A Case Formulation Approach to Cognitive Processing Therapy

Posttraumatic stress disorder (PTSD) is common and complicated. Recent estimates suggest that over 610,000 US Veterans treated by the Veterans Health Administration (VHA) suffer from posttraumatic stress disorder (PTSD), a disorder that can be chronic and debilitating. The heterogeneity of the 20 symptoms of PTSD; comorbidity with disorders such as depression, panic, and substance use; high rates of concurrent and lingering effects of physical injury, and suicidality all contribute to complex clinical presentations and can exact a significant toll on functioning, quality of life, and well-being decades after exposure to the trauma. The complex and enduring challenges inherent in PTSD and their effect on patients' functioning pose significant hurdles for patients and clinicians.

Impairments in psychosocial functioning are an important, but less well-attended, facet of PTSD. While significant impairment in functioning is clearly a requirement for the diagnosis of PTSD as indicated by Criterion G of the diagnostic criteria for PTSD, resolution of functional impairment is not considered to be a primary therapeutic target in evidence-based PTSD treatment protocols. Improvements in domains of functioning and, more broadly, quality of life and well-being, are most typically considered secondary outcomes in RCTs, if they are reported at all. This seeming lack of attention to impairments in functioning stands in stark contrast to patients' reports of the meaningfulness of these impairments in their lives. In fact, it is often precisely these types of impairments that drive patients suffering from PTSD to seek treatment, arguably more so than the 20 core symptoms of the disorder.

Researchers and providers alike recognize the importance of well-being and seek to maximize functional recovery. It has theoretically been difficult to directly target impairment in functioning (PTSD Criterion G) in manualized therapies, perhaps because "functional impairment" is quite variable across patients in breadth and scope. Including explicit and manualized instruction on "treating functional impairment" is impossible and is likely the reason that no single PTSD psychotherapy exists that is specifically designed to "treat" functional recovery. Trauma-focused EBPs instead are specifically designed to directly target the core symptoms of PTSD. Functional impairment is targeted indirectly through relief in the core symptoms. For example, improvement in occupational functioning may be achieved through decreases in avoidance, mood or anger symptoms that might be interfering with job performance. Gains in functioning (e.g. improving marital relations or workplace functioning) and large effects in well-being and quality of life have historically been difficult to operationalize and are often considered more longterm therapeutic goals, perhaps beyond the scope of brief therapies. However, given the importance of functioning to patients with PTSD, intentionally and thoughtfully building on the success of the skills acquired in evidence-based psychotherapies (EBPs) and expanding those skills to specifically target functional outcomes warrants further exploration.

Cognitive Processing Therapy (CPT) is effective at reducing PTSD symptoms and is widely used. CPT is one of the therapies with the most accumulated empirical support to date. Recent meta-analyses indicate that CPT has the largest effect size of existing evidence-based treatments for PTSD in soldiers and Veterans and across trauma populations (mean ES = 1.69). The effectiveness of CPT extends across patient populations suffering from a range of comorbid conditions including TBI, chronic child abuse histories, comorbid psychiatric and substance use disorders, and ongoing peritraumatic situations. CPT has been designated a first-line treatment in a number of clinical guidelines, including the VHA / Department of Defense (DoD) PTSD Clinical Practice Guidelines. Since 2007, VHA has dedicated substantial resources and effort to the historic, large-scale dissemination of CPT via the Mental Health Dissemination Initiative. To date, over 4,150 VHA providers have been trained in CPT and at least 20,774 Veterans have begun CPT in the last year. Yet the impact of CPT on both PTSD symptoms and psychosocial functioning can be improved.

Veterans with PTSD present to treatment with challenges in psychosocial functioning and other clinical complexities; the failure of manualized treatments for PTSD to provide guidance on how to address these concerns is negatively impacting outcomes. Nearly all (87%) Veterans with PTSD presenting to VHA primary care have at least one comorbid psychiatric condition and, on average, Veterans with PTSD had 2.95 comorbid mental health diagnoses and over 50% reported suicidal ideation. Across trauma populations, PTSD is associated with severe impairments in social and occupational functioning, exerting a more deleterious effect than most mental health diagnoses. Clinical presentations in Veterans suffering from PTSD are further complicated by a host of psychosocial stressors, impairments in major domains of functioning, and comorbid physical conditions. Veterans are more likely to be homeless, be under- or unemployed, and have poor physical health status as compared to non-Veterans. Prior work suggests that unaddressed difficulties in these domains contributes to suboptimal outcomes and premature dropout from CPT among Veterans. The investigators posit that frontline treatments for PTSD must be expanded to directly and intentionally target functional impairments and improve holistic outcomes.

While fidelity to the CPT protocol is essential, skillful divergences can be clinically wise. Administering EBPs in a standardized manner is critical to ensuring that patients receive an accurate and adequate dose of the intervention. Yet, as outlined above, strict protocol adherence (e.g., failing to assess and address psychosocial stressors and other clinical complexities) may result in premature abandonment of trauma-focused therapy and/or suboptimal outcomes. Modifications to the CPT protocol can enhance therapy outcomes. The next step in this program of research is to leverage the success of CPT in treating core symptoms of PTSD and expand the protocol to directly target functional outcomes. Thoughtfully and intentionally enhancing the latitude of the CPT protocol to target clinical complexities that pose a risk to holistic outcomes personalizes the delivery of care to best meet the individual patient's needs.

There is little to no guidance in trauma-focused therapy protocols to address clinical complexities and modify therapy accordingly. The perception of a forced choice between 1.) adherence to a manualized therapy and 2.) personalizing the protocol to best meet the patient's needs, leads to higher levels of dropout and poorer outcomes. Overly rigid adherence to the treatment protocol is a common contributor to CPT dropout. In the face of treatment challenges, one-half of patients received a modified trauma-focused therapy. However, patients who completed modified trauma-focused treatment had significantly worse outcomes than Veterans who completed standard trauma-focused therapy. The rationale for and content of the modifications was unknown. The other half of complex cases were switched to a different type of therapy and PTSD was no longer the target of treatment. When faced with clinical complexities, there is no guidance for therapists to expand and enhance the protocol to accommodate patient needs, leaving therapists to initiate strategies lacking in evidence or abandon treating PTSD altogether to target the clinical complexity.

Integrating a case formulation (CF) approach into the existing CPT protocol will enable providers to simultaneously address Veterans' clinical complexities that interfere with CPT delivery and enhance functional outcomes, while maintaining fidelity to effective CPT principles. CF is a patient-centered, collaborative process between providers and patients. CF allows providers to tailor cognitive-behavioral treatments to specific patients' unique clinical complexities within clear parameters of what justifies divergence from the standard protocol. Integrating a case formulation approach into CPT provides therapists with the tools to effectively navigate the fine line between maintaining the trauma-focus necessary to treat PTSD and attending to the clinical complexities and functional impairments that contribute to suboptimal doses of therapy and/or poorer outcomes. This study seeks to improve the clinical effectiveness of CPT by integrating a CF approach that will enable providers to directly target impairments in functioning and flexibly address clinical complexities that arise during the delivery of the CPT protocol.

Methods To accomplish the study aims, this randomized controlled trial will utilize a national sample of VHA CPT providers. This study will increase the external validity of the project by including: (a) broad provider and Veteran inclusion criteria, (b) using field-based providers, (c) employing CPT delivered in accordance with VHA's CPT Dissemination Initiative as the comparison condition, (d) avoiding the use of study-team strategies to improve Veterans' engagement, and (e) utilizing ITT analyses. A national sample of CPT providers (n = 50) will be randomized to deliver either CPT or CF-CPT. Each provider will deliver the treatment to 4 consecutive Veterans presenting for CPT treatment in his/her clinic who consent to study participation (n=200 Veterans. Key features of this pragmatic design include substituting centralized study processes and reliance on the existing network of CPT providers instead of relying on recruitment sites and local site investigators. Methods will include (1) centralized recruitment, enrollment, \& data collection, (2) enrollment \& consent of study providers, and (3) consent of Veteran participants by phone with a waiver of documentation of informed consent.

Description of Treatments. Cognitive Processing Therapy: CPT is a brief therapy for PTSD predominantly based on cognitive theory. Traditionally delivered over 12 one-hour sessions weekly or twice weekly, CPT is now variable length depending on patient's recovery from PTSD. CPT is delivered in three phases: education, processing, and challenging and focuses on challenging beliefs and assumptions related to the trauma, oneself, and the world. Changing dysfunctional beliefs alters negative emotions emanating from those beliefs.

Case Formulation + CPT (CF+CPT): The CF approach builds on the success of CPT in reducing core PTSD symptoms and alters the protocol in two important ways: expanding the protocol to intentionally and systematically address impairment in functioning, and enhancing the providers' latitude to navigate challenges to optimal therapy outcomes (COTOS) that threaten patient engagement. CPT protocol expansion will occur for all patients, intentionally increasing the focus of the intervention to explicitly assess, track and intervene with impairments in functioning (Criterion G of PTSD). To accomplish this, CF-CPT begins with a formal CF assessment session; elements of CF are then integrated throughout CPT. CF modifications to the original CPT protocol occur in each session by intentionally attending to cognitions that are impeding the patient's functional recovery. The goal of the case formulation approach to CPT is to provide guidance to the therapist in increasing the patient's active involvement in CPT and personalizing the approach to best meet the patient's needs while ensuring that the trauma-focused work is prioritized.

The second modification to the original CPT protocol includes enhancing the provider's latitude to diverge from the protocol when clinically wise. The clinical complexities commonly observed during care delivery and identified as risk factors in empirical research present challenges to optimal therapy outcomes (COTOs) during recovery from PTSD. COTOs constitute any and all potential patient-level challenges or clinical issues that might arise during therapy and present obstacles to standardized administration of a manualized therapy protocol such as CPT. COTOs are central to functional outcomes and, as such, present risks to holistic recovery insofar as they 1.) increase the likelihood of premature drop-out resulting in inadequate doses of treatment, and 2.) are often considered outside the scope of trauma-focused treatment and are subsequently left unattended. The universe of possible COTOs cannot be quantified in any single protocol as they are idiosyncratic to the patient, but they can be categorized into five larger domains for ease of assessment: 1) emotional dysregulation, 2) significant avoidance behaviors 3) ambivalence and beliefs about therapy, 4) comorbid conditions, and 5) significant psychosocial and environmental stressors and crises.

Intentional assessment of these domains and identification of COTOs provides the opportunity to monitor and address the COTO before the protocol is irreparably ruptured or abandoned. As COTOs present themselves during the protocol-driven delivery of CPT, therapists face multiple decision points. In weighing competing patient needs, therapists struggle with balancing fidelity to the treatment manual and simultaneously ensuring that the patient's multifaceted needs are met. As with other single disorder protocols, CPT offers little guidance for navigating existing and emerging COTOs during treatment. By default, the global recommendation for clinicians is to stay on CPT protocol and maintain the focus on PTSD or administer a different treatment. However, too much rigidity negatively impacts patient engagement and modifications to the protocol have significantly enhanced CPT effectiveness and are now considered the standard of care. CF-CPT thus provides guidance around the identification, monitoring and management of COTOs, and, importantly, the expedient return to the CPT protocol with continued attention to the patient's idiosyncratic COTOs.

Aim 1 Methodology: Compare the effectiveness of CF-CPT as compared to CPT in improving Veterans' functioning and quality of life.

Provider Inclusion / Exclusion Criteria. Providers will include up to 56 licensed VHA clinicians who have been trained in CPT, are listed on the CPT National Provider Roster, and who self-report administering CPT to at least 7 Veterans during the prior twelve months.

Provider Recruitment. Provider recruitment and participation will occur nationally over 4 waves. From the pool of interested providers, a random, stratified sampling frame of providers who meet inclusion criteria will be developed. Stratification will include oversampling for providers who engage female patients to have an adequate representation of both genders and will be stratified by VISN to increase the likelihood that the study sample will be representative of the racial/ethnicity diversity of Veterans with PTSD nationwide. Using phased provider recruitment will enable enrollment of providers and patients at a manageable rate, adequately staff the study, and begin participation as soon after provider recruitment and training as possible to reduce the chances of therapist attrition Provider Randomization. Using a 1:1 allocation ratio, a computer-generated randomization sequence will randomize eligible providers to each study condition. Randomization will be blocked within recruitment wave. Providers will only be delivering the therapy in the condition to which they have been randomized. This will minimize therapist drift and possible contamination effects.

Veteran Inclusion / Exclusion Criteria and Recruitment. Each of the 50 providers will treat 4 CPT patients (total sample = 200) with either CF-CPT (n = 100) or CPT (n = 100), as dictated by provider randomization. To avoid selection bias in patient population, all consecutive Veterans with whom the provider is intending to begin a course of individually-delivered CPT will be offered study participation until 4 Veterans are enrolled. Thus, in order to be invited to participate, Veterans will have been identified as PTSD positive per clinic screening processes. Once Veterans express interest in participation to the CPT provider, they will complete a diagnostic interview with study staff blinded to condition in order to confirm a full diagnosis of PTSD. Study exclusion criteria include active suicidal ideation with intent, homicidality, current mania, psychosis, or serious drug or alcohol abuse that requires immediate medical attention (e.g. inpatient care). Patients should not be participating in another trauma-focused therapy at the time of enrollment but can continue any psychiatric medications (dose must be stable for one month prior to enrollment). Medication changes will not be prohibited for Veterans in either treatment arm following baseline assessment and randomization.

Study Procedures and Data Collection. Patient participants will be assessed prior to treatment (pretreatment), mid-treatment (after session 6 or 6 weeks post-randomization for treatment dropouts), 2 weeks post- treatment or 14 weeks after study initiation for treatment dropouts (posttreatment), and 12 weeks after the posttreatment assessment (follow-up). Over the course of the study, patient data will be collected by a) interviews delivered via phone, b) mailed paper and pencil survey, and c) retrieval from EMR. Veterans will complete the phone interview and survey at each assessment interval (pre, mid, post, and follow-up) and will be paid for each completed interview and for each completed packet of questionnaires.

Phone Interviews. All diagnostic clinical interviews will be conducted by an independent evaluator (IE), blinded to study condition, by telephone. IEs participate in four stages of training: relevant readings, expert-led classroom instruction, mock interviews with national experts, and co-rating exercises with previously taped assessments. After training, all IEs will engage in weekly calibration exercises to ensure maintenance of high quality standards and prevent drift in scoring.

Self-report surveys. Self-report measures will be collected at all assessment intervals online via Qualtrics or by paper and pencil. Depending on participant preference.

Measures. In addition to outcomes specified in the study specific aims, the investigators will measure constructs and conditions associated with each COTO domain. While not able to formally assess the universe of possible COTOs for each participant, the investigators selected standardized measures for those believed are most likely to arise. For those randomized to CF-CPT, the investigators will also utilize the Daily Monitoring Diary completed throughout the course of treatment as an idiosyncratic measure of all relevant COTOs for the individual patient.

Treatment Delivery. All treatment will be delivered in in-person individual therapy sessions by rostered CPT providers and will be audio-recorded.

Delivery of CPT. Providers randomized to the CPT intervention will be asked to continue to deliver the treatment in accordance with the CPT manual.

Delivery of CF-CPT. The CPT protocols will be modified with the CF-CPT approach in three ways. First CF-CPT will begin with the case formulation assessment. Feedback from the case formulation session will be incorporated into the traditional CPT session 1, allowing time to continue developing the individualized monitoring tool (daily diary). Second, patients in CF-CPT will monitor identified COTOs throughout therapy via the daily diary. Third, patients in CF-CPT will also be instructed to apply CPT specific skills to COTOs and functioning related cognitions during practice assignments, etc. Some portion of the CF-CPT group's identified COTOs will increase during the course of CPT. For only those patients for whom it is necessary, the content of the therapy may be altered accordingly.

Clinical Case Consultation and Fidelity Assessments. To ensure that each condition is delivered in the way in which it was intended, case consultation will be provided to all providers on a weekly basis. Adherence and competence will be determined by independent and expert raters who are not otherwise involved in the project.

Power Analysis. Power calculations based on mixed model tests for two means in a two-level hierarchical design with level-2 randomization to intervention group confirmed that the proposed sample (200 Veterans) adequately powers the study to detect meaningful between group differences. With an estimated sample size of 200, the study would have 83% power to detect an 8-point difference (d = 0.47) on the total IPF score.

Data Analysis: Analyses will follow ITT methodology. The investigators will test Aim 1 hypotheses using generalized linear mixed models. Continuous patient Aim 1A \& 1B outcome measures (all but treatment completion), within the assumed and appropriate fitted distributions, will be modeled with treatment and training wave as fixed effects and a random provider effect (clustering by provider). The time (pretreatment, mid-treatment posttreatment, three-month follow-up) by treatment interaction term will provide the test of the investigators' primary hypothesis that CF-CPT will be superior as compared to CPT.

For the dichotomous measure of treatment completion (Aim 2), logistic mixed models with a random provider effect will be used. Given that the amount of divergence from the CPT protocol will vary across patients, the investigators will augment the primary analysis with secondary analyses comparing subsamples of the CF-CPT group (e.g., those with and without divergences) to the CPT group. This secondary analysis will require adjustments for possible covariate imbalance, as the investigators will lose the balance induced by randomization. The investigators will use propensity analysis and will estimate average treatment effect on the treated (ATT) and the corresponding sensitivity analysis to gain more insight on the magnitude of the effect. Post hoc power will also be calculated to assess the generalizability of these estimates.

Finally, the investigators will examine the COTO reduction score as a predictor of improvements in functioning and symptoms - i.e., testing whether more improvement in COTO scores is associated with greater improvement in study outcomes (Aim 3). The investigators propose adapting an idiosyncratic method of assessment via locally constructed diaries used in previous studies to track PTSD symptoms to monitor idiosyncratic COTOs and provide data to evaluate the effectiveness of the CF intervention on the individual challenges in patients' lives by calculating a Composite Primary COTO Reduction (CPCR) score. This score is an index of overall change in COTO level and can be conceptualized as a percentage of improvement. The CPCR score also provides a means for describing clinically significant improvement in symptomatology, functioning, etc.