Abstract

Background: Approximately 6,000 Veterans per year die by suicide. Many Veterans with mental health conditions go untreated due to access barriers. Telehealth may improve access to mental health care but VA’s evidence synthesis program and multiple VA program offices have noted the need for stronger evidence on the effectiveness of telehealth for mental health care, particularly for suicide prevention. As telehealth requires access to digital technology, the health disparities impact of telehealth’s rapid adoption should also be evaluated. Causal inference methods offer enormous untapped opportunities for rigorous evaluations using VA’s rich data. Recent advances in these methods, including use of machine learning, offer new advantages but also require in-depth understanding and careful use. Use of mixed methods and an equity lens can enhance causal inference. Significance/Impact: This CDA will use enhanced causal inference methods to provide robust evidence to inform VA programs. Few VA researchers are using causal inference methods to examine VA’s virtual care or mental health initiatives. My quantitative background, prior research, and operational ties position me well to provide such robust evidence to VA and share lessons from using novel causal inference and mixed methods. Innovation: The CDA will evaluate three recent telehealth initiatives that have not been evaluated previously at a large scale with novel and rigorous methods. It will offer advances in methods for observational evaluations across VA through its: 1) use of mixed methods to transparently inform sophisticated causal inference methods, innovatively expanding both mixed methods and causal inference fields and 2) use of machine learning with an equity lens to improve causal inference, offering long-term benefits for using machine learning methods. Specific Aims: I will further develop my causal inference skills and augment them through the incorporation of qualitative methods, suicide prevention expertise, and a health equity lens. I will examine the suicide prevention and disparities impact of three recent telehealth initiatives aimed at increasing Veterans’ access to care: 1) VA’s Anywhere to Anywhere (A2A) policy which enabled virtual care provision across state lines; 2) VA’s expansion of video care; and 3) VA’s distribution of video-enabled tablets. I will pursue the following aims: Aim 1: Examine the impact of telehealth across state lines, via VA’s Anywhere to Anywhere (A2A) policy, on psychotherapy frequency, suicide outcomes (suicide-related ED visits, hospitalizations, deaths), and costs. Aim 2: Examine the impact of video vs. in-person or phone therapy on suicide outcomes and care costs. Aim 3: Examine the impact of VA tablets on the frequency of suicide risk screens and on suicide outcomes. Methodology: For each aim, I will use an exploratory sequential mixed methods design. In Aims 1a, 2a, and 3a, I will conduct interviews with Veterans and providers in rural and urban areas about factors affecting their use of out-of-state therapy, video therapy and VA tablets. Rapid qualitative analyses will inform causal inference study designs, variable selection, secondary analyses and interpretation of results in Aims 1b, 2b, and 3b. In Aims 1b, 2b, and 3b, I will leverage difference-in-difference (DiD) study designs which can account for selection bias due to unobserved reasons for patient/provider selection into VA telehealth initiatives. I will innovatively use causal decomposition methods with machine learning to highlight the extent to which health disparities are explained through observed variables vs. unexplained or due to potentially inequitable treatment of disadvantaged groups. Next Steps/Implementation: I will work with my operational partners in VA’s Office of Connected Care, Office of Rural Health and Office of Mental Health & Suicide Prevention to ensure that my research aligns with their implementation priorities. Aim 1 will inform how to leverage out-of-state therapy relationships to improve access in under-resourced areas. Aims 2 and 3 will inform OMHSP’s decisions on how to integrate VA’s video telehealth and tablets with VA’s routine suicide prevention efforts to improve suicide risk detection and reduce suicides. Aims 2 and 3 will also inform OCC’s and ORH’s decisions on whether and how to expand VA’s tablet program.

NIH Reporter Project Information: https://reporter.nih.gov/project-details/10864883


PUBLICATIONS:
None at this time.