A Scoping Review of Trauma-Informed Pediatric Interventions in Response to Natural and Biologic Disasters

A scoping review was performed of trauma-informed psychological interventions to treat anxiety, depression, and posttraumatic stress symptoms in youth in response to natural/biologic disasters. The specific aims were to identify psychosocial interventions used in response to natural/biologic disasters, report the interventions’ effectiveness, describe limitations, and provide treatment recommendations and future directions. Of the 45 studies extracted, 28 were on natural disasters and 17 on biologic disasters with the majority related to the COVID-19 pandemic. The most commonly implemented interventions were Cognitive Behavioral Therapy (CBT), Trauma-Focused Cognitive Behavioral Therapy (TF-CBT), and eye movement desensitization and reprocessing (EMDR). The UCLA Posttraumatic Stress Disorder Reaction Index (UCLA PTSD-RI) and the Strengths and Difficulties Questionnaire (SDQ) were the most frequently used measures. Methodological rigor was varied, with 60% randomized, controlled trials. Overall, there was a significant decrease in posttraumatic stress symptoms, distress, anxiety, and depression regardless of whether the participant received CBT, TF-CBT, or EMDR. Generally, there was not a significant decrease in anxiety and depression with yoga, cognitive fear-reduction, emotion-based drawing, and community health education. Recommendations for future directions include larger-scale studies with group and on-line interventions that include younger children with moderation analyses by gender and race/ethnicity.


Introduction
Disasters can be categorized as either manmade or natural/biologic with the majority of studies exploring the efficacy of intervention in response to manmade disasters such as terrorism and war. There has been a significant increase, however, in natural disasters (e.g., floods, wildfires, hurricanes, earthquakes, and tsunamis) and biological events (e.g., pandemics) during the 21st century [1]. Millions of children were affected by the COVID-19 pandemic with over 15.5 million testing positive for this infectious disease and over 100,000 US children losing a primary caregiver to the pandemic [2]. Such traumatic exposures can be associated with a higher risk of developing post-traumatic stress disorder (PTSD), depression, anxiety, behavior problems, substance use disorders, suicidal ideation/self-harm, poorer academic and occupational achievement, and adverse physical outcomes [3,4].
Professionals in emergency medicine have identified five phases of the disaster cycle: prevention, mitigation, preparedness, response, and recovery. As it relates to behavioral health, the prevention phase consists of identifying threats/risks and sources of community resilience that could serve to minimize the psychological impact of a disaster. The mitigation phase involves taking steps to prevent and reduce the cause, impact, and potential consequences of future disasters. In the preparedness phase, behavioral health clinicians assist with evaluation and quality improvement to ensure everyday readiness. The response phase is focused on restoring optimal psychological well-being and functioning through both shortand long-term behavioral health interventions. And finally, in the recovery phase, the goal is to stabilize and restore all behavioral health supports, returning children and their families to their pre-disaster level of functioning, and/or creating a new normal [5].
The intent of this review is to identify interventions that can be used in the response phase of a natural/biologic disaster and to use the information obtained to inform recommendations for preparedness. To date, there has been a lack of specific focus on the efficacy and effectiveness of child disaster mental health interventions in response to such disasters.
Mental health treatments shown to be effective in the treatment of trauma-exposed youth include trauma-focused cognitive behavioral therapy (TF-CBT), trauma and grief component therapy for adolescents (TGCT-A), and eye movement desensitization and reprocessing (EMDR). TF-CBT is an evidence-based manualized treatment that consists of nine cognitive and behavioral components [6]. These components include (1) psychoeducation, (2) parenting skills, (3) relaxation skills, (4) affective modulation skills, (5) cognitive processing skills, (6) trauma narration and processing, (7) in-vivo mastery of trauma reminders, (8) conjoint parent-child sessions, and (9) enhancing future safety. TGCT-A is an evidence-based modular approach to treat trauma, bereavement, and traumatic bereavement. The four modules in this intervention include providing foundational knowledge about trauma and skill building (Module 1), working through traumatic experiences (Module 2), working through grief experiences (Module 3), and looking to the future to identify developmental disruptions and set realistic expectations and plans for the future (Module 4) [7]. EMDR has been established as an effective intervention in the treatment of children/adolescents. EMDR is an eight-phase approach rooted in the adaptive information-processing model with the objective of desensitizing the patient to an unpleasant memory by reliving the trauma without experiencing the sensations of imminent threat [8]. Although all are empirically supported interventions, it is unclear as to the frequency with which they are implemented in response to distress associated with experiencing a natural/biologic disaster. Other possible interventions should be explored due to the potential lack of access to clinicians trained in these approaches and possible barriers to accessing services.
To the authors' knowledge, this is the first scoping review to focus solely on natural/biologic disasters and to include a specific focus on interventions used in response to the COVID-19 pandemic. It is also the first review to include evaluation of the efficacy of interventions on internalizing symptoms in addition to posttraumatic stress in response to a disaster. Our specific aims of this review are to: (1) identify psychosocial interventions that are used in response to natural/biologic disasters, (2) report on the efficacy and effectiveness of the identified interventions, (3) identify gaps/limitations, (4) provide treatment recommendations based on this review, and (5) identify strategies for enhancing research in this field.

Search Strategy
A scoping review approach was chosen as it was anticipated that there would be a diverse body of literature pertaining to the broad topic of natural disasters. A search strategy was generated to capture the concepts of psychological interventions used to treat children in response to natural disasters. The search included database-controlled vocabulary and text word searching with Boolean operators. Unpublished literature, dissertations, conference proceedings, and registered trials were included in the results from the databases. A hand search of the most relevant articles reviewed in the full text was completed. Terms for natural disasters were weather or COVID-19 pandemic related. Pediatric terms were used to limit results to youth and transitional-aged youth. The search was limited to English. Studies published between 2000 and 2023 were included.
Titles and abstracts of articles identified via the search terms were uploaded to the online platform Covidence. See Table 1 for abstracts imported to Covidence for review. Once uploaded, two reviewers followed the built-in protocols to screen identified articles in the following manner: (1) Screen title and abstract, (2) Full text review screen, and 3) Article extraction. A third rater on the research team addressed disagreements as to the inclusion/exclusion criteria.

Study Selection
The study selection PRISMA flowchart is presented in Figure 1. The combined searches initially yielded 1979 articles. Duplicates were removed (N = 427). Whereas a total of 1552 studies were screened, 45 studies met criteria to be included in the review.
Up2-D2 intervention (12 cognitive-behavioral and positive psychology sessions for 45-50 min each) • designed as a prevention program targeting internalizing and externalizing problems • Improved anxiety, but not depression or anger • At 2 month follow-up, anxiety was shown to be significantly decreased • At 6 month follow-up, anxiety, general difficulties, emotional symptoms, and anxiety were significantly decreased • General difficulties domain was also significantly decreased at post-assessment • More improvement for those with higher baseline anxiety, especially at post-assessment • PTSD improved faster and more dramatically for CBITS group than the TF-CBT group • Significant decrease in depressive symptoms in CBITS group • CBITS is more accessible to families • More children at risk for PTSD in TF-CBT group than CBITS • Both significantly decreased PTSD symptoms even after 10 month follow-up    The CBT group evidenced significant improvement in anxiety, depression, and dissociation at the 6 month follow up, and in PTSD and dissociation at the 12 month follow up    • Attending the intervention was predictive of significantly higher self-efficacy but was not associated with differences in self-esteem, depression, or emotional/behavioral problems • Double parental death exerted a powerful effect on child psychosocial health, eliminating the association between intervention attendance and higher self-efficacy • An interaction was found between bereavement status and intervention attendance on child self-efficacy Pityaratstian (2015)      • Distress experienced by the children and adolescents decreased over the sessions as measured by the coding system that evaluated coherence (structure, orientation of events, normative elements), level of elaboration of the story (description, emotion, sense of self), and the evaluation of the story (causality, insight, and integration) • Analysis using the coding system identified that the internal logic of the stories were maintained as the detail diminished, and the level of evaluation of causality, self-reflectivity, and integration into their life increased

Demographics of Participants
The age of children and adolescents included in these studies ranged from 5 to 19 years, including three studies with emerging adults who were as old as 20 to 24 [27,29]. The majority of the studies included only adolescents (33/45, 73.3%). This is interesting as the recommended age for the therapies tested includes younger children. For example, TF-CBT has been rated as "probably efficacious" for preschool-aged children as young as 3 years old who have at least some memory of their trauma [53]. In addition, self-report measures of trauma typically start between ages 6-8 years. This may partially explain why very young children were not included. The age range varied between studies, with some including children within a 2-year age range and others including a large age range [14,21,27,38]. For studies that reported sex, females were included at slightly higher percentages than males (overall, 5284/8383 = 63%). This seems reasonable considering that females tend to have a higher risk of PTSD after exposure to a disaster, such as found in a systematic review of children and adolescents after earthquakes and floods [54].
In the U.S., the race/ethnicity of the participants was reported as African American/Black, White/non-Hispanic White/Caucasian, Latino/Latinx/Hispanic, with fewer studies reporting percentages of Asian, Native American/American Indian/Alaska Native, Native Hawaiian/Other Pacific Islander, or biracial/multiracial/other racial/ethnic backgrounds. Three U.S. studies did not include race/ethnicity data, with one in Hawaii, the other in Puerto Rico, and the third in New Orleans. Reporting standards have recently been proposed for ethnicity/race for journal articles, which include not collapsing across categories and avoiding the assumption that whiteness is the norm [55]. In terms of geographic region, the studies took place in multiple countries in 6 of the 7 continents, including Africa (South Africa), Asia (China, Iran, Japan), Europe (Italy, Spain), North America (Canada, Mexico, United States), Australia/Oceania (Australia, New Zealand), and South America (Argentina).

Study Design Features
Sample sizes varied greatly, with 25/45 (55.5%) having fewer than 100 participants. One of these small n studies was of 31 children affected by the tsunami in Sri Lanka [11]. There were two large-scale research studies of more than a thousand adolescents. One was a group after-school intervention of 1030 adolescents (545 intervention/550 control) following Hurricane Hugo in which the control group did not receive an intervention [10]. The other was an on-line single-session randomized controlled trial of more than 2000 adolescents during the COVID-19 pandemic [15].
The studies varied in their methodological rigor. Slightly more than half (26/45; 58%) were randomized controlled trials. Some of these trials compared two experimental groups (n = 10, 38%), others including control groups received either a standard intervention (n = 8, 31%) or were a no-treatment control (n = 8; 31%); somewhat surprisingly, only four studies reported using a wait-list control group [9,50].

Description of the Constructs and Measures
Constructs most commonly assessed included screening for PTSD symptoms, reactions to trauma, depressive symptoms, anxiety, general psychological adjustment, and selfefficacy. The most commonly assessed construct was PTSD symptoms, with 20/45 studies (44%) including such measures. Constructs less often explored in these studies were subjective units of distress, social support, hopelessness, cognitive functioning, healthrelated quality of life, psychological flexibility, experiential avoidance, post-traumatic growth, negative life events, and overall level of fearfulness.
Multiple measures were used in at least three studies. One of the most often used measures, albeit not focused specifically on trauma symptoms, was the Strengths and Difficulties Questionnaire (SDQ) [56]. The SDQ is a 25-item questionnaire that assesses emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems, and prosocial behavior. The SDQ has been normed in both Western (such as Denmark and Italy) and Eastern (Japan) cultures. This measure of general psychological adjustment was employed in 7 studies, most with children ages 11 and older. The other most commonly used self-report scale was the UCLA Posttraumatic Stress Disorder Reaction Index (UCLA PTSD-RI), which was used in 7 studies to screen for children's and adolescents' reactions to trauma. More specifically, the UCLA PTSD-RI is a self-report questionnaire that assesses the frequency of occurrence of PTSD symptoms over the past month and maps directly onto DSM intrusion, avoidance, and arousal criteria [57]. Measures that were employed for children ages 6 and older included the Center for Epidemiologic Studies Depression Scale (CES-D) and the Depression Self Rating Scale for Children (DSRS-C). These measures were used in 3 studies. For children ages 7 and older, the Child Depression Inventory (CDI) was used in 5 studies (one was the short form). For children ages 8 and older, scales included the Impact of Event Scale-Revised (IES-R), which was used in 5 studies and the Child PTSD Symptom Scale (CPSS) as well as the Child Revised Impact of Events Scale-13 (CRIES-13) used in 3 studies. Numerous other measures were used in only one study (e.g., the Derogatis Brief Symptom Inventory (BSI), the Multidimensional Anxiety Scale for Children Second Edition™ (MASC 2™), the Screen for Child Anxiety Related Disorders (SCARED), the Patient Health Questionnaire modified for adolescents (11)(12)(13)(14)(15)(16)(17) (PHQ-A), and the Short Cognitive Assessment System).
Of note is that certain commonly measured constructs such as depression and anxiety were measured using multiple different instruments (e.g., CDI, CES-D, PHQ-A, DSRS-C, and the Mood and Feelings Questionnaire to measure depression and the Anxiety Disorder Interview Schedule for Children, MASC-2, SCARED, and the Spence Children's Anxiety Scale to assess anxiety). Self-efficacy was also measured in multiple studies using different instruments. Working definitions of constructs such as anxiety and depression varies from tool to tool. As a result, one should be cautious about making cross-comparisons of diagnostic terms and findings. Moreover, cultural/racial validity of assessment tools needs to be considered as the majority are established using population representation processes in which Whiteness is the norm. It is also important to note that these instruments are being used in various cultural environments such as collectivistic, communal/intergenerational, and individualistic. Particularly when studies are conducted with children in the same age range and culture, it is vital for researchers to select the same measures to more easily compare results. As the UCLA PTSD-RI and the SDQ were in the most studies, researchers are encouraged to consider these for future work in this area and to conduct additional norm-based studies.

Features of the Treatments
Approximately 31% of the studies evaluated the efficacy of CBT or its components (e.g., thought stopping; behavior activation), 16% evaluated TF-CBT or its treatment components (e.g., narrative exposure), and 16% of the studies used EMDR to treat associated posttraumatic stress symptoms and/or internalizing concerns associated with exposure. Additional interventions included Acceptance and Commitment Therapy (ACT), Catastrophic Stress Intervention (CSI), Cognitive Behavioral Intervention for Trauma in Schools (CBITS), Grief and Trauma Intervention (GTI) for Children, Narrative Exposure Therapy (NET), Psychological First Aid, Solution Focused Brief Therapy, various mindfulness strategies, and yoga and/or aerobic exercise education [10,11,20,26,28,34,47]. Finding common core elements across these therapies in order to evaluate their effectiveness to reduce traumatic stress in children and adolescents exposed to natural or biologic disaster is a crucial next step.
The interventions were primarily delivered in a group versus individual format, which makes practical sense when trying to reach as many effected children and adolescents as possible right after a disaster has occurred. TF-CBT was the most common intervention to be provided to individual children, such as in a treatment study conducted in Puerto Rico for trauma-exposed children after experiencing a hurricane or earthquake [22]. An example of a group format was in the CSI study following Hurricane Hugo in South Carolina [10]. Students met in a larger group of approximately 150 and broke out in smaller groups of 10-16 students each.
The dose of treatment varied greatly across studies. Specifically, the number of sessions ranged from one to 12 sessions, with many offering six. The length of sessions ranged from as short as 20 min to as long as 4 h, also the treatments commonly included sessions lasting for 45 min to 2 h. Studies varied greatly in intensity, i.e., how often the intervention material was presented, ranging from multiple times per week to every few months. One Chinese study provided treatment as often as 4 times per week for a month, with significant positive effects during the COVID-19 pandemic [28]. Comparison across studies would be quite valuable on this aspect of treatment delivery as it relates to cost effectiveness. The type of facilitator for treatments included psychologists, advanced practice psychiatric nurses, other mental health professionals as well as trainees.

Findings
Overall, there was a significant decrease in posttraumatic stress symptoms, distress, anxiety, and depression regardless of whether the patient received CBT, TF-CBT, or EMDR. Effect sizes were large in some of these studies, such as when providing individual therapy using TF-CBT. A large group after-school intervention also reported significant reductions in mental distress compared to students in the control condition. With regard to developmental considerations, at least one study found the opposite effect for EMDR among younger children with increased distress and anger, but other studies that included young children using TF-CBT or Psychological First Aid were determined to be therapeutic [16,22,46]. As there were so few studies with young children as participants, further research is needed to determine any differences in findings based upon developmental level.
There was not a significant decrease in anxiety and depression with yoga, cognitive fear-reduction, emotion-based drawing, and community health education. It seems as if a lack of selection of a standardized and well-validated intervention such as TF-CBT or even of less well-studied standardized treatments was associated with lower likelihood of significant effects.
Effect sizes were not reported for many studies, making it difficult to evaluate the impact of the intervention, i.e., small, medium, or large. An exception was for the Puerto Rican study of the individual intervention of TF-CBT for children exposed to a hurricane or earthquake; this study reported a large effect size for their sample of 56 [22]. There is also the issue of comparing disparate outcome constructs and measures so that conducting meaningful meta-analyses is not feasible with this collection of studies in the area thus far.

Follow-Up for Outcomes
These studies all contained immediate post-intervention assessments, but the number and timing of their follow-up varied. Follow-up data was provided for 44% of the studies included in this review. Follow-up data collection ranged from 1 month to 4 years with the modal duration being between 3 and 6 months. Of the 44%, 85% of the studies indicated that treatment gains were maintained. In the Hardin et al. study, treatment effects for CSI in response to a hurricane between the intervention and control groups dissipated after 24 months (i.e., were not found at 30 and 36 months) [10]. However, in the Giannopoulou et al. study evaluating the use of CBT in response to an earthquake decreases in posttraumatic stress symptoms were maintained at a 4-year follow up [37].

Discussion
Forty-five studies were identified that investigated the efficacy of intervention in response to natural and biologic disasters in children and adolescents. The most commonly implemented interventions were CBT, TF-CBT, and EMDR. Almost one-third of the studies evaluated the efficacy of CBT or one of its components (e.g., thought stopping; behavior activation). A sizable percentage (16%) evaluated TF-CBT or its treatment components (e.g., narrative exposure), and 16% used EMDR to treat associated posttraumatic stress symptoms and/or internalizing concerns associated with exposure. There were numerous other standardized interventions, including Acceptance and Commitment Therapy (ACT), Catastrophic Stress Intervention (CSI), Cognitive Behavioral Intervention for Trauma in Schools (CBITS), Grief and Trauma Intervention (GTI) for Children, Narrative Exposure Therapy (NET), Psychological First Aid, and Solution Focused Brief Therapy [10,11,20,26,28,34,47]. Twenty studies evaluated the efficacy of the intervention on posttraumatic stress symptoms with seven using the UCLA PTSD Index to measure posttraumatic stress symptoms. The UCLA PTSD Index and SDQ were the most commonly used measures. Overall, there was a significant decrease in posttraumatic stress symptoms, distress, anxiety, and depression regardless of whether the child or adolescent received CBT, TF-CBT, or EMDR. There was not a significant decrease in anxiety and depression with less standardized interventions such as yoga, cognitive fear-reduction, emotion-based drawing, and community health education.

Recommendations
Core principles of disaster interventions need to be identified, as there seems to be no single best practice approach. This will provide children and adolescents with a broader range of possible treatment options and providers. Considerations for choice of intervention are guided by many factors, including the specialty training of available clinicians, a focus on prevention vs. intervention, individual vs. group treatment, and accessibility. When referring professionals or the behavioral/mental health providers are seeking to choose an intervention in response to natural/biologic disasters, they could utilize this review's table for comprehensive lists of intervention used by type and outcome. Additional investigation is needed for other types of therapies such as ACT and mindfulness strategies, etc., as there does not seem to be enough research that has been done to determine their effectiveness for the treatment of children who have post-traumatic stress symptoms following a natural disaster. In terms of study populations, future research should seek to include younger children (i.e., including preschool-aged children) as evidence exists that trauma-informed treatments such as TF-CBT can be effective for children as young as three years old who are able to remember details of the event. Also, studies should be conducted in continents other than North America, such as Africa, for which only one study was located for our review that provided treatment for children affected by AIDS. Ethnicity/race data should be collected for all studies and should not be collapsed across categories. Larger samples will provide the degrees of freedom to explore any racial/ethnic group differences in response to the elements of interventions to treat/prevent PTSD, i.e., to allow for moderation analyses. Finally, in order to reach a greater number of children and adolescents, future research should also consider the methodology of the larger-scale studies, such as group interventions and on-line therapy options.
It is recommended that future studies examine and dismantle specific treatment components that are shared across empirically-supported treatments to determine what is essential. Additional research is needed to determine the treatment modality (individual vs. group), intervention setting, length of therapy, and timing of intervention delivery that is most effective and compatible given the culture, resources, and type of disaster.
From a public health standpoint, effort should be focused on strategies that mitigate the impact of disasters on youth and families and on preparedness. This review underscores the importance of using the mitigation phase to identify the most cost-effective and culturally appropriate way to reach the greatest number of youth building on a community's strengths and removing barriers. This review assists with everyday readiness as it provides a single source of interventions that has been used in the response and recovery phases and serves as a roadmap for needed considerations for methodological design. It is recommended that a repository of disaster-related resources exist for the behavioral health clinician to expedite post-disaster response. Such a repository will allow for the efficient design of developmentally and culturally appropriate interventions and measurement of treatment response with consideration of the disaster type.

Limitations
Despite some broad-sweeping generalizations drawn from this body of research, more fine-grained conclusions cannot yet be made due to the vast methodological differences, limited consensus on how constructs of interest should be measured, and lack of validated instruments in various populations and cultures. These studies demonstrated marked variability in passage of time between exposure and intervention. Some studies took a modular approach (e.g., thought stopping; behavior activation) while others evaluated the intervention effects as a whole. There were both individual and group interventions. The treatments were conducted in a variety of locations (e.g., school, on-line), and the provider type delivering the intervention was mixed. The authors encountered difficulty in conducting any meaningful meta-analyses due to lack of consistent measures used across studies. As many of the studies had small sample sizes, testing more complex models including mediation and moderation was not possible. Furthermore, self-selection bias must be considered when interpreting the findings of these studies. Those who participated in these studies might not represent the vast majority of those affected by natural disasters as these individuals were able to navigate barriers such as challenges with transportation, access to the internet, connection with mental health providers, and stigma that might be related to accessing mental health services. This might also have influenced the lack of inclusion of very young children and the infrequency of including younger children in general. Not all young children are in formal preschool programs in which traumainformed interventions could be offered. The generalizability of this collection of studies is therefore limited by child age, geographic region, and by racial/ethnic groups with lack of representation of minorities across multiple countries. Due to the nature of the randomness of natural and biologic disasters, it is difficult to plan to replicate even the most promising methodologies. However, this is why preparedness is so important in effective behavioral health pediatric disaster response. Finally, a limitation of all systematic reviews is selection bias because of publication bias. Data from studies with statistically significant results are more likely to be published, limiting the scope of research in the area. This is particularly a problem in research with children and adolescents conducted after a natural disaster due to smaller samples reducing power to detect significance when effect sizes are not large.

Conclusions
This review provides compelling evidence for the use of CBT, TF-CBT, and EMDR for youth in response to natural/biologic disasters. Despite limitations of a research area that is broadly defined with inconsistent measurements and a range of treatment approaches, this review underscores the importance of devoting efforts and resources to mitigate the psychological impact of a natural disaster, engage in everyday readiness, and prepare for future catastrophic events. The review's findings have notable implications not just for clinical practice, but also for policy development and future research studies. A take-home point is that evidence-based trauma-informed treatments for children and adolescents implemented via large group and internet-based interventions are viable for real-world applications with the broadest reach. While designing these response and recovery interventions, limitations of past research can be minimized by selecting empirically validated treatments, reaching out to the broadest possible age range and families from all affected racial/ethnic groups, selecting measures with established reliability/validity, and planning for follow-up assessments. Careful prior preparation will maximize the potential benefits of future trauma-informed interventions for children and adolescents affected by natural and biologic disasters.

Conflicts of Interest:
The authors declare no conflict of interest.