The Veterans Health Administration (VHA), an agency of the US Department of Veterans Affairs (VA), began using physician assistants (PAs) in 1968.1 PA employment in the VHA has grown steadily and they are deployed in most VA medical centers (VAMCs) and outpatient clinics.2 Although PAs function in multiple roles in VHA settings, little is known about their responsibilities except as primary care providers.3 One component of VHA service is the community-based outpatient clinic (CBOC) system, small clinics (often storefront arrangements in malls) that are geographically separate as satellites of a VAMC 30 to 100 miles distant.
The development of CBOCs began in 1995.4 The policy and intent was to deliver decentralized care that was conveniently close to the veteran.5 As of 2015, the 150 VAMCs and about 1,400 CBOCs enrolled 8.9 million veterans.6 Family medicine physicians, PAs, and NPs staff most CBOCs and tend to be skilled in adult primary care along with some specialty care. A panel size per provider is targeted at 1,200 patients, consistent with part of the Patient Aligned Care Team (PACT) initiative implemented in 2010-2014. PACT is intended to achieve team-based care, improve access, and improve care management for more than 5 million primary care VA patients.7 Typically the PA or NP in the CBOC, remote from the main facility, works as a primary care provider with a small set of support staff. The size of a CBOC ranges from 3 to 10 providers.
Access to healthcare facilities means the timely use of personal health services to achieve the best health outcome.8 CBOCs were designed to improve access by being more convenient for veterans, reducing travel time, travel expense, and waiting times for appointments. Before this project, most orthopedic care was in the main hospital setting and relatively little was managed in the CBOCs. However, CBOCs are well suited for initial evaluation of nonurgent problems such as arthritis or for administering joint injections. Appointments in CBOCs are, for the most part, scheduled and not set up for same-day visits without appointments.
In 2012, a project was undertaken by the Houston VAMC to expand the repertoire of formally trained and nationally certified PAs in CBOCs to deliver more decentralized orthopedic care. The reasoning was that care would be more efficient and effective if patients who needed musculoskeletal evaluations or treatments could be assessed and managed in the CBOC. Patients needing more in-depth evaluation or surgery would be referred to the VAMC.
In the Houston VAMC, PAs in the orthopedic service work in a variety of outpatient clinics: joint clinic (mainly knee, hip, and shoulder replacement), sports clinic (focus on shoulder, hip, and knee arthroscopy), spine clinic, foot and ankle clinic, pre-op clinic (taking histories and physicals, obtaining signed consents), and joint injection clinic (primarily shoulder, knee, and hip injections). The orthopedic subspecialty clinics were on different days or half days so each PA in training attended every clinic in a week. The chief of orthopedics, various orthopedic subspecialists, and staff PAs in orthopedic medicine and surgery provided the training.
In 2009, a PA with experience in the orthopedic clinics was transitioned to a CBOC in Conroe, Tex. Initially, the position was for 2 days a week in Conroe and 3 days in Houston. Gradually, over half a year, the Conroe volume of cases grew and a full-time position was justified. The Conroe CBOC medical providers appreciated the help and patients appreciated avoiding a long drive to the Houston VAMC. An administrative decision was made to expand medical orthopedics to four additional CBOCs.
Four PAs were recruited from outside the Houston VAMC to participate in training and permanent deployment to a CBOC. The prototype PA remained in the Conroe CBOC, for a total of five PAs for this project.
During the orthopedic training period, all PAs rotated through each orthopedic subspecialty clinic over 6 to 12 weeks. In addition, 3 to 4 weeks were spent in hand clinic (plastic surgery) staffed by orthopedic residents and a PA in hand surgery. Three weeks were spent in podiatry clinic with two podiatrists and a PA in podiatry. For each clinic, a reading list of appropriate articles from UptoDate.com and/or Wheelessonline.com and electronic orthopedic texts was suggested for diagnoses likely to be encountered. Total time for training for the CBOC position was a median of 4 months (range, 3 to 9 months) and adjusted in accordance with the PA's previous experience. PA careers varied from 1 to 30 years.
After deployment, the chief of orthopedics at the Houston VAMC visited each CBOC monthly for difficult cases, to consult on surgeries, and to maintain overall communication with the CBOC's PAs in orthopedics. The more common forms of communications were telephone calls or countersignature of charts. Issues of medical questions that the PA could not resolve were directed to the CBOC medical director. Administratively, the PAs reported to the medical director of the CBOC for time scheduling and routine yearly evaluation. In the absence of citable literature or reports about CBOC diversification into medical specialties to draw upon, an administrative research study was selected as the mode of assessment.
In 2015, centralized data in the main facility (Houston VAMC) was probed, downloaded, and partitioned for hospital and CBOC encounters for orthopedics from May 2012 to December 2012 and May 2014 to December 2014. This pair of dates was selected to compare the effect on workload or access by the addition of the four new PAs added in 2013 to the CBOC orthopedic staffs. In addition, the number of OR orthopedic cases, numbers of orthopedic injections, and time to complete a consult at the main facility were included, spanning a 4-year period. All patient names were made anonymous and no effort was made to identify any patient or provider. The institutional review board for Baylor College of Medicine and Affiliated Hospitals approved the project.
The data consisted of 300 sequential charts downloaded from the computerized patient record system for each of the five CBOCs under study. A total of 1,500 charts was considered adequate for administrative assessment. Diagnoses were grouped by body area (knee, shoulder, etc.) and subgrouped by primary and secondary diagnosis (osteoarthrosis, tendinitis, fracture, etc.). Primary and secondary procedures were noted for each encounter; bilateral procedures counted as two conditions. Referral to other orthopedic services at the main facility, a visit at the CBOC by an orthopedist, or telephone contact/chart countersignatures for patient disposition was noted. Data were analyzed using descriptive statistics.
The project involved five CBOCs in the Houston VAMC region (Table 1). Before the project (2012), the number of providers per CBOC ranged from four to seven (mean, 5.2). Veteran enrollment averaged 7,862 per CBOC. The average distance from the CBOCs to the VAMC was 36 miles (range, 29 to 45 miles). Wait times at the CBOC varied between 1 and 28 days. At times, some patients enrolled in primary care at the VAMC were sent to the CBOC for orthopedic care when it was more convenient.
An assessment of orthopedic visits was made before and after the project. During a period from 2013 to 2014, the number of orthopedic surgical cases rose by 11% over the previous year but also tended to fluctuate by year (Table 2). During the observation period of an added PA in orthopedics in the CBOC, the wait time for an orthopedic consultation at the VAMC decreased from 30 days (SD ± 25) to 10 days (SD ± 10).
Orthopedic encounters pre- and postevaluation were assessed (Table 3). The number of orthopedic encounters at the VAMC increased by 994 (11%) during the study period and overall encounters were up 31%. The Conroe CBOC also increased its orthopedic encounter count by 10% during the same period of observation. A total of 3,909 orthopedic encounters took place in the five CBOCs during this study period—a 383% increase over previous CBOC orthopedic encounter volume.
During the study period, 106 orthopedic diagnoses were made and were grouped by body area (Figure 1). Knee conditions accounted for 42% of all orthopedic encounters, with osteoarthrosis of the knee making up 33% of all diagnoses.
Orthopedic procedures in terms of injections (soft tissue and intra-articular) pre- and postproject were summarized and compared with the central VAMC data during the same period (Figure 2). Specifically, knee joint injections accounted for 66% of all injections and shoulder joint injections accounted for 26%. Two-thirds of all orthopedic visits in the CBOC resulted in an injection. During this observation period, the total number of injections for the VAMC and CBOCs rose from 1,528 to 2,028, a 33% increase. Before the orthopedic CBOC project was started, the total number of CBOC injections was 200 (all in the Conroe CBOC). In the postproject evaluation, that number rose to 828; injections at the main facility decreased by 10%.
In examining how many times in the first 300 visits the PA sought assistance in determining the final disposition of a patient, the most common occurrences were sending a patient to a subspecialty orthopedic clinic in the Houston VAMC (3.2%), having the orthopedic surgeon see the patient at the CBOC (2.7%), discussing cases on the phone with the orthopedist (1.7%), or asking the orthopedist to countersign the chart to validate the treatment plan (1.7%) (Table 4). The number of referrals outside of orthopedics, such as to podiatry, pain clinic, neurology, or rheumatology, was small.
From May 2014 to December 2014, a total 13,885 orthopedic encounters occurred, 9,976 at the main hospital and 3,909 at the five CBOCS, representing a 31% volume increase from the year before. Although a 10% increase occurred at the main hospital, the remaining upturn was from CBOC PA orthopedic encounters. The CBOC project managed 28% of all the orthopedic encounters in the Houston VAMC-wide system with five out of seven CBOCs staffed by a PA in orthopedics. The PA addition at the CBOC sites demonstrated that adding a decentralized medical PA in orthopedic services does not necessarily decrease the volume of patients at the medical center; total volume of patient visits in this VAMC increased. The interpretation was that while volume was increasing overall, the PA was improving access for orthopedic services.
Most of the orthopedic encounters (about 40%) were for knee complaints. The total percentage of diagnoses for three body areas (knee, shoulder, and lumbar spine) accounted for 75% of all orthopedic diagnoses in this study.
In the aggregate, the PA in orthopedics at a CBOC appears to have managed the vast majority of musculoskeletal cases in this study. Aside from the convenience factor, from the patient's point of view, value was added in less congestion at the main hospital orthopedic clinics, and shorter waiting times for new consults and routine care. Although wait times varied at the CBOC, the perception was that patients with chronic problems would rather be seen locally at their medical home (in this case the CBOC) if possible.
In terms of procedures, the number of orthopedic injections given in 2012 compared with 2014 increased by 33%, but the number of injections given at the main facility decreased by 10%. One interpretation is the improved CBOC access to procedures permitted more patients getting injections within a reduced time period. The type of procedures and services performed appear consistent with other studies of orthopedic services employing PAs.9,10
Overall, CBOCs appear to have a role in VHA patient management and their use seems to be successful. However, these clinics are small, have limited staff, and the number of veterans who are accessing them is small as well. Therefore, most veterans can only access the VAMCs. Expanding more CBOCs or adding more specialty PAs will require more analysis as to whether this should be a CBOC enhancement or a replacement of a primary care provider. Trade-off evaluations and cost benefit analyses are needed.
In summary, this administrative research and demonstration project took advantage of unique characteristics of American PAs—the ability to change roles, ramp up new skill sets, and be mobile.11 The adaptability and flexibility of new roles has been identified as one of the defining characteristics of PAs.12 That the VHA can take advantage of this role malleability is a testimony to its employment of PAs.
A number of limitations are embedded in this type of administrative study. Not all the CBOCs in the study could add a PA at the same time. Nor did the new PAs have the same skill set, years of experience, or the same complement of VAMC orthopedic rotations. Nor did they rotate through the VAMC at the same time. Bias could have been introduced in the PAs of interest knowing that this was a demonstration project that would be judged on its merits.
Another limitation was that this was administrative research; an activity that takes its name from the organization that supports it. It differs from critical analysis guided by research questions that tend to be hypothesis-driven. When administrative research is undertaken in healthcare, some of the objectives are to look for centralization, quality assurance, department review, program evaluation, and human values. However, as a demonstration project, administrative research often is the first step for laying the groundwork for more probing types of analyses such as financial, social, outcomes, economic, and other health services research activities. In that sense, the project met its goal of improving healthcare service delivery of veterans.
Although the role of the PA in a CBOC has been in primary care/adult medicine, the large number of orthopedic diagnoses in the veteran population makes it worthwhile to consider a newer concept, teaming up a PA in medical orthopedics. The Houston VAMC demonstrated the practical application of adding a PA in orthopedics to many of its CBOCs; administratively, it appears successful. During the period of observation, the PA handled most cases with little need to confer with surgeons. Referral for further evaluation or surgery at the main VAMC accounted for a small percentage of cases. All of these factors together contribute to increased access to care demonstrated by increased surgical cases being completed, increased volume of outpatient musculoskeletal injections, an overall increase in outpatient orthopedic visits, and a decrease in time for new consult completion at the main facility. If the improved efficiency and access to care remains durable, this concept has the potential for widespread application within the VHA system.