Effectiveness RCT of a CBT Intervention for Youths Who Lost Parents in the Sichuan, China, Earthquake

The study was conducted two years after the earthquake (2010–2011). Forty adolescents from two secondary schools were initially recruited to participate in this study. Their mean±SD age was 14.50±.71, with 13 boys (32%) and 27 girls (68%). Each adolescent had lost at least one parent in the earthquake. All of them scored ≥18 points on the 13-item Children’s Revised Impact of Events Scale (CRIES-13) and thus were considered to have PTSD symptoms. The parents or guardians and schools provided written informed consent on behalf of the youths. Approval of the study was obtained from the institutional review board of West China Hospital and Sichuan University.

The 40 adolescents were randomly divided into three groups: the short-term CBT group (N=16), the general support group (N=12), and the nontreatment control group (N=12). Only 32 adolescents completed the entire study. Six students were dropped from the CBT group, and two students were dropped from the general support group.

Chinese versions of three psychological instruments were used to evaluate the outcomes of the interventions: the CRIES-13 was used to assess PTSD symptoms (10), the Center for Epidemiologic Studies Depression Scale (CES-D) (11) was used to assess depressive symptoms, and the Connor-Davidson Resilience Scale (CD-RISC) (4) was used to measure psychological resilience.

The CRIES-13 was designed for children older than eight years who have experienced traumatic events and who can read independently. The CRIES-13 assesses 13 trauma symptoms in three areas: flashback, avoidance or numbness, and high vigilance. Higher scores represent more severe trauma.

In a pilot study with 252 students, the Chinese version of CRIES-13 showed good internal consistency (Cronbach’s α=.90). Using 18 points as the cutoff for PTSD, the Youden index was 58%, sensitivity was 81%, specificity was 77%, and diagnosticity was 81%.

The CES-D scale rates 20 symptoms of depression, with emphasis on the affective component and depressed mood. Each symptom is rated on a Likert scale ranging from 0, never, to 3, often. Cronbach’s α=.94 was obtained in this study.

The CD-RISC includes 25 items spread across five factors: personal competence, affect tolerance, acceptance of change, sense of internal control, and spirituality. Each symptom is rated on a 5-point Likert scale (from 0, never, to 4, always). The range of possible scores is 0–100, with the higher scores indicating more resilience. Cronbach’s α=.90 was obtained in this study.

Psychological assessments of the adolescents in the three intervention groups were conducted by trained mental health professionals before treatment, at the end of the last session, and three months after treatment.

The short-term CBT program was completed by ten youths. The CBT group intervention offers participants the opportunity to share their experiences, develop a narrative, and learn skills to cope with PTSD and depression. We designed the CBT program as a six-week intervention with a one-hour session each week.

We adapted the program from the manual Children and Disaster: Teaching Recovery Techniques (7), basing our revisions on Chinese cultural relevance and on our experiences working with earthquake victims. Specifically, two of the sessions were modified from the original technique. We omitted avoidance for several reasons. Our previous clinical experiences have shown that PTSD symptoms are exacerbated when avoidance is addressed, and to do so usually takes longer than allowed for with a six-week intervention. Also Chinese cultures generally believe that acceptance and avoidance are the most adaptive approaches, and school staffs are not trained to deal with avoidance and are discouraged from addressing this issue. The other modification was to integrate cognitive reframing throughout the CBT program. Each session began with a review and included a homework assignment to practice the techniques learned.

Sessions addressed participants’ fears by providing training in relaxation and coping techniques. Session 1 included an introduction to normalization concepts and relaxation training. Session 2 focused on helping participants to establish a safe place in their imaginations and included a guided meditation on imagining a safe place that was secure and trouble-free. Session 3 further developed imagery skills. It included imagery exercises designed to improve the adolescents’ control over the intrusion of painful images, thereby building coping skills. Session 4 included a dream intervention exercise to help the adolescents cope with and control nightmares. Session 5 focused on dual-attention therapy, which included an exercise, similar to some EMDR techniques, in which the patient moves his or her eyes back and forth while concentrating on a problem. Session 6 engaged participants in a group discussion and activity to look ahead to the future.

The general support intervention consisted of general support provided on an individual basis by three local volunteers who had received basic training in counseling techniques such as listening, reflection, and empathy. The volunteers visited the ten adolescents’ homes weekly for six weeks to provide support and assistance to them in dealing with problems at home and at school.

No intervention or services were provided to the 12 students in the control group.

The changes from baseline in CRIES-13, CD-RISC, and CES-D scores posttreatment and at the three-month follow-up were analyzed with repeated-measures analysis of variance (ANOVA), with time as a within-group factor and intervention as the between-groups factor. The slope of lines (scores as ordinate and time as abscissa) was determined by subtracting the baseline from the posttest (or subtracting the baseline from the follow-up) to find the change. The changes in CRIES-13, CD-RISC, and CES-D scores across groups were compared by one-way between-factors ANOVA followed by Scheffé’s post hoc tests. Three repeated-measures ANOVAs were conducted on dependent measures. In all three analyses, Mauchly’s sphericity test (to validate the repeated-measures ANOVA) was not significant, and therefore sphericity was assumed for interpretation of results.

Thirty-two students completed the study (CBT, N=10, including two boys and eight girls 14.70±.68 years old; general support, N=10, including three boys and seven girls 14.60±.70 years old; control group, N=12, including four boys and eight girls, 14.17±.39 years old). There were no significant differences among the three treatment groups in terms of demographic characteristics and baseline CRIES-13, CES-D, and CD-RISC scores.

The CRIES-13, CES-D, and CD-RISC repeated-measures ANOVAs each showed a significant within-subject main effect of time and a time × group interaction (CRIES-13: time, F=14.36, df=2 and 95, p<.001, and time × group, F=5.54, df=4 and 95, p<.01; CES-D: time, F=10.86, df=2 and 95, p<.001, and time × group, F=3.48, df=4 and 95, p<.05; and CD-RISC: time, F=17.97, df=2 and 95, p<.001, and time × group, F=4.55, df=4 and 95, p<.01).

Six one-way between-factor ANOVAs, followed by Scheffe’s post hoc tests, were conducted, one each for the slopes created by the respective CRIES-13, CES-D, and CD-RISC scores at posttreatment and the three-month follow-up (Table 1).

ANOVAs for the CRIES-13 scores between baseline and posttreatment and between baseline and three-month follow-up showed a significant difference among the three groups (F=5.82, df=2 and 31, p<.01), especially between the CBT and control groups at the end of treatment (p<.01). There was also a significant difference among the three groups (F=10.31, df=2 and 31, p<.001), especially between the CBT and control groups (p<.01) and between the CBT and general support groups (p<.05) at the three-month follow-up.

ANOVAs for the CES-D scores between baseline and posttreatment and between baseline and three-month follow-up showed a significant difference among three groups (F=7.40, df=2 and 31, p<.01), especially between the CBT group and control group (p<.01) and between the CBT group and general support group (p<.05) at the three-month follow-up.

Between baseline and posttreatment and between baseline and three-month follow-up, ANOVAs for the CD-RISC scores showed a significant difference among three groups (F=7.18, df=2 and 31, p<.01), especially between the CBT group and control group (p<.05) and between the general support group and control group (p<.01) at the three-month follow-up.

The findings of this first RCT of the use of CBT for disaster-traumatized youths dealing with parental loss showed that the effect of short-term group-based CBT in alleviating PTSD symptoms was robust. Furthermore, the group-based CBT intervention was more effective than either general support or nontreatment. Neither the general support intervention nor nontreatment had any effect in alleviating PTSD symptoms. Similar results were obtained by Giannopoulou and colleagues for children with PTSD symptoms after the 1999 earthquake in Athens, Greece (8). They used an intervention based on the Children and Disaster manual (7). Shooshtary and colleagues (9) used the same intervention to ameliorate PTSD symptoms among adolescents after a catastrophic disaster.

Children who are caught in a natural disaster often suffer from long-lasting psychological problems such as PTSD if no clinical intervention is provided. Persons exposed often experience extreme fear, elicited by stimuli associated with the traumatic event. Naturally, to reduce pain and suffering, adolescents may want to avoid the scene of the event and thoughts, feelings, or discussion about the earthquake. However, avoiding trauma-related stimuli may offer only temporary relief and often creates a barrier against successful long-term recovery. Thus early confrontation and management of intrusive fearful memories are very important. Confronting and learning to handle effectively these intrusive images can help adolescent trauma victims cope with difficulties and gain an early recovery.

The short-term CBT intervention in this study dealt with the fears of these adolescents through relaxation techniques and learning to establish an imaginary safe place or refuge; participants also learned to cope with flashback symptoms through imagery, dream intervention, and dual-attention techniques. The findings of this study showed that short-term CBT was effective in ameliorating PTSD symptoms of traumatized adolescents, and it was significantly better than the general support intervention.

Depression occurs frequently after exposure to a traumatic experience (12). This study showed that the short-term CBT intervention not only reduced PTSD symptoms but also ameliorated depressive symptoms.

This study was limited by the small sample size, and the use of multiple ANOVAs for comparisons may have inflated the alpha and led to a higher type I error. Also, the differences in treatment approach between CBT (group based) and general support (individual based) and differences in training and preparation of providers between the two groups could account for some of the variance of the results. Furthermore, many other treatments may have performed better than the general support intervention provided in this study.

Psychological resilience is the ability to adapt to and cope with change and enables people to recover and thrive against adversity (13). Among adolescents, resistance to psychological injury caused by negative events is enhanced if the youths receive nurturing and stable care from others (14). In this study, the reliable structured care provided by CBT and general support interventions improved the adolescents’ psychological resilience.

The funding sources of the study included grants CNTSP 2007BAI17B04 and CSCTP 2012SZ0139 and the Hong Kong Youth Foundation. The authors express appreciation to all adolescents who participated in this research. The interpretation and conclusions of this report are solely the authors’ and do not necessarily represent those of the Chinese government or the Hong Kong Youth Foundation.