Funding stopped early

Suicide is a national crisis, especially among older Veterans for whom evidence-based suicide prevention efforts are lacking. This proposal responds to the national priority to develop and improve interventions for suicide prevention, with a focus on at-risk older Veterans. The randomized control trial will compare VA usual care, which is suicide safety planning, with brief Problem Solving Therapy and suicide safety planning. This study uses Problem Solving Therapy because it has support from our pilot data and from secondary data analysis from other studies for reducing late life suicide risk. This treatment also has support for alleviating two key risk factors for late life suicide risk, functional disability and executive dysfunction, and thus this study will examine how older Veterans with varying levels of functional disability and executive functioning respond to treatment to inform future targeted implementation. In accordance with national priorities, existing infrastructure in Problem Solving Training could be expanded to support more rapid VA-wide implementation.

The proposed randomized control trial will examine the efficacy of a brief behavioral treatment for reducing suicidal ideation and for increasing reasons for living in older Veterans at risk for suicide. Two common issues in older adults, functional disability and executive dysfunction, increase suicide risk and will be examined as moderators and mediators of treatment outcomes. The proposed study will examine Problem Solving Therapy (PST) using a protocol that teaches skills to address emotional and practical barriers to effective problem solving. The VA usual care practice of collaborative safety planning will be the control condition, herein referred to as enhanced usual care (EUC). For participant safety, Veterans enrolled in PST will also receive EUC. The study design is random assignment to the treatment, PST plus EUC, or the control condition, EUC only. Both the treatment and the control are evidence-based interventions delivered in six sessions. The study will occur at VA Palo Alto (primary site) and VA Syracuse/VA Canandaigua (secondary site).

The assessments and treatments will be conducted entirely by phone. Longitudinal assessment of suicidal ideation and reasons for living will occur at eleven timepoints: baseline, after each of the six weekly treatment sessions, posttreatment (7 weeks), at 1-, 3-, and 6-month follow-up. The recruitment goal is 150 Veterans over a five-year period with 75 participants randomized to each treatment. Veterans must be 60 years or older, report suicide ideation, and meet diagnostic criteria for a depressive disorder, anxiety disorder, and/or posttraumatic stress disorder to be eligible. Veterans will be excluded and referred for other services if they have psychotic symptoms, bipolar disorder, or severe OCD; certain alcohol/substance use disorders; severe or unstable medical conditions; recent head injury (past year); terminal illness; positive cognitive screen indicative of possible dementia; or if they are unable to participate in the study or follow-up sessions. Recruitment will occur through multiple strategies at VA, Vet centers, and in the community to ensure the feasibility of recruiting a total of 75 Veterans at each site. To meet this recruitment goal, the primary and secondary site will each enroll 2 to 3 Veterans per month.

Fidelity ratings of audiotaped treatment sessions will be made by independent raters on 20% of sessions. Mixed effects modeling will be used to estimate the effect of treatment versus the control for the outcomes of suicidal ideation and reasons for living. Models will covary for demographics, medical and psychiatric comorbidity, and psychotropic use. For the primary aim, models will determine whether PST plus EUC is associated with a larger treatment effect compared with EUC only for suicidal ideation and reasons for living. Secondary aims will determine if baseline functional disability and executive dysfunction moderate treatment outcomes, and whether a change in these two variables from baseline to posttreatment are significantly associated with (mediate) change in suicidal ideation and reasons for living. The tertiary aims will use qualitative methods to assess Veteran satisfaction with the treatment and control including the types of problems with which treatment helped them, and suggestions for improving it for future Veterans. The current VA/DoD Clinical Practice Guidelines (CPGs) for suicide do not list treatment options specifically for older Veterans despite the majority of VHA patients being 55 years or older. Thus, evidence-based outcomes from the proposed work have potential to inform the CPG recommendations for using PST with Safety Planning and/or Safety Planning alone with older Veterans. Dissemination of the treatment could be supported by expanding existing VA training infrastructure in problem solving based interventions.

Inclusion Criteria:

• >= 55 years old

• SI (past month) on the C-SSRS

  • No SI or imminent risk (will be excluded)
  • Eligibility now includes non-specific active suicidal thoughts to active SI with some planning (on C-SSRS item 3: "Have you been thinking about how you might do this") and/or some intent to act (on C-SSRS item 4: "Have you had these thoughts and had some intention of acting on them?") provided that those endorsing item 4 have an established relationship with a psychiatrist aware of this intent level (who is not also performing psychotherapy)
  • Those endorsing C-SSRS item 5: "Do you intend to carry out this plan?" would be considered at imminent risk and excluded

• Presence of DSM-5 depressive disorder or anxiety disorder, specified or unspecified, or PTSD or subthreshold PTSD.

• No diagnosis of dementia; evidence that cognitive impairment is not indicative of possible dementia based on a negative cognitive telephone screen (<10 errors on the Blessed cognitive screen)

• No diagnosis on MINI or medical record for psychotic symptoms or disorders, bipolar disorder, or severe OCD (mild or moderate OCD will be eligible)

• No history of head injury past 12 months

• No severe or unstable chronic medical illness or other severe or unstable respiratory, cardiovascular, neurologic, hepatic, hematopoietic, gastrointestinal or metabolic dysfunction

• AUDIT Total score <15 for men or <13 for women (no current alcohol use disorder)

• No substance use disorder of any type for illicit substances, no moderate or severe substance use disorder for cannabis/marijuana - on the MINI.

• No prominent homicidal ideation

• English language proficiency to engage in treatment

• Sensory abilities sufficiently intact to engage in assessment and treatment

• Not currently enrolled in individual psychotherapy for a mental health issue. Participants who are receiving mental health care are eligible if their current care includes treatment from a psychiatrist, supportive therapy, peer counseling or support, or participating in psychoeducational groups.

• No current prescription for anti-psychotics if prescribed for a psychotic disorder

  • Anti-psychotics for depression without psychosis are eligible
  • Psychotropic medications (e.g., SSRIs) are also eligible

Exclusion Criteria:

• < 55 years old
• No SI (past month) or imminent risk on the C-SSRS will be excluded. Those endorsing C-SSRS item 5: "Do you intend to carry out this plan?" would be considered at imminent risk and excluded
• No DSM-5 depressive disorder or anxiety disorder, or PTSD or subthreshold PTSD
• Diagnosis of dementia; evidence that cognitive impairment is indicative of possible dementia based on a positive cognitive telephone screen (>=10 errors on the Blessed cognitive screen)
• Diagnosis on MINI or medical record for psychotic symptoms or disorders, bipolar disorder, or severe OCD (mild or moderate OCD will be eligible)
• History of head injury past 12 months
• Severe or unstable chronic medical illness or other severe or unstable respiratory, cardiovascular, neurologic, hepatic, hematopoietic, gastrointestinal or metabolic dysfunction
• AUDIT Total score >14 for men or >12 for women (no current alcohol use disorder)
• Substance use disorder of any type for illicit substances, moderate or severe substance use disorder for cannabis/marijuana - on the MINI.
• Prominent homicidal ideation
• Lacking English language proficiency to engage in treatment
• Sensory abilities not sufficiently intact to engage in assessment and treatment
• Currently enrolled in individual psychotherapy for a mental health issue. Participants who are receiving mental health care are eligible if their current care includes treatment from a psychiatrist, supportive therapy, peer counseling or support, or participating in psychoeducational groups.
• Current prescription for anti-psychotics if prescribed for a psychotic disorder, and not prescribed for depression without psychosis

A de-identified, anonymized dataset will be created and shared. The data set will be shared upon written request and will be shared pursuant to a Data Use Agreement (DUA) once our planned manuscripts have been accepted to peer-reviewed journals. At that time, we will send researchers who have completed a DUA a electronic copy of the database in a password protected database file. Final datasets will be maintained locally until VA enterprise-level resources become available for long term storage and usage.

Publications from this project will be made available through the National library of Medicine PubMed Central website 1 year of the date of publication. The final de-identified datasets underlying these publications will be available by written request to the PD/PI. Data will be available by written request until March 31, 2046. This date will be extended if it is determined that availability of the data to researchers would be in the interest of the scientific literature on suicide prevention, or if additional research questions can be developed the data to address important gaps in late life suicide prevention assessment and treatment.

Protection of privacy, confidentiality, and proprietary data is ensured using the following mechanisms: 1) a de-identified dataset, 2) signed data user agreement that will prohibit researchers requesting the data from identifying or attempting to re-identify individuals in the dataset ,3) "90+" age category, but not an exact age for adults aged 90 years or older to prevent identification, 4) sharing only of quantitative data, but not qualitative data to further prevent identification.

The participant's assignment to treatment will not be shared with the outcomes assessor. A "randomizer" will provide treatment assignment to the participant after the assessment.