Psychosocial interventions after crises and accidents
Systematic review
|Updated
Key message
The purpose of this health technology assessment was to evaluate the preventive effects of psychosocial interventions delivered within one year following accidents and crises, on trauma-related psychological disorders, functional impairments and behaviour problems. The assessment was commissioned by the Directorate for Social and Health Affairs in September 2004. We searched international scientific databases, selected studies according to preset criteria, appraised the methodological quality using checklists, and summarised the results narratively, in tables and in meta-analyses. We included 10 systematic reviews, 29 randomised controlled trials (RCTs) and 10 non-randomised effect evaluations. The systematic reviews and the nonrandomised studies were accounted for only, while we performed more thorough analyses of the RCTs. The 29 RCTs were published in 34 different articles. Interventions were: Psychological debriefing (PD) delivered in one or two sessions less than 3 weeks after trauma (7 studies) Other one- or two-session interventions, such as information, practical assistance and memory structuring therapy (4 studies) Cognitive behaviour therapy (CBT) delivered in 4-14 sessions less than 6 months after trauma (9 studies) Other multiple-session interventions, such as group interventions, counselling programs and individual support (4 studies) Pharmacological interventions (2 studies) Interventions for traumatised children (3 studies) Meta-analyses showed that there was no effect of PD compared to no intervention. The outcomes were posttraumatic stress disorder (PTSD), posttraumatic stress symptoms, anxiety and depression after 3-6 months and 1-3 years. Meta-analyses also showed that CBT was effective, both in comparison with no intervention and with other interventions. Relative risk for a PTSD diagnosis for CBT versus other interventions was 0.54 after 3-6 months, 0.38 after 9 months and 0.25 after 3-4 years. Conclusions Cognitive behaviour therapy for 4 weeks or more may prevent trauma related psychological disorders There is no evidence of preventive effects of psychological debriefing There is not sufficient research evidence on other types of interventions to conclude about effects.
Summary
Background This report is the second part of a health technology assessment of the effects of psychosocial interventions after crises, catastrophes and accidents. The first report was published in September 2006 (Knowledge Centre Report no 8-2006), and focused on events that were too demanding to be managed by local health services. Examples of such events were large transport and industrial accidents, natural disasters and war-related situations. The present report focuses on smaller scale events that may traumatise individuals, and that can be managed by the regular support system. This kind of events includes road traffic accidents, fires, violence, sexual abuse and unexpected/ unnatural death. The health technology assessment was commissioned by the Directorate for Social and Health Affairs with an aim to provide an evidence base for national guidelines for psychosocial crisis interventions. The objective of the assessment was to evaluate the preventive effects of psychosocial interventions delivered within one year of the event, on trauma-related psychological disorders, functional impairment and behaviour problems. Methods We searched systematically for systematic reviews (SRs), randomised controlled trials (RCTs) and non-randomised effect evaluations in The Cochrane Library, Centre for Reviews and Dissemination, OVID MEDLINE, OVID EMBASE, OVID PsycINFO, OVID CINAHL, The Pilots Database and SveMed. In addition, we examined reference lists in relevant systematic reviews to identify single studies that we might otherwise have missed. We included studies where participants were exposed for potentially traumatising events involving serious danger and/or death, where at least one psychosocial intervention was delivered after the event, and where outcomes were trauma-related physical or psychological symptoms, or level of functioning. Two independent reviewers evaluated studies for inclusion using a standardised inclusion form, and appraised the methodological quality of the included studies with appropriate check lists. We summarised the studies narratively, in tables and in meta-analyses. Results We identified 2461 unique references in the search for SRs, and 3536 references in the search for RCTs and non-randomised effect evaluations. After appraisals of titles, abstracts and full text articles we included ten SRs, 34 RCT articles and ten non-randomised effect evaluations. We chose, however, not to use the SRs and the non-randomised effect evaluations in the assessment of effects of interventions. The reasons for this were that most of the SRs were of low methodologically quality, they were overlapping to a great extent, and they were not up to date. Further, given the considerable number of relevant RCTs it was not necessary to supplement with nonrandomised studies. The 34 included RCTs were based on 29 different studies. The interventions were: Psychological debriefing (PD) delivered in one or two sessions less than 3 weeks after trauma (7 studies) Other one- or two-session interventions, such as information, practical assistance and memory structuring therapy (4 studies) Cognitive behaviour therapy (CBT) delivered in 4-14 sessions less than 6 months after trauma (9 studies) Other multiple-session interventions, such as group interventions, counselling programs and individual support (4 studies) Pharmacological interventions (2 studies) Interventions for traumatised children (3 studies) Meta-analyses showed that there was no effect of PD compared to no intervention. The outcomes were posttraumatic stress disorder (PTSD), posttraumatic stress symptoms, anxiety and depression after 3-6 months and 1-3 years. Meta-analyses also showed that CBT was effective, both in comparison with no intervention and with other interventions. Relative risk for a PTSD diagnosis for CBT versus other interventions was 0.54 after 3-6 months, 0.38 after 9 months and 0.25 after 3-4 years. We found marginal differences only between intervention groups and control groups in the studies focusing on other interventions than PD or CBT. These studies were also small and isolated and would be difficult to use as an evidence base. Conclusions Cognitive behaviour therapy for 4 weeks or more may prevent trauma-related psychological disorders. There is no evidence of preventive effects of psychological debriefing. There is not sufficient research evidence on other types of interventions to conclude about effects.