Emergency Psychiatry: Critical Incident Stress Management: I. Interventions and Effectiveness

One of the most often cited studies evaluating the perceived helpfulness of debriefings described the opinions of 172 emergency service, welfare, and hospital personnel in Australia (5). Subjects reported that the debriefings were of considerable to great value to themselves and their peers. The greater the impact the event had on subjects, the more likely they were to value the debriefings. Seventy-seven to 96 percent of the subjects reported that the debriefing had contributed to a reduction in their stress symptoms. The effectiveness of the debriefing seemed to derive primarily from affording subjects an opportunity to talk about the incident and to develop an increased "understanding of one's self."

Criticism of the debriefing process centered primarily on procedural concerns, such as exclusion of some persons who should have been present, the length of the debriefing session, and ineffective leadership.

Similarly, Burns and Harm (6) studied emergency nurses' perceptions of critical incidents and stress debriefings, finding that the majority (88 percent) of those who had participated in a debriefing found it helpful in reducing their stress reactions. Other studies reported similar positive evaluations of debriefings (7,8,9,10).

Despite the generally favorable view of debriefings, studies that have actually compared persons who have been debriefed with persons who have not been debriefed have not found evidence of a protective effect or reduction in symptoms of distress (8,9,11,12,).

Kenardy and colleagues (9) studied 195 emergency service personnel and disaster workers after an earthquake in Newcastle, Australia, 62 of whom were debriefed and 133 of whom were not. Assessments of symptoms of psychological distress and of PTSD were taken 27, 50, 86, and 114 weeks after the event. Results suggested that distress tended to decrease over time for both groups. Subjects in the debriefing group did not report significantly fewer symptoms than those in the nondebriefed group at any point after the event. In fact, they tended to show the opposite pattern: debriefed subjects showed less improvement than nondebriefed subjects.

The Kenardy study did not provide evidence in support of symptom reduction or a protective effect of debriefings, even though debriefed subjects rated the value of the debriefings positively. The major limitation of the study, however, was that the groups were not randomized. Furthermore, the groups differed significantly in demographic characteristics—the debriefed group was more likely to be female and had more education on average—as well as in occupational characteristics—the debriefed group was more likely to have higher occupational prestige, to have participated in nonthreatening situations, and to be employed as counselors and coordinators of services. Although the groups did not report different levels of event-related stress, it is possible that the debriefings were not conducted randomly and that some selection criteria that may have confounded the interpretation of the results were used.

MacFarlane (13), studying firefighters responding to a bushfire disaster, found that debriefings were associated with reduced acute posttraumatic stress but increased delayed posttraumatic stress. Hytten and Hasle (8) studied firefighters who responded to a hotel fire and found no difference in distress and psychological symptoms between those who had been debriefed and those who had not. A study conducted on troops responsible for burying the dead in the Gulf War found that although participants valued the debriefings, the debriefings appeared to have no significant effect on their psychological adjustment nine months after their tours (11).

Leonard and Alison (12) studied how debriefings affected coping and anger among a sample of Australian police officers involved in shootings. They found that the group receiving debriefings showed significant reductions in anger levels and greater use of some specific adaptive coping strategies. However, whether an officer was debriefed was correlated with several other factors that could have accounted for these favorable differences. For example, the debriefed group contained a higher proportion of officers involved in incidents in which someone was shot or killed. Hence it is not known whether the effect was due to the debriefing or resulted from the nonrandom manner in which debriefings were conducted.

Other researchers also have found that debriefing is not a random event but is associated with the severity of the event (10), suggesting that research that does not use random assignment may be critically flawed.