Original Article

Journal of Perinatology (2009) 29, 310–316; doi:10.1038/jp.2008.228; published online 15 January 2009

Analysis of enacted difficult conversations in neonatal intensive care

G Lamiani1,2, E C Meyer1,3, D M Browning1,4, D Brodsky1,5 and I D Todres1,6

  1. 1Institute for Professionalism and Ethical Practice, Children's Hospital, Boston, MA, USA
  2. 2Department of Medicine, Surgery and Dentistry, University of Milan, Italy
  3. 3Department of Psychiatry, Harvard Medical School, Boston, MA, USA
  4. 4Education Development Center Inc., Newton, MA, USA
  5. 5Neonatal Intensive Care Unit, Beth Israel Deaconess Center, Boston, MA, USA
  6. 6Medical Ethics Unit, Massachusetts General Hospital for Children, Boston, MA, USA

Correspondence: G Lamiani, CURA Research Center, University of Milan, Via Di Rudiní 8, 20142 Milan, Italy. E-mail: Giulia.lamiani@unimi.it

Received 18 June 2008; Revised 2 November 2008; Accepted 3 December 2008; Published online 15 January 2009.





To analyze the communicative contributions of interdisciplinary professionals and family members in enacted difficult conversations in neonatal intensive care.

Study Design:


Physicians, nurses, social workers, and chaplains (n=50) who attended the Program to Enhance Relational and Communication Skills, participated in a scenario of a preterm infant with severe complications enacted by actors portraying family members. Twenty-four family meetings were videotaped and analyzed with the Roter Interaction Analysis System (RIAS).



Practitioners talked more than actor-family members (70 vs 30%). Physicians provided more biomedical information than psychosocial professionals (P<0.001), and less psychosocial information than nurses, and social workers and chaplains (P<0.05; P<0.001). Social workers and chaplains asked more psychosocial questions than physicians and nurses (MD=P<0.005; RN=P<0.05), focused more on family's opinion and understanding (MD=P<0.01; RN=P<0.001), and more frequently expressed agreement and approval than physicians (P<0.05). No differences were found across disciplines in providing emotional support.



Findings suggest the importance of an interdisciplinary approach and highlight areas for improvement such as using silence, asking psychosocial questions and eliciting family perspectives that are associated with family satisfaction.


difficult conversations, neonatal intensive care, interdisciplinary communication, RIAS, family-centered care



Practitioner–family communication in the neonatal intensive care unit (NICU) can be particularly difficult and presents unique challenges. It is often difficult for clinicians to accurately predict neonatal outcomes, leaving parents with uncertainty about the infant's survival and developmental outcomes. Parents whose infants require NICU care may experience loss of control over the most basic issues such as their infant's protection and well-being.1 Conversations about end-of-life care are especially emotionally charged because of the difficult if not ‘impossible’ decisions that are required of parents.2, 3, 4 Given these unique challenges, family meetings in the NICU often involve an interdisciplinary approach, with attendance by physicians, nurses, and psychosocial professionals. Understanding the contribution that different practitioners and families make during these discussions is important to the process of improving the quality of care.

The aim of this study was to analyze the communicative contributions of physicians, nurses, psychosocial professionals, and family members in enacted family meetings in the NICU. Using the Roter Interaction Analysis System (RIAS),5 we analyzed difficult conversations that focused on communication of serious medical complications and the possibility of withdrawal of life-sustaining therapy. As examining these conversations in actual clinical encounters is an ethically complex and sensitive undertaking,6 we analyzed video-recorded realistic enactments between actual practitioners and family members portrayed by professional actors as part of educational workshops offered by the Program to Enhance Relational and Communication Skills (PERCS) at Children's Hospital Boston. Although there are limitations to the use of realistic enactments, by providing experiential realism7 the enactments permitted an initial exploration of the contribution of different practitioners and family members to difficult conversations.





Groups of 10–15 physicians, nurses, social workers, chaplains, and medical interpreters with varied levels of experience attended one of eight PERCS-NICU workshops at Children's Hospital Boston from November 2005 to June 2007. PERCS, a program of the Institute for Professionalism and Ethical Practice, offers day long workshops featuring realistic enactments to teach communication and relational skills involved in difficult conversations (www.ipepweb.org). A detailed description of the program pedagogy and evaluation outcomes is described elsewhere.8, 9

During the workshops, the clinical course of a preterm infant, Sylvia Ortiz, and conversations with her family were presented (Appendix 1 is available online). Interdisciplinary participants, usually a physician, nurse, and/or a social worker or a chaplain, volunteered to meet with Sylvia's parents, portrayed by actors. Each workshop consisted of three consecutive meetings with Sylvia's family. In the first meeting, the team shared that Sylvia had developed a serious complication, bilateral intraventricular hemorrhages with ventricular dilation and unilateral intraparenchymal brain hemorrhage. In the second meeting, after Sylvia's cranial hemorrhage had progressed, the team met again to update parents and maternal aunt about the worsening of her medical condition and introduced the options of continuing or withdrawing life-sustaining therapy. In the third and final meeting, practitioners and the actor-family members discussed withdrawal of life support, as requested by the family. The enacted family meetings were video-recorded. Participants completed pre and post-questionnaires that included demographic characteristics and self-appraisal of communication skills and relational abilities.

Family members

Family members were portrayed by professional actors. The scenario and the roles of actor-family members were defined in advance; however, the unfolding of each conversation was unique because practitioners approached the conversation differently. Unlike other programs where actors portray standardized patients or adhere to a specific script, to increase experiential realism, the actors were encouraged to improvise responses based on the approach that the individual practitioner had taken.10

Research ethics

The Institutional Review Board of Children's Hospital, Boston, reviewed the study and considered it exempt according to criterion no. 1 (that is, research conducted in established educational settings involving normal educational practices) under the Health and Human regulations 45, Code of Federal Regulation 46. All PERCS participants gave their written consent for their videotaped meetings and questionnaire data to be used for research purposes.

Data analysis

Video-recorded family meetings were coded by trained RIAS coders at Johns Hopkins University, who were external to the research team and unaware of the purpose of the study. RIAS is one of the most widely used quantitative systems to analyze provider–patient interaction (http://www.rias.org/manual). To our knowledge, the RIAS has never been used to analyze difficult conversations in NICU settings. The RIAS has demonstrated strong reliability and predictive validity to a variety of patient outcomes such as satisfaction, comprehension and compliance11, 12 and provided the capacity to analyze clinical encounters with more than two participants involved. Inter-rater reliability has ranged across categories from 0.7 to 0.9 based on Pearson correlation coefficients.13, 14 Coding was performed directly from audiotapes without transcription.

In accordance with the RIAS coding criteria, the communication flow was divided into utterances defined as the ‘smallest discriminable speech segment’5 to which a meaning and a classification could be assigned. All practitioner and actor-family member utterances were coded and classified according to one of the 41 mutually exclusive RIAS categories, depicting the content of the communication exchange. To facilitate the analysis of the family meetings’ content and allow for a comparison of the results with other studies,15, 16 we aggregated the 41 RIAS categories into the following 10 composite categories: Biomedical information (information on medical condition and treatment/procedures), Psychosocial information (information on psychosocial, lifestyle and spiritual issues), Biomedical questions (questions on medical condition and treatment/procedures), Psychosocial questions (questions on psychosocial, lifestyle and spiritual issues), Orientation (directions and instructions), Social talk (non medical chit-chat), Positive talk (approvals and agreements), Negative talk (criticism and disagreement), Emotional talk (empathy, concern and reassurance) and Facilitation and family activation (asking for family opinion, paraphrasing and asking for understanding). Table 1 summarizes the RIAS composite categories with illustrative excerpts from NICU enactments.

Statistical analysis

Statistical analyses were performed with SPSS for Windows 14.0. Participants were grouped into the following disciplines: physician, nurse, and social workers and chaplains. Medical interpreters did not participate in enacted conversations. One-way ANOVA was performed to detect differences in the RIAS composite categories by discipline and by nature of the meeting (conveying bad news, introducing withdrawal of life support, and discussion of withdrawal of life support). Statistical significance was set at Pless than or equal to0.05. Where significant differences were detected, post hoc multiple comparisons were conducted using the Tukey method to determine which professional group and meeting differed significantly from the others. To account for the different number of utterances per discipline, ANOVA and post hoc multiple comparisons were conducted on relative frequencies.




Of the 74 participants who attended the workshops, 50 (68%) participated in at least one enacted family meeting. No significant demographic differences were found between participants who did or did not participate in enactments (Table 2). Eight actors (two men, six women; mean age=36.7 years) portrayed family members.

Family meetings

During each workshop, three meetings with Sylvia Ortiz's family were enacted, for a total of 24 meetings. Practitioners rated the scenario as very realistic (mean=4.8; s.d.=0.5) on a 5-point Likert scale (1=not at all realistic; 5=very realistic). The average duration of the enacted family meetings was 21.6min [range=13.1–31.8] with involvement of 2–3 actor-family members and 1–4 practitioners. The mean number of practitioners’ utterances per meeting was 239; for 70% of the total. The mean number of actor-family members’ utterances was 101; for 30% of the total. Neither the nature of the meeting (that is, conveying bad news, introducing withdrawal of life support, and discussion of withdrawal of life support) nor the gender of practitioners had any significant impact on the duration or verbal dominance of the meetings. The communicative content of practitioners’ and actor-family members’ communication during the family meetings is summarized in Table 3.

Of the total number of practitioners’ utterances, physicians spoke the most (56%), followed by nurses (33%) and social workers and chaplains (11%). The communicative contribution of physicians, nurses and social workers and chaplains was compared and the differences across disciplines are reported in Supplementary Table 1.

Differences in the RIAS categories of both the practitioners and actor-family members were found based on the nature of the meeting. Practitioners provided more biomedical information when first conveying the difficult news, than when discussing withdrawal of life support (P=0.019). Practitioners expressed more empathy, concern, and reassurance when explaining the worsening of Sylvia's medical conditions and introducing the possibility of withdrawal of life support, than when discussing withdrawal of life support (P=0.011). Practitioners asked more psychosocial questions when discussing withdrawal of life support than in the other meetings (first P=0.007; second P=0.005). When being introduced to the possibility of withdrawal of life support, actor-family members used more negative statements, such as criticism and disagreement, than in the other meetings (P<0.001). Actor-family member also asked more psychosocial questions than when first learning the difficult news (P=0.018), and they asked more biomedical questions (P=0.022) and used more facilitation statements (P=0.011) than when discussing withdrawal of life support. Actor-family members provided more psychosocial information when discussing withdrawal of life support than in the other meetings (first P=0.001; second P=0.045).



Patient and family members highly value clear communication and empathic relationships with healthcare providers and often base their trust and perception of the quality of and satisfaction with care on such indices.17, 18, 19, 20 In our study, practitioners verbally dominated family meetings talking substantially more than actor-family members (70 vs 30%). Our results using realistic enactments are strikingly consistent with the findings of McDonagh et al.,21 who reported that in actual end-of-life conferences in adult intensive care units (ICUs), clinicians spoke 71% and family members 29% of the time. Greater opportunity for family members to speak in ICU family meetings has been positively associated with family satisfaction and less conflict with staff.21 The unequal proportion of family-to-practitioner speech observed in our study highlights an area of clinical practice worthy of further exploration and improvement.

Analysis of the content of the conversations revealed that practitioners, especially physicians, devoted most of their statements to providing biomedical information. Actor-family members used statements to ensure that they correctly understood the information provided, such as recapping what the practitioners had said and paraphrasing, more frequently than practitioners. Previous studies have found that clarity and honesty of information is of pivotal importance for families in pediatric17, 22 and neonatal intensive care settings,23 and is associated with higher parent ratings of the quality of care. Unfortunately, information regarding patient's diagnosis, treatment and prognosis is often not communicated effectively to family members in ICU24 and NICU.25 Our study seems to confirm that despite practitioners’ provision of biomedical information, the burden of clarification rests with the family.

NICU practitioners provided more emotional support (28%), such as expressions of empathy, partnership, concern and reassurance, than practitioners of other medical specialties such as anesthesia, where it accounted for just 7% of the physicians’ utterances, and general (4%) or orthopedic surgery (3%).26, 27 Our findings confirm the complex nature and emotional engagement that NICU conversations often entail. Interestingly, no differences were found across disciplines in offering emotional support suggesting that this is perceived as a pivotal component of NICU care and that physicians, nurses, social workers and chaplains all contribute to families’ emotional well-being. Providing communication and relational trainings, and debriefing opportunities may be especially valuable for physicians and nurses, who historically receive limited education in this area of practice.

Reflecting their characteristic role as the bearer of difficult news,28 physicians generally talked the most in conversations, offering primarily biomedical information. Nurses’ communication style bridged the medical and the psychosocial worlds as their communicative contributions differed neither from those of physicians in providing biomedical information, nor from those of social workers and chaplains in providing psychosocial information and engaging families in social talk. Nurses played an important part in providing parents information about treatment and procedures and offering psychosocial information and counseling, thus representing a knowledgeable resource for parents. Nurses’ communicative style highlighted here, including formal information, emotional support as well as social talk, has been identified by parents as characterizing good interactions with nurses.29

Compared to the other practitioners, social workers and chaplains asked more psychosocial questions, sought more family opinion and understanding, and more frequently expressed agreement and approval. This finding is especially meaningful because integration of psychosocial aspects of care and the opportunity for parents to express their opinion and understand information have been highly valued by parents.15, 22, 29 Wolfe et al.30 reported that the inclusion of psychosocial professionals in pediatric oncology family meetings was associated with greater concordance in the timing of physicians’ and parents’ understanding that the child had no realistic chance for cure, and also with greater integration of palliative care. Further studies may be valuable to determine whether the unique communication contributions of social workers and chaplains highlighted here, such as asking psychosocial questions, eliciting the family perspective, and engaging in positive talk, enhance patient and family outcomes.

The communicative differences noted across disciplines highlight the advantage of an interdisciplinary approach when holding difficult conversations because each discipline contributes in unique ways to the unfolding of these conversations. Interdisciplinary communication is a key component of good end-of-life care for patients and families.31 Including nurses, social workers and chaplains in family meetings may enhance the quality of communication exchanges and meet the different family needs.

The evolving nature of the difficult conversations also influenced practitioners’ and family members’ communication. During meetings where difficult news of a serious complication first had to be conveyed, practitioners focused on providing medical information. In meetings where the options of continuing or withdrawing life support were raised, actor-family members were more active in asking medical and psychosocial information that would help them to make decisions, and expressed more criticism of health care professionals. In our clinical experience, parents may be more likely to express disagreement or criticism in discussions about end-of-life if they feel that their opinions are not sought or if the news and timing of the discussion are unexpected. Finally, meetings in which withdrawal of life support was described were characterized by practitioners asking and by actor-family members providing more psychosocial information including spiritual issues. These findings suggest that as end-of-life conversations shift in focus from cure to care, consideration of psychosocial issues takes on greater importance than the conveyance of biomedical information.

Limitations of the study

Participants were largely self-selected and the data regarding the communicative content of family members’ talk derived from realistic enactments repeated a number of times, thus limiting their generalizability. That said, practitioners rated the scenario as very realistic and the data generated from the enacted family meetings were strikingly similar to published accounts of actual family meetings.21 Participants were aware of being videotaped within the program, which may have affected their performance; however, studies have demonstrated that the effect on performance is minimal.32, 33 Although RIAS allowed for a quantitative and comparable description of the communicative contribution of practitioners and family members, it did not capture potentially important aspects of communication such as non-verbal gestures and moments of silence. Finally, the small number of practitioners for each professional group did not permit analysis of the effects of other variables, such as participants’ experience, which may affect the way in which difficult conversations are conducted.


During the last years, many efforts have been made to implement better practices for family-centered care in the NICU.34, 35 This study highlighted areas for improvement at a communicational level. Practitioners in the NICU are encouraged to listen more, integrate psychosocial aspects of care, seek family's opinion and ensure that the information regarding medical condition and prognosis of their newborns is understood. An interdisciplinary approach, with the richness of contributions and viewpoints that it entails, could better serve families of critically ill newborns to meet their needs during their journey in the NICU.



  1. Meyer EC, Bigsby R. Family-centered and relationship-based care in the NICU In: Paul H. Brookes, Developmental and therapeutic interventions in the NICU. Brookes Publishing: Baltimore, 2004, pp 163–176.
  2. Jellinek MS, Catlin EA, Todres DI, Cassem EH. Facing tragic decisions with parents in the neonatal intensive care unit: Clinical perspectives. Pediatrics 1992; 89: 119–122. | PubMed | ChemPort |
  3. Javier A. How much emotion is enough. J Clin Ethics 2007; 18: 362–365. | PubMed |
  4. Browning DM, Meyer EC, Brodsky D, Truog RD. Reflection on love, fear and specializing in the impossible. J Clin Ethics 2007; 18: 373–376. | PubMed |
  5. Roter DL. Roter Interaction analysis system: Coding manual and training materials. Johns Hopkins University: Baltimore, 1989.
  6. Ziv A, Wolpe PR, Small SD, Glick S. Simulation-based medical education: an ethical imperative. Acad Med 2003; 78: 783–788. | Article | PubMed |
  7. Dieckmann P, Gaba D, Rall M. Deepening the theoretical foundations of patient simulation as social practice. Simul Healthc 2007; 2: 183–193. | PubMed |
  8. Browning DM, Meyer EC, Truog RD, Solomon MZ. Difficult Conversations in Health Care: Cultivating Relational Learning to Address the Hidden Curriculum. Acad Med 2007; 82: 905–913. | Article | PubMed |
  9. Meyer EC, Sellers DE, Browning DM, McGuffie K, Solomon MZ, Truog RD. Difficult conversations: Improving communication skills and relational abilities in healthcare. Pediatric Critical Care Medicine, in press.
  10. Mckoul G. Communication skills: How simulation training supplements experiential and humanist learning. Acad Med 2006; 81: 271–274. | Article | PubMed |
  11. Inui TS, Carter WB, Kukull WA, Haigh VH. Outcome-based doctor-patient interaction analysis: I. Comparison of techniques. Med Care 1982; 10: 535–549. | Article |
  12. Bensing JM, Dronkers J. Instrumental and affective aspects of physician behavior. Med Care 1992; 30: 283–298. | Article | PubMed | ChemPort |
  13. Roter D, Hall J, Katz N. Relations between physicians' behaviors and patients' satisfaction, recall, and impressions: An analogue study. Med Care 1987; 25: 399–412. | Article | PubMed |
  14. Roter DL, Stewart M, Putnam S, Lipkin M, Stiles W, Inui T. Communication patterns of primary care physicians. JAMA 1997; 277: 350–356. | Article | PubMed | ChemPort |
  15. Bertakis KD, Roter D, Putnam SM. The relationship of physician medical interview style to patient satisfaction. J Fam Pract 1991; 32: 175–181. | PubMed | ISI | ChemPort |
  16. Roter DL, Larson S, Fischer CS, Arnold RM, Tulsky JA. Experts practice what they preach: A descriptive study of best and normative practices in end of life discussions. Arch Intern Med 2000; 160: 3477–3485. | Article | PubMed | ChemPort |
  17. Mack JW, Hilden JM, Watterson J, Moore C, Turner B, Grier HE et al. Parent and physician perspectives on quality care at the end of life in children with cancer. J Clin Oncol 2005; 23: 9155–9161. | Article | PubMed |
  18. Abbott KH, Sago JG, Breen CM, Abernethy AP, Tulsky JA. Families looking back: one year after discussion of withdrawal or withholding of life-sustaining support. Crit Care Med 2001; 29: 197–201. | Article | PubMed | ChemPort |
  19. Contro N, Larson J, Scofield S, Sourkes B, Cohen H. Family perspectives on the quality of pediatric palliative care. Arch Pediatr Adolesc Med 2002; 156: 14–19. | PubMed | ISI |
  20. Azoulay E, Pochard F, Barnes NK, Chevret S, Aboab J, Adrie C et al. Risk of post-traumatic stress symptoms in family members of intensive care unit patients. Am J Respir Crit Care Med 2005; 171: 987–994. | Article | PubMed |
  21. McDonagh JR, Elliott TR, Engelberg RA, Treece PD, Shannon SE, Rubenfeld GD et al. Family satisfaction with family conferences about end-of-life care in the intensive care unit: increased proportion of family speech is associated with increased satisfaction. Crit Care Med 2004; 32: 1484–1488. | Article | PubMed |
  22. Meyer EC, Ritholz MD, Burns JP, Truog RD. Improving the quality of end-of-life care in the pediatric intensive care unit: parents' priorities and recommendations. Pediatrics 2006; 117: 649–657. | Article | PubMed | ISI |
  23. Kowalski WJ, Leef KH, Mackley A, Spear ML, Paul DA. Communicating with parents of premature infants: who is the informant? J Perinatol 2006; 26: 44–48. | Article | PubMed | ChemPort |
  24. Azoulay E, Chevret S, Leleu G, Pochard F, Barboteu M, Adrie C et al. Half the families of intensive care unit patients experience inadequate communication with physicians. Crit Care Med 2000; 28: 3044–3049. | Article | PubMed | ChemPort |
  25. Harrison H. The principles for family-centerd neonatal care. Pediatrics 1993; 92: 643–650. | PubMed | ISI | ChemPort |
  26. Levinson W, Chaumeton N. Communication between surgeons and patients in routine office visits. Surgery 1999; 125: 127–134. | PubMed | ChemPort |
  27. Kindler CH, Szirt L, Sommer D, Häusler R, Langewitz W. A quantitative analysis of anesthetist-patient communication during the pre-operative visit. Anaesthesia 2005; 60: 53–59. | Article | PubMed | ChemPort |
  28. Meyer EC, Burns JP, Griffith JL, Truog RD. Parental perspectives on end-of-life care in the pediatric intensive care unit. Crit Care Med 2002; 30: 226–231. | Article | PubMed |
  29. Jones L, Woodhouse D, Rowe J. Effective nurse parent communication: A study of parents' perceptions in the NICU environment. Patient Educ Couns 2007; 69: 206–212. | Article | PubMed |
  30. Wolfe J, Klar N, Grier HE, Duncan J, Salem-Schatz S, Emanuel EJ et al. Understanding of prognosis among parents of children who die of cancer. JAMA 2000; 284: 2469–2475. | Article | PubMed | ChemPort |
  31. Curtis JR, White DB. Practical guidance for evidence-based ICU family conferences. Chest 2008; 134: 835–843. | Article | PubMed |
  32. Pringle M, Stewart-Evans C. Does awareness of being video recoded affect doctors' consultation behavior? Br J Gen Pract 1990; 40: 455–458. | PubMed | ChemPort |
  33. Coleman T, Manku-Scott T. Comparison of video-recorded consultations with those in which patients' consent is withheld. Br J Gen Pract 1998; 48: 971–974. | PubMed | ChemPort |
  34. Dunn MS, Reilly MC, Johnston AM, Hoopes RD, Abraham MR. Development and dissemination of potentially better practices for the provision of family-centered care in neonatology: the family-centered care map. Pediatrics 2008; 118: S95–S107. | Article |
  35. Cooper LG, Gooding JS, Gallagher J, Sternesky L, Ledsky R, Berns SD. Impact of a family-centered care initiative on NICU care, staff and families. J Perinatol 2007; 27: S32–S37. | Article | PubMed |


We honor and dedicate this work to I David Todres, our friend and colleague, who passed away during the final revisions of this article. We acknowledge the Argosy Foundation for financial support of this work. We thank Anne Hansen, MD, Susan Larson, MS, Debra Roter, MPH. DrDh, Robert Truog, MD, and Allyson Wall, BA, for their many contributions. The authors express their gratitude to Pam Varrin, PhD, Erin Ward, MEd, Meg Comeau, MHA, who enriched the PERCS workshops offering parent perspectives, and to the actors and PERCS participants.

Supplementary Information accompanies the paper on the Journal of Perinatology website (http://www.nature.com/jp).