This study will determine long-term suicide related outcomes of a study of suicide-specific group therapy following psychiatric hospitalization. Results will be determined over a three-year follow-up through administrative records for the 134 participants plus qualitative interviews with a subset of 30 participants. The study will also examine group cohesion, working alliance, and number of therapy sessions as possible mechanisms of action.

Therapists have found group therapy to be effective for treating Veterans for depression, chronic traumatic stress, and various other mental ailments, but it has generally not been used to deal with suicide-related issues. The reason, according to Lora Johnson of the Robley Rex VA Medical Center in Louisville, Kentucky, was the concern that putting together a number of potentially suicidal patients in a group setting could lead to interactions that might actually increase the risk of suicide. Thus, the standard approach for dealing with Veterans at risk of suicide has been one-on-one counseling. This is effective for many of the Veterans who receive it but, because of limited resources, it is impossible to provide to all the Veterans who could benefit from it. Additionally, some Veterans who experience suicidal thoughts express an interest in addressing the issue in a group setting where they can help one another and learn strategies from their peers. 

For the past several years Johnson, with funding from the MSRC, has been exploring the use of group therapy for treating suicidal Veterans, including those who have attempted suicide. In a study completed in 2015, Johnson and colleagues investigated the value of formalizing the assessment process in an already established group therapy in use at one VA medical center. Although no main effects between groups in different assessment conditions were found, participants in both groups reported significant improvements in suicidal ideations and symptom distress over the course of the study. These findings suggested that, although changes in risk assessment had little impact on the suicide specific group therapy, the group itself was associated with a positive recovery trajectory for suicidal Veterans, which suggested that it would be valuable to take a closer look at the potential benefits of group therapy for those with suicidality. 

Now, in a new study, Johnson plans to examine how the Veterans in the original study have fared over the three years following the original study’s time period. The purpose of this three year follow-up study is to investigate the long-term safety of a group therapy designed specifically to help suicidal patients as well as to study how some of the mechanisms found to be correlated with clinical symptoms in the original study are associated with outcomes like suicide attempts and service utilization. Ultimately, the overall goal of this line of research is to help as many suicidal Veterans as possible deal with their issues and live happy and productive lives.

The group therapy approach developed by Johnson and her colleague Barbara Kaminer, also of the Louisville VA Medical Center, was grounded in the work of two psychologists, Thomas Joiner of Florida State University and David Jobes of the Catholic University of America. According to Joiner’s interpersonal theory of suicide, those individuals who attempt or complete suicide generally share three characteristics: thwarted belongingness, a sense of burdensomeness, and an acquired capacity for suicide. Jobes’ work focuses on treating suicidal ideation directly by identifying each individual’s personal drivers of suicidality. In his Collaborative Assessment and Management of Suicidality (CAMS) approach, the clinician and patient work collaboratively to address these drivers in therapy. This approach has proved effective in treating individual patients at risk of suicide. Considering Joiner’s theory and Jobes’ treatment framework together, Johnson says, suggests that a group setting focused specifically on suicidality and its drivers, offers a unique opportunity for collaboration not only between therapist and patient, but also between patient and peers. Furthermore, the setting will provide patients with a chance to build relationships with one another and increase their sense of belongingness. Finally, to the extent that the Veterans feel that they help others in the group, it offers opportunity to increase their sense of value to others and decrease feelings of burdensomeness. All of these factors may act to decrease the Veterans’ suicide risk, while also matching many Veterans’ preferences to address suicidal ideation with others who they feel can understand what they are going through.

The group therapy approach that Johnson and Kaminer developed is in line with the CAMS approach in the sense that the Veterans in a group are tackling suicide directly and looking at the factors that influence it—the “drivers”—instead of focusing on general diagnostic conditions, such as depression, which may or may not be driving the individual’s suicidality, but which have been the more typical historic focus of suicide treatments. But unlike CAMS, which was developed for use in one-on-one counseling, there is now an entire group of patients working with the clinician, collaborating in assessments and talking about treatment.

The goal of the previous MSRC-funded work was to add a formal CAMS-like assessment to this group therapy approach and observe its effects. The Veterans were recruited from patients at an in-patient mental unit. Of those who agreed to take part in group therapy after their discharge, half were assigned to groups that would use the formal assessment, while the others took part in the typical group therapy. For those Veterans in the formal assessment arm, the initial assessment would be performed by a therapist at the in-patient unit before the Veteran’s discharge. The assessment included such things as numerical ratings of various issues related to suicide status. Once the Veteran was working with a therapy group, an assessment was given each session, and the assessment was shared with the group, whose members discussed the responses.

Based on the effectiveness of performing such assessments in one-on-one CAMS therapy, Johnson had hypothesized that performing these formal assessments would increase the effectiveness of the group therapy. In particular, she thought that the Veterans who participated in group therapy with the formal assessments might be more likely to attend the group therapy sessions than the Veterans who were part of the no-formal-assessments arm.

That did not happen, Johnson says. There were no major differences in outcomes between the two groups of Veterans. In retrospect, Johnson says, she believes that may be because the formal assessments used in the study were just a fraction of the full CAMS assessments. “We’re thinking we didn’t do enough CAMS for it to show differences between the groups.”

But that’s okay, she says, because the study was a clear success in one way: It showed that, group therapy may be an effective way of treating suicidal Veterans. “Regardless of which group they were in, Veterans on average showed better outcomes,” she says. The Veterans in both groups showed a decrease in suicidal ideation (thinking about suicide), and there was no sign of “group contagion,” that is, of negative outcomes from putting a number of suicidal Veterans in a group together. That is the most important message of this earlier study.

Now Johnson has received additional funding from MSRC to follow the Veterans in her earlier study and examine the long-term effects from the original three months of group therapy. She and her colleagues, including co-principal investigator Stephen O’Connor of the University of Louisville, will examine the medical records of those Veterans to determine how they fared in the three years following the completion of the original study. They will examine such things as suicidal ideation, suicide attempts, and suicide deaths; the use of acute care services for psychiatric emergencies; and the use of outpatient mental health services for such things as suicide risk and medication management. 

They will also delve more deeply into the data to investigate how mechanisms that were found to be significant in the first study were associated with longer term outcomes. For example, in their original study they found that higher group session attendance was associated with lower thoughts of suicide after the first month of therapy. They also found that increased feelings of group cohesion were associated with a lower sense of thwarted belongingness. The current study will look at relationships between variables like these and long-term outcomes such as suicide attempts and service utilization. In this way Johnson and her colleagues hope to learn more about how these aspects of the group therapy may affect group safety and outcomes.

Finally, Johnson and her colleagues will interview thirty of the Veterans to learn what they think were the most valuable parts of the group therapy sessions. They will ask, for instance, about how important the Veterans feel that group cohesion was to their recovery and how valuable the formal assessments were.

Ultimately, Johnson says, the goals of the study are to show that group therapy is safe, to lay the groundwork for establishing its efficacy in treating suicidal Veterans, and to identify which particular aspects of that therapy are important to emphasize in order to help these Veterans live longer, better, and safer lives.