Mobile Menu - OpenMobile Menu - Closed

Michael Shepherd, M.D.

Michael Shepherd, M.D., Office of Inspector General, Physician, Office of Healthcare Inspections, U.S. Department of Veterans Affairs

Mr. Chairman and Members of the Committee, thank you for the opportunity to testify today on suicide prevention and the Office of Inspector General (OIG) report, Implementing VHA’s Mental Health Strategic Plan Initiatives for Suicide Prevention


In 2004, suicide ranked as the 11th leading cause of death with a rate of 11.1 per 100,000 in the general U.S. population and the 3rd leading cause of death within the 15-24 age range.  It is estimated that each suicide intimately affects the lives of at least six other people.

In any particular suicide, individual and collective proclivities tend to combine.  Consequently, the attempt to make sense of the multiple potential contributions from identifiable psychiatric disorder(s), co-morbid medical illness and functional impairments, specific personal events, and sociocultural factors has been the work of and an ongoing challenge to mental health professionals, sociologists, and epidemiologists.  This effort has increased knowledge about suicidal patients and provided information for utilization in their treatment.  However, there has been little reduction in overall rates through the years, indicating there is more to learn. 

Suicidologists have struggled with standardization issues for many years.  While it has long been held that the pursuit of valid and reliable suicide statistics is important to public health policy, establishing the validity and reliability of suicide rates has been a notable source of concern.  In the U.S. it is widely assumed by mental health professionals that the actual suicide rate is higher than officially reported rates.  Establishing the validity and reliability of suicide rates is complicated by stigma.  Other sources of variability include limitations of death certificates, variability in the training of those tasked with certifying cause of death, use of differing guidelines for suicide determination, and the presence of equivocal causes such as single car accidents and drug overdoses. 

The 2001 Surgeon General’s National Strategy for Suicide Prevention identifies steps in a public health model for suicide prevention.  Collecting data on rates of suicide and suicidal behavior is typically referred to as medical surveillance.  Data may include information on how suicide rates vary by time, geography, age or special populations.  In addition, data collection may include information on characteristics of individuals who suicide, circumstances surrounding suicide events, the presence and absence of possible precipitants, and the adequacy or accessibility of supportive factors and health services. 

For example, the National Violent Death Reporting System is a Centers for Disease Control and Prevention (CDC) effort to develop a nationwide, state-based monitoring system for violent deaths. State and local agencies use this system to input data from medical examiners, coroners, death certificates, police reports, toxicology studies, and other sources.  At present 17 states are designated to participate in the system.  Veteran status is one of several uniform data elements recorded for input into the system.  The data is pooled with the hope that it can ultimately be used to answer fundamental questions about suicide and to aid participant states in the design and implementation of tailored suicide prevention and intervention efforts. 

Suicide is not a single illness with one true cause; it is a final common outcome with multiple potential antecedents, precipitants, and underlying causes.  Interventions that may be more effective for one set of patients may differ from those of greatest benefit for a different set of patients.  Comprehensive suicide prevention programs, those employing a portfolio of intervention elements, and particularly those that incorporate a range of services and providers, are thought to have a greater likelihood of reducing suicide rates.  Selecting which interventions to implement includes consideration of the needs and characteristics of the target population, ways to integrate interventions into existing programs, efforts to strengthen collaboration, and an analysis weighing the resource requirements versus the potential effectiveness of individual interventions.

Veterans Health Administration’s Mental Health Strategic Plan

In 2003, a VA mental health work group was asked to review the President’s New Freedom Commission on Mental Health’s 2002 report, to determine the relevance to veteran mental health programs of the Commission’s goals and recommendations, and to develop an action plan tailored to the special needs of the enrolled veteran population.  A 5-year action plan with more than 200 initiatives was ultimately developed and finalized in November 2004.   Among the action items were a number specifically aimed at the prevention of suicide.  In addition, endorsement and implementation of the goals from the Surgeon General’s 2001 National Strategy for Suicide Prevention, and recommendations from the Institute of Medicine’s 2002 report Reducing Suicide: A National Imperative, were incorporated into the VA Mental Health Strategic Plan (MHSP).

OIG Report on VHA’s Implementation of Suicide Prevention Initiatives

In response to a request from this committee, the OIG undertook an assessment of VHA progress in implementing initiatives for suicide prevention from the MHSP.  In our May 2007 report, individual MHSP initiatives for suicide prevention were categorized and consolidated into the following domains: 

  • Crisis Availability and Outreach.
  • Screening and Referral.
  • Tracking and Assessment of Veterans at Risk.
  • Emerging Best Practice Interventions and Research.
  • Development of an Electronic Suicide Prevention Database.
  • Education.

We recommended that: 

  • VHA make arrangements for 24-hour crisis and mental health care availability, either in person, or via a crisis line, and that at each facility an on-call mental health specialist should be available to crisis staff either in person or by phone.
  • All non-clinical staff who interact with veterans receive mandatory training about responding to crisis situations involving at-risk veterans inclusive of suicide protocols for first contact personnel. 
  • Health care providers receive mandatory education about suicide risks and ways to address these risks. 
  • The requirement of sustained sobriety should not be a barrier to treatment in specialized mental health programs for returning combat veterans.
  • VHA should facilitate bi-directional information exchange between VA and DoD for patients with mental illness coming into VHA health care and/or leaving VHA health care for re-deployment to active duty status.
  • VHA should establish a centralized mechanism to review ongoing suicide prevention strategies, to select among available emerging best practices for screening, assessment, and treatment, and to facilitate system-wide implementation, in order to ensure a single VHA standard.

Crisis Availability

Although we found that most facilities reported availability of 24-hour mental health care either through the emergency room, a walk-in clinic, or a crisis hotline, this initiative had not achieved system-wide implementation and a coordinated toll free hotline was not in place at the time of our report.  On July 25, 2007, the Department of Veterans Affairs subsequently began operation of a 24-hour national suicide prevention hotline for veterans.   The hotline has reportedly received greater than 9000 calls.  Callers include veterans who previously would have called a non-VA suicide hotline, veterans who would not have utilized a non-VA hotline, family members and friends of veterans, and other distressed non-veterans.  Several of the veteran calls have resulted in 911 emergency rescues and admission to VA hospitals.  Hotline personnel facilitate referral of distressed non-veterans to a non-VA suicide prevention hotline through a partnership with the Substance Abuse and Mental Health Services Administration.

I recently visited the hotline, located at the Veterans Integrated Service Networks (VISN) 2 – Center of Excellence at Canandaigua, New York, on less than 24 hours notice.  During my visit with hotline staff, the phone lines were in use throughout the duration.  I observed a call from a young veteran who told the hotline clinician that she planned to take the bottle of pills that she had next to her.  After assessment and a lengthy discussion with the caller, the hotline line clinician arranged for an emergency rescue.  I also observed a call from a discouraged Vietnam era veteran who had recently become homeless and was calling from his car in which he was living.  Hotline staff arranged for him to be met by the suicide prevention coordinator at the local VA facility.

Suicide Prevention Coordinators

The VA Office of Mental Health Services has been in the process of implementing suicide prevention coordinators at all VA medical centers. At present, dedicated staff are reportedly in place at approximately 85 percent of facilities and “acting” suicide prevention coordinators are in place at remaining sites.  Hotline clinical staff told me that after requesting a consult for a caller at a VA facility, they contact the facility suicide prevention coordinator electronically and/or by phone.  If they do not hear back within 24 hours, they contact the coordinator again.  Within 48 hours of the call to hotline, an update on the patient’s disposition is to be reported by the suicide prevention coordinator to hotline staff.  At 2 weeks post call, hotline staff contact the suicide prevention coordinator for an update as to whether the caller has remained engaged in follow-up in the VA system. 

Education and Training of VA Personnel

In terms of initiatives for education on suicide prevention, at the time of our May report, we found that only 50 to 60 percent of facilities provided programs to train first contact non-clinical personnel about crisis situations involving veterans at-risk for suicide.  Only one-fifth of these programs included mandatory presentation of suicide response protocols.  Almost all facilities provide education to health providers on suicide risks, ways to address these risks and best practices for suicide prevention.  However, at only a small percentage of facilities were these programs mandatory.  Since that time, the VISN 2 Canandaigua Center of Excellence has developed a CD and guide for training VA non-clinical personnel and a second CD and guidebook for community based training.  The training, titled Operation S.A.V.E. (Signs of suicidal thinking; Ask questions; Validate the veteran’s experience; Encourage treatment and Expedite referral) will reportedly be carried out by the facility suicide prevention coordinators.  A copy of the CDs and guide were provided to me on my recent visit.  The VISN 2 Center of Excellence leadership report plans to subsequently develop a guide and CD for VA clinicians. 

Treatment for Co-Morbid Mental Health and Substance Use Disorders

In terms of eliminating sustained sobriety as a barrier to treatment in specialized mental health programs for returning combat veterans, on November 23, 2007, the Deputy Under Secretary for Health for Operations and Management issued a memorandum to Network Directors that states that “VHA facilities and providers must never take the position that a patient is untreatable because substance use or dependence precludes treating mental health conditions while mental illness makes it impossible to treat abuse or dependence.  Instead, services must be designed and available to provide care for veterans with substance use disorders and mental health conditions, alone or together, regardless of acuity or chronicity.”

Facilitation of Emerging Best Practice Implementation

The OIG report recommended that VHA facilitate establishment of a centralized mechanism to select among emerging best practices for suicide prevention, the VISN 2 Center of Excellence has subsequently been organized into a clinical core, an education and training core, a VACO initiatives core, and a research core.  The clinical core group is responsible for the organized development of pilot and demonstration projects.  The initiative core is responsible for implementation of VA Central Office suicide prevention initiatives.  The research core is focused on performing program evaluation, health services research, and intervention effectiveness research in order to expedite the dissemination of promising approaches throughout VA.   

Bi-Directional Exchange of Health Information

Bi-directional information exchange between VA and DoD which includes patients with mental illness coming into VHA health care and/or leaving VHA health care for re-deployment is an ongoing issue that has been discussed at other hearings.

VHA Development of a Veteran Suicide Database

At the time of our inspection, researchers at the VHA Serious Mental Illness Treatment Research and Evaluation Center (SMITREC) in a joint effort with researchers at the University of Michigan School of Public Health in Ann Arbor, Michigan, had been developing a methodology by which to create a data base of veterans who had utilized VHA care in an index year and then stopped utilizing VHA care in subsequent years.  This database would then be matched with data from the CDC National Death Index (NDI), to determine which of these veterans were deceased.  This data would then be matched with an enhanced version of the National Death Index to determine which veterans no longer accessing VHA care had died from suicide.  In early October, SMITREC researchers reported that they have subsequently calculated suicide rates for 2001 and 2002 among veterans who obtain care in VHA.  In recent weeks, they reported working on data received from the NDI for calculation of rates from 2003-2005. 

At the time of the May OIG report, a template of data elements pertaining to suicides and suicide attempts had been piloted in Rocky Mountain Network (VISN 19) facilities.  In the past few months, VHA has reportedly been expanding use of the template to VHA facilities nationwide. Clinical providers at VHA facilities nationwide have been asked to input data regarding attempts or completed suicides by their patients using a template which contains prompts for data elements including age, gender, diagnosis, date of attempt, method used, outcome, date last seen at VA prior to attempt, among others. The facility suicide prevention coordinator is responsible for receiving and collating data inputted into the template by clinical providers and submitting a spreadsheet to the Center of Excellence at the Canandaigua VAMC on the 10th of each month. October was the first month for which data was submitted to the Canandaigua Center of Excellence.  Most but not all VHA facilities submitted data and the extent of provider compliance with filling out the templates is presently unclear.  

Since October 2003, the Department of Defense (DOD) Defense Manpower Data Center has sent the VA Environmental Epidemiology Service a periodically updated personnel roster of troops who participated in Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF), and who had separated from active duty and become eligible for VA benefits.  This data however does not include recently discharged or retired veterans who were not deployed in support of OEF/OIF or veterans who have served in other eras.

The OIG LC Database

During the past year, colleagues at the OIG Office of Healthcare Inspections have diligently pursued creation of a data base to quantitatively characterize the care transition process from DOD to VHA.  A September 2007, OIG Informational Report entitled Quantitative Assessment of Care Transition:  The Population-Based LC Database, describes the creation of an analytical database derived from more than 30 data files acquired from VA and DOD that incorporates details about all service members discharged from July 1, 2005 to September 30, 2006.  The database includes veterans who were deployed, those who were not deployed, members of the Reserves and National Guard, those who have accessed care in VHA and those who have not.  The paper discusses the methodology used to create the database, data confidentiality issues, its limitations, and analytic potential for research and other applications. This unique database may provide background for understanding and interpreting ongoing and planned studies pertaining to select medical conditions, causes of mortality, and/or health care access. 


Suicide is an unequivocally tragic and often incomprehensible event.  Preventing suicide is a complex, multifaceted challenge to which there is not one best practice but several promising but not proven approaches and methods.  Since 2004, progress had been made toward implementation of the MHSP initiatives for suicide prevention.  Progress has continued with greater integration and at an accelerated pace since the time of the OIG report in May and the enactment of the Joshua Omvig Suicide Prevention Act.  The full array of suicide prevention initiatives has not yet attained system-wide implementation.  It is therefore incumbent upon VA to continue moving forward toward full deployment of suicide prevention strategies for our Nation’s veterans.  

Mr. Chairman, thank you again for this opportunity to testify on this important issue.  I would be pleased to answer any questions that you or other members of the Committee may have.