dilemmas that began with questions about what was right for a patient became questions about whether the trainee was a good physician or even a good person. This distress was hard for many trainees to leave at work and bridged personal and professional domains with implications for trainees’ sense of themselves as members of their own families, as human beings, and as moral agents.

The narrative lens used in this study demonstrates how patients’ stories become important parts of physicians’ stories. Trainees struggled with a sense of role morality (Gibson 2003) as they approached challenging medical decisions by juxtaposing the perspective of the NICU parent with the perspective of the physician versus the theoretical perspective of the physician who could one day be a NICU parent. Writing about the different perspectives seemed to bring trainees both a respect for the different moral decisions people make depending on their role as a stakeholder but also a sense that a “common morality” (Alexandra and Miller 2009) could—or should—balance the different perspectives into the “right” thing to do. Rarely did trainees describe receiving help with these conflicts from senior clinicians or from ethics consultants. Without guidance, the narratives often included mistrust and dehumanization of parents. Blame and anger at parents were not uncommon feelings among trainees, feelings that are rarely expressed among health care professionals. Patient blame has been described as a symptom of physician burnout, but perhaps it is not just the feelings of blame per se but also the stigma against expressing and processing it that exacts its toll.

The ethics of care would suggest that physicians have a moral obligation to care about their patients and that engaging in relationships is an essential component of medical care and medical ethics (Tong 1998; Groenhout 1998). In our study, it was concerning that, in response to conflict, trainees admittedly blunted their emotions and emotional attachments to the patients. Trainees talked of feeling numb at work and at home in response to overwhelming experiences with patients and families. The empathy scores in our sample decreased steadily with greater training, as has been shown in other areas of medicine (Chen et al. 2007; Neumann et al. 2011). Compassion fatigue, or the lessening of clinician compassion in the face of repeated experiences with patient suffering, often results in significant clinician distress and burnout (Bernhardt et al. 2009, 2010). Though our sample size was small, it is interesting to note that the narratives of those participants with the lowest empathy scores were most often focused on the irritations and frustrations of being a trainee in the NICU, while those with the highest empathy scores wrote more about interactions with families and related internal conflicts. We had hypothesized that the act of writing about their NICU experiences would increase participants’ empathy scores, but we found no change. Future work should examine whether debriefing with NICU staff after journaling might have a more marked impact on empathy. Providing additional supports and coping mechanisms for pediatric trainees may be particularly important during their ICU rotations, when they may be most vulnerable (Tschudy 2008).

Learning to manage one’s feelings about patient death is an important means of professional coping. Several studies highlight the need for increased clinician training, coping strategies, and emotional support when faced with patient death (Moores et al. 2007; Geller et al. 2010). Trainees with the least experience may need the most emotional support when faced with patient deaths (Redinbaugh et al. 2003). In pediatrics, the seeming untimeliness and injustice of child death can be particularly distressing to families and clinicians alike. Pediatric trainees are often young parents themselves and so can identify readily with families of critically ill and dying children. The literature suggests significant variability in trainee education related to pediatric death (McCabe, Hunt, and Serwint 2008) and that many trainees experience residual guilt and responsibility after a child’s death (Serwint, Rutherford, and Hutton 2006). Given that so many pediatric deaths occur in infancy, it is not surprising that trainees often find neonatal intensive care to be particularly distressing. The emotional toll of stillbirth and neonatal death also has been shown to lead a significant minority of obstetricians to consider leaving practice (Gold, Kuznia, and Hayward 2008).

Although not a major theme, some trainees wrote about the support they felt from fellow residents who were sharing their experiences. Team dynamics may be particularly important in the intensive care environments where the heavy workload and brief duration of the rotation hinders residents from making meaningful connections with other staff. In fact, the lack of time to foster meaningful relationships may contribute to the physician–nurse tension highlighted in the narratives (Rushton et al. 2013). Physician–nurse communication is further complicated by issues of professional identity, power, and institutional culture (Wanzer, Wojtaszczyk, and Kelly 2009) and may be even more complex when the physician is a trainee. Collaborative practice training may help physicians and nurses build cross-discipline team identities (Bartunek 2011), diminishing differences, establishing support mechanisms, and improving patient care.

Our study reflects the use of narrative as one strategy for support and coping. The unique benefits of narrative medicine are increasingly recognized in medical training (Cohn et al. 2009; Gaufberg et al. 2010; Hatem and Ferrara 2001; Kind, Everett, and Ottolini 2009; Kumagai 2008; Ogur and Hirsh 2009; Pearson, McTigue, and Tarpley 2008; Wear 2002). Enhancing physician empathy is one notable impact of narrative writing (Charon 2001). Narrative writing also has been used to effectively engage trainees around topics such as professionalism (Karnieli-Miller et al. 2011) and ethics (Verkerk et al. 2004). Trainees often appreciate the opportunity to reflect upon and write about their experiences with patients and other clinicians (Hatem and Ferrara 2001). Evidence-based strategies for providing feedback to trainees about their narrative writing have been developed with the goal of enhancing the reflective experience (Wald et al. 2012). Helping trainees to find greater meaning in their work through reflective writing may be an antidote to burnout. We found that trainees in a busy clinical rotation often noted that an obligation to write prompted them to process their reactions to the NICU in a way that would otherwise not have occurred.

Teaching mindfulness practice (Baer et al. 2008; Epstein 1999)—or how to be fully present in the moment—and guided self-awareness exercises have been shown to positively influence physician well-being and patient-centered care attitudes (Krasner et al. 2009). Understanding one’s own biases, trigger points, communication style, and stress level gives physicians the opportunity to modulate responses, understand patient experiences more fully, and attend to self- care needs. Debriefing residents at the end of each intensive care rotation, as we do, gives them time to reflect on the “human response” to challenging and rewarding cases and gives them permission to explore these responses in a safe environment.

Developing methods to enhance empathic communication by clinicians also is likely to enhance patients’ experiences. Our previous work has shown that parents feel more trusting of physicians who communicate with compassion and empathy (Boss et al. 2008). Others have shown that parents’ feelings of trust in their physician may be more important to their long-term outcome after an infant death than the process of end-of-life decision-making itself (Brosig et al. 2007). Because of the importance in neonatology of engaging with families to determine what is in the best interests of the infant and family, more systematic preparation of trainees to do this must be incorporated into these rotations. Narrative medicine should be explored as a way to apply quality improvement to the therapeutic relationship in this context (Greenhalgh and Heath 2010).

We achieved high enrollment and compliance rates with this research. Anecdotally, several residents suggested that narrative writing be part of the NICU rotation as a learning experience. Residents may have passed this attitude on to potential participants. Incentives also may have increased participation. Our study has several potential limitations. The data were gathered from a single training center; trainee concerns may differ in other neonatal intensive care settings. Trainees who participated in the study may differ from those who did not. Participants may not have been entirely forthcoming in their writings due to concerns regarding confidentiality or negative repercussions.

Our work adds to the existing literature by describing the evolving NICU experiences of trainees as their patients’ illnesses unfold. The writings offer a prospective, real-time picture of the training curriculum as experienced in the NICU. The longitudinal writing allowed participants to focus not just on the “most memorable” experiences of their clinical rotation but on the more consistent everyday experiences. The writings also offered a sense of how caring for critically ill and dying patients impacts a trainee’s own sense of identity and sense of moral agency.