A Descriptive Study of Decision-Making Conversations during Pediatric Intensive Care Unit Family Conferences

Abstract

Background:

Little is known about how decision-making conversations occur during pediatric intensive care unit (PICU) family conferences (FCs).

Objective:

Describe the decision-making process and implementation of shared decision making (SDM) during PICU FCs.

Design:

Observational study.

Setting/Subjects:

University-based tertiary care PICU, including 31 parents and 94 PICU healthcare professionals involved in FCs.

Measurements:

We recorded, transcribed, and analyzed 14 PICU FCs involving decision-making discussions. We used a modified grounded theory and content analysis approach to explore the use of traditionally described stages of decision making (DM) (information exchange, deliberation, and determining a plan). We also identified the presence or absence of predefined SDM elements.

Results:

DM involved the following modified stages: information exchange; information-oriented deliberation; plan-oriented deliberation; and determining a plan. Conversations progressed through stages in a nonlinear manner. For the main decision discussed, all conferences included a presentation of the clinical issues, treatment alternatives, and uncertainty. A minority of FCs included assessing the family's understanding (21%), assessing the family's need for input from others (28%), exploring the family's desired decision-making role (35%), and eliciting the family's opinion (42%).

Conclusions:

In the FCs studied, we found that DM is a nonlinear process. We also found that several SDM elements that could provide information about parents' perspectives and needs did not always occur, identifying areas for process improvement.

Introduction

Quality communication is essential in the pediatric intensive care unit (PICU), particularly when facing life-changing decisions. Parents of dying children may need to consider withdrawing life-sustaining therapies, avoiding, or providing cardiopulmonary resuscitation, or approaches to end-of-life care.¹ Parents of medically complex patients may decide to place a tracheostomy tube; initiate chronic ventilation; and/or accept palliative care or limits on life-prolonging therapies.¹ Understanding how parents and physicians approach decision making (DM) in the PICU could identify opportunities for improvement.

The components of healthcare DM, analysis and judgement,² are often depicted in three stages: (1) information exchange; (2) deliberation; and (3) plan determination.^3,4 Three prevailing models describe how these stages are operationalized: (1) paternalistic DM (emphasizing physician authority); (2) “autonomous”/informed DM (emphasizing patient/family authority); and (3) shared DM (emphasizing shared authority reflecting patient/family preferences).^3–5 Experts endorse shared decision making (SDM) for preference-sensitive healthcare decisions.⁶ Recommendations for implementing SDM involve introducing choice, describing options, and making a decision.⁷

Family conferences (FCs) are commonly used when clinicians discuss difficult decisions with patients and their families, especially in intensive care units.^8–12 FCs help bring families and patients together with healthcare team members (HTMs). Effectively conducted FCs for critically ill adult patients can improve family understanding, increase satisfaction, and lessen the burden of caregiver bereavement when patients die.^13–15 While many studies about FC decision-making conversations for adult patients exist,^11,13–18 the literature about PICU FCs is less robust.^19–23

This report provides information about decision-making conversations using a subset of data from a PICU FC study.²³ This analysis focuses on FCs involving DM discussions. We explored if and how decision-making conversations integrated the three DM stages described above (information exchange; deliberation; and plan determination), and if they included previously defined SDM elements.¹⁶

Methods

Setting

This study took place at a university-based, tertiary care PICU from January 1, 2011 to January 1, 2012, a timeframe determined a priori.²³ Patients in the 42-bed “closed” PICU (the attending for all patients was a pediatric intensivist) included medical and surgical (general or cardiac) patients. The PICU teams included an attending, PICU fellow, and pediatric residents or PICU advance practice nurses (APNs) and subspecialists. The project was approved by the hospital's Institutional Review Board.

Identifying FCs

We defined FCs as planned meetings involving parents, HTMs (e.g., physicians, APNs, nurses, social workers, chaplains), and when present, non-HTMs providing parent support (e.g., family or friends).²³ We did not include impromptu parent updates, one-on-one bedside conversations, or discussions held during daily family-centered rounds. We identified FCs by weekday chart review and communication with the medical team. We included only weekday FCs secondary to staffing limitations. We excluded conferences when any participant did not provide written consent or did not speak English or Spanish. Further details have been described previously.²³

Data collection

We audio and video recorded FCs using unobtrusive recording devices managed by a study team member present during the meeting. Some conferences were only audio recorded at participants' requests. The study team member identified the FC “leader” as the person moderating the FC. After the conference, parents and HTMs completed a written questionnaire requesting demographics.²³ We collected patient information via medical chart review. The audio recordings were transcribed verbatim and de-identified. For FCs with Spanish speakers involving an interpreter, only English dialogue was transcribed. We did not analyze video data because our focus was on dialogue, not behaviors.

Data analysis

Identifying FCs, including DM discussions

Three reviewers (K.M., N.H.B., and C.R.) reviewed the transcripts and ascribed goals to each FC based on previously described FC goals.²⁴ Ascribed goals were compared and discrepancies resolved by consensus. Goal categories included medical update (medical information regarding patient status), care planning (future medical plans during the hospitalization or discharge planning), and DM (patient care decisions).²³ In this analysis, we include FCs with DM as an ascribed goal.

Qualitative analysis of FC transcripts

We combined the inductive process of grounded theory with content analysis.²⁵ Two authors (K.M. and M.S.) read and labeled the transcripts identifying themes, concepts, and patterns to understand the conversation content and flow. Using these labels, we created a preliminary coding dictionary. We reread the transcripts to identify excerpts (a sentence or contiguous-related sentences) that conveyed similar content. We tagged the excerpts with the preliminary codes or created new codes as needed. We achieved saturation (no new codes created) by the third transcript. With our final coding dictionary, two researchers (M.S. and M.A.) separately assigned codes to the excerpts. Agreement between the two coders within a transcript ranged from 67% to 84% (average: 75.64%). Discrepancies between the coders were discussed and resolved through consensus. Excerpts were categorized as either spoken by a HTM or a parent. Family members and support individuals spoke infrequently and thus were grouped with parents (hereafter, “parents” refers to parents, family, or support individuals).

After coding each FC transcript, two authors (M.S. and K.M.) organized codes into themes and then grouped the themes into categories corresponding to previously defined decision-making stages—information exchange, deliberation, and determining a plan.³ We created new categories if themes did not clearly fit into one of these predefined stages. We present the final categories as DM stages in the results below. To analyze the flow of conversation between stages, we color-highlighted the excerpts according to their coded decision-making stage. This provided a visual representation of how HTMs and parents navigated through the decision-making stages. We used Dedoose (Dedoose Version 5.2.0 (2014). Los Angeles, CA: SocioCultural Research Consultants, LLC [www.dedoose.com]) for qualitative analysis.

Quantitative analysis of FC transcripts

We modified a previously validated measure of SDM elements.¹⁶ Two reviewers (C.R. and N.H.B.) separately reviewed parts of the transcripts that focused on the main decision discussed (as determined by consensus between the two reviewers), and noted the presence or absence of each SDM element. Initial agreement between the two coders ranged from 50% to 100% per conference (average: 81.82%). Discrepancies were resolved through consensus, or by a third reviewer (K.M.) in cases without consensus.

General quantitative analysis

We present quantitative information with descriptive statistics analyzed with Microsoft Excel (Microsoft Office 2013, version 15.0.4937.1000).

Results

We recorded 23 of the 39 FCs for which participation was requested (59%). Technical problems prohibited use of one FC recording. We reviewed transcripts from 22 FCs discussing 20 patients. Of the 22 transcripts analyzed, 14 involved DM discussions. The eight excluded conferences involved medical update and care planning discussions. (Fig. 1) Decisions included the following: introducing palliative care; withdrawal of life-sustaining therapies; “do not resuscitate” orders; agreeing to a new surgical or medical therapy; placement of a tracheostomy or gastrostomy tube; use of chronic noninvasive positive airway pressure ventilation; transfer to a chronic care facility; and initiation of long-term ventilation.²³ Some FCs involved discussions about multiple decisions.

FIG. 1.

Family conference inclusion process.

Of the 14 analyzed FCs, 11 were video and audio recorded, and 3 were only audio recorded. Five FCs involved a Spanish-language interpreter. The FCs lasted an average of 48 minutes (median 45.5 minutes, range 13–93 minutes) and occurred, on average, 45.4 days after PICU admission (median 30.0 days, range 3–172 days). On average, 6.8 HTMs (median 6.5, range 4–10) attended the FCs. FCs were led by a PICU fellow or attending physician (n = 5, 36%) or by a pulmonary (n = 6, 43%), cardiology (n = 1, 7%), neurology (n = 1, 7%), or hematology/oncology (n = 1, 7%) subspecialist. The FCs included 25 parents. Eleven FCs included two parents and four included at least one family support person. Table 1 shows the demographic data for the 20 parents (involved in 10 FCs) who completed a postconference questionnaire. Table 2 provides patient demographics and clinical characteristics.

Table 1.

Parent Demographics

Variable	N = 20^a (%)
Sex
Male	9 (45)
Female	11 (55)
Race
Black	6 (30)
White	13 (65)
Other	1 (5)
Ethnicity
Hispanic/Latino	8 (40)
Not Hispanic/Latino	12 (60)
Primary language
English	10 (71)
Spanish	4 (29)
Marital status
Single	4 (20)
Married	16 (80)
Education
High school	12 (60)
College	5 (25)
Graduate school	1 (5)
Missing	2 (10)

Missing data from five parents involved in four conferences.

Table 2.

Patient Demographics and Clinical Characteristics

Variable	N = 14 (%)
Sex
Male	8 (57)
Female	6 (43)
Race/ethnicity
Black	3 (21)
White	4 (29)
Hispanic/Latino	7 (50)
Age (years)
Average (median, range)	6.3 (5, 1–19)
Time in PICU before FC (days)
Average (median, range)	45.4 (30.0, 3–172)
Insurance
Public	7 (50)
Private	7 (50)
Previous palliative care involvement
Yes	2 (14)
Required respiratory support
Yes	7 (50)
No^a	7 (50)
Required pressors
Yes	2 (14)
New involvement with PC
Yes	3 (21)
Required CPR
Yes	1 (7)
New tracheostomy or chronic invasive ventilator support/new GT placed
Yes	2 (14)
New DNAR order
Yes	4 (29)
Died
Yes	2^b (14)
Required ECMO
Yes	1 (7)

Could have been given FiO2 via NC or mask.

one patient had active withdrawal of life-sustaining therapies; one died despite continuing all therapies.

CPR, cardiopulmonary resuscitation; DNAR, do no attempt resuscitation; ECMO, extracorporeal membrane oxygenation; FC, family conference; GT, gastrostomy tube; PC, palliative care; PICU, pediatric intensive care unit.

Qualitative analysis of FCs

Figure 2 depicts our DM stages and associated themes. Discussion with codes pertaining to deliberation seemed inherently linked to either information exchange—understanding the details of the decision and their impact on the patient—or determining a plan—defining the plan and how to operationalize a decision. Therefore, we redefined the DM stages as information exchange; information-oriented deliberation; plan-oriented deliberation; and determining a plan. Table 3 presents these stages, the themes within each stage, the codes assigned to each theme, and example quotes from parents and HTMs. Below, we elaborate on the DM stages and themes.

FIG. 2.

Decision-making stages and associated themes.

Table 3.

Stages of Decision Making

Decision-making stages ^a	Code definition	Example quotes
Information exchange
Current patient status
Reason for conference	Comments about the conference goals and plans	“We want to give you an update on where X stands, and what we think we can anticipate…” HTM
		“That's why we have meeting. We need to know what everybody thinks about.” Parent
Opinion of medical status	Personal interpretation of patient's medical status	“He's not requiring 80 percent oxygen. He's requiring between 40 to 25 which is a pretty good…” HTM
		“I don't feel that she needs ventilator now.” Parent
Provides medical/patient information	Giving information about how the patient is doing or what is occurring with the patient, medically or otherwise (before making a decision)	“She's in pain. She was in pain. Yesterday was more comfortable than—probably it's time of the day. Some of the day she was like not comfortable” Parent
		“So the one you're looking at there is the one that was done last week…everywhere that you see that is very bright to the video, that is brain that is damaged or dead…that part is the stroke part, but you can see that for the most part the rest was unaffected.” HTM
Asks for medical/patient information	Asking for information about how the patient is doing or what is occurring with the patient, medically or otherwise (before making a decision)	“How do you guys feel about the way that she's breathing?” HTM
		“Now the sleep apnea of the babies, is that the same thing as sleep apnea of adults?” Parent
Prognostication
Prognosticates future outcomes	Explaining long-term prognosis (not related to decision being discussed)	“We don't expect her to get any better or have any more interaction with the environment …” HTM
		“I know that he is very sick. At any time he's not going to be with us anymore.” Parent
Asks for prognosis	Asking about long-term prognosis (not related to decision being discussed)	“How long she can live the way she is now with her brain fading?” Parent
Presenting the decision
Intro to decision	Beginning a conversation about a decision that needs to be discussed or made.	“D3 and I both agree that this bronchiolitis is likely to continue to get worse. And, as that happens, we need to make some decisions.” HTM
Indication for decision	Explaining why a decision must be made or considered	“…she has episodes, periods where she's not able to breathe on her own and the nurses have to bag her.” HTM
Asks doctor for options	Asking the healthcare team for potential courses of action pertaining to a decision	“So, I was just wondering is there something else that can be done?” Parent
		“Is there another way for him to go home not on a vent? Are there other options?” HTM (Social Worker)
Presents options	Explaining potential courses of action/options pertaining to a decision	“Now, it is possible to escalate pain medication high enough that you can get pain control.” HTM
		“So even if I say no DNR, I can still–”Parent
Information-oriented deliberation
Outcomes
Unknown outcome	Indicating that the outcome of a decision is unknown	“It may be that we take the breathing tube out and he breathes on his own…” HTM
Predicting/estimating outcome	Providing an estimate or guess of outcome	“I do think he's going to need a breathing machine for like a year.” HTM
Explains decision-related outcomes	Explaining an outcome of a particular course of action/decision that is known to occur	“And so the likelihood is that any of those things would not significantly prolong his life.” HTM
		“Okay. But it would be—if he gets that trach, you know, the bottom line would be an infection.” Parent
Asks for immediate outcomes	Asking for a prediction of what will occur as an immediate consequence of a certain decision	“What would happen if he were to not be on a ventilator right now?” Parent
		“You think if you do that…she's going to come back, and she's going to do the same thing again?” HTM
Processes/procedures
Details of decision process/procedure	Explaining care changes pertinent to the decision and what specific care processes/procedures the decision entails	“A tracheostomy is where they make a little cut right here, and they put a plastic breathing tube in here, and then we can hook him up to a breathing machine.” HTM
		“If there's a problem, so we could like crank up the lipids on his enteral feeds and back it off in order to see that he gets what he needs.” Parent
Asks for details of decision process/procedure	Asking for an explanation of care changes pertinent to the decision and what specific care processes/procedures the decision entails	“What's the different between CPAP now, the pressure and ventilator?” Parent
		“But he'll be you know more sedated. I think they do it, he'll just be sedated, right?” HTM
Pros/cons
Discusses the cons of the decision	Explaining potential negative consequences of a decision or Asking for an explanation of potential negative consequences of a decision	“But the consequence of that will be that she will be less able to communicate…” HTM
		“His sister is always scared when he goes back on the oxygen, and he got very scared.” Parent
		“I mean it's, that's why I asked if any more fluid's going to come down. Is it possible?” Parent
Discusses the pros of the decision	Explaining potential positive consequences of a decision or Asking for an explanation of potential positive consequences of a decision	“The advantage of that is that we can take him off the sedation, and we can let him wake up.” HTM
		“What would be the positive of that, of doing that?” Parent
Plan-oriented deliberation
Recommendations
Asks for physician recommendation	Asking for a HTM's recommendation on the current decision	“I'm basing this with what you're saying and what would you recommend?” Parent
		“Did you want to comment on what you think a good plan and strategy would be?” HTM (to another HTM)
No recommendation	HTM defers or avoids offering any recommendation about a decision	“Nobody can know what they would do if they were you because until you walk in those shoes you really don't know how it feels.” HTM
Subtle recommendation	A HTM subtly supports or encourages a particular choice	“So if the family wants to, can live with the risk that something could badly go wrong…” HTM
Direct recommendation	A HTM or parent explicitly recommends a particular choice	“And I think we should have our palliative care team involved.” HTM
Considering a decision
Asks for decision	HTM asks parent to make decision or asks for parent's opinion about a decision	“Would you want CPR or the breathing tube placed back in?” HTM
Justifying decision	Explaining the rationale for making a particular choice (include only comments about a decision)	“I don't want her suffering. I know she's suffering and we are suffering.” Parent
		“And the idea of photopheresis is there is no hope that it would make his bronchiolitis obliterans better. The only hope is that it would stall it where it is.” HTM
Determining plan
Expression of decision
Makes a decision	Statements indicating that the parent has made a decision and states what that decision is, or makes a comment revealing their opinion	“If we get to the point that he gets like that, I don't want to revive him.” Parent
Asks for decision clarification/review	Asking to restate, review, or summarize the decision made	“I want to make sure that I heard you properly because sometimes words are hard to interpret.” HTM
		“So do you agree that you're going to do what XX wants?” HTM
Clarifies/summarizes decision	Restating, reviewing, or summarizing decision made	“…you're telling us that the ventilator is not something that you want to do…” HTM
		“For now full code, for now full code. If she was, if she is still the same stage as she was last week, we definitely would think maybe differently.” Parent
Operationalizing the decision
Timing of decision	Comments about the timeframe in which the decision must be made	“He might get to that point where we need a decision within a month, if not sooner than that.” HTM
		“And we don't have to have all of our questions answered today?” Parent
Discusses medical plans	Discussing how medical care will proceed for patient after a decision has been made	“If you make the decision to do the trach and then the ventilator, you will be in the hospital for anywhere from 3–6 months.” HTM
		“We'll see. We'll see what sleep study says and then we go from there.” Parent
Information about support needs	Explaining assistance parents may need in providing care after decision (includes issues pertinent to care after hospital discharge and financial/social barriers to providing care)	“If she goes with ventilator home, I'm afraid to work with ventilator.” Parent
		“I wonder if she's eligible through the link card to qualify for an Advocate from the ADA. She's going to get big.” HTM
Asks about support needed	Any question regarding assistance parents may need in providing care after decision (includes issues pertinent to care after hospital discharge and financial/social barriers to providing care)	“What would be most helpful for you, as far as taking care of him at home?” HTM
		“So what equipment, do we need to go out and do things for a long period of time?” Parent
Family/friends planning	Discussions about communicating with and supporting family/friends after making a decision	“Did you want to have your daughter come in first today and speak with one of the child life specialists?” HTM
		“So that's my sister. She's in medical field in Germany. That's my twin brother and that's my sister-in-law from Germany. So I ask if they can attend” Parent

Stages of decision making are bolded, the themes within each stage are underlined, and the codes assigned to each theme are indented.

ADA, Americans with Disabilities Act; CPAP, continuous positive airway pressure; DNR, do not resuscitate; HTMs, healthcare team members.

Information exchange

The information exchange stage included discussions about the patient's status, comments about prognosis based on the patient's current situation, and a presentation of decisions. HTMs always discussed the reason for the FC and provided information about the patient's current status. Parents always discussed their perceptions of the patient's current condition. HTMs and parents both asked questions about the patients' current medical situation in the majority of FCs. In about half of the FCs, parents asked about the patients' prognoses. HTMs provided prognosis information more often than it was asked for. In all FCs, HTMs used a transition or introduction to conversations about a decision, and this often involved HTMs describing why a decision needed to be addressed. In all conferences HTMs talked about the different options available.

Information-oriented deliberation

The information-oriented deliberation stage described conversation involving information about the patient and decisions that helped parents and HTMs deliberate. It included three themes: potential decision outcomes; details about processes/procedures relevant to the decision; and the pros/cons of different choices. Almost all conferences included discussion from HTMs about known, unknown, and predicted outcomes. Most HTM statements involved details about processes and procedures related to a decision, including explaining pros and cons. In contrast, most parent speech involved asking questions. Although fewer questions came from HTMs, typically this was a HTM asking a question of another HTM.

Plan-oriented deliberation

Themes in the plan-oriented deliberation stage included recommendations and considering a decision. Although asked for by parents explicitly in less than half of the FCs, HTMs provided recommendations in most FCs. The recommendations were typically explicit and directive. However, providers often also made “subtle” recommendations, using indirect language without explicitly endorsing a course of action (e.g., “… if you choose to have her on the machine 24/7 … that would be pretty uncomfortable in the long run”). In a few instances, HTMs used language to avoid or defer making a recommendation. Also included in this plan-oriented deliberation stage were comments justifying a decision. Following such comments were statements weighing the presented options, thus we considered them related to the deliberation process.

Determining a plan

Plan determination included two themes, making and operationalizing the decision. A final decision was made in most, but not all, FCs. Most dialogues during this stage focused on operationalizing decisions and discussing the plan—when and how care would be provided to the patient. Operationalizing decisions also often included a discussion of support needs (social, structural, or financial) following hospital discharge or death.

Conversation flow

We found that conversations rarely flowed in a linear manner through the four decision-making stages. Figure 3 depicts an example portion of text with excerpts labeled according to their coded decision-making stage. This example was typical of how discussions flowed unpredictably between all the stages, with physicians and parents frequently working through different stages multiple times.

FIG. 3.

Example conversation with excerpts labeled by decision-making stage. This example highlights the unpredictable flow of conversation through the different stages of decision making.

Results from quantitative analysis of FCs

Table 4 shows the 11 SDM elements adapted from previous work¹⁶ and the percentage of conferences, in which each activity occurred. The number of elements present in the FCs ranged from 4 to 10 with an average of 6.9.

Table 4.

Shared Decision-Making Activities during Family Conferences

Item	Activity	Present in FC N (%)
1	Assess family understanding	3 (21)
2	Assess need for input from others	4 (28)
3	Explore family's role in decision	5 (35)
4	Elicit family's opinion about decision	6 (42)
5	Discuss patient values and preferences	3 (60)^a
6	Discuss parent values and preferences	11 (78)
7	Explore effect on family's life	11 (78)
8	Explore effect on patient's life	12 (85)
9	Clinical issues presented	14 (100)
10	Treatment alternatives given	14 (100)
11	Discuss uncertainty	14 (100)

Only assessed in five conferences involving patients for whom preferences might be known or obtainable.

Discussion

This study provides insight into the content and flow of DM conversations during PICU FCs. We found that deliberation was integrated into information exchange and plan-making discussions. In these FCs, HTMs infrequently obtained information about parents' perspectives about decisions discussed, missing important SDM elements. We also found DM to be a nonlinear process that flows unpredictably between the decision-making stages. In the nonlinear flow of DM, we describe echoes reflections of other researchers—that SDM is dynamic, provisional, and iterative.^26–28 Our results demonstrate the complexity of these discussions and highlight the need for HTM patience and flexibility with communication during FCs.

Our results also highlight important aspects of the approach to and content of DM discussions. For example, how HTMs made recommendations was notable. While, many providers made direct recommendations, some providers used less explicit, indirect language. In a few instances providers failed to make any recommendation. Admittedly, HTMs may feel more or less comfortable providing recommendations, depending on the type of decision discussed or the degree of uncertainty associated with a decision. Where uncertainty is high, providers might feel less comfortable providing a recommendation. The latest policy statement on SDM from the American College of Critical Care Medicine and American Thoracic Society states that many critically ill patients and their surrogates prefer to make decisions after considering HTM recommendations, even in “value-laden” decision-making scenarios.⁶ This suggests that clinicians should make direct recommendations. However, in one study, White et al. showed that 42% of ICU surrogates preferred not to receive recommendations regarding the continuation of life support.²⁹ Perhaps HTMs should ask patients/surrogates about their preferences regarding clinician recommendations. Acknowledging differences in surrogate preferences about how or if HTMs provide a direct recommendation accords with the SDM approach. We cannot determine if the HTMs' approach to providing recommendations matched parent preferences in the FCs studied. Such questions could be considered in future research.

We found that assessing parental understanding occurred throughout most FCs. In this study, we did not explore how such understanding was obtained, for example, whether HTMs accepted a “yes/no” response to questions about understanding or if they requested that parents describe the issues and situation in their own words. In looking for the presence of SDM elements, we found assessing family understanding present in only 21% of discussions related to the main decision considered. This suggests that while assessing understanding may occur throughout FCs, it may not occur regularly for all decisions discussed. Further work into the process of assessing family understanding is warranted.

Finally, our analysis of SDM elements present or absent in these conferences identifies potential areas for improvement. Less frequently occurring SDM elements included assessing family understanding of decisions, asking families if input from others would help them make decisions, exploring the family's preferred DM role, and eliciting the family's opinion about a decision. Addressing each of those elements may not always have been warranted. For example, perhaps HTMs wanted to introduce a decision, but did not necessarily need to determine a final choice, obviating the need to elicit a family's preference. However, these less frequent SDM elements represent efforts to obtain information from families. That these elements occurred inconsistently in the FCs suggest that HTMs might consider directing more effort toward encouraging families to share their views, perspectives, and needs.

This study has a number of strengths. We analyzed actual FCs, not simulated FCs or comments provided retrospectively about FCs. Conferences were convened to discuss patients with a broad range of medical conditions and decision topics and were not focused on one particular kind of decision, allowing us to learn about discussions related to a range of DM challenges. Parent participants represented people from mixed racial, ethnic, and educational backgrounds. We included Spanish speakers and enrolled a high proportion (45%) of fathers.

We acknowledge some limitations. First, FCs represent one component of a complex decision-making process that also involves family-centered rounds, bedside conversations, input from family and friends, previous experiences, and so on. Second, we analyzed a relatively small number of conferences, limiting generalizability. The small sample also obviates our ability to statistically analyze associations between SDM elements, types of decisions, and parent/HTM outcomes as well as to explore other interesting areas within FC communication, particularly the role of supportive statements and emotion in FCs. The small sample partly reflects our refusal rate of 41%. Third, we analyzed audio data and do not comment on the impact of body language. Fourth, we acknowledge our somewhat low initial interrater agreement rates for the content analysis coding and coding of SDM elements, as low as 67% and 50%, respectively, in some FCs. To ensure the validity of our results, all coding discrepancies were resolved among the group. Finally, recording FCs and having a study team member present during the FCs may have influenced the conversations, although we feel this is unlikely based on similar work.^30–32

Conclusions

We found decision-making discussions during PICU FCs nonlinear with deliberation integrated into information exchange and plan-making conversation. SDM elements meant to obtain parents' perspectives were not always present. We highlight needed areas of study related to concordance of HTM recommendations with family preferences for such information, approaches to assessing parent understanding, use of supportive language during FCs, and the relationship between SDM elements and parent/HTM outcomes.

Footnotes

Acknowledgment

This work was supported by a grant from The National Palliative Care Research Center (NPCRC) entitled, “Evaluating Family Conferences in the Pediatric Intensive Care Unit.”

Author Disclosures Statement

No competing financial interests exist.

References

Michelson

, Patel

, Haber-Barker

, et al.: End-of-life care decisions in the PICU: Roles professionals play. Pediatr Crit Care Med, 2013; 14:e34–e44.

Eddy

: Clinical decision making: From theory to practice. Anatomy of a decision. JAMA, 1990; 263:441–443.

Charles

, Gafni

, Whelan

: Decision-making in the physician-patient encounter: Revisiting the shared treatment decision-making model. Soc Sci Med, 1999; 49:651–661.

Charles

, Whelan

, Gafni

: What do we mean by partnership in making decisions about treatment?. BMJ, 1999; 319:780–782.

Charles

, Gafni

, Whelan

: Shared decision-making in the medical encounter: What does it mean? (or it takes at least two to tango). Soc Sci Med, 1997; 44:681–692.

Kon

, Davidson

, Morrison

, et al.: Shared decision making in ICUs: An American College of Critical Care Medicine and American Thoracic Society Policy Statement. Crit Care Med, 2016; 44:188–201.

Elwyn

, Frosch

, Thomson

, et al.: Shared decision making: A model for clinical practice. J Gen Intern Med, 2012; 27:1361–1367.

Cypress

: Family conference in the intensive care unit: A systematic review. Dimens Crit Care Nurs, 2011; 30:246–255.

Boyle

, Miller

, Forbes-Thompson

: Communication and end-of-life care in the intensive care unit: Patient, family, and clinician outcomes. Crit Care Nurs Q, 2005; 28:302–316.

10.

Billings

: The end-of-life family meeting in intensive care part II: Family-centered decision making. J Palliat Med, 2011; 14:1051–1057.

11.

Curtis

, White

: Practical guidance for evidence-based ICU family conferences. Chest, 2008; 134:835–843.

12.

Davidson

, Aslakson

, Long

, et al.: Guidelines for family-centered care in the neonatal, pediatric, and adult ICU. Crit Care Med, 2017; 45:103–128.

13.

Stapleton

, Engelberg

, Wenrich

, et al.: Clinician statements and family satisfaction with family conferences in the intensive care unit. Crit Care Med, 2006; 34:1679–1685.

14.

McDonagh

, Elliott

, Engelberg

, 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.

15.

Lautrette

, Darmon

, Megarbane

, et al.: A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med, 2007; 356:469–478.

16.

White

, Braddock CH

III

, Bereknyei

, Curtis

: Toward shared decision making at the end of life in intensive care units: Opportunities for improvement. Arch Intern Med, 2007; 167:461–467.

17.

Gay

, Pronovost

, Bassett

, Nelson

: The intensive care unit family meeting: Making it happen. J Crit Care, 2009; 24:629.e621–612.

18.

Cahill

, Lobb

, Sanderson

, Phillips

: What is the evidence for conducting palliative care family meetings? A systematic review. Palliat Med, 2017; 31:197–211.

19.

October

, Watson

, Hinds

: Characteristics of family conferences at the bedside versus the conference room in pediatric critical care. Pediatr Crit Care Med, 2013; 14:e135–e142.

20.

Van Cleave

, Roosen-Runge

, Miller

, et al.: Quality of communication in interpreted versus noninterpreted PICU family meetings. Crit Care Med, 2014; 42:1507–1517.

21.

Michelson

, Clayman

, Haber-Barker

, et al.: The use of family conferences in the pediatric intensive care unit. J Palliat Med, 2013; 16:1595–1601.

22.

Michelson

, Emanuel

, Carter

, et al.: Pediatric intensive care unit family conferences: One mode of communication for discussing end-of-life care decisions. Pediatr Crit Care Med, 2011; 12:e336–e343.

23.

Michelson

, Clayman

, Ryan

, et al.: Communication during pediatric intensive care unit family conferences: A pilot study of content, communication, and parent perceptions. Health Commun, 2016:1–8.

24.

Billings

: The end-of-life family meeting in intensive care part I: Indications, outcomes, and family needs. J Palliat Med, 2011; 14:1042–1050.

25.

Pope

, Ziebland

, Mays

: Qualitative research in health care. Analysing qualitative data. BMJ, 2000; 320:114–116.

26.

Epstein

, Gramling

: What is shared in shared decision making? Complex decisions when the evidence is unclear. Med Care Res Rev, 2013; 70(1 Suppl):94S-112S.

27.

Clayman

, Gulbrandsen

, Morris

: A patient in the clinic; a person in the world. Why shared decision making needs to center on the person rather than the medical encounter. Patient Educ Couns, 2017; 100:600–604.

28.

Clayman

, Makoul

, Harper

, et al.: Development of a shared decision making coding system for analysis of patient-healthcare provider encounters. Patient Educ Couns, 2012; 88:367–372.

29.

White

, Evans

, Bautista

, et al.: Are physicians' recommendations to limit life support beneficial or burdensome? Bringing empirical data to the debate. Am J Respir Crit Care Med, 2009; 180:320–325.

30.

Albrecht

, Blanchard

, Ruckdeschel

, et al.: Strategic physician communication and oncology clinical trials. J Clin Oncol, 1999; 17:3324–3332.

31.

Ickes

: Methods of studying close relationships. In: Weber

, Harvey

, (eds). Perspectives on Close Relationships. Boston: Allyn & Bacon, 1994, pp. 18–44.

32.

Ickes

, Tooke

: The observational method: Studying the interaction of minds and bodies. In: Duck

, Hays

, (eds). Handbook of Personal Relationships: Theory, Research and Interventions. Chichester, England: John Wiley & Sons, 1988, pp. 79–97.