Implementing Group Physical Therapy for Veterans with Knee Osteoarthritis (Group PT): Function QUERI 3.0 aims to compare two implementation approaches for implementing Group PT in 20 VA facilities using a type III effectiveness-implementation hybrid study framework and parallel cluster randomized trial (CRT) design.

Background/Purpose. Exercise-based PT is a key component of guideline concordant knee OA management. However, PT is underutilized for knee OA, resulting in missed opportunities to improve outcomes for the many Veterans with this health condition. A key challenge is the high demand for PT services for knee OA, paired with limited availability of PT services, which can lead to long wait times for Veterans or high community care costs for the VA. Group-based delivery of PT for knee OA can enhance access and efficiency. Following an RCT that supported the effectiveness of Group PT relative to individual PT, the Durham Veterans Affairs Healthcare System (DVAHCS) offered Group PT as a clinical service, and found that patients achieved clinically relevant improvements in pain and functional outcomes that were comparable to those observed in the RCT. Most recently, the investigators conducted a 19-site trial comparing implementation strategies to support delivery of Group PT in the VA. Overall, 16 sites successfully delivered a Group PT program, enrolling a total of 366 Veterans. There were clinically relevant improvements in pain interference and functional outcomes among participating Veterans.

The overall goal of this trial is to evaluate implementation of Group PT with a foundational arm versus the Reach Equity and Collective Health (REACH) arm.

Methodology. To evaluate implementation of Group PT, the investigators will randomize sites (n=20) 1:1 to either the foundational arm or the REACH arm. Implementation activities will occur over a 12-month period. The Group PT program includes 6 sessions that include exercise and educational content. Patients will be eligible for Group PT if they have a clinician diagnosis of symptomatic knee OA and ineligible if they have a substantial fall risk or co-occurring health conditions that would make participation in a group exercise class unsafe. The primary follow-up outcome time-point will be 18 months; metrics will also be evaluated at 12 months (reach, patient engagement). Implementation outcomes include 1) Reach (primary), defined as the number of patients initiating Group PT, 2) Patient Engagement, defined as the average number of sessions attended by patients who initiate the EBP, and 3) Time to Adoption, defined as the time from baseline to launching a Group PT program and enrolling 5 Veterans. Generalized linear models will be used to examine the effect of Foundational vs. REACH arms on reach and patient engagement.

Inclusion Criteria:

Sites are eligible to participate if they have:

• 2 outpatient PT clinicians (one primary and one back-up) for delivering Group PT
• space to conduct group sessions (if delivering Group PT in person) or ability to conduct visits via telehealth (video)

Patients will be eligible for Group PT if they have a clinician diagnosis of symptomatic knee OA.

Exclusion Criteria:

• Site will be ineligible to participate if they are already conducting a group physical therapy class for patients with knee OA.
• Patients will be ineligible for Group PT if they have a substantial fall risk or co-occurring health conditions that would make participation in a group exercise class unsafe.

A de-identified, anonymized dataset will be available upon request. Prior to distribution, a local privacy officer and study statistician will certify that the dataset contains no protected health information (PHI). Data will be provided to requestor in electronic format. Sufficient data and descriptors will be made available to duplicate statistical analysis and confirm conclusions in publication. No data or statistical code that could lead to re-identification of individuals will be released. Data will be stored & maintained in an approved, secured location as described in the VA Research Data Inventory Form. The statistician will create de-identified, publication-specific datasets that include variables from statistical models presented in publication. Data will only be released per appropriate authorizations or agreements. Written agreements will specify that recipients are prohibited from taking steps to re-identify any individual whose data is included in the dataset.

Available upon request.

Data will only be released per appropriate authorizations or agreements. Written agreements will specify that recipients are prohibited from taking steps to re- identify any individual whose data are included in the dataset.