• 
• 

Three Part Question

In [pediatric patients under-going trauma resuscitation] does [the presence of family members] have an inmpact on the [efficiency of pediatric trauma resuscitations]?

Clinical Scenario

An eight-year-old female patient arrives at the Emergency Department with her mother after being struck by a car while riding her bicycle. Her respirations are strained and she has an obvious deformity of the right shoulder. She is evaluated immediately by the trauma service. Possible interventions include the placement of a chest tube. Should her mother be removed from the room?

Search Strategy

Medline 1950-05/11 using OVID interface, Cochraine Library (2011), PubMed clinical queries
[(exp family) AND (exp resuscitation) AND (exp wounds and injuries OR trauma.mp). Limit to English language and all child (0 to 18 years).

Search Outcome

35 papers were identified, two were relevant to the clinical question

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Dudley et al, 2009, USA	Children evaluated at an American College of Surgeons Level 1 pediatric trauma center after having been categorized as a trauma code activations.	This was a prospective study spanning 28 months. It was non-blinded. Initial enrollment was randomized to odd-versus-even days. The intervention was yes-family-presence. The control was no-family-presence. The primary outcome measure was time from arrival in trauma room to departure for CT scan. The secondary outcome measure included time to completion of all laboratory tests, emergency procedures, portable radiographs and the secondary survey.	Time from arrival in the trauma room to departure for CT scan between the yes-family-presence and no-family-presence groups.	There was no significant difference.	This study was not fully randomized since actual family presence, or lack thereof, was not strictly limited by odd-even days. Therefore enrollment bias is a possible confounder. The yes-family-presence group (average ISS 9, 14% as highest trauma priority) was less severely injured than the no-family-presence group (average ISS 11, 27% as highest trauma priority). This possible confounder is related to both group assignment and outcome. For example, unstable patients may have been more likely to arrive via flight EMS (37% of yes-family-presence versus 71% of no-family-presence) without family members or may have been involved in initial events with a more significant mechanism of injury, possibly with family members as co-victims and therefore not available to be present. Additionally, this patient population was skewed toward blunt trauma, a mechanism of injury for which significant intervention during resuscitation is less likely than penetrating trauma. Finally, because this study was non-blinded and because some of the data were obtained by surveys of medical providers, it is possible that self-awareness influenced the results.
			Time from arrival in the trauma room to completion of resuscitation between the yes-family-presence and no-family-presence groups.	There was no significant difference.
			Survey response by trauma team to yes-family presence.	Trauma team members strongly agreed that yes-family-presence increased their stress level. Trauma team members tended to agree that yes-family-presence was helpful for both the patient and the family. A significant minority agreed that professionalism was increased by yes-family-presence. Very few members of the trauma team perceived prolonged resuscitation secondary to yes-family-presence. Nearly all trauma team members reported being aware of family presence.
			Survey response by family to yes-family-presence.	Family members very strongly agreed that yes-family-presence was helpful to the patient, to the family members, and that they would choose to be present again. Family members very strongly disagreed that yes-family-presence was upsetting.
O'Connell et al, 2007, USA	Children evaluated at an American College of Surgeons Level 1 pediatric trauma center after having been categorized as trauma code activation.	This was a non-random, non-blinded cross-sectional study spanning 19 months. It combined prospectively obtained data with retrospective chart review. The first group was yes-family-presence during trauma resuscitation. The second group was no-family-presence. Group assignment was not an active intervention by the study authors; it was simply a reflection of whether family members presented with the patient and whether they remained present during resuscitation. A third group was evaluated as well. This so-called "missed group" consisted of patients who presented during over-night and early-morning hours when prosepctive data was not being collected. In effect, this group was used as the control. Primary outcomes included elapsed time for completion of various resuscitation components including log-roll, first radiograph, IV access, central line, chest tube and intubation.	Elapsed time for completion of log-roll.	No significant difference between yes-family-presence and no-family-presence groups.	This was an observational study with no active randomizable intervention. As such, it was vulnerable to confounders between the two groups. No significant demographic differences were noted between the two groups with respect to gender, race, trauma score, GCS, admission rate, operation rate and death. The no-family-presence group was older (average 9.7 years versus 7.8 years) but this seems unlikely to have affected the results. The "missed group" consisted of patients who presented at times when prospective data was not being collected. This third group functioned as a mutual control to the prior two groups. The only significant difference between the missed group and the other two groups was GCS (average 12.7 versus 13.5 for yes-family-presence and 12.9 for no-family-presence). Again, this seems unlikely to have affected the results. Of course, this third group was also not randomized and so confounders may still exist. For example, the geographical site of injury in the no-family-presence group and the missed group, such as school, may have affected both the type and severity of injury. Additionally, severe pediatric trauma that places high demands on the trauma team is a rare event. It is possible that not enough cases of this maximum trauma were included to detect any differences in this sub-group. Finally, because this study was non-blinded and because some of the data were obtained by surveys of medical providers, it is possible that self-awareness influenced the results.
			Elapsed time for completion of first radiograph.	No significant difference between yes-family-presence and no-family-presence groups.
			Elapsed time for establishment of intravenous access.	No significant difference between yes-family-presence and no-family-presence groups.
			Elapsed time to completion of central venous access.	No significant difference between yes-family-presence and no-family-presence groups.
			Elapsed time to completion of intubation.	No significant difference between yes-family-presence and no-family-presence groups.
			Elapsed time to insertion of chest tube.	No significant difference between yes-family-presence and no-family-presence groups.
			Occurrence of interference with medical care by family members.	There were no cases of interference with medical care by family members.
			Pre-mature departure of family membes during trauma resuscitation.	A total of seven cases of 197 yes-family-presence were converted to no-family-presence for two instances of medical provider comfort during intubation, two instances of family member overwhelmed but appropriate, two instance of family member overwhelmed and inappropriate, and one instance of suspected child abuse.
			Trauma team response to yes-family-presence during resuscitation.	A large majority reported that yes-family-presence made no change in medical-decision-making, institution of care or team communication. A large majority reported that yes-family-presence either improved or did not change communication between trauma team and family.

Comment(s)

Family presence should be encouraged during trauma resuscitation in the absence of a compelling reason against inclusion. Compelling reasons include concerns for patient privacy during simultaneous trauma resuscitations, evaluation for suspected child abuse, demonstrations of inappropriate behavior by family members, and trauma team discomfort during highly technical procedures. A structured program to prepare trauma team members for family presence and to guide family members through the experience is beneficial.

Editor Comment

GCS, Glasgow Coma Scale; ISS, Injury Severity Score.

Clinical Bottom Line

Family presence does not adversely effect efficiency during trauma resuscitation and should be encouraged.

References

1. Dudley NC. Hansen KW. Furnival RA. et al. The effect of family presence on the efficiency of pediatric trauma resuscitations. Annals of Emergency Medicine June 2009; 53(6):777-84.e3
2. O'Connell KJ. Farah MM. Spandorfer P et al. Family presence during pediatric trauma team activation: an assessment of a structured program. Pediatrics September 2007; 120(3):e565-74