The harmful effects of Critical Incident Stress Debriefing (CISD): Why police departments should stay up to date on evidence-based practices

I attended the 2018 International Association of Chiefs of Policing Conference in Orlando, FL where I had the good fortune to sit in on a Critical Incident Peer Support session led by Dr. Bradley S. Feuer, the Chief Surgeon of the Florida Highway Patrol. I was looking forward to this class because I have read several books that mentioned there was empirical evidence to show that Critical Incident Stress Management (CISM) or Critical Incident Stress Debriefing (CISD) has no effect on PTSD symptoms or may even increase PTSD symptoms, something social scientists call a backfire effect. These books piqued my curiosity about what the empirical research showed, as books will cover some of the research that supports their point of view, but not necessarily all the research in the field. I was unable to access some of the most pertinent articles, leading me to email the researchers directly and asking for the articles. I explained that CISD is often a requirement for police officers involved in a critical incident with some agencies going so far as having a general order mandating, “The involved employee’s commander/manager will ensure the Department psychologist is notified within 24 hours after the critical incident. (a) The Department psychologist will conduct a debriefing session within 72 hours of the incident. The employee’s immediate family may be included in the debriefing session.” The psychologist I spoke with, Dr. Scott Lilienfeld of Emory University, sent me several studies and a graph visually depicting that CISD has either no effect or increases PTSD symptoms. He was disappointed to hear that policing was still using CISD as an intervention when his opinion was that this method was debunked in psychology over ten years ago. By not staying up to date on current research, police agencies are potentially causing harm to their employees by endorsing an intervention that has been shown to either be ineffective or harmful to participants and potentially exposing the city to legal liability by using CISM/CISD interventions.

I attended Dr. Feuer’s session because I wanted to know if he was reading what I was reading, that there is overwhelming research that supports the discontinuation of CISM/CISD as a therapeutic intervention for individuals involved in a critical incident. That was exactly what his session was about; showing how the development of CISM/CISD was created and almost entirely supported by research from the founder of the program and/or affiliates of the International Critical Incident Stress Foundation, Inc. (ICISF) and that studies done by unbiased researchers demonstrated no efficacy of the intervention. The best article I read that reviews CISM/CISD was written by McNally, Bryant, and Ehlers (2003) Does early psychological intervention promote recovery from posttraumatic stress, where they not only review the extant literature but analyze Jeffrey Mitchell’s (the co-creator of CISM/CISD with Dr. George Everly) arguments refuting the research that does not support CISM/CISD. McNally et al. (2003) demonstrate that Mitchell consistently contradicts his earlier arguments using the same factors when arguing unsupportive research is ‘invalid,’ as when he argues the research is ‘valid’ when it supports CISD. Even the CISM primer on the ICISF website refutes the research that shows CISD is ineffective and/or harmful where seven of the nine articles cited are written by Mitchell or Everly.

What is Critical Incident Stress Management/Critical Incident Stress Debriefing?

For those who have not heard of CISM/CISD, it is an intervention where a trained therapist (usually trained through ICISF) conducts a debriefing session either individually or with a group of individuals who have experienced a critical incident. Mitchell now advocates for CISM/CISD to be performed in groups. The debrief has seven components as described by McNally et al. (2003): 1. explanation of the debriefing process, 2. fact gathering (what happened during the critical incident), 3. thought phase (participants describe cognitive reaction to the event) 4. reaction phase (expressing their feelings about the event), 5. symptom phase (asking what psychological or physical reactions they are experiencing), 6. teaching phase (demonstrating that stress reactions are normal), and 7. reentry phase (facilitators’ attempts to attain closure). According to the ICISF website “CISM is a comprehensive, integrative, multicomponent crisis intervention system” also consisting of seven steps: 1. pre-crisis preparation, this includes stress management education, stress resistance, and crisis mitigation training for both individuals and organizations. 2. disaster or large-scale incident, as well as, school and community support programs including demobilizations, informational briefings, “town meetings” and staff advisement, 3. defusing, this is a 3-phase, structured small group discussion provided within hours of a crisis for purposes of assessment, triaging, and acute symptom mitigation, 4. Critical Incident Stress Debriefing (CISD), 5. one-on-one crisis intervention/counseling or psychological support throughout the full range of the crisis spectrum, 6. family crisis intervention, as well as, organizational consultation, and 7. follow-up and referral mechanisms for assessment and treatment, if necessary (International Critical Incident Stress Foundation, Inc. Primer, 2018). CISM is a comprehensive program starting before the crisis begins, including community and family, where CISD is a onetime debriefing for the individuals directly involved in the critical incident.

What does the research show about CISM/CISD?

Barlow (2003) and Lilienfeld (2007) have written about psychological interventions that have backfire effects and both specifically mention CISD as a psychological intervention that has increased the participants PTSD symptoms. Hobbs, Mayou, Harrison, and Warlock (1996) conducted a randomized controlled trial on victims of road traffic accidents. The victims were randomly assigned to either a single debriefing session or an assessment-only control condition. The debriefing occurred between 24 and 48 hours after the incident. At a four-month assessment, neither group experienced a reduction in symptoms of PTSD, anxiety, or depression. However, the debriefing group experienced worse scores on two areas in the Brief Symptom Inventory (BSI). The BSI is a questionnaire concerning symptoms of emotional distress. The research team then did a three-year follow-up with both groups and found that relative to the control group, the debriefing group had significantly more PTSD symptoms (Mayou, Ehlers, & Hobbs, 2000). Further, they found the individuals who had initially scored high for PTSD symptoms and were debriefed remained highly symptomatic compared to individuals who also had high PTSD scores but were only assessed. The hypothesis was that debriefing interferes with the natural recovery of PTSD. This is just one example of one study.

What if research articles find different outcomes?

I was once asked by an officer, “What do you do when one research article demonstrates a positive outcome and another study examining the same intervention demonstrates a negative outcome?” One way of deciding is to look at where the studies land on the Maryland Scientific Methods Scale (Farrington, Gottfredson, Sherman, & Welsh, 2002). I do not want to get off on a tangent explaining how to evaluate the quality of a research article, so the best way of determining the overall impact of an intervention is to find a systematic review or meta-analysis which takes all of the high-quality research in the field and examines the quality of the research, the effect size (how large of a finding discovered from the study) and statistically examines the outcomes to determine the overall effects. Rather than trying to make a decision from reading 10-20 studies, a meta-analysis or systematic review gives an objective review of those research articles to date. A meta-analysis article will explain what studies the authors have included and excluded, how they completed their search of databases, what databases they searched, and how they performed their statistical analysis. The process by which the authors come to their conclusions will also be clearly laid out in the methods section.

Examining the totality of CISD research: What does the meta-analysis show?

Two meta-analyses have been done on CISD and psychological debriefing. This first meta-analysis examined seven studies, five studies specifically evaluated CISM and three non-CISD studies interventions, and six no-intervention controls (van Emmerik, Kamphuis, Hulsbosch, & Emmelkamp, 2002). The studies chosen for the analysis were those that evaluated single session debriefings within one month after the trauma where symptoms were assessed with accepted psychological measures, and the psychological assessments were performed before (pre-test) and after (post-test) the intervention. The overall finding was that CISD and non-CISD interventions do not improve natural recovery from psychological trauma (van Emmerik et al., 2002). The second meta-analysis examined eleven studies at either a level 4 or 5 of the Maryland Scientific Methods Scale, meaning either a randomized control trial (the gold standard of research) or a quasi-randomized trial of individuals who received a single debriefing session within one month of the traumatic incident (Rose, Bisson, & Wessley, 2002). This review was a Cochrane Review. Cochrane is “internationally recognized as the benchmark for high-quality information about the effectiveness of healthcare” (Cochrane About Us, 2018). Cochrane does not accept commercial or conflicted funding, their level of research is the highest in the world, and the researchers conducting a review for Cochrane must perform the analysis following a strict protocol. In sum, Cochrane reviews are the best empirical evidence on any healthcare intervention in the world. The Cochrane review Rose et al. (2002) stated:

“This review concerns the efficacy of single session psychological “debriefing” in reducing psychological distress and preventing the development of post-traumatic stress disorder (PTSD) after traumatic events. Psychological debriefing is either equivalent to, or worse than, control or educational interventions in preventing or reducing the severity of PTSD, depression, anxiety and general psychological morbidity. There is some suggestion that it may increase the risk of PTSD and depression. The routine use of single session debriefing given to non-selected trauma victims is not supported. No evidence has been found that this procedure is effective.

In this review, Rose et al. (2002) discussed the fact that debriefing may ‘medicalise’ normal PTSD symptoms, meaning that avoidance and disassociation with little affect are a natural occurrence after a critical incident and may be a healthy protective reaction to a critical incident. CISD/CISM forces an individual to talk about or face the incident before they are psychologically ready. This is apparent in individuals where the incident resulted in elevated PTSD symptoms initially and when involved in debriefing sessions, those individuals were worse off four months later than individuals who were allowed to process the event naturally (Mayou et al., 2000).

CISM/CISD is ineffective and can be harmful

There is little evidence, other than supplied by the creators of CISM/CISD and ICISF, that CISM/CISD is an effective intervention for individuals involved in critical incidents to prevent symptoms of PTSD. Studies have been conducted on police officers (Carlier, Voerman, & Gersons, 2000; Carlier, Lamberts, van Ulchelen & Gersons, 1998) and firefighters (Harris, Balolu, & Stacks, 2002; Harris & Stacks, 1998) where both the firefighters and police officers were worse off than the control group. The evidence has piled up and demonstrated that at best CISM/CISD is ineffective and at worst, it leaves our employees worse off than if we would have just left them alone. This research has led several top agencies in the world to recommend the discontinued use of CISM/CISD as an intervention for individuals involved in critical incidents.

            During his talk, Dr. Feuer listed out all the international entities who have issued statements discontinued using CISM/CISD as an intervention. This list was laid out in Bledsoe (2003): The World Health Organization (WHO), the National Institute of Mental Health (NIMH), the British Health Service, the North Atlantic Treaty Organization (NATO), the Australasian Critical Incident Stress Association, and the British Navy. Yet, policing across the United States still engages in this intervention and hails it as a ‘best practice.’ ‘Best practice’ in policing usually means that there are enough agencies engaging in the practice and in agreement that the practice is useful. ‘Best practice’ is typically founded upon nothing other than collective opinion, rather than research and science. This experience is what DiMaggio and Powell (1983) explained when they paraphrased Schelling (1978:4) to describe organizations in a structured field such as policing, “respond to an environment that consists of other organizations responding to their environment, which consists of organizations responding to an environment of organizations’ responses.”

            We have been replicating these same types of mistakes with programs like Scared Straight, D.A.R.E, and now CISM/CISD. Police organizations have a history of adopting social interventions without testing them. It took policing 20 years before a properly led randomized controlled trial showed Scared Straight increased the likelihood of offending rather than reduced it. If we do not follow the evidence-base, we can potentially be making people worse off than they would have been had we left them alone. This is part of the reason for the creation of the American Society of Evidence-Based Policing; to advocate, educate, and facilitate the use of quality research in policing. ASEBP strives to educate the police profession on up-to-date research to prevent the use of outdated practices. Policing has a responsibility to both the community and its employees to be cognizant of emerging research and alter policies and practices that do not align with the research base. I believe, as does Dr. Feuer and a myriad of other respected organizations, that this is the case for CISM/CISD interventions. Other international agencies have already discontinued this practice, and I believe policing should take the same stance and discontinue the use of CISM/CISD as a therapeutic intervention for individuals involved in a critical incident.

Where do we go from here?

            CISM/CISD forces individuals to discuss a critical incident during a period when they may not be psychologically ready. They may not feel comfortable talking about the incident in front of other people. They may find other ways to deal with negative emotions such as exercise, journaling, prayer, mediation, or just talking with a spouse, coworker, or close friend about the incident. People are resilient and there is evidence that people naturally recover on their own without a prescribed intervention (Pennebaker & Harber, 1993). Conventional wisdom would argue that people who bottle up their emotions will be worse off than people who talk about their feelings and release emotions. Research supports this part of conventional wisdom with Pennebaker & Beall (1986), demonstrating that journaling about an “upsetting personal event” has positive long-term benefits on an individual’s health. It appears it is not the releasing the emotion, it is when and how the emotion is released. We also do not want to leave officers with no social support, as lack of perceived social support is linked to increased risk of PTSD (Brewin, Andrews, & Valentine, 2000). Practical help rather than psychological help improve PTSD symptoms after a critical incident and can be considered “psychological first aid” (Litz et al., 2002; Raphael et al., 1996).

            Dr. Feuer’s talk gave a nice break down of the list of psychological first aid components that is listed out in McNally et al. (2003, p.67): listening, conveying compassion, assessing needs, ensuring that basic physical needs are met, not forcing someone to talk, providing or mobilizing company from family or significant others, encouraging but not forcing, social support, giving information, protecting from additional harm, ventilation of feelings as appropriate for the individual, and when appropriate, refer to mental health specialist. What I have seen repeatedly mentioned across the majority of studies is that if a person shows increased signs of PTSD, then refer to a mental health specialist. Intervening before a person shows signs of PTSD is ineffective and, in some cases, damaging. Dr. Feuer broke the idea down into simpler components for his Critical Incident Peer Support (CIPS) program – Listen, Protect, and Connect. Peer support volunteers receive training on psychological first aid. The training is free and online provided by several different agencies: Behavioral Health Professionals, American Red Cross, World Health Organization, and the US Department of Health and Human Services. In summary, Dr. Feuer’s Critical Incident Peer Support teams provide peer support (listening and comforting), normalize reactions (information), DO NOT provide referrals to debriefing, no mandatory interventions by the team, are trained to appreciate individual coping styles, and will refer officers to a competent mental health specialist if they are showing a negative reaction to the critical incident.

            At the end of his talk, I asked if Dr. Feuer had conducted a randomized controlled trial (RCT) on his method. He stated that he had not. He explained that the program was evidence-informed and he was working with academic researchers to first establish the efficacy of the training and planned on conducting a randomized controlled trial (RCT) in the future to determine if individuals receiving CIPS were better or worse off after receiving the intervention. According to the Society of Prevention Research, the cycle of research should be: 1. Conducting research to understand the predictors of problem and positive developmental outcomes and understanding the epidemiology and natural history of the problem, 2. Developing interventions to motivate changes in individuals and environments, based on theories of behavior and our understanding or mechanisms for behavior change, 3. Testing the efficacy of these preventive interventions, and 4. Testing the effectiveness of efficacious interventions in contexts under realistic delivery conditions (Biglan, Domitrovich, Ernst, Etz, Mason, & Robertson). Dr. Feuer’s program is at step three, and once he demonstrates efficacy, then he will be proceeding to step four, demonstrating CIPS is effective at reducing PTSD outcomes.

Why policing needs evidence-based information?

            The psychology field has known for over ten years that CISM/CISD is an ineffective intervention for PTSD symptoms and may be potentially harmful. When I spoke with Dr. Lilienfeld, he expressed dismay that policing was uninformed about the lack of empirical support for CISM/CISD. This quote was from 2003, “Although psychological debriefing is widely used throughout the world to prevent PTSD, there is no convincing evidence that it does so. RCTs of individualized debriefing and comparative, nonrandomized studies of group debriefing have failed to confirm the method’s efficacy. Some evidence suggests that it may impede natural recovery. For scientific and ethical reasons, professionals should cease compulsory debriefing of trauma-exposed people” (McNally et al., 2003 p. 72) And this is not the first time that policing practice fell behind current research. Scared Straight was commonly practiced before it was tested in a randomized controlled trial 20 years later (McCord, 2003). McCord stated in her seminal paper, Cures that Harm, “Social programs deserve to be treated as serious attempts at intervention with possible toxic effects, so that a science of intervention can prosper” (p.17). Although good intentions are behind our social interventions, good intentions do not equal positive outcomes. We have an ethical duty to our public and our officers to be up to date on empirical research concerning our interventions (Mitchell & Lewis, 2017) and when creating a new program such as CISP, to rigorously evaluate the program at the highest research standard possible in order to understand the impacts our interventions have on the individuals involved. To do otherwise fails to protect and serve.

References:

Barlow, B. (2010). Negative effects from psychological treatments: A perspective. American psychologist65(1), 13.

Biglan, A., Domitrovich, C., Ernst, J., Etz, K., Mason, M. J., & Robertson, E. (2011). Standards of knowledge for the science of prevention. Fairfax, VA: Society for Prevention Research.

Bledsoe, B.X. (December, 2003) EMS Myth #3: Critical Incident Stress Management (CISM) is effective in managing EMS-related stress. Retrieved from https://www.emsworld.com/article/10325074/ems-myth-3-critical-incident-stress-management-cism-effective-managing-ems-related-stress

Brewin, C. R., Andrews, B., & Valentine, J. D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of consulting and clinical psychology68(5), 748.

Carlier, I.V.E., Lamberts, R.D., van Uchelen, A.J., & Gersons, B.P.R. (1998). Disaster-related post-traumatic stress in police officers: A field study of the impact of debriefing. Stress Medicine14, 143–148.

Carlier, I.V.E., Voerman, A.E., & Gersons, B.P.R. (2000). The influence of occupational debriefing on post-traumatic stress symptomatology in traumatized police officers. British Journal of Medical Psychology73, 87–98.

Cochrane About Us, (n.d.) In Cochrane. Retrieved from https://www.cochrane.org/about-us

DiMaggio, P. J., & W. W. Powell (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review. la, 147-160.

Farrington, David P., Denise C. Gottfredson, Lawrence W. Sherman, and Brandon C. Welsh. “The Maryland scientific methods scale.” Evidence-based crime prevention(2002): 13-21.

Harris, M. B., & Stacks, J. R. (1998). A Three Year Five State Study on the Relationships Between Critical Incident Stress Debriefings, Firefighters’ Disposition, and Stress Reactions. Texas A & M University.

Harris, M. B., Baloğlu, M., & Stacks, J. R. (2002). Mental health of trauma-exposed firefighters and critical incident stress debriefing. Journal of Loss &Trauma7(3), 223-238.

Hobbs, M., Mayou, R., Harrison, B., & Worlock, P. (1996). A randomised controlled

trial of psychological debriefing for victims of road traffic accidents. British Medical Journal313, 1438–1439.

A Primer on International Critical Incident Stress Foundation, Inc. (n.d.) In International Critical Incident Stress Foundation, Inc. Retrieved from https://icisf.org/a-primer-on-critical-incident-stress-management-cism/

Lilienfeld, S. O. (2007). Psychological treatments that cause harm. Perspectives on psychological science2(1), 53-70.

Litz, B. T., Gray, M. J., Bryant, R. A., & Adler, A. B. (2002). Early intervention for trauma: Current status and future directions. Clinical psychology: science and practice9(2), 112-134.

Mayou, R.A., Ehlers, A., & Hobbs, M. (2000). Psychological debriefing for road traffic accidents: Three-year follow-up of a randomised controlled trial. British Journal of Psychiatry176, 589–593.

McCord, J. (2003). Cures that harm: Unanticipated outcomes of crime prevention programs. The Annals of the American Academy of Political and Social Science587(1), 16-30.

Mitchell, R. J., & Lewis, S. (2017). Intention is not method, belief is not evidence, rank is not proof: Ethical policing needs evidence-based decision making. International Journal of Emergency Services6(3), 188-199.

McNally, R. J., Bryant, R. A., & Ehlers, A. (2003). Does early psychological intervention promote recovery from posttraumatic stress?. Psychological science in the public interest4(2), 45-79.

Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: toward an understanding of inhibition and disease. Journal of abnormal psychology95(3), 274.

Pennebaker, J. W., & Harber, K. D. (1993). A social stage model of collective coping: The Loma Prieta earthquake and the Persian Gulf War. Journal of Social Issues49(4), 125-145.

Raphael, B., Wilson, J., Meldrum, L., & McFarlane, A.C. (1996). Acute preventative interventions. In B.A. van der Kolk, A.C. McFarlane, & L. Weisath (Eds.), Traumatic stress: The effects of overwhelming experience on mind, body, and society (pp. 463–479). New York: Guilford Press.

Rose, S., Bisson, J., Churchill, R., & Wessely, S. (2002). Psychological debriefing for preventing post traumatic stress disorder (PTSD). Cochrane Database Syst Rev2(2).

Schelling, T. (1978). Micromotives and Macrobehavior. New York: W.W. Norton.

Van Emmerik, A. A., Kamphuis, J. H., Hulsbosch, A. M., & Emmelkamp, P. M. (2002). Single session debriefing after psychological trauma: a meta-analysis. The Lancet360(9335), 766-771.