Article Text
PDF
Abstract
Despite the resources dedicated to specialised mental healthcare for patients with post-traumatic stress disorder (PTSD) within the US Veterans Health Administration, evidence-based psychotherapies (EBPs) for PTSD have been underutilised, as evidenced by low EBP reach to patients. A research-operation collaboration evaluated whether implementation facilitation delivered by regional PTSD mentors as part of a national mentoring programme improved EBP reach compared with less-intensive quality improvement interventions. We used a non-equivalent comparison-group design that included all PTSD clinics with low EBP reach at baseline (n=51). Clinics were grouped into one of four quality improvement conditions according to self-selection by regional PTSD mentors: facilitation (n=6), learning collaborative (n=15), mentoring as usual in the regions that had facilitation-target clinics (n=15) and mentoring as usual in other regions (n=15). The primary outcome was EBP reach among therapy patients with PTSD at preintervention baseline and postintervention sustainment periods. We used the ratio of odds ratios (ROR) between the two time periods to evaluate the effectiveness of facilitation compared with the other conditions, adjusting for patient-level and clinic-level confounders. 26 126 veterans with PTSD received psychotherapy in one of 51 low-reach PTSD clinics during preintervention baseline and postintervention sustainment periods. The odds of a patient receiving an EBP increased over time across conditions. The adjusted ORs of a patient receiving an EBP from baseline to sustainment were 1.35–1.69 times larger in clinics that received facilitation compared with the three comparison conditions (adjusted RORs of comparison condition versus facilitation ranged from 0.59 (95% CI 0.47 to 0.75) to 0.74 (95% CI 0.58 to 0.94)). Implementation facilitation can be integrated into a national programme for quality improvement for PTSD specialty care and may be particularly useful when less-intensive approaches are not sufficiently effective.
- Implementation science
- Evidence-based medicine
- Mental health
- Healthcare quality improvement
Data availability statement
The datasets analysed for this evaluation are available from the corresponding author on reasonable request. These data will remain within the VA firewall and be housed on VINCI data servers. Outside investigators can follow VA procedures and receive training and approval for access within VA firewalls.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Despite the resources dedicated to specialised mental healthcare for patients with post-traumatic stress disorder (PTSD) within the US Veterans Health Administration, evidence-based psychotherapies (EBPs) for PTSD have been underutilised, as evidenced by low EBP reach to patients. There is a need for sustainable and scalable strategies to improve EBP access in low-performing clinics.
WHAT THIS STUDY ADDS
Implementation facilitation delivered by staff with regional mentoring responsibilities and tailored to specialty care for PTSD increased uptake of EBPs in low-performing clinics more than less-intensive quality improvement interventions.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Implementation facilitation can be integrated into a national programme for quality improvement for PTSD specialty care and may be particularly useful when less-intensive approaches are not sufficiently effective.
Background
Post-traumatic stress disorder (PTSD) is a potentially disabling psychiatric disorder resulting from trauma exposure that is particularly prevalent among veterans who receive medical care through the US Veterans Health Administration (VHA).1 2 To ensure that veterans have access to high-quality PTSD care, in 2005, VHA began nationwide rollout of two of the evidence-based psychotherapies (EBPs) recommended as frontline treatments across PTSD clinical practice guidelines—cognitive processing therapy (CPT) and prolonged exposure (PE).3 4 Chart note templates were developed and performance measures based on template data are used to monitor and incentivise delivery of CPT and PE to patients with PTSD, particularly those seen in outpatient clinics that specialise in PTSD treatment, known as PTSD Clinical Teams (PCTs). A national PTSD mentoring programme provides guidance to PCTs on policy and practice through a network of regional mentors.5
Despite the resources dedicated to PTSD care within VHA, CPT and PE have been underutilised.6–8 The Promoting Effective, Routine, and Sustained Implementation of Stress Treatments (PERSIST) programme, a multiproject collaboration between VHA health services researchers and operational partners in VHA’s Office of Mental Health and Suicide Prevention (OMHSP), was designed to address this implementation gap by improving PCT clinician uptake and veteran access to CPT and PE. The first PERSIST project identified team and organisational factors associated with high versus low reach of CPT and PE to patients with PTSD.9 The practices associated with the high reach PCT model of care included establishing a team mission that prioritised delivery of EBPs; engaging the team in the mission; establishing clinic processes that enable the mission (eg, monitoring patient outcomes); and enlisting external clinics (eg, general mental health clinics) and facility leadership to support EBP specialisation within the PCT. The next PERSIST project piloted implementation facilitation (IF) tailored to low EBP reach PCTs. In IF, a facilitator supports individuals and teams to identify organisational barriers and enabling factors and make changes to implement evidence into practice.10–12 For PERSIST, IF was guided by a specialised toolkit and informed by audit on EBP reach. The toolkit bundled strategies associated with high reach in the first PERSIST study9 along with resources to develop an EBP-focused mission, build engagement in this mission, implement procedures to facilitate EBP delivery, foster positive perceptions of EBPs, and strengthen facility leadership support for a team structured to promote EBP implementation. The facilitator was part of the PTSD mentoring programme leadership team and external to the implementation sites. The evaluation showed a much larger increase in EBP reach in intervention compared with matched control PCTs,13 with continued improvement over the year after facilitation (unpublished data).
Results from the above implementation project solidified the decision among the involved researchers and VHA operational partners to scale up the PERSIST implementation strategies through VHA’s PTSD mentoring programme. To accomplish this, the specialised toolkit was integrated into existing resources for regional PTSD mentors and reports to audit EBP reach, referred to as ‘reach reports’, were expanded and modified to better align with the VHA’s regional structure and performance monitoring systems. Concurrently, the PTSD mentoring programme established a unified set of principles of PTSD specialty care that contained and reframed the elements of high-reach teams identified through the first PERSIST study so that they were congruent with mental health policy and meaningful to clinicians. Central to these principles was prioritisation of delivery of EBPs for PTSD, which are time-limited (usually 8 to 15 sessions) psychotherapies to reduce PTSD symptoms, and use of measurement-based care (MBC), which is required for patients seen in PCTs, regardless of intervention delivered. MBC involves the use of patient-reported outcomes to track progress in mental health treatment and has been shown to improve patient outcomes.14 15 Last, the PTSD mentoring programme leadership team piloted an expanded role for the regional PTSD mentors involving the option to: (1) use IF to improve EBP reach; (2) participate in a learning collaborative (LC) organised around the principles of PTSD care; or (3) continue in mentoring as usual (MAU) without added responsibilities.
This evaluation was conducted to determine whether IF integrated into the national mentoring programme and delivered by regional PTSD mentors without researchers’ involvement in decision-making improved EBP reach to patients more than the other concurrent but less-intensive quality improvement interventions. The secondary objective of the evaluation was to determine whether IF also increased uptake of MBC. While MBC is not an EBP-specific therapeutic technique, the CPT and PE protocols indicate that PTSD symptom monitoring should be implemented.16 17 Therefore, interventions that increase uptake of EBPs may also increase uptake of MBC, at least for EBP patients. This evaluation will inform decisions regarding the use of regional mentors as internal facilitators to address quality gaps within PCTs and may serve as a model for integration of an evidence-based implementation intervention into a national field-based programme within a healthcare system.
Methods
The evaluation was designed for internal purposes in support of the VHA mission and findings were to be used for VHA programme improvement. Researchers were not involved in intervention delivery; they conducted the evaluation using data collected through routine care. OMHSP provided documentation that the project involved non-research operations activities pursuant to Department of Veterans Affairs Office of Research and Development Program Guide 1200.21. The Minneapolis VA Healthcare System IRB determined that the activities did not meet the definition of research. The Revised Standards for Quality Improvement Reporting Excellence (SQUIRE 2.0) guidelines provided the framework for this article.18
Context and design overview
VA facilities are organised into regional networks referred to as Veterans Integrated Services Networks (VISNs). At baseline, there were 118 PCTs across VHA’s 18 VISNs (3–11 PCTs per VISN). We excluded the six PCTs for which data were inconsistently available over the evaluation because of transitioning to a new medical record system or losing or gaining a PCT. Because IF was customised for PCTs needing assistance to improve patient access to EBPs, this evaluation included the 51 PCTs that were providing CPT or PE to less than 25% of therapy patients with PTSD in the preimplementation baseline period; 25% was the median and mean of EBP reach across all PCTs at that time. PCTs were assigned to condition based on mentors' self-selection.
This programme evaluation used a non-equivalent comparison group design. The preimplementation baseline and postintervention sustainment periods each covered 6 months (1 July 2020–31 December 2020 and 1 October 2021–31 March 2022, respectively). The IF and LC implementation periods also covered 6 months (1 April 2021–30 September 2021). During the 3 months before implementation, IF and LC mentors were engaged in activities to prepare for implementation. Other IF projects have used comparable evaluation periods.13 19
It was not appropriate or possible to involve patients or the public in the design, conduct, reporting or dissemination of this evaluation.
Improvement interventions
Mentoring as usual
All 112 PCTs received MAU. MAU involved the PTSD mentoring programme national leadership team and 36 regional mentors (2 for each of VHA’s 18 VISNs) who were clinical staff with expertise in EBP delivery and PTSD specialty care. As part of MAU, the national leadership team provided education on best practices and policy and reviewed PCT-relevant data in monthly group meetings. The mentors then shared this knowledge with and provided consultation to PCT leadership in their VISNs in separate monthly meetings. In the quarter prior to baseline, the PTSD mentoring programme augmented MAU to include education and consultation on unified principles of PTSD care that were consistent with the characteristics of high EBP reach PCTs,9 access to the toolkit expanded from prior work,13 and quarterly audit and feedback on EBP reach in PCTs and other mental health clinics.
Thirty of the 51 low-reach PCTs received MAU only, meaning that they did not also receive IF or LC. However, 15 of these 30 PCTs were in the same VISN as a PCT targeted for IF. Because these PCTs were subject to the same regional influences as PCTs in IF, we labelled them IF_VISN PCTs and considered them as a separate group from the 15 PCTs that received MAU only in the 7 VISNs that did not have an IF or LC PCT.
Implementation facilitation
Six PCTs within five VISNs received IF. The VISN that had two IF PCTs had the lowest EBP reach across VHA. The mentors providing IF had 10% protected time for facilitation. IF was preceded by the training of PTSD mentors in facilitation during a 2.5-day workshop through the VHA Implementation Facilitation Learning Hub.20 The mentors who were to serve as facilitators identified a PCT within their VISN to receive IF based on the following criteria: (1) low EBP reach (< 25%) as identified through PERSIST EBP reach reports; (2) a potential champion for the project at the PCT; and (3) facility leadership and network chief mental health officer support for the project. The PTSD mentoring programme used the IF model adopted by VHA10 and customised for low-reach PCTs,13 with reach reports to monitor progress. In the 3-month preimplementation, the facilitator worked with the site champion to identify goals for improving EBP reach, barriers and facilitators, and key stakeholders to include during the implementation phase. The 6-month implementation phase began with a site-visit followed by weekly consultation between the facilitator and project champion at the PCT to enact a structured implementation plan. The PTSD mentoring programme leadership team provided consultation to the facilitators on a weekly basis.
Learning collaborative
Mentors for 15 low-reach PCTs across 6 VISNs participated in LC. The LC model was based on recommendations for effective LCs developed by the Agency for Healthcare Research and Quality.21 Mentors in LC participated in monthly learning sessions led by the PTSD mentoring programme leadership team. These calls were structured to provide training on available tools and data sources and to facilitate exchange of ideas for strategies to address organisational barriers to quality improvement, taking into account contextual factors within each VISN. While the overall goal of LC was to implement the principles of PTSD specialty care, the agreed on priority was improvement in MBC because OMHSP had recently developed benchmarks for MBC within PCTs.
Measures
The data for all measures were available through VHA’s repository of clinical and administrative data, the Corporate Data Warehouse (CDW), or dashboards used for monitoring performance.
Outcomes
The primary outcome was EBP reach at the patient-level (yes or no EBP receipt). We used CDW data to retrospectively identify all patients who had psychotherapy for PTSD as outpatients in the 51 PCTs. Each qualifying visit during the baseline and sustainment time periods was classified as an EBP (CPT or PE), or ‘other’ using health factors generated by structured EBP templates that clinicians were required to use for EBP documentation. By the time of this work, the CPT and PE templates were identifying the vast majority of CPT and PE cases.22
The secondary outcome was VHA’s metric of MBC in PCTs. VHA operationalises MBC for PCTs as the proportion of PCT patients with any diagnosis who have at least two patient reported outcome measures using the PTSD Checklist-5.23 MBC is calculated each quarter of the fiscal year and available on an internal dashboard designed to support PCT management. For this evaluation, operational partners provided the MBC values for the seven quarters from baseline through sustainment.
Other measures
Clinic-level EBP reach, extracted from the quarterly reach reports disseminated to all PCTs, is operationalised as the proportion of therapy patients with PTSD who received at least one session of CPT or PE in the PCT. EBP reach data were available for the year prior to baseline through the end of sustainment.
We extracted the following variables for patients with PTSD who received psychotherapy through the 51 low-reach PCTs during the two evaluation periods: age, sex (gender is inconsistently available in administrative data), race, ethnicity, marital status, period of military service, driving distance from home to the nearest VHA medical centre, psychiatric hospitalisation, disability status for PTSD, and type and number of psychiatric comorbidities in addition to PTSD. Additionally, we constructed a measure of baseline clinic workload by dividing the number of patients with PTSD seen by PCT clinicians by the number of clinicians in the clinic, as done in prior research.24
Analysis
Preliminary analyses identified confounding variables to include in adjusted models of EBP reach. We defined patient-level confounders as patient characteristics that were imbalanced across conditions and associated with a patient receiving an EBP for PTSD at baseline at p<0.05. We used analysis of variance or Pearson’s χ2 tests, depending on the variable type and distribution to identify these confounders. Age, sex, race, marital status, PTSD disability status and the presence of the following diagnoses met inclusion criteria: depression, anxiety disorder, alcohol use disorder and other substance use disorders.
A priori identified clinic-level confounders were baseline workload, baseline EBP reach and the linear time trend for EBP reach. To estimate the linear time trend, we predicted what sustainment (post) period EBP reach would have been in the 51 PCTs in the absence of the quality improvement interventions using the EBP reach data from the 4 quarters prior to baseline. We developed a prediction model using logistic regression with a random slope (time), a random intercept (site), time (in quarters) as the independent variable, and EBP receipt during each of the four quarters prior to baseline as the dependent variable. We used this model to predict EBP reach during the sustainment period.
Primary analyses examined the effect of condition on the change in the odds of a patient receiving an EBP from baseline to sustainment. First, we implemented simple and multiple logistic regression to obtain unadjusted and adjusted ORs, respectively, of a patient receiving an EBP during each time period by condition. Adjusted models included the above identified patient-level and clinic-level confounders. To avoid deletion of patients seen in only one of the two evaluation periods, we ran the regressions separately for each evaluation period. Next, we used ratio of odds ratios (RORs) to compare the change in the odds of EBP receipt by condition. The ORs for each period were the odds of receiving an EBP versus not with IF as the reference condition. The ROR was the sustainment period ORs divided by baseline period ORs. We used 500 bootstrap replicates to obtain the 95% CIs for the RORs. We did not include random effects for the clustering of PCTs within VISNs because including random effects did not improve model fit (Akaike Information Criterion=10 619.72 versus 10 621.72 without and with random effect, respectively).
We had limited data (number of cases and eligible patients) on MBC. We used these data to assess the effect of conditions on MBC using repeated measure logistic regression. We also implemented alternative models using MBC values for the quarter at baseline and sustainment closest to implementation and the average of MBC values across the quarters at each evaluation period. The alternative models produced inferentially equivalent RORs and did not make use of all the available data. Therefore, we report findings from the repeated measures logistic regression. Due to the lack of patient-level and prebaseline MBC data, we could not adjust for confounders or the time trend in MBC.
Analyses were preformed using SAS 9.4 and R V.4.2.1.
Results
Over the 2 evaluation periods, 26 126 unique patients with PTSD received psychotherapy in one of the 51 PCTs with low reach at baseline. This included 2829 in IF, 8073 in LC, 8464 in IF_VISN and 6760 in MAU PCTs. Characteristics of patients by condition are presented in table 1. The only characteristic that was not imbalanced across conditions was psychiatric hospitalisation (p=0.45). Out of the remaining patient characteristics, ethnicity, driving distance, bipolar disorder and psychotic disorders were not related to the receiving an EBP for PTSD and thus not treated as confounders.
Characteristics of patients with post-traumatic stress disorder (PTSD) who received psychotherapy during the evaluation periods
Figure 1 presents EBP reach by condition per quarter beginning the year before baseline through the end of sustainment. As can be seen, EBP reach was low and largely stable in the four quarters before baseline, even after the onset of the COVID-19 pandemic which resulted in a reduction in the number of patients with PTSD receiving psychotherapy. Importantly, EBP reach increased across all conditions from baseline to sustainment.
Primary and secondary outcomes over time by condition. Bars represent 95% CIs. The X-axis shows the month/year per quarter and the number of eligible patients for each outcome. The number of eligible patients for the first quarter of the baseline was not available (NA) for measurement-based care. EBP, evidence-based psychotherapy; MBC, measurement-based care; IF, implementation facilitation; IF_VISN, implementation facilitation Veterans Integrated Service Network; LC, learning collaborative; MAU, mentoring as usual.
Table 2 presents the ORs for receiving an EBP at each evaluation period and the RORs for the change in the odds from baseline to sustainment with IF as the reference group. At baseline, the odds of a patient receiving an EBP for PTSD were lower in IF than in the other three conditions. During sustainment, the odds of a patient receiving an EBP for PTSD were lower in MAU and IF_VISN than in IF. The RORs demonstrated that the improvement in the odds of a patient receiving an EBP for PTSD were greater for IF compared with each other condition. In unadjusted analyses, the magnitude of improvement in IF was almost twice that in the other conditions. As shown in the lower half of table 2, when we adjusted for patient-level confounders, baseline clinic workload, baseline EBP reach, and the time trend in EBP reach from the four quarters preceding baseline, the odds of improvement in reach in PCTs that received IF was 1.35–1.69 times greater than in PCTs in the other conditions.
Estimated effect of implementation facilitation on changes in the odds of receiving of an evidence-based psychotherapy (EBP) for post-traumatic stress disorder (PTSD) from baseline to sustainment evaluation periods
Figure 1 presents MBC over the seven quarters from baseline through sustainment. The RORs showed that the OR for a patient receiving MBC from baseline to sustainment was greater in IF compared with other quality improvement conditions. Specifically, with IF as the reference condition, the ROR for LC was 0.71 (95% CI 0.55 to 0.90), the ROR for IF_VISN was 0.53 (95% CI 0.42 to 0.67) and the ROR for MAU was 0.62 (0.48 to 0.80). Thus, the odds of improvement in MBC in PCTs that received IF were 1.41–1.89 larger compared with those in the other conditions.
Discussion
A research-operation collaboration evaluated whether IF delivered by regional PTSD mentors as part of a national mentoring programme improved EBP reach compared with less intensive quality improvement interventions. Our primary finding was that while on average EBP reach improved in all low-reach PCTs, the PCTs that received IF experienced the largest improvement. The magnitude of improvement in EBP reach among the IF PCTs was 1.35–1.69 times greater than in the other conditions after adjustment for time trends and confounders. The effect of IF was specific to the targeted low-reach PCTs and did not spread to other low-reach PCTs within the same VISN (ie, the IF_VISN PCTs). We conclude, therefore, that it is unlikely that the improvement in EBP reach in IF was due to changes in policies or procedures implemented regionally. The fact that the effects of IF did not spill over to other PCTs within the same VISN demonstrates that IF requires focused activities with a target clinic. Overall, our evaluation’s results are consistent with research evidence that facilitation can improve implementation outcomes in low-performing clinics.25 26
We find it interesting that EBP reach increased from baseline to sustainment across all conditions. By the time of this evaluation, MAU had been enhanced to include unified principles of PTSD care that were consistent with the characteristics of high EBP reach PCTs,9 access to a specialised toolkit and regular audit and feedback on EBP reach. While we cannot derive definitive conclusions as to cause, we conjecture that these enhancements to MAU contributed to the average level of improvement in EBP reach in PCTs over time. This premise is supported by the observation that before these enhancements to MAU, EBP reach did not increase in PCTs over a comparable 21-month time frame.13 It is further supported by the finding that there was little or no change in the proportion of therapy patients with PTSD who received an EBP in general mental clinics from baseline to sustainment. Specifically, data from the quarterly reach reports showed that the across the 51 involved facilities, the proportion of therapy patients with PTSD who received an EBP in general mental health clinics was 4% and 3% during baseline and sustainment, respectively, regardless of the quality improvement condition for the PCT in the same medical centre. The increase in EBP reach over time was specific to PCTs, all of which received MAU, and not part of a larger trend within the involved medical centres.
Mirroring the pattern observed with EBP reach, the magnitude of improvement in MBC was greater in IF than in the other conditions, including LC which had prioritised implementation of MBC. This is not surprising given that the EBP treatment manuals call for regular administration and review of PTSD symptom measures.16 17 However, we were not able to determine the degree to which the improvement in MBC was attributable to use of MBC as part of EBP delivery.
IF was enacted by regional PTSD mentors, whereas in prior work, it was enacted by an external facilitator who was part of the PTSD mentoring programme leadership team.13 The PTSD mentoring programme made this change to enhance IF scalability and the capacity for IF across PTSD specialty care. If successful as facilitators, PTSD mentors could apply facilitation skills to improve reach and other outcomes among PCTs in their VISN. The PTSD mentors may have been uniquely positioned to assume the facilitator role after brief training because of their existing relationships with PCTs within their region. Regardless, this evaluation demonstrated the feasibility of using clinical staff with regional responsibilities to expand the cadre of internal facilitators for quality improvement.
The six facilitators had 10% (4 hours per week) protected time; mentors in LC and MAU did not have protected time for these quality improvement activities. While the IF PCTs experienced a greater level of improvement in EBP reach than PCTs in the other conditions, this evaluation does not answer the question as to whether this difference in improvement is worth the resources required. Given that IF is more intensive than MAU and LC, we propose a stepped-care approach27 28 to improving the quality of care offered in PCTs, with the clinic being stepped to increasingly intensive implementation interventions depending on clinic performance. For example, it seems reasonable to use IF when a PCT does not show the desired improvement from MAU or other less-intensive interventions. A disadvantage of the stepped approach is that clinics that need a higher level of support for change would have to wait until they have demonstrated continued need after failing to improve. Alternatively, it may make sense to use IF when a clinic is known to have a greater need for support to overcome barriers to making structural changes just as it makes sense to offer more intensive clinical interventions to patients who are experiencing high levels of clinical need. At the same time, effective IF requires a champion for change within the clinic10 and not all low-performing clinics have such a potential change agent. A different approach would be needed to improve clinic performance when there is no clear site-level change leader.
This evaluation used a quasi-experimental observational design. As such, clinics were grouped into condition based on their mentors’ self-selection and the mentors and clinics in each condition may have differed on unmeasured factors. We were, however, able to restrict the evaluation to PCTs with largely comparable EBP reach at baseline and to adjust for patient case-mix, baseline clinic reach, workload and the time trend for reach in our primary analyses. Unfortunately, we did not have patient-level or prebaseline MBC data and therefore cannot rule out the influence of confounders or the prebaseline time trend for this secondary outcome. We also acknowledge that this project took place during the COVID-19 pandemic and that the number of therapy patients dropped considerable during the first phase of the pandemic and did not fully recover over the evaluation periods. While we are not aware of the COVID-19 pandemic having affected PCTs in one condition more than another, this possibility cannot be ruled out. The main change necessitated by the COVID-19 pandemic was the addition of information on use of telemedicine for EBP delivery in the reach reports that all PCTs received. A final and important limitation is that we did not interview mentors or clinical staff from all conditions or track the time mentors and champions spent in the different quality improvement activities. A formative evaluation involving staff in each condition would have helped to understand the reasons for these findings and ways to improve or tailor the quality improvement strategies for different clinic contexts.
Conclusions
IF enacted by regional mentors improved access to EBPs for PTSD in low-performing PCTs more than other, less-intensive quality improvement interventions. IF focused on improving EBP reach was also associated with greater improvement in MBC. These findings support the training and deployment of clinical staff with regional mentoring responsibilities as facilitators for quality improvement in PTSD specialty care when less-intensive approaches are not sufficiently effective. Finally, this evaluation illustrates the successful integration of an evidence-based quality improvement intervention tailored for a specific context through a research-operation collaboration into a national programme within a healthcare system. This integration has the potential to improve intervention scalability and sustainment.
Data availability statement
The datasets analysed for this evaluation are available from the corresponding author on reasonable request. These data will remain within the VA firewall and be housed on VINCI data servers. Outside investigators can follow VA procedures and receive training and approval for access within VA firewalls.
Ethics statements
Patient consent for publication
Ethics approval
This evaluation was designed for internal purposes in support of the VHA mission and findings were to be used within VHA for program improvement. The data were collected as part of routine care and VHA operations. The project’s primary operational partner (VHA’s Office of Mental Health and Suicide Prevention) provided documentation that the project’s aims and this report involved non-research operations activities pursuant to US Department of Veterans Affairs Office of Research and Development Program Guide 1200.21. The Minneapolis Veterans Affairs Health Care System IRB determined that the activities did not meet the definition of research. This was a quality improvement project (see above). We used data collected as part of routine care to understand provider behaviour (the provision of evidence-based psychotherapies for PTSD). Data were retrospectively collected.
Acknowledgments
The authors would like to thank Sean Nugent for procuring the data and all PTSD mentors and PCT staff.
References
Footnotes
Contributors Concept and design: NAS, KPM, SN. Acquisition, analysis, or interpretation of data: all authors. Drafting of the manuscript: all authors. Obtained funding: NAS. Administrative, technical, or material support: RJO. The authors read and approved the final manuscript. NAS is the guarantor who accepts full responsibility for this work, had access to the data and controlled the decision to publish.
Funding This quality improvement project was funded through a U.S Department of Veterans Affairs (VA), Health Services Research and Development (HSR&D) grant (RVR 19-479) awarded to the first author.
Disclaimer The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the VA, HSR&D or U.S. Government. Funders had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
Competing interests The authors are VHA employees. NAS received VHA research funding. The authors have no other competing interests to declare.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.