OBJECTIVES: Our objectives were to determine the impact of family presence during PICU rounds on family satisfaction, resident teaching, and length of rounds and to assess factors associated with family satisfaction.
METHODS: This was an observational study of a convenience sample of morning work rounds in a PICU, followed by surveys of family members of patients in the unit and residents who had been present for rounds.
RESULTS: A total of 411 patient encounters were observed, 98 family questionnaires were fully completed, and 33 resident questionnaires were completed. Ninety-eight percent of family members liked to be present for rounds. On the first day of admission, family members were less likely to understand the plan (P = .03), to feel comfortable asking questions (P = .007), or to want bad news during rounds (P = .009). They were more likely to have privacy concerns (P = .02) and to want 1 individual to convey the plan after rounds (P = .01). Higher education level was associated with decreased privacy concerns (P = .002) but did not affect understanding of the plan. Fifty-two percent of residents perceived that teaching was decreased with families present. Time spent with individual patients was not increased by family member presence (P = .12).
CONCLUSIONS: Family satisfaction is high, but families of patients on the first day of admission may need special attention. The medical team should conduct rounds in a manner that addresses the privacy concerns of families. Residents often think that teaching is decreased when families are present.
Family-centered care has become the standard care model in pediatrics.1 The American Academy of Pediatrics is committed to family-centered care and, in a 2003 policy statement, recommended that “conducting attending physician rounds (ie, patient presentations and rounds discussions) in the patients' rooms with the family present should be standard practice.”2 In 2007 clinical practice guidelines on the support of the family in the ICU, the American College of Critical Care Medicine Task Force stated that “family participation in rounds is beneficial” but “the topic of family presence on rounds is the least studied of any section [of the guidelines].”3
Previous studies with adults reported patients having positive perceptions of rounds and the medical team when they were included in rounds,4,5 but evidence in pediatrics is limited. A randomized study on a pediatric oncology ward demonstrated that parents preferred bedside rounds and rated them higher with respect to information received and the ability to ask questions.6 A small, randomized, controlled trial in a PICU also found that parents preferred bedside rounds and felt more well-informed when they were present for rounds.7 A recent observational study found no significant differences in time spent on rounds and time spent teaching when families were present.8 Pediatric residents are more comfortable asking and receiving questions in conference room rounds7 but do think that bedside rounds are better for patient care.6,7
With these studies as our background, our objectives were to determine family satisfaction, understanding of the daily care plan, and confidence in the medical team when present for rounds and to assess resident perceived amount of teaching. We also evaluated whether the preferences of family members were affected by day of admission, previous PICU admission, or education level and whether family presence was associated with longer rounds.
This was a prospective observational study of a convenience sample of weekday morning work rounds in the PICU. Data were collected by an observer not involved in direct medical care, and available family members and residents were approached after rounds to complete a questionnaire. The study was conducted at a tertiary care, freestanding, academic children's hospital, in a 45-bed PICU that has ∼3000 medical and surgical admissions per year (cardiac patients are cared for in a separate cardiac ICU). The PICU is divided into 2 separate wards, one with 24 single rooms and the other with 9 single rooms and 6 double rooms. Each ward is covered by 1 pediatric critical care team, consisting of 1 attending physician, 1 pediatric critical care medicine, emergency medicine, or anesthesia fellow, 3 or 4 pediatric residents, and 1 or 2 emergency medicine or anesthesia residents. The team conducts daily morning work rounds at the bedside for all patients in the unit. The team typically is joined on rounds by a pharmacist, a nutritionist, a case manager, and the nurse caring for each patient that day. Representatives of subspecialty services are variably present during rounds for patients for whom they are consulting or for whom they are long-term care providers. Rounds have been conducted at the bedside, with families, in this unit since 1998. Families are invited routinely to join rounds, but there is no standardized preparation or debriefing of the families about the rounds process. Rounds observation occurred 2 weekdays per week (1 day observing each team) between August 2007 and January 2008.
The observer completed a standardized observation checklist that included total length of rounds, length of rounds for individual patients, length of stay for each patient, family member presence, whether introductions were performed, attending physician examination of the patient, number of subspecialty services present, whether new information was learned from the family, and time spent teaching; 24-hour Pediatric Risk of Mortality (PRISM) scores also were recorded. The family questionnaire was completed by 1 family member per patient and consisted of objective data regarding length of admission, number of previous admissions to the PICU, and family member demographic information, as well as 17 statements that elicited family member satisfaction with being present for rounds, confidence in the medial team, understanding of the discussion on rounds, preferences for receiving bad news and for having 1 individual return after rounds to convey the plan, and privacy concerns. Non–English-speaking families and families in the process of having life-sustaining measures discontinued were not approached to complete a questionnaire. Families were approached to complete a survey only once during a hospitalization.
The resident questionnaire included resident specialty and year of training, as well as 6 statements assessing resident preference for family presence and perceived amount of teaching. Family member and resident questionnaire responses were on a 5-point Likert scale from strongly disagree to strongly agree.
The institutional review board of the Children's Hospital of Philadelphia approved the study, and verbal consent for participation was obtained from family members and residents completing the questionnaires. Explicit consent was not required from family members or staff members observed during the usual conducting of rounds.
Proportions were compared by using Fisher's exact test, and continuous data (eg, time spent on rounds) were compared by using the Wilcoxon rank sum test, because of nonnormal distribution. Linear regression analyses were performed for time spent in rounds for individual patients, with adjustment for attending physician physical examination, subspecialty service presence, and 24-hour PRISM score, and for total length of rounds, with adjustment for total number of patients and average 24-hour PRISM score. Logistic regression was used to compare families that were present on rounds with those that were not. We analyzed Likert scale responses by using ordered logistic regression, to find the odds ratio (OR) of a particular group progressing 1 level up on our 5-point scale, from strongly disagree toward strongly agree.
Our a priori sample size calculation estimated that 100 family member interviews would give us 90% power to detect a 2-point rating difference on a 5-point scale, with a 2-tailed α of .05. We estimated that an 8-week period of rounds observation would yield >300 patient encounters, giving us 97% power to detect a 3-minute difference in rounding time per patient with families present versus not present, with a 2-tailed α of .05.
A total of 411 patient encounters during morning work rounds were observed on 25 different days. Family members were present during 52% of the observed patient encounters (213 of 411 encounters). Introductions of the members of the medical team were performed 11% of the time (46 of 410 encounters; not recorded for 1 encounter). The critical care attending physician examined the patient on rounds 44% of the time (170 of 386 encounters; not recorded for 25 encounters), and ≥1 additional subspecialty service was present on rounds 5.4% of the time (22 of 411 encounters). New information was discovered from the family 46% of the time when a family member was present on rounds.
Families in a double room were less likely to be present on rounds (OR: 0.6 [95% confidence interval [CI]: 0.4–0.99]; P = .045). Higher PRISM scores, indicating greater severity of illness, also were associated with lower odds of a family member being present (OR: 0.97 [95% CI: 0.94–0.99]; P = .048). Families that were absent from rounds did not differ significantly from families that were present with respect to patient age or length of admission.
In unadjusted analyses, the time spent in rounds for individual patients was slightly greater when family members were present (10.6 vs 9.5 minutes; P = .008). After adjustment for attending physician physical examination, subspecialty service presence on rounds, and PRISM score, family presence added 1 minute to rounding time, on average, but this difference was no longer significant (95% CI: −0.3 to 2.2 minutes; P = .12). Attending physician physical examination increased the length of rounds for individual patients by 2.4 minutes (95% CI: 1.2–3.7 minutes; P < .001), and each subspecialty service present increased rounds by 2.8 minutes (95% CI: 1.1–4.5 minutes; P = .002). The total length of rounds was not affected by the number of families present after adjustment for the number of patients (P = .89).
Ninety-eight family questionnaires were fully completed, 2 were partially completed, and 4 families declined to participate, which yielded a 94% response rate. The demographic features of the participating family members and their children (patients) are presented in Table 1.
Family satisfaction with being present for rounds was high (Table 2); 98% of family members liked to be present for rounds, and 97% thought it was helpful to hear the entire presentation and discussion of their child's case. Ninety-two percent had confidence in the residents caring for their child, and 91% reported that being present gave them more confidence in the medical team. It is important to note that 44% of family members still preferred 1 individual to return after rounds to convey the plan.
Table 3 presents the preferences of family members on the first day of admission, compared with >1 day, after adjustment for 24-hour PRISM scores. Family members on the first day of admission were less likely to understand the plan, to feel comfortable asking questions, or to want bad news during rounds. They also were more likely to find the discussion on rounds more confusing than helpful and to want 1 individual to return after rounds to convey the daily plan. Family members on the first day of admission had greater odds of having privacy concerns, even after adjustment for being in a double room (OR: 2.6 [95% CI: 1.2–5.9]; P = .02).
Higher education level for the family member did not affect reported understanding of the daily plan (OR: 0.8 [95% CI: 0.5–1.5]; P = .55), comfort asking questions (OR: 0.8 [95% CI: 0.4–1.5]; P = .46), or preference for the plan to be conveyed by 1 individual after rounds (OR: 0.9 [95% CI: 0.5–1.3]; P = .5). Higher education level was associated with decreased privacy concerns (OR: 0.4 [95% CI: 0.3–0.7]; P = .002), which remained significant after adjustment for race and the patient being in a double room (OR: 0.5 [95% CI: 0.3–0.8]; P = .005). There was a nonsignificant trend toward families with a higher education level having less confidence in the residents (OR: 0.6 [95% CI: 0.4–1.0]; P = .06).
Family members with a previous admission to the PICU were less likely to report understanding the daily plan on rounds (OR: 0.3 [95% CI: 0.1–0.8]; P = .02) but did not find rounds less frightening than did those experiencing a first admission to the PICU (OR: 0.4 [95% CI: 0.2–1.1]; P = .08). Previous admission to the PICU was associated with less confidence in the residents (OR: 0.4 [95% CI: 0.1–0.9]; P = .03).
Thirty-three resident questionnaires were completed, 26 by pediatric residents and 7 by nonpediatric residents (6 emergency medicine residents and 1 anesthesia resident). No residents who were approached refused to participate. Eighty-five percent preferred family members to be present for rounds, agreeing that communication and patient care were improved when families were present; 82% perceived that rounds took longer when families were present. Fifty-two percent reported that teaching was decreased when families were present, although the majority did not feel self-conscious.
The American Academy of Pediatrics states that family-centered care is “grounded in collaboration among patients, families, physicians, nurses, and other professionals” and conducting rounds with families present should “facilitate the exchange of information between the family and other members of the child's health care team and encourage the involvement of the family in the decisions that are commonly made during rounds.”2 At our institution, a multidisciplinary team approach to PICU rounds, including the presence of family members, was established to promote this collaboration.
A growing body of literature, including our current study, has demonstrated consistently that family members' satisfaction is high when they are included in morning work rounds.7–9 This study is the first to report that, on the first day of admission, family members were less likely to understand the daily plan and felt less comfortable asking questions. A study of adult patients reported that 46% thought that too much medical terminology had been used during bedside rounds.5 In a PICU with complex medical care, the team should tailor the discussion to minimize confusion and should pay special attention to ensuring that family members early in an admission feel comfortable asking questions. A concern in the PICU is the discussion of unfavorable results or prognoses during rounds with families present. The medical team should consider withholding bad news until 1 individual can return to have a discussion with the family. This also would provide an opportunity to clarify the daily plan and to elicit questions. Marking family member preferences for participation in rounds on a card posted on the patient's door, as described by Muething et al,10 may be a useful method of eliciting family member preferences before rounds. Delivering bad news, however, is a skill that needs to be taught.11,12 For educational purposes, trainees should be afforded the opportunity to witness senior physicians delivering bad news. If this does not occur during rounds, then other methods (such as simulation) could be used.13–15
Privacy concerns are important when rounds are conducted at the bedside in a double room or in the hallway outside the patient room, as is often the practice in our PICU. Landry et al7 reported parents thought that confidentiality and intimacy were more respected during bedside rounds than during conference room rounds. Our results indicate that special attention to confidentiality may be important for some families, particularly those with a lower educational level or on the first day of admission.
Our finding that previous PICU admission decreased the odds of understanding the plan may be a result of patients with previous admissions representing more-complex cases, necessitating more-complicated discussions. The medical team should be mindful during these discussions that previous experience in a PICU does not necessarily equate to better understanding of the medical terms used during rounds. Although higher education level was associated previously with lower preference for bedside rounds6 and finding medical terminology less confusing,5 we did not observe these differences.
Physicians often worry that family presence lengthens rounds.16 Our study, in agreement with others,8,16 suggests that this concern is unfounded. Interestingly, attending physician examinations increased the length of rounds for individual patients in our study. Children in a pediatric oncology ward felt that the physical examination on rounds, in front of a group, was more upsetting than the discussion on rounds.6 Attending physicians performed a physical examination during nearly one half of the patient encounters in our study, which indicates that this is a frequent practice on rounds. Given the added time and the potential emotional impact on patients, it may be optimal for attending physicians to examine some children either before or after rounds, unless there are particular physical findings to be demonstrated.
Our study has several limitations. The first is that our study did not compare bedside rounds with conference room rounds or rounds outside the room without family involvement. Because bedside rounds with family participation are a long-established tradition at our institution, we thought it would be too challenging to prepare families for even a temporary change. We also did not observe rounds on weekends, when the team composition is smaller. Rounds are otherwise conducted in the same manner, including family presence, and it is likely that our results would have been similar. In addition, because rounds in our PICU have been conducted with families present for 10 years, the staff is experienced with this format; therefore, our findings may not be generalizable to rounds in other ICUs or in locations other than the PICU, where families are not routinely present. Studies on non–ICU wards have shown benefits of bedside rounds,6,17,18 but further study on general pediatric wards would be useful.
We had difficulty with our aim of quantifying teaching time on rounds with families present versus not present, because teaching was so often interspersed in patient care discussions that it was difficult to time the teaching separately. Residents' perception of decreased teaching with families present may reflect this difficulty. The attending physician stating, “I am going to teach now,” is one way of identifying teaching for the residents, while also clarifying for families when the discussion is focused on education rather than patient care. Phipps et al8 found no significant difference in the time spent teaching in the presence or absence of family members, but further evaluation of the quality and quantity of teaching on rounds is needed. Surveys of residents may not represent the opinions of the entire multidisciplinary team, although previous studies showed other health care providers to have positive perceptions of bedside rounds.8,18
Although we found that new information frequently is learned from families on rounds, we did not assess the quality of this information or whether it changed patient management, and objective outcomes such as these are important areas for future study. In addition, to assess fully the impact of family presence on patient care, future studies should evaluate mortality rates and other clinical outcomes with family presence on rounds. Studies that evaluated the presence of parents during cardiopulmonary resuscitation and painful procedures showed no negative impact on physician performance,19,20 although physicians, especially residents, have expressed concern regarding this practice.20 Physicians have also expressed apprehension regarding family presence on PICU rounds,21 and randomized, controlled trials are needed to evaluate the objective impact on the medical team of families being present on rounds. We did not ask family members who were absent from rounds why they were not present; further studies could ask family members to delineate logistic or psychosocial barriers to family presence on rounds.
Family satisfaction is high when families are present for rounds, but this study suggests that families of patients on the first day of admission may need special attention, to optimize their understanding of the plan and their comfort in asking questions. The medical team should conduct rounds in a manner that addresses the privacy concerns of families. Although the number of families present does not increase the length of rounds, attending physicians should be aware that the residents often think that teaching is decreased when families are present.
- Accepted May 28, 2009.
- Address correspondence to Paul L. Aronson, MD, Children's Hospital of Philadelphia, Division of Emergency Medicine, Department of Pediatrics, 34th Street and Civic Center Boulevard, Philadelphia, PA 19104. E-mail:
Financial Disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on this Subject:
Family members prefer bedside rounds and feel more informed when they are present for rounds.
What This Study Adds:
This study is the first to show that families on the first day of admission have different preferences for family-centered rounds than do families later in admission. Family presence does not affect the length of rounds.
- ↵American Academy of Pediatrics, Committee on Hospital Care. Family-centered care and the pediatrician's role. Pediatrics.2003;112 (3):691– 697
- ↵Landry MA, Lafrenaye S, Roy MC, Cyr C. A randomized, controlled trial of bedside versus conference-room case presentation in a pediatric intensive care unit. Pediatrics.2007;120 (2):275– 280
- ↵Kleiber C, Davenport T, Freyenberger B. Open bedside rounds for families with children in pediatric intensive care units. Am J Crit Care.2006;15 (5):492– 496
- ↵Muething SE, Kotagal UR, Schoettker PJ, Gonzalez del Rey J, DeWitt TG. Family-centered bedside rounds: a new approach to patient care and teaching. Pediatrics.2007;119 (4):829– 832
- ↵Kolarik RC, Walker G, Arnold RM. Pediatric resident education in palliative care: a needs assessment. Pediatrics.2006;117 (6):1949– 1954
- Greenberg LW, Ochsenschlager D, O'Donnell R, Mastruserio J, Cohen GJ. Communicating bad news: a pediatric department's evaluation of a simulated intervention. Pediatrics.1999;103 (6):1210– 1217
- ↵Kuck MA, Morris MC. The effect of parental presence on pediatric intensive care unit (PICU) rounds. Crit Care Med.2007;35 (12 Suppl):A122
- ↵Rosen P, Stenger E, Bochkoris M, Hannon MJ, Kwoh CK. Family-centered multidisciplinary rounds enhance the team approach in pediatrics. Pediatrics.2009;123 (4). Available at: www.pediatrics.org/cgi/content/full/123/4/e603
- ↵Bauchner H, Vinci R, Bak S, Pearson C, Corwin MJ. Parents and procedures: a randomized controlled trial. Pediatrics.1996;98 (5):861– 867
- Copyright © 2009 by the American Academy of Pediatrics