More than 25% of Veterans meet criteria for a current diagnosis of posttraumatic stress disorder (PTSD). To ensure that Veterans with PTSD receive appropriate services, the Veterans Health Administration (VHA) has mandated PTSD screening for all Veterans. Despite this, nearly half of Veterans who receive new positive PTSD screens in primary care receive no follow-up VA mental health care. To increase timely VA mental health care access – a priority area for both HSR&D and VHA – understanding access pathways (the series of options offered to, and choices made by, Veterans after a positive PTSD screen that may lead them to VA mental health care) is essential. Mapping these pathways will provide information about both which Veterans are not receiving this care and where in the process they are being lost. Further, by examining system-, provider-, and patient-level (SPP) factors that may predict who and where we are losing Veterans, we can begin to ascertain why we are losing them. Correctly classifying the first step of the access pathways – the immediate response to a positive PTSD screen in primary care – is arguably the most important, because if the initial classification is incorrect, any subsequent comparisons will be misleading. In this pilot, we focus on validating a method for identifying the initial access step for Veterans with new PTSD screens. Further, we will use the pilot to begin to develop a dataset of accessible SPP variables and use this to provide preliminary descriptions of each initial access step. Aim 1: Develop a feasible and efficient method for identifying the initial access step for Veterans with a new positive PTSD screen in primary care. Aim 1a: Use VA electronic medical record (EMR) data to identify the initial access steps for Veterans with a new positive PTSD screen in primary care during the most recent year available. We hypothesize that these initial access steps – defined as the immediate response to a positive PTSD screen in primary care – will include: (1) consult to a non-primary care VA mental health clinic; (2) consult to the VA Primary Care-Mental Health Integration (PC-MHI) program; (3) consult to VA inpatient mental health care; (4) prescription of psychotropic medication by the VA primary care provider; and (5) other. Aim 1b: Conduct a chart review to validate our method for classifying Veterans and to identify other initial access steps by: (1) ensuring the completeness of the first four initial access steps identified in Aim 1a; (2) determining if there are additional initial access steps that should be considered beyond those identified in Aim 1a; and (3) recording language for detecting initial access steps that cannot be identified using the EMR which can be used for Natural Language Processing (NLP) in a future IIR. We expect that the chart review will reveal at least three additional steps that would otherwise be classified in the “other” category: (a) referral to non-VA mental health care; (b) patient was offered, but refused, a referral to mental health care; (c) referral to a VA non-primary care, non-mental health care clinic. Aim 2: Develop a dataset of SPP variables and begin to describe each initial access step identified in Aim 1. Aim 2a: Mine VA data sources to build a dataset with available variables describing SPP factors in our conceptual framework. Aim 2b: Conduct descriptive analyses with available SPP variables to provide preliminary descriptions of each initial access step. This pilot will lay crucial groundwork to inform a follow-up IIR that will use the validated method to map the remainder of the access pathways. Further, the information gleaned about SPP variables will guide a mixed methods approach for understanding the factors associated with access to care following a positive PTSD screen. This will ultimately improve access to VA mental health care for all Veterans with PTSD.
External Links for this Project
Grant Number: I21HX002613-01A1
None at this time.
Mental, Cognitive and Behavioral Disorders
Treatment - Observational, TRL - Applied/Translational
None at this time.