Development of a Recovery Oriented Treatment for Post Acute Suicidal Episode (PASE) Veterans
Development of a Recovery Oriented Treatment for Post Acute Suicidal Episode (PASE) Veterans
Mental health care for Veterans with suicidal symptoms is of paramount import to the VA. Unfortunately, VA suicide reports show suicide rates increasing, suggesting a need for enhancing current VA suicide mental health care efforts. While several psychotherapeutic treatments exist for acute suicidality, there are few treatments designed to help Veterans following an acute suicidal episode (Post-Acute Suicidal Episode; PASE), particularly after acute risk declines but when they still have ongoing mental health needs and, at times, long-term suicidal symptoms. Available suicide treatments are not designed to promote the recovery and rehabilitation of PASE Veterans. This is a significant gap in comprehensive suicide-focused mental health care. One avenue to close this gap lies through the development of a recovery-focused psychotherapy for PASE Veterans. Developing recovery-oriented care, "a process of change through which individuals improve their health and wellness, live a self-directed life, and strive to reach their full potential" is a VA priority; however, available treatments for suicidal Veterans do not place a strong focus on recovery. Decades of research have shown the importance of increasing Veterans hopefulness about the future, developing a positive self-identity, promoting Veterans' sense of self-empowerment and improving relationships. Continuous Identity-Cognitive Therapy (CI-CT) is a promising new manualized suicide intervention focused on improving Veterans sense of their life story and personal future, with goals similar to recovery-oriented care.
The proposed study will assess and enhance the CI-CT treatment materials using Veteran feedback and acceptability and feasibility data. Then, with the guidance of scientific and Veteran consumer advisory boards, use these results to update the treatment. Findings will be used to make an updated adaptation of the treatment materials and to develop a research protocol for a pilot RCT of CI-CT for PASE Veterans. This study will develop and pilot test a well-specified, group-based intervention tailored to the unique needs of PASE Veterans. The results of the proposed study will provide data to 1) identify adaptations needed to optimize CI-CT for PASE Veterans: 2) identify possible benefits of CI-CT; 3) inform development of a pilot RCT of CI-CT for PASE Veterans.
Research Design and Methods
Phase 1
CI-CT has not, to date, been used for the PASE Veteran population. To optimize CI-CT's effectiveness for a PASE population and continue to align it as closely as possible with a recovery orientation, a dynamic adaptation approach will be used to run multiple treatment development trials of the group therapy to continue to improve CI-CT prior to running a pilot RCT. In this first phase the treatment materials will be used to run 3 one-arm treatment development trials of CI-CT (each with an n of 4-6 Veterans) to test and improve the therapy protocol. One goal of these trials is to identify CI-CT features or materials that appear functional in theory but do not execute as expected in a clinical setting. Following each treatment development trial, the feedback of the members of the group about the treatment as well as the assessment materials will be used to assess acceptability and feasibility. Upon completion of the trials, through consultation with the SAB and CAB, these results will be used to update treatment materials, instruments used for assessment, and delivery form. The experience of running these treatment development trials will also be used to develop a CI-CT adherence measure which will then be tested during the pilot RCT.
Veteran Participants
To have a final sample of 15 Veterans, 19 Veterans will be recruited (estimating 20% attrition) with a recent (within 1 year) acute suicidal episode, for participation in treatment development trials of CI-CT. The 20% estimate attrition rate is based on the initial CI-CT pilot data. As suggested by the aforementioned Post-Acute focus, treatment will be limited to PASE Veterans after the first week following their suicidal episode and without current acute suicidality (see assessment strategy below). This clinical decision was based on; (1) Veteran feedback from the initial pilot of CI-CT asking for screening to be used to limit the group to those ready to work on recovery, (2) empirical evidence suggesting that the first week after a suicidal episode is a period of distinctly high suicide risk with men 175, and women 583, times more likely than matched controls to die by suicide (N > 20,000 suicide attempters). During this 'acute suicidality' period, treatments exist for Veterans that are more focused on their immediate safety needs while CI-CT focuses on the recovery of Veterans following this 'acute' period. Furthermore, participants will be screened for current acute suicidal symptoms using two validated measures of suicidal symptoms the Suicide Behavior Questionnaire - Revised (SBQ-R) and the Beck Suicide Intent Scale (BSI). Due to the historical nature of some of these measures' items, as have other studies, the SBQ-R item 4 will be used, focusing on self-perception of likelihood of future suicide, with scores above 3 (i.e. at least 'likely') considered current acute suicidality. For the BSI, a cut off score >21 (i.e. 1SD from the mean BSI score) will be used among suicide attempt survivors (Mean=16, SD=5.7).
Recruitment, Screening and Enrollment Procedures
Investigators at the VISN 2 MIRECC maintain close collaborative relationships with mental health providers across the JJPVA. The standard operating procedure is to notify JJPVA mental health leadership when recruitment is being conducted for a study. Presentations will be made about ongoing studies at clinical team meetings and flyers will be sent to individual clinicians to facilitate recruitment. Potential participants identified through these avenues will be approached in-person at scheduled appointments. After initial contact, potential participants will subsequently be contacted by phone. Flyers will also be posted in waiting room areas. These recruitment procedures have yielded substantial success; investigators at this MIRECC have completed large clinical trials of Veterans with suicidal behavior with sample sizes of over 100 participants. Participants who meet eligibility will be scheduled for a consent and screening visit. Research staff will obtain informed consent using IRB approved procedures including assessment of the prospective participant's capacity to fully understand study procedures and related risks and benefits. Veterans who complete informed consent will be screened for eligibility criteria including a thorough medical record review, confirmation of clinical stability from a VA mental health provider, and confirmation of medical stability from a medical provider and completion of suicide symptoms assessments. If participants discontinue the study or are excluded before the first intervention session, a new participant will be recruited to maximize sample size. Detailed records recruitment process will be kept for reporting in a CONSORT chart.
Subject Characterization
Treatment
The trial will use the CI-CT intervention of the proposed project. CI-CT is planned to be a weekly, 90-minute, 12-session group treatment and to be run by two clinicians using the final version manual and workbook. The study will include 3 cycles of the CI-CT intervention with a projected average of 5 Veterans/group. The groups will be led by YS (co-PI) and a co-therapist (a TBD MIRECC staff interventionist who will be trained by YS to deliver CI-CT). YS will be a co-group facilitator to aid in the further tailoring and development of CI-CT following each of these developmental trials. The treatment is adjunctive, and Veterans will continue with their psychiatrist, case manager and services by the suicide prevention coordinator if needed.
CI-CT Training and Fidelity
CI-CT training will include one full-day training with the PI, weekly supervision with therapists, audiotaping and review of sessions for fidelity to the CI-CT manual. An adherence scale will be developed during this phase to assess core features of its structure, contents and treatment principles along with general clinical competence (e.g., building rapport, crisis management, etc.). The adherence scale will have three sections addressing different areas of adherence; 1) general CI-CT requirements, 2) session specific requirements, and 3) general group psychotherapy requirements. Clinicians will be required to maintain an average score of 4 on each session, rated on a 5-point Likert scale (where 0 = unacceptable and 5 = excellent) to demonstrate adequate adherence. If ratings fall below this criterion, supervision will be increased and adherence will be closely monitored until adequate adherence is regained.
Management of Disruptive Participants
Significant disruptive behavior by Veterans is rare but possible, therefore plan will be developed to manage or discontinue disruptive Veterans including a guide for clinicians to decide whether a Veteran should be immediately discontinued (e.g., for making threats) or if redirecting the Veteran is likely to help the Veteran participate in an appropriate manner.
Developmental Trial Assessments and Assessment Strategy
There will be four assessment time points: (TP-1) baseline, (TP-2) post-intervention, (TP-3) follow-up at 3 months post-intervention and (TP-4) 6 months post-intervention; to track Veterans until the end of the 1-year PASE period. Veterans will receive compensation for assessment completion but not for intervention attendance to minimize potential impact of compensation on treatment engagement.
Measurement of Feasibility and Acceptability of CI-CT
To examine Feasibility, the following will be tracked: 1) ease of implementation by recording the number of hours CI-CT clinicians spend in preparation, delivery of the intervention, and supervision. While the target is 60 hours/12 session cycle, beginning cycles are anticipated to take longer while the intervention is still being learned; 2) recruitment by measuring rates of successful referral to CI-CT and 3) attendance/retention, by tracking the number of sessions and specific sessions each Veteran attends. Based on rates in studies conducted by Dr. Sokol's primary mentor, the predetermined criteria for retention will be "adequate" if 70% of participants attend at least 9 of the 12 sessions, and "inadequate" otherwise. A similar strategy will be used for recruitment rates, with "adequate" feasibility if at least 65% of Veterans approached for participation in the study agree. To examine Acceptability, participants and CI-CT therapists will complete a brief survey after each session, and upon completion of the intervention. The Client Satisfaction Questionnaire (CSQ-8) will also be used to measure client satisfaction with the treatment (post-intervention). The CSQ-8 is an 8-item widely used measure in mental health studies containing items about quality and helpfulness, and provides a single score. Homework adherence will also be monitored.
Preliminary Efficacy Outcomes
Key outcomes of interest include: personal recovery and suicidal ideation. Secondary outcomes include recovery related constructs including quality of life, life satisfaction, hopelessness, disability, future self-continuity, and suicide related constructs including isolation and burdensomeness. Two primary outcomes were chosen, with recovery as a primary outcome, in addition to suicide ideation (of paramount import in all suicide treatments), in line with the significance given to personal recovery in this treatment.
Primary Outcome Measures
Personal recovery will be measured using the widely used Recovery Assessment Scale (RAS; Corrigan et al., 2004): The RAS is a 41-item self-report measure using a 5-point Likert scale. Sample items include: 'I have a desire to succeed.' and 'I have goals in life that I want to reach.' Factor analysis identified five factors: personal confidence and hope, willingness to ask for help, goal and success orientation, reliance on others, and not being dominated by symptoms. Narrative reviews of recovery measures have been highly favorable toward the RAS which has been described as the most acceptable and valid measure of personal recovery available. A review study found that across 28 studies, means ranged between 3.14 and 4.12, Cronbach's alphas from .76 and .97 and test-retest reliability from .65 to .88. The RAS has been widely used in VA and non-VA clinical trial studies with diverse populations, a range of mental illnesses and has demonstrated sensitivity to change over time. Suicidal ideation will be assessed using the Columbia Suicide Severity Rating Scale (CSSRS) a widely used semi-structured interview in VA clinical studies measuring suicide ideation and behavior has demonstrated sensitivity to change. The first subscale, (severity scale) is a 6-point ordinal scale, ranging from 1 (wish to be dead) to 5 (suicidal intent with plan). The second subscale, (intensity scale) is comprised of five items (i.e., frequency, duration, controllability, deterrents, reasons for ideation), rated on an ordinal scale (total scores ranging from 2 to 25). The behavior scale is a 5-point scale assessing interrupted, aborted, and actual suicide attempts; preparatory behavior for a suicide attempt; and non-suicidal self-injurious behavior.
Exploratory/Secondary Outcome Measures
Quality of life will be measured using the WHOQOL-BREF a 26-item instrument assessing four domains of quality of life: physical health, psychological health, social relationships and environment. Items are rated from 1 to 5. Calculated scores range from 0 (poorest quality of life) to 100 (best quality of life). The WHOQOL-BREF has high internal consistency, test-retest reliability, discriminant validity, and construct validity for general, mental health, and physical health populations. Life satisfaction will be measured using the Satisfaction With Life Scale (SWLS). Past studies have found the SWLS to be reliable, valid, and associated with measures of recovery. Scores on the scale can range from 5-35 with increasing scores indicating increased satisfaction with life. Hopelessness will be measured with the Beck Hopelessness Scale (BHS), a 20-item self-report measure that gives a single score with increased values indicating increasing hopelessness. Adequate reliability and concurrent validity data exist for this measure which has been shown to be predictive of eventual suicide in psychiatric inpatients. Disability will be measured using The World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0), a 36-item questionnaire assessing disability across physical and mental health disorders in both clinical and non-clinical populations. Items are scored on a 5-point likert scale ranging from none to extreme and are summed to create total and domain scores. Future self-continuity will be assessed using the Future Self-Continuity Questionnaire (FSCQ) a 10-item self-report measure assessing overall future self-continuity as well as three domains of future self-continuity, similarity to the future self, vividness of the future self, positive affect towards the future self. The FSCQ has high levels of internal reliability, test-retest scores, convergent and discriminant validity, and had been used with clinical and non-clinical populations. Scores range from 0-6 with increasing scores indicating increased future-self continuity. Social network will be assessed using the Lubben Social Network Scale-Revised (LSNS-R), a 12-item measure assessing two domains of social networks, family and friends. The LSNS-R scores range from 0-5 with increasing scores indicating larger social network. The LSNS-R has high internal consistency and has been used in clinical and non-clinical populations. Isolation and burdensomeness will be assessed using the Interpersonal Needs Questionnaire (INQ). INQ scores range from 1-7 with increasing scores indicating increased perceived burdensomeness and thwarted belongingness. The above measures will be given at each assessment point. A post-intervention questionnaire will also be included requesting a written description of their experience, what they felt they gained and would take away from the experience, and how the experience could be improved.
Data Analytic Plan
Data from the three 1-arm trials will be entered into an SPSS database at the JJPVA. Tenets of the data system include double data entry, range checks and exclusion of identifiers that can be traced to individuals. Computers are accessible only to authorized study personnel. Preliminary analyses, performed by the PI or co-PI, will include computing descriptive statistics and inspecting features and patterns of data to determine whether data transformations are needed. Feasibility of the following will be assessed: 1) ease of implementation, 2) recruitment, 3) attendance/retention, and 4) treatment fidelity. Collected data on ease of implementation will be compared to the predetermined feasibility criteria. Due to the small sample size (N=15) null-hypothesis testing is not feasible and assessment data will be examined using descriptive statistics rather than inferential statistics. Due to error in effect size estimation in small sample studies effect sizes will not be computed. The satisfaction questionnaires will be reviewed and summarized. Following the three planned developmental trials, feasibility and acceptance data will be reviewed and a written summary will be sent to the SAB and CAB to ascertain the need for changes in the CI-CT treatment materials or design. Any suggested changes will be implemented before moving to the pilot RCT.
CI-CT Adherence Measures Development
Using information gathered from the final CI-CT materials an objective scale will be developed to assess core features of its structure, contents and treatment principles along with general clinical competence (e.g., building rapport, crisis management, etc.) This measure will be piloted during phase 2 in a pilot RCT to assess treatment fidelity.
Phase 2
In this second phase, the treatment materials will be used to run a pilot RCT (N =30) to assess the acceptability and feasibility of using the current research methods, as a preliminary step for a future large-scale RCT.
Treatment Adaptation of CI-CT
Following all trials and revisions of treatment materials, the CAB and SAB will receive the updated version and be consulted as to whether the final version is ready for the pilot RCT. Any final suggestions will be implemented, and the updated version sent out again. Following agreement by the CAB and SAB that the treatment is ready for piloting, participant recruitment will begin. The pilot RCT treatment group will use the CI-CT intervention.
Comparison Condition - Health Education Active Control (AC) Group
The control condition will receive General Health Education (GHE) a structured manualized group health education intervention previously developed by VISN 2 MIRECC investigators as a control condition for group psychotherapy RCTs. It has 12 1.5-hour weekly group sessions focusing on health and wellness topics such as Sleep, Physical Activity, Impact of Stress, Relaxation Techniques, Substance Use, Nutrition, Managing Daily Activities, Medication Benefits and Side Effects. GHE was chosen for the AC because it aligns in many respects with CI-CT (e.g., group format, length of sessions, similar expectations) while diverging in specific topics and skills targeted allowing for control of common factors like attention without causing confounding due to overlap in concepts.
Participants, Recruitment, and Screening
Recruitment procedures are identical to those outlined for the development trials including the same inclusion and exclusion criteria but may be adjusted based on lessons learned from the development trials. To have a final sample of 30 Veterans combined across conditions, 38 Veterans will be recruited (estimating 20% attrition). There will be 3 cohorts, with 10 Veteran per cohort. This will be a continuous recruitment process, beginning a new group whenever sufficient participants are recruited. The PI does not anticipate any difficulty recruiting 38 subjects over 18 months, across JJPVA mental health service.
RCT Randomization and Treatment Procedure
Following enrollment and baseline assessment, Veterans will be randomized to one of two treatment conditions (1:1 allocation): CI-CT or AC, using Proc Plan in SAS by an independent research staff member from another project who will place treatment assignments in separate envelopes according to the randomization sequence. Individual subjects will be randomized using a computer generated permuted blocked randomization, with condition order permuted within blocks of varying size. Randomization will be stratified by SB history (none vs. >1 past suicide attempts) because multiple suicide attempt history is associated with risk of subsequent SB. The interventionists will not be blind to condition but the RA who completes the assessments conducted after randomization will be blind.
Subject Characterization
Process with be the same as in phase 1 with any revisions based on difficulties encountered while running the three 1-arm developmental trial groups.
Intervention Training and Fidelity
CI-CT training process for GHE interventionists (project RA hired for this purpose and MIRECC fellows) will be the same as in phase 1. For both CI-CT and GHE a structured supervision format will be utilized that includes the audio recording of all sessions, weekly supervision conducted the PI (licensed psychologist), and objective assessment of fidelity. For CI-CT the interventionist adherence rating scale will be used. For GHE an existing adherence rating scale will be used and will assess a randomly selected 20% of CI-CT and GHE session recordings for fidelity and competence. The interventionist will not know which sessions are to be selected. Fidelity problems will be addressed, if any arise, as soon as they are known and will note any fidelity challenges for future implementation. After study completion, overall percentages will be calculated to assess overall fidelity. Interventionist fidelity will be demonstrated if at least 85% of interventionist behaviors in the CI-CT and GHE sessions have been rated acceptable to excellent.
Assessments and Assessment Strategy This will be identical to phase 1 unless one or more assessments were shown to not be reasonably feasible during the developmental trials. If this occurs, the infeasible assessment(s) will be removed and replaced by another assessment following consultation with the SAB. Regardless of study arm, all participants will complete the same assessments. Veteran compensation procedure will follow that of the developmental trial.
Data Analytic Plan
Data management will be identical to phase 1. Summary statistics of demographic, clinical, and baseline outcome scale scores will be calculated for the total sample and compared between treatment groups to check the success of the randomization and for outliers. If any covariates are out-of-balance and are potential confounders of treatment effect estimation, they will be adjusted for between-group comparisons. In accordance with intent-to-treat principles, regardless of level of participation, the research team will attempt to assess all randomized participants at each time point and analyses will include all available data.
The purpose of this pilot RCT is to assess feasibility and acceptability of the research protocol. With this in mind, following consultation with Dr. Clayton Brown, biostatistician and mentor on this project, a sample size (N=30) was selected as sufficient for this task. The focus is assessing feasibility of: 1) ease of implementation, 2) recruitment, 3) attendance/retention, and 4) treatment fidelity. Collected data on each of these areas will be compared to the predetermined criteria. The small sample size of this feasibility pilot does not allow for statistically powered (at .80) null-hypothesis testing. Due to error in effect size estimation in small sample studies effect sizes will not be computed. Exploratory analyses will be run to examine change over time using mixed models to estimate the treatment effect (difference in mean change) on primary and secondary scale outcomes at post-treatment and over the two follow-up assessments with 95% confidence intervals. Only available scores are required as mixed models do not require imputation for missing data and are unbiased when outcome data are 'missing at random' (MAR). For each outcome analysis there will be a single categorical predictor (assessment time), group indicator variable, and a random participant effect. The mixed model approach is recommended in small sample longitudinal clinical trials, as it uses all available data and increases statistical power. Restricted maximum likelihood and, Heterogeneous first-order autoregressive covariance structure (which assumes larger covariance among points closer in time) will be specified for the covariance matrix. Confidence intervals for the primary outcomes will be adjusted to account for multiplicity.
Power Consideration
After considering time and resources available with this CDA-2, it appears that it will be feasible to randomize 19 individuals per condition. 20% of those randomized in each condition are expected to be missing assessments at time point 2, and 27% at time point 2 and 34% at time point 4. The sample size will not be sufficient to detect minimal clinically important differences (MCID's) or medium-sized differences (Cohen's d=.50) with 80% power as would be expected of a larger confirmatory RCT.
6) Protection of Human Subjects
Sources of Materials Sources of research material for the proposed research include the patient interviews, and questionnaires involved in the project. Clinical and demographic data are derived from interview ratings of patients specifically for research purposes. Past clinical records will, however, be utilized to supplement patient's reported past symptoms in determining their diagnoses and confirm history of suicidal behavior including attempt, suicide planning or preparatory action. All relevant data is collected in separate books for each patient and stored in locked areas to ensure confidentiality.
Protection from Risks
Data and Safety Monitoring Plan
A)The Principal Investigator will be responsible for study monitoring to ensure the safety of participants and the validity and integrity of the data. Data will be collected using standardized forms and will only be identified with the ID of the study participant. The codes that link the name of the participant and the study ID will be kept confidential. Study staff will have weekly contact with the PI for care review conferences and supervision, and the PI will be available in-person for additional monitoring as needed. Should a mental health crisis arise, several steps will be followed according to the safety management plan detailed in the sections 'protection against risk' above and 'risk management plan' below.
B)Medical Monitor: The study will also include a medical monitor at each site. The medical monitor will review all unanticipated problems involving risk to subjects or others, serious adverse events and all subject deaths associated with the protocol and provide an unbiased written report of the event. At a minimum, the medical monitor will comment on the outcomes of the event or problem and in case of a serious adverse event or death, will comment on the relationship to participation in the study. The medical monitor also will indicate whether he/she concurs with the details of the report provided by the principal investigator.
C)Risk Management Plan: The PI will be available to perform a suicide assessment at any point a rater suspects that a subject is at elevated risk for suicide as determined through clinical questioning or a research instrument assessing suicidal intent. The senior clinicians will ask questions regarding suicidal ideation and suicide plan and arrange for transfer to the psychiatric emergency room and/or inpatient stabilization at the JJPVA should imminent suicidal behavior manifest itself. All suicide attempts and deaths will be reported to the JJPVA.
Inclusion Criteria:
Exclusion Criteria:
Yosef.Sokol@va.gov(718) 584-9000