The Youngest Victims: Bereaved Children After 9/11

Dr. Cynthia Pfeffer is professor of psychiatry and program director of the Childhood Bereavement Program at Weill Cornell Medical College.  Her clinical work, research, and teaching center on child development, anxiety, depression, suicidal behavior, and bereavement. At ACIA’s conference, she spoke about a major empirical study on children who lost a parent on 9/11.

I got swept into 9/11 practically the moment after the disaster happened.  I had started a Childhood Bereavement Program the year before, which developed primarily as a result of my research with children who lost a very close relative, a brother, sister or parent, to suicide. I had decided to develop a treatment program in a group format for children from the ages of 6 to about 14.  Of course whatever is going on with the child is affected by what is happening with the parents, and the parents of these children obviously were grieving also. This was a traumatic loss for all of them.  So our work focused on group intervention based on grief, loss, and trauma.  We wrote a manual and tested it with some families and then launched an empirical study.

The program was a 16 week intervention. There was a parent group and a child group that met simultaneously every week. We found that that approach did decrease children’s depression and anxiety and to some degree decreased the trauma factor too, but not entirely.  So that was really yet an area left as an open problem. We follow these children.  They seem to have done much better, those who were able to be in the group.

Around 2000 a donor donated money to us to develop a bereavement program. I decided to write a grant for work with children who were bereaved, not from a loss from suicide but children who lost a parent primarily as the result of illness. The medical center was willing to help. They would not have done it just for a research project but because we were helping the families.  The grant was due October 1, and I was ready to send it in and suddenly 9/11 occurred.  I was at the Westchester division of New York Presbyterian Hospital and one of my research assistants came running into my office. She had just driven from New York City and as she was driving she heard that an airplane hit one of the World Trade Center towers. We turned the TV on and I watched and I was in shock as I saw both towers burning and then suddenly I saw them both coming down and my first reaction was what about all the people.

I wondered if there was a way that I could help, since I had this bereavement program. I contacted some people and people started to contact me because of the program. The first step was in helping the companies that had lost many people. They set up sites at hotels and the families of people who worked for that particular company came there and were given help.  One of the first types of help was “debriefing.” It was discussed as a panacea to help these people. We have learned subsequently that that is not necessarily the best approach, but that was what was discussed.

9/11 was not an event that had much research. Most of the companies were not willing to help out a researcher but they were more than willing to help out people who were offering an intervention.  So I was well suited because I had the bereavement program.

I met one little girl whose parent died at Cantor Fitzgerald.  She was in this room, lot of people were milling around.  She was 5 years old.  I walked past and I said hello to her and I figured she lost someone. When I said hello she said to me, “my daddy died, my daddy just died.”  At the age of 5. She just blew me over, I almost did not know what to say to her because she was so small and this was such a horrendous situation and that was her first comment to me.  I knelt down and talked to her a little bit and then her mother came over. The mother was very busy trying to get assistance from the company, there were lots of lines. This child had an older sister who was about 7. Later on this family did come into our study program and so we followed that family for little bit. I changed my grant around, because NIH developed a rapid process for funding studies on 9/11. I was awarded a grant but the grant did not come through until four months after 9/11 occurred.  But we had started preparing and I had my program so I was able to start meeting with people.

And now ten years after 9/11 the question is what have we learned and what can we offer at this point and what else have we learned subsequently in many other major disasters, the tsunamis, what is happening in Japan, Katrina and many others that may have been on a smaller scale. There is a lot of discussion and literature on bereavement, lots of books by people who were either therapists or people who had lost someone, but not a lot of empirical research to know what really helps, what doesn’t help and what the outcomes are. The study  that I proposed to do focused on children who lost someone on 9/11.  It was more or less a naturalistic study, in which you follow whatever is happening to people without necessarily offering an intervention as part of the research. But we could offer help, we were willing to help them also. If they needed help and intervention, we felt it would be unethical not to help them. And we felt that having them come and talk and react was also a way of helping.

In the study we had children who lost their parents or a very close relative and were getting interventions that were offered in the community, and we had a control group of children who were not bereaved but were living in the same community as the children we were studying.  That’s an interesting control group because one would wonder how they were affected by 9/11.  But it’s an adequate control group because our aim was to find out what is the difference in outcomes between the children who lost someone and other children who were living in the community. So if they were all somewhat traumatized that’s still part of the naturalistic study.

It was to be a longitudinal study. We hoped to follow the families as long as we could or as long as our funding held out, but the actual plan was for a two-year followup, with a reassessment of the children and the surviving parents every six months. We took every child in the family, not just one child, and the main focus of our research was on stress responses. We were interested in two main features relating to stress: what type of psychological or psychiatric symptoms did the children have, and did they have physiological responses as well.

When we met with the children, we had lots of materials to play with and to draw, and we made it very comfortable.  We had a lot to ask but we did it in a very comfortable way for them.  I had several research assistants so if someone came with several children, one research assistant was with one child, the next was with another and another one was with the parent.  We also had volunteers who were helping us.  We established five sites, in New York City, Westchester County, and in New Jersey.

On the psychological aspect, we were looking for psychiatric symptoms, and whether these people had a disorder.  We used standard measures. For the children we used the K-SADS* and for the adults we used the SCID.**  We also used self reports on depression, anxiety, PTSD as well as an interview for PTSD.  We also got family histories. We had measures on social adjustment for the parents and a variety of other measures.  It was a massive study.  The biggest task was to get people to come because different companies had all kinds of mixed issues about the whole thing, but we did publicize our study and we got people.

I think the biggest thing that they found helpful was to be able to talk.  The mothers would come and they would talk about what happened, how their husbands died. We had some fathers who lost their wives also, and I think the fathers were more affected than the mothers.  Some of the fathers were from foreign countries.  I remember one man whose family is from India.  He and his wife worked with one of the companies but he wasn’t at the world trade center that day. The wife died.  He had a little boy and the father did not know what to do about raising his child because in his culture the women raise the children, not the men, and he was tremendously grieving and he was also depressed.  So we worked with him and the child, that was just one example.

There was another family where the father was from Israel. He was a technology person; he did not work at the World Trade Center but he was going to the World Trade Center that day for a conference. He died. There were two children and the mother came into the study with the children and she stayed with us for about a year and then she moved back to Israel with the children. The father of the man who died was a general in the Israeli army and this grandfather in Israel developed memorials in many of the cities of Israel to 9/11 victim families.  This is how intense the effects of 9/11 were.

What we found was that among the children as well as the parents, the majority developed symptoms of PTSD.  One can say how can that be, they weren’t there at the site, a lot of the parents rightly did not allow their children to watch TV.  But the children heard about things in school, they heard about 9/11, they knew their parent died. The majority of the bereaved children had PTSD. Of the children in our control group, there were only two who developed PTSD, and most of them did not have any major disorders.  There were some children in the bereaved group who also had developed symptoms of depression and some with major depression.

Since we had a followup time, we were interested in knowing also whether over time, did these symptoms disappear? What we found is that the PTSD seemed to stay high in terms of symptomatology.  For at least two years PTSD symptoms remained, over two years.  The depression diminished, gradually but we could definitely see there was a significant decrease in depression symptoms.

A majority of the parents had PTSD and many of them had depression and anxiety.  Anxiety was a big issue, more so than depression — anxiety disorder and other types of anxiety, generalized anxiety which some of the children had also.

On the physiological aspect, we worked with cortisol, which is one of nature’s stress hormones. We  got samples of salivary cortisol three times a day, in the morning, around noon, and in the evening. They collected the cortisol at home. We had the mothers make sure that the children were doing it; we had the parents do it also and we evaluated the parents as well as the children. We gave them tubes, everything was labeled. We gave instructions. A research assistant was assigned to call every single family every day.  “How is it going, remember to do it.”

We had them do it for three days in a row, and on the third night they took a pill called dexamethasone. Dexamethasone is a synthetic cortisol-like compound. It inhibits the adrenal gland’s secretions of cortisol because the brain thinks it has enough cortisol in the system and shuts off the secretion of cortisol in the adrenal gland. There is other research showing that when dexamethasone is working that way in the normal population it suppresses the level of cortisol in the body and that is what we were measuring, the level of cortisol in the saliva. If people had a suppression that would be a normal reaction; if they didn’t suppress, other research has shown that may be associated with disorders such as depression.

So on the fourth day, they again collected the saliva to measure cortisol after the dexamethasone. We were looking for the effect, if there was suppression or not.  That was the model for studying the physiological effects.

The results were very interesting. The children and the parents had higher baseline cortisol levels, that is in the first three days without the dexamethasone, than the control group.  That’s a typical stress response. With stress there’s more cortisol produced by the adrenal gland as a result of feedback from the brain.

There were also differences in the post-dexamethasone levels. The study group seemed to have suppression of cortisol that was greater than the suppression among the non-bereaved children in the control group. They not only suppressed; they over-suppressed. In the literature you’ll find that this over-suppression is often associated with PTSD. We had the same findings for parents too, the over-suppression and the higher rate of cortisol before dexamethasone.

Among the bereaved children there were subgroups. There were some children who had really no other pathology. They were grieving, they had lost a parent, but didn’t show any psychopathology.  When we analyzed our cortisol data, we found that the ones who suppressed most had only PTSD, the ones who suppressed next most were those who had PTSD and depression, and the ones who suppressed least in the bereaved group were the ones who were just grieving without psychopathology. And then there were the control children who were within normal range of cortisol levels and post-dexamethasone suppression.

So what does this mean?  I think the children who had PTSD looked very much like the literature shows, very much suppression of cortisol.  The children who had PTSD and depression had less suppression and that goes along with what’s in the literature too. And those who had no pathology but were just grieving had some suppression but not as much as the other two groups. They were primarily within the normal range.  The children who were grieving had baseline high levels of cortisol. All the bereaved children had high levels. They were reacting to stress.  But post-dexamethasone we saw there were different levels of suppression.  So that made me realize that grief is not just grief and trauma is not just trauma, one has to look at comorbidity and one has to look at not only psychiatric disorders and symptoms but underlying hidden issues like physiological concerns.

In all categories, the cortisol abnormalities persisted over the two years. Unfortunately we were not able to follow the children up until now or even five years ago, because there was no more funding left. And also, the children wanted to blend into the community.  Some of them didn’t want to come any more for research, they didn’t want to be different.  Children think being different is not good.

And I want to add something: I’ve focused today on children who lost a parent. That’s a small group, but the remaining population of children in the United States were affected also by 9/11. There are studies of that, some of them are epidemiological, some are smaller studies that point out that there was  an increase of anxiety in children.  You can live in Oregon and they were affected, you can live in Boston and they were affected, and we can’t lose sight of that. When you talk about educators and storytellers and others who work with the children in different capacities, we can’t forget those children too.

* Schedule for Affective Disorders and Schizophrenia for School-Aged Children, a diagnostic interview instrument often called Kiddie SADS

** Structured Clinical Interview for DSM-IV Axis I Disorders