Measurement-based care in psychology practice with veterans: Why we should and how we can

Veterans comprise approximately 23.4 million of the U.S. population. Many of these individuals face myriad mental health issues including, but not limited to, post traumatic stress disorder, major depression, traumatic brain injury, and substance use disorders. Moreover, suicidality rates among veterans has increased steadily since 2001. In response to these concerns, the national VA Healthcare System has sought to focus on the use of evidence-based treatments in the hopes of providing the most effective treatments. However, given the limited (and often shrinking) resources available to VA behavioral health providers, there is not only pressure to demonstrate effectiveness on an aggregate basis, but also on a case-to-case basis to ensure the resources being consumed are being used effectively and efficiently. To this end, VA mental health providers are tasked with examining outcome on a session-to-session basis to ensure treatment is working. The purpose of this paper, which is a review of the current state of the literature, is to examine the rationale and methods for the use of measurement-based care, and to provide recommendations for how this approach may be applied in practice. Journal of Addiction and Recovery


Introduction
According to the Substance Abuse and Mental Health Services Administration (SAMHSA) [1], there are approximately 23.4 million veterans in the United States. A large percentage of this population is at risk for serious mental health problems; in fact, results of a recent study found that one in three veterans is diagnosed with a mental health disorder [2]. In particular, veterans are at increased risk for developing posttraumatic stress disorder (PTSD), major depression, and/or traumatic brain injury (TBI), among other conditions compared to the general population. Mental health issues such as these may also be linked with substance use disorders and/or homelessness.Perhaps most alarming, the U.S. Department of Veterans Affairs (2016) reports a steady increase in suicide rates among veterans since 2001.
In an effort to address these issues and provide the best possible care, the VA Healthcare System has implemented a variety of behavioral health treatment options including the use of evidence-based treatments and varying therapeutic modalities (e.g., individual, group, couple, family), and flexible service delivery models (telehealth, inpatient, etc.). Depending on the setting and presenting concerns, veterans may receive psychiatric care, psychotherapy, or a combination of the two. For example, a veteran battling depression may receive antidepressant medications in conjunction with ongoing cognitive behavioral therapy (CBT) for depression (i.e., an evidence-based treatment); the antidepressant medication may aid in reducing the individual's biological symptoms, while CBT for depression may provide assistance in altering the veteran'smaladaptive patterns of thinking into healthier, more positive ones [3]. Unfortunately, limited federal resources has resulted in renewed emphasis on accountability for VA behavioral healthcare managers and providers. More specifically, the current era of cost containment has increased the pressure on VA therapists to demonstrate that what they are doing is working (i.e., effective) on a regular basis (i.e., session-to-session). They are expected to determine overall effectiveness and satisfaction at the end of treatment using aggregate data, and engage in clinical decision-making on a weekly basis as a way to improve efficacy and increase retention. By examining progress on a weekly basis over time, therapists can identify those cases that are not benefitting from therapy and modify the current approach as needed or refer to a more appropriate level of care. The purpose of this paper is to (1) describe the rationale for using measurement-based care (MBC) to inform clinical decision-making, (2) offer possible ways for applying MBC in practice (including measures), and (3) provide recommendations for future directions in this emerging area.

Rationale for use of Measurement-Based Care
The concept of measurement-based care (MBC) may be best conceptualized at a meta level of abstraction that can be applied with any existing therapy model; thus, it is not a model per se but an overarching guideline or principle for service delivery that can be used in making decisions about clinical resource allocation [4]. Regular monitoring and measurement of outcome requires a paradigm shift on the part of the practitioner from exclusively privileging the therapist's point of view of the treatment process to a willingness to partner with clients and explore their perceptions of progress in a clinically meaningful way. This effort to enlist client's view of therapy creates a culture of feedback in which clients play an active role in shaping the direction of treatment based on their progress (or lack thereof).
In keeping with the notion that the therapeutic alliance in and of itself is a strong predictor of success and is fundamental to the change process [9,10], participation in MBC can potentially enhance rapport building and the working relationship through a more deliberate attempt to join with clients [10]. MBC necessitates collaborative progress-monitoring, whereby the client is encouraged to make joint or shared decisions about treatment interventions with thetherapist [10]. Whether the client is moving in the direction of his/her desired goals or not, he or she acquires ownership over treatment progress and comes to understand his/her own authority and power in creating change. Consequently, by delivering objective and tangible data to clients that depict how they are doing from session-to-session, clients can identify what is working versus what is not [10], and guide their therapists on a path that may be more aligned with their stated goals; therefore, increasing the likelihood of outcome attainment, the personalization of treatment, and the improvement of quality of care. In this way, to the client's benefit, MBC is "effective", "empowering", and permits "changes you can see" [10].
It is important to note that the results of numerous studies reveal that a small percentage of clients (approximately 10%) tend to account for a substantial amount of resources (therapist and clinic) consumed [11,12]. These clients are typically characterized by little to no progress and therefore serve to clog up caseloads with little return on the investment; this results in frustration for both the therapist and client. These clients also impact the therapist's ability to take on new clients and inadvertently decrease the therapist's availability. As a result, the therapist's availability for and access to new patients may be limited, causing new patients to end up on waitlists. Moreover, results of studies examining self-assessment bias reveal average drop-out rates of 25%, 1 out of 10 clients accounts for 60-70% of expenditures, and that mental health providers frequently fail to identifying those cases that are either failing or not making progress [13][14][15]. Thus, examining outcome on a regular basis offers a mechanism for providers to identify those clients not benefitting from therapy and begin the discussion about modifying treatment or the possibility of referral or transfer to a more appropriate level of care and ultimately allowing for increased access to services among veterans seeking treatment.

Methods for Measuring Outcome in VA
Given that no one model, method, or clinician is sufficient for treating all problems [16], soliciting feedback is critical to effectively address the diverse problems among veterans seeking treatment. Simply stated, despite our best efforts to identify a single model that could be applied with diverse clients with a wide array of presenting concerns, treatment is not a onesize-fits all approach. Moreover, therapists consistently fail to recognize those clients that do not seem to be getting better or in fact, getting worse, thereby increasing the chance of a failed clinical experience [17]. That said, it is incumbent on clinicians to make sure the chosen theoretical approach makes sense to clients and serves to increase their engagement in the therapy process. Without a mechanism for monitoring client progress and engagement, therapists run the risk of clients not engaging or complying with treatment, and thereby being conceptualized as resistant or non-compliant.
Although there may be some variation in measures used across VAs depending on the clinical setting, we minimally recommend the use of brief, psychometrically sound measures which can be easily administered and interpreted as a way to assess progress. Moreover, given the relationship between strength of alliance (based on clients' perceptions), we also suggest collecting data about client's perception of the alliance as a way to gauge engagement in the therapy process.
To that end, Miller,et al. [18] have developed a pan-theoretical approach to examining outcomes (i.e., Feedback Informed Treatment) focused on assessing the two main areas found to consistently be linked with positive outcomes: 1) early change and 2) engagement. Within VA, the use of measurement-based treatment may be best conceptualized as an attitude or treatment philosophy (rather than a distinct model) which may be applied to any model or technique.
At present, VA behavioral health programs (see: https:// www.myhealth.va.gov/mhv-portal-web/depression-screening) typically use the Patient Health Questionnaire (PHQ-9), an instrument used to assess depressive symptoms and their severity [19]. The PHQ-9 is valuable in that it can be used to inform clinical decisions about treatment, as it outlines recommended actions based on a patient's obtained score [19,20]. Additionally, the instrument includes a measure of functional impairment, where patients are asked how their emotional difficulties or problems are impacting their work and home life, as well as their relationships with others [21]. Upon the commencement of treatment, the PHQ-9 can be routinely used for session-tosession planning and treatment monitoring; accordingly, clinicians are able to assess patient outcomes and overall improvement.
The Short Post-Traumatic Stress Disorder Rating Interview (SPRINT), developed by Connor and Davidson (2001), is an eight-item self-report measure designed to assesses cardinal symptoms of PTSD; for example: intrusion, avoidance, arousal, somatic concerns, and social impairments. The measure possesses sound psychometric properties, and in addition to being used as a screening tool, can be used to determine illness severity and symptom change or global improvement over time [21][22][23]. Sample items include: "In the past week, how much have you been bothered by unwanted memories, nightmares, or reminders of the event?", "In the past week, how much have you been bothered by pain, aches, or tiredness?" [21][22][23]. In this way, SPRINT can be deemed an advantageous tool for progress monitoring and assessing treatment outcomes in patients who present with post-traumatic stress disorder.
Two additional brief outcome and process instruments which are simple to administer and easy to score are the Outcome Rating Scale [24] and Session Rating Scale (SRS; [25]). Each measure contains 5-items and provides a method for assessing progress (ORS) and engagement (SRS) in an ongoing manner. The particular outcome of interest may vary depending on the type of service offered but should provide an overall sense of progress and strength of alliance, and should be assessed on a session-to-session basis to maximize benefit. Although some have recommended using an assessment schedule (e.g., 1st 3rd, 5th session), we argue that the use of this type of intermittent assessment is not sensitive enough and runs the risk of clients dropping out of treatment before an alliance rupture or lack of progress is identified and can be remedied.

Recommendations
Given the increased demand for behavioral health services among veterans, coupled with fiscal constraints and limited resources, it is critical that behavioral health providers are providing effective services. To be clear, offering evidence-based treatments, although theoretically helpful, is not a panacea and does not guarantee successful or efficient treatment. Thus, the challenge for clinicians (as well as VA Healthcare System) is to deliver MBC as a mechanism for improving retention, delivering more efficient and effective services, and a tool for identifying those clients not benefiting from service sooner. The results of these efforts will allow veterans to receive more effective services and clear up therapist caseloads to allow for greater access to services and ultimately clients to be seen in a timelier manner. The following section provides some brief recommendations for implementing measurement-based care. This list is not meant to be exhaustive, but rather to serve as important considerations in planning the transition to outcomes-based care.

Selecting Measures
In choosing outcomes measures, it is important that the instrument is psychometrically sound, and easily administered in practice. The ease of administrating, comprehending, and interpreting the instrument is critical to effective implementation and continued use over time.

Creating a Culture of Measurement-Based Care
Routinely assessing progress from the client's perspective may be an unfamiliar concept to many therapists used to privileging their point-of-view over their clients. Thus, introducing client feedback into the therapy process may feel uncomfort-able for some. Yet, given what we know about psychotherapy outcome (i.e., importance of client's perception of the alliance -early in treatment as well as that the fact that change seems to occur early on in the process), we argue that assessing outcomes not only makes scientific and clinical sense, but should be required component of treatment to ensure quality veteran care. That said, it is incumbent on VA leadership (managers, supervisors, etc.) to create an environment conducive to client feedback and as such, be supportive of staff in their efforts to elicit and obtain feedback. Along these lines, it also critical that client feedback on progress should not be used by supervisors or program leadership in a punitive manner, but instead viewed as way to provide more efficient treatment to veterans and possibly identify therapist growth edges or areas for skill improvement. Therapists must be willing to share and learn from their failures, and supervisors must be willing to create an atmosphere conducive to this type of risk-taking.
Moreover, therapists must also convey a genuine interest in the client's feedback. They should attempt to create an environment in which clients feel safe in honestly sharing their thoughts about the therapy process (good and bad) and where they believe it will be helpful in guiding the process. Therapists can facilitate this process by explaining the purpose of the measures, how the results will be used, and use the client's feedback (i.e., data) in shaping the direction of treatment.

Resistance to Measurement-Based Treatment
Some clinicians may be reluctant to engage in measurementbased treatment, whether out of inertia or insecurities related to clinical skills and ability. In these cases, it is important that supervisors and leadership communicate clearly and transparently the purpose of the approach and collaboratively workthrough the perceived barriers. Moreover, many therapists may not be used to soliciting client input and based on training and theoretical beliefs, may not believe in the therapeutic value of partnering with clients in this manner; they may be reluctant to relinquish their 'expert' role. It is important that while supervisors and leadership must be supportive of therapists' efforts, they do so in terms of their expectations and follow-through in holding staff accountable as needed.

Using Measurement Data to Inform Clinical Practice
While collecting the data is an important step, this information must then be used to inform treatment-decision making. If data are gathered but not examined in the context of the current direction of therapy, then the process is merely an academic exercise and of no value to the current treatment episode. For example, a therapist collects the progress data prior to the session, but does not review the scores and thus, does not recognize that the client is reporting that current direction of treatment is not helping with the presenting issues -in fact, things have gotten significantly worse. Consequently, the therapist is unaware, and does not alter the course of treatment, but continues with more of the same which has not worked to this point. It is critical that client feedback be used in the treatment planning process.

Conclusions
Given the increasing demand among veterans for access to psychological and behavioral services, it is critical that VA provides efficient and effective treatments and begins monitoring progress on a consistent basis to ensure services provided are of therapeutic value. Thus, the process of obtaining outcome data to monitor progress should be viewed as a standard of care, held up by the VA in the treatment of its veterans. By routinely collecting data on outcomes, therapists can identify those clients not progressing, alter treatment plans accordingly, and provide more efficient services. Furthermore, the use of session-to-session outcome ratings allow the therapist to identify sooner those clients not benefitting from treatment (and clogging up caseloads), permitting earlier discussions about clinical options that better meet client needs (e.g., transfer to another provider, higher level of care), thereby improving access of services to veterans. The use of measurement-based care is important step in the VAs continued quest to provide the highest quality services to veterans and their families.