Family centered eco-system model of traumatic stress and coping: Structural equation modeling and their clinical/policy implications*)

(2nd Edition, 2000/12/12)

Shigeo Tatsuki, Ph.D.

This is a review of relief activities following the 1995 Kobe Earthquake and subsequent practice/research on post-traumatic stress management. During the first three months after the quake, the author organized Kwansei Gakuin University Relief Volunteer Center and managed its crisis response activities. This was continued until April 15, 1995. The Center has been granted a new mandate since then and it has become the Kwansei Gakuin Human Service Center. One of the Center's new mandates includes research on disaster relief for the earthquake victims. In June 1995, with help from the American Jewish World Service, the Center organized a workshop on post-traumatic stress management. 24 helping professionals including medical doctors, public health nurses, social workers, clinical psychologists, and volunteer managers participated in this 7 day overnight workshop. Dr. Robert J. Lifton of New York City University outlined the psychology of disaster victims citing his works about atomic bomb survivors in Hiroshima and holocaust survivors (Lifton, 1967; Lifton & Mitchell, 1995). Dr. Reuven Gal of Carmel Institute in Israel presented an integrated framework based on Lazarus's stress and coping model (Lazarus & Folkman, 1982) and on an Israel-based coping style model (Lahad & Cohen, 1989). Dr. and Mrs. Gal introduced several stress management techniques such as the Critical Incident Stress Debriefing, Relaxation, Scribble Therapy, Movement Therapy, and Biblio Therapy (Gal, 1995). The Center further helped to organize three more workshops sponsored by the Japanese Organization for Crisis Response (JOCR) from March 1996 to March 1997.

Utilizing knowledge, skills and the personal network gained from the first workshop, the author participated in community-based debriefing group work projects organized by the Kobe City Child Guidance Clinic from November 1995 to March 1996. The project aimed at mothers of preschool age children. A team consisting of a social worker, a clinical psychologist and a psychiatrist instigated an out-reach program to local preschools.

The Kobe Mother-Child Stress-Coping Research (Noguchi, Tsubokura, Taniguchi, & Tatsuki, 1997;  Tsubokura, Noguchi, Taniguchi, & Tatsuki, 1997) was conducted during the same period as the out-reach group work project. This study examined the impact of the Great Hanshin Earthquake on preschool children and their mothers, and the way the families coped with traumatic stress.


Subjects: The subjects were collected from 10 public and 8 private preschools where the Kobe Child Guidance Clinic team conducted debriefing group work sessions. Those 18 preschools are located in 5 wards (Higashi-Nada, Nada, Chuou, Hyogo, and Tarumi), which were heavily hit by the earthquake. In total, 865 questionnaires were handed by preshool teachers during the first week of Novermber, 1995. 445 or 51.4% of the subjects returned the questionnaire to preschool teachers by the end of the month. Among the 445 responses, 7 subjects showed too many missing responses or acquiscent responses and therefore omitted. In total, 438 subjects responses were used for further analyses.

Instruments: The instruments include 1) the DSM-IV based Children's Stress Symptom Scale, 2) a Japanese translation of the Impact of Event Scale (Horowitz, Wilner, & Alvarez, 1979), 3) BASIC-Ph coping style scale, 4) the Family Adaptability and Cohesion Evaluation Scale at Kwansei Gakuin version 3 (Tatsuki, 1993), and 5) the Post-Earthquake Daily Hardship Scale. 1) The Children's Stress Symptom Scale contains 21 Likert-type scale items asking mothers about their children regarding major PTSD symptoms as described in DSM-IV. The internal consistency reliability for the present study is 0.79. 2) Japanese translation of 15 item Impact of Event Scale was administred to mothers. Based on Horowitz et al. (1979), the scale measures the degree of avoidance and intrusion as well as the total score. Internal consistency reliability for avoidance was 0.78, that for intrusion 0.87, and that for the total scale was 0.88. 3) BASIC-Ph coping style scale evaluates the degree of prefered coping resources and is based on Lahad and Cohen's (1989) model, which delineates 6 different coping resources, Belief, Affect, Social, Immagination, Cognitive, and Physical. For the current study, only Belief, Social, Cognitive, and Physical subscales were used due to the fact that Affect and Immagination subscales did not show adequate internral consistency reliabilities. The four subscales reliability estimates ranged from 0.48 to 0.87. 4) The Family Adaptability and Cohesion Evaluation Scale at Kwansei Gakuin version 3 or FACESKGIII is based on David Olson's Circumplex model of marital and family systems and measures cohesion and adaptability dimensions of family system. According to the Circumplex model, a moderate level of both cohesion and adaptability dimensions promote healthy family functioning. On the other hand, extreme levels of cohesion or adaptability hinders family functioning. Unlike Olson group's FACESIII, FACESKGIII is a Thurstone equal appearing distance scale and was specifically designed to evaluate the curviliner hypothesis between cohesion/adaptability on one hand, and the level of familiy functioning on the other. FACESKGIII consists of 11 cohesion, and 15 adaptability equal appearing distance scale yes-no items. The items were designed to capture very low (1 point) to very high (8 point) level of cohesion and adaptability dimensions. The R-squared was used as the reliability estimate and that for cohesion is 0.45 and that for adaptablity is 0.45. 5) The Post-Earthquake Daily Hardship Scale consists of 15 items asking the magnitude of hardship that were caused by the earthquake. The scale covers the degree of physical damage to the household and to the immediate neighborhood, loss of relatives, friends, job, and precious items, and the length of life line recovery.


Daily Hardship: The majority of the respondents went through major hardships in their daily lives after the earthquake. Only 11.4% (50 ) of the respondents reported that their houses were not damaged by the earthquake. 55.6% (243) answered light (green tag) damage, 21.7% (95) modarate (yellow tag) damage, and 11.0% (48) severe (red tag) damage. About 45 % (193) respondents reported that the damage in the immediate neighborhood was light, while 31.9 % (138) reported moderate (many yellow tag damges), and 15.9% (69) reported severe (most houses were red-tagged) damage. 7.6 % (33) reported fires in the immediate neighborhood. In terms of days evacuated from their houses, almost half of the respondents (227) spent not more than 9 days. 12.8% (56) spent 10 to 29 days. 18.3% (80) spent 30 to 59 days. 10.1 % (44) spent 60 to 89 days. 6.8 % (30) evacuated for more than 90 days. In terms of lifeline desruption, about three out of five respondents were able to use telephone and electricity one day after the earthquake. Meanwhile, only a third of the respondents enjoyed city water and city gas within two weeks and 40 days respectively after the earthquake, and two thirds had to wait more than two weeks for city water and 40 days for city gas.

Children's Stress Symptom: The Children's Stress Symptom Scale is a 4 point Liker-type scale, where none is scored 0, sometimes 1, often 2, and always 3. The stress symptoms most frequently reported as "often" or "always" occurring were item #5 "Child cannot go to bed alone or without a room light" (30.5%), #21 "Child often insists on taking care of other children" (23.4%), #4 "Child looks nervous when left alone" (21.4%), #15 "Child sometimes becomes excessively dependent or regresses"(13.7%). The responses to each of the 21 items were added and the total score was computed for each child. The mean and median of the child stress symptom scale scores were 11.2 and 9, respectively within a theoretical range of 0 to 63. This suggests that the majority of the children were not presenting severe stress symptoms about 10 months after the earthquake. However, exploratory data analysis of the distribution outliers suggested about 2.5 % (11 out of 438 children) showed extremely high stress symptom scores. Those 11 children were suspected for seriously suffering from PTSD.

Mother's Stress Symptoms: The Mother's Impact of Event Scale is a 4 point Liker-type scale, where none is scored 0, sometimes 1, often 2, and always 3. The stress symptoms most frequently reported as "often" or "always" occurring were #14 "When I see anything that reminds me of the earthquake, the sensation of that time is revived." (29.7%), #1 "The memory of the moment of the earthquake intrudes my thoughts" (19.8%), #5 "When I think about the earthquake, I am overcome with waves of emotion"(18.4%). The responses to each of the 15 items were added and the total score was computed for each mother. The mean and median of the mother's Impact of Event Scale scores were 10.9 and 7, respectively within a theoretical range of 0 to 45. Like children's stress symptom scores, however, exploratory data analysis of the distribution outliers suggested about 8.2 % (36 out of 438 mothers) showed extremely high stress symptom scores. Those 36 mothers who scored more than 33 points were suspected for seriously suffering from PTSD. It should be noted that three times as many PTSD symptom scale outliers (i.e., extremely high scores suggesting immediate clinical intervention) were found among mothers (8.2 %) as compared with children (2.5%). This suggests that more mothers rather than their children needed clinical help 10 months after the earthquake.

Structural equation modeling: The causative relations were investigated among the post-earthquake hardship, mother's stress symptoms, children's stress symptoms, mothers preferred coping resources, and family cohesion and adaptability variables. Structural equation modeling of these variables identified that 1) hardships caused by the earthquake damage were responsible for an increase in mothers' psychological stress, which in turn caused a higher stress reaction among their children; 2) mothers reacted to the hardships by empowering internal and external resources of their own but those empowered resources were used solely to alleviate their children's stress level; 3) the family adaptability balance alleviated the children's stress.; 4) family cohesion increases as a response to the hardship stressor, but it does function as a coping resource to alleviate stress levels of child or mother.


As previously mentioned, the author partcipated in outreach debriefing groupwork programs for mothers of preschool children. From these experiences, the author felt that preschoolers were well protected by their mothers both physically and mentally, however, a greater number of mothers were still suffering from post-traumatic stress 10 months after the earthquake. The Kobe Mother-Child Stress-Coping Research substantiated these practice findings. Lessons learned from this research for future disaster stress management include the importance of treating families as a whole rather than focusing only on children, the role of the family's ability to change its power structure as a coping resource, and the need to encourage mothers to use their internal resources to alleviate their own stress rather than directing all of their resources to benefit their children.

As previously mentioned, the author participated in the out-reach group work program in order to help mothers of preschool children cope with their post-traumatic stress. The out-reach program started with a public lecture on stress and coping, followed by the participation of the audience in the "six piece story" technique. The six pieces of the story consist of 1) a main character, 2) a mission, 3) a thing or person that will assist in the accomplishing of the mission, 4) an obstacle, 5) a means to overcome the obstacle, and 6) an ending. This is a projective technique to identify individual coping styles based on Lahad and Cohen's BASIC-Ph (Belief, Affect, Social, Imagination, Cognition, and Physical) model. Lahad and Cohen (1989) suggest that the most favored coping resources were projected to each individual's story writing. The story writing exercise and scoring of coping styles helped to reveal the strength and positive resources that each mother mobilized in order to alleviate the impact of disaster-related stressors. Debriefing group work followed the six piece story exercise. During the debriefing session, information concerning the coping styles was explained as evidence of each participant's strength and healthy functioning. Thus, this debriefing group work not only ventilated the participants' affect and provided psychoeducation regarding stress responses, but also facilitated the positive reappraisal of the situation and the self. Most importantly, mothers were taught that it is OK to express stress and that it is important that they themselves receive stress care. They did a good job and now is the time for them to give something nice for themselves. This was the basic message conveyed to the mothers who attended the debriefing sessions.

Several government surveys on gender role stereotypes, Kobe residents are known to be very conservative about their gender role expectations. Full time house wife until childs become teenagers are considered to be still the dominant life style for most of urban dwellers in Kobe. High mother/nurturer expectation seems to have played a major role to facilitate mothers to use their own coping resources in order to alleviate not their own but rather their children's stress. The media also promoted the traditional mother role of the protector of their children during the post-earthquake disaster periods. The companies also took it for granted that their male employees to come back to workplace as soon as possible. Those societal demands seem to have functioned to force mothers to overpoure all nesessary coping resources for the protection of their children.

Responsible care givers need to be well assisted by formal and informal helpers. This was the key implications for future disaster management policy makers.

*) Acknowledgment

The Kobe Mother-Child Stress-Coping Research, which was supported by a grant from the Asahi Glass Foundation for the year of 1996 to the author, was conducted with adminstrative assistance from the Kobe Child Guidance Clinic. Their help  was most appreciated. Mr. Keiji Noguchi, a graduate student at the time of the research, was involved with the questionnaire design, administration, data entry, and data analyses towards the completion of his masters thesis under the supervision of the current author.


Gal, R. (1995). Trip Report: Training Workshop for Japanese Mental-Health Professionals in Kobe, Japan - June, 1995. Carmel Institute for Social Studies, Zikhron Yaakov, Israel.

Horowitz, M., Wilner, N., & Alvarez, W.(1979). Impact of Event Scale: A Measure of Subjective Stress. Psychosomatic Medicine, 41(3), 209-218.

Lahad, M., & Cohen, A.(1989). BASIC-Ph: The Study of Coping Resouces. In M. Lahad (Ed.) Community Stress Prevention, Vol. 2. Kiryat Shmona, Israel: Center For Emergency.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. NY: Springer

Lifton, R. J. (1967). Death in life: Survivors of Hiroshima. NY: Random House.

Lifton, R. J., & Mitchell, G. (1995). Hiroshima in America: Fifty years of denial. NY: G. P. Putnam's Sons.

Noguchi, K., Tsubokura, Y., Taniguchi, H., & Tatsuki, S. (1997). A Family centered eco-system model of traumatic stress and coping I: Structural equation modeling on stress and coping, Kwansei Gakuin University Department of Sociology Studies, 76, 101-115

Tatsuki, S. (1993). The construct validity of the Circumplex Model of marital and family systems(VII): Confirmatory factor analytic inter- and intra-cultural cross validation, Kwansei Gakuin Sociology Department Studies, 66, 143-165.

Tsubokura, Y., Noguchi, K., Taniguchi, H., & Tatsuki, S. (1997). A Family centered eco-system model of traumatic stress and copingII: Qualitative analysis of free-romatted responses, Kwansei Gakuin University Department of Sociology Studies, 76, 117-131