Introduction
Total hip arthroplasty (THA) and total knee arthroplasty (TKA) are among the most commonly performed elective surgical procedures in the United States (Kremers et al., 2015; Ruiz et al., 2013; United States Arthritis Foundation, 2018). In 2015, 4.7 million people in the United States were estimated to be living with at least one TKA and 2.5 million with at least one THA (Kremers et al., 2015; Park et al., 2018; United States Arthritis Foundation, 2018). Utilization of total joint arthroplasty (TJA) continues to be high with projections of increasing use throughout the next decade (Inacio et al., 2017; Kurtz et al., 2007; Kurtz et al., 2014; Sloan et al., 2018).
As TJA procedures increase, hospitals are working to continually demonstrate improvement in quality outcomes while limiting costs and mitigating risks of complications. An innovative approach to healthcare that has arisen from this is the destination center of excellence (DCOE) program. These DCOE programs are typically contractual between a self-insured large corporation and hospitals that have shown consistent success in key outcome metrics for a given diagnosis or procedure. The programs are designed to ensure a high quality of care provided to the corporation employees while standardizing and containing costs among participating hospitals (Badlani et al., 2012; Hamid et al., 2014; Lansky et al., 2012; Mehrotra et al., 2013; Robinson, 2013; Smith et al., 2021; Wyles et al., 2021).
As with most referrals to an orthopaedics office, the patient is initially seen by a primary care provider or a general practice provider who then makes the referral to an orthopaedics specialist. In DCOE programs (based on the patient's insurance and employer contract) patients are filtered to higher volume practices with multidisciplinary care integration and clearly outlined postoperative rehabilitation pathways that help improve the patient experience and outcomes (Etzion et al., 2013; Munoz & Samet, 2004; Nwachukwu et al., 2015; Smith et al., 2021). Medical tourism usually refers to travel to another country for healthcare (Nwachukwu et al., 2015), but this concept is more broadly defined in this article to imply travel for the purpose of healthcare. Despite potential advantages to DCOE programs, numerous drawbacks exist in traveling for a major surgical procedure including limitations in the preoperative evaluation and postoperative follow-up. Currently, evidence is limited on patient satisfaction, complications, and outcomes when participating in the DCOE programs.
In 2017, we created a DCOE program for TJA through the work of a multidisciplinary task force. The present study reviewed the early experience and outcomes of patients undergoing either THA or TKA through our DCOE program. Herein, we detail the program development and the resources required for successful implementation. The primary aims were to (1) evaluate the rates of surgical patients through the DCOE program; (2) review acute perioperative course trajectories; and (3) evaluate clinical outcomes affected by the DCOE program's implementation.
The vision for the DCOE program began with the formation of an Office of Practice Specialty Collaborations and Contracts. The Office comprised key clinical leadership and administrative stakeholders who performed current state assessment and gap analysis to identify existing strengths and barriers of the implementation of a DCOE program. A cost analysis was also performed to determine any applicable costs to the institution associated with additional required staff. Simultaneously, the orthopaedics department identified best practices for postoperative TJA care. High priority was placed on educational resources for patients to guide them from the decision to undergo surgery, through the surgical process, and returning home afterward.
DCOE Program Development and Design
Barriers initially encountered included (1) absence of outpatient staff dedicated to the DCOE program, (2) limited inpatient physical therapy services after hours and on weekends, and (3) provider concern about consistent postoperative follow-up care. To address these barriers, our institution hired one full-time clinic registered nurse (RN) and one patient care navigator dedicated solely to the DCOE program. The RN's primary responsibilities were facilitating intake of patients, interviewing patients, and assisting the care navigator with patient travel arrangements as needed. In addition, the physical therapy department expanded its inpatient, outpatient, and weekend hours of operation to facilitate rehabilitation care. To ensure postoperative follow-up, the RN called the patient at 3 weeks postoperatively to evaluate pain control, wound healing, rehabilitation progress, and their ongoing primary care provider medical management. If any concerns arose, a telemedicine visit was scheduled or pictures were sent to the surgical team through an online patient portal, or both. The use of telemedicine in this process allowed for patients to receive high-level care and avoid potential hospitalizations (Gajarawala & Pelkowski, 2021).
For a patient to participate in the DCOE program, the patient's local orthopaedic surgeon identified the patient's need for surgery, which triggered a flag when the requests were submitted through the patient's insurance carrier. After contractual approval, the patient's candidacy was evaluated by the patient care navigator through an in-depth review of pertinent health records and radiographs. Specific attention was directed to medical inclusion criteria for participation in the DCOE program (see Table 1). If the patient did not meet the criteria, their application for the DCOE TJA program was denied until medical optimization was completed. Patients who had chronic pain were eligible to participate in the program. An agreement was required by the patient's pain management provider or primary care provider that they would continue to manage all chronic pain. After the patient traveled home, care was resumed by their local primary care provider and pain management provider, if applicable.
After a patient met defined medical criteria (see Table 1), the RN interviewed the patient and completed a formalized DCOE intake form (see Figure 1). The form and appropriate imaging were sent to the surgeon for review and determination of surgical candidacy. Once approved, the patient care navigator scheduled all perioperative appointments and provided an itinerary to the patient. In addition, assistance with travel arrangements (driving or flying) and lodging accommodations (a hotel is located on campus) were provided for the patient and the caregivers. Most patients were from outside the greater Jacksonville area and utilized the on-campus hotel accommodations (98% of study patients). Patients typically arrived 1 or 2 days in advance of the surgery date to undergo preoperative assessment by the surgeon and anesthesia clinicians. After this clearance, the patient underwent surgery and was admitted for overnight hospitalization. After discharge, the patient remained near the clinic for 1 week to receive outpatient physical therapy. Before the patient returned home, an in-person postoperative appointment was scheduled with the surgical team at 1 week. After the patient traveled home, care was resumed by their local primary care provider.
DCOE Surgeons, Techniques, and Multidisciplinary Care Team
Surgeons were required to fulfill requirements to be eligible to perform surgeries as part of the DCOE program. Program requirements included performing a minimum of 100 primary THA or TKA cases per year. Quality metrics such as complications, readmissions, and patient-reported outcomes were continually monitored and reviewed to provide surgical teams real-time opportunities for practice improvement.
Although surgical techniques were at the discretion of the surgeons, all operations were performed with contemporary approaches (posterolateral or direct anterior for THAs and medial parapatellar for TKAs) and implants (uncemented THAs and cemented posterior-stabilized TKAs). Spinal anesthesia was routinely performed with the addition of single-injection regional nerve blocks (adductor canal in TKAs and paravertebral in THAs) and intraoperative periarticular injections. Postoperatively, patients were initiated and educated on a graduated multimodal pain regimen beginning with over-the-counter analgesia (acetaminophen and anti-inflammatory agents unless contraindicated) and followed with tramadol or a stronger narcotic pain medicine, or both with the expectation that narcotics are for acute postoperative pain only. In addition, nonpharmacologic methods (such as cold packs, meditation, and aromatherapy) may have been used. All patients received 6 weeks of venous thromboembolism prophylaxis with either 81-mg aspirin twice daily or low-molecular-weight heparin, based on risk stratification and surgeon discretion.
The entire multidisciplinary heath care team had a crucial role in development of the DCOE. Outpatient nursing staff focused on patient education, medication reconciliation, and ensuring adequate home support and patient flow through the perioperative process. Inpatient nursing staff provided standardized wound care, medication regimens (such as analgesic administration), and physical therapy. Postoperative physical therapy began on the day of surgery and continued at a minimum of two formal sessions per day until the patient was safely discharged from the hospital. Rehabilitation continued with daily outpatient therapy after discharge for 1 week on campus. When the patient returned home after 1 week, physical therapy continued at two to three times weekly until the physical therapist deemed the patient suitable for a home exercise program. Case management collaborated with the clinic RN and the care navigator to facilitate consistent postoperative planning including delivery of durable medical equipment. Overall, the integrated team stayed in constant communication with the patient and with each other to provide a safe, high-quality experience.
Study Design and Methods
Study Description
After approval by the Mayo Clinic institutional review board (IRB), the first 100 patients in the DCOE program who had undergone either primary THA (n = 33) or TKA (n = 67) at Mayo Clinic in Jacksonville, FL, from January 1, 2018, through December 31, 2019, were included. The sample size of 100 was chosen to represent the exploratory and feasibility nature of this study. During this period, there were 213 patients referred to the program but of these, only 100 met defined inclusion criteria. Informed consent was not required by the Mayo Clinic IRB as there were no deviations of standard of care for all patients. Retrospective record review was performed by a nurse practitioner in the orthopaedics department to obtain routine patient demographic characteristics, perioperative outcomes (such as length of stay, readmissions, reoperations), and validated patient-reported outcome measures. The outcome measures were reviewed with the use of Knee Society Score (KSS) and Knee Injury and Osteoarthritis Outcome Score (KOOS) for TKA, and Harris Hip Score (HHS) and Hip Disability and Osteoarthritis Outcome Score (HOOS) for THA (Banaszkiewicz, 2013; Nilsdotter et al., 2003; Roos & Lohmander, 2003; Scuderi et al., 2012). All outcomes were verified through telemedicine or a telephone visit, including readmissions or reoperations on patient return home. The mean timing of follow-up (telemedicine or phone visit) was 1.2 years (range: 1.0-1.3 years) following the joint arthroplasty surgery.
Outcome Measurement
Objective perioperative outcomes were collected and validated with subjective and objective patient-reported outcome measures. The outcome measures were reviewed with the use of KSS (including KSS function) and KOOS for TKA and HHS and HOOS for THA (Banaszkiewicz, 2013; Nilsdotter et al., 2003; Roos & Lohmander, 2003; Scuderi et al., 2012). All outcomes were verified through telemedicine or a telephone visit, including readmissions or reoperations upon the patient's return home. These outcome measures were selected because of their proven validity and reliability, ease of use, and application to both acute and chronic conditions (Lee et al., 2016; Nilsdotter et al., 2003).
For TKA, the KSS and KSS-F were used. The KSS consisted of three domains: pain (three items worth 50 points), stability (two items worth 25 points), and range of motion (one point for every eight degrees with a maximum of 25 points). Deductions to the possible 100 points were given for extension lag, flexion contracture, malalignment, and pain at rest. The maximum possible score was 100 points, with a higher score indicating less disability. Scores of 80-100 were considered excellent, 70-79 were considered good, 60-69 were considered fair, and scores below 60 were considered poor. The knee functional activity score was reported separately as the KSS-F. The KSS-F was also worth a maximum 100 points, with higher scores indicating higher levels of function. Negative scores were reported as zero. The items in the KSS-F were walking distance (50 points maximum), ability to ascend and descend on stairs (50 points maximum), and gait aides used (deducts from total score; Culliton et al., 2018; Lee et al., 2016; Scuderi et al., 2012).
In conjunction with the KSS and the KSS-F, the KOOS was used. This was a 42-item questionnaire that allowed patients to self-report outcome measures in the five domains of pain, symptoms, activities of daily living, function for sports and recreation, and knee-related quality of life. The maximum score a patient could achieve was 100, indicating no knee problems. The minimum score was zero, indicating severe knee problems. The KOOS score has documented use for a variety of knee pathologies and has demonstrated use in patients of various ages (Hung et al., 2018; Roos & Lohmander, 2003).
For THA, the HHS was used. This was a clinician-based outcome measure (not patient-reported) commonly used for evaluation of patients after THA. The HHS was composed of four domains: pain severity (possible 44 points), function (includes activities of daily living and gait, possible 47 points), deformity (includes hip flexion, adduction, internal rotation, leg length discrepancy, possible four points), and range of motion (possible five points). The HHS takes less than 10 minutes total for patient completion and scoring completed by the provider. The HHS has also been proven to measure outcomes after nonsurgical interventions such as physical therapy (Harold et al., 2021; Nilsdotter & Bremander, 2011).
In conjunction with the HHS, the HOOS Scale was used to measure THA outcomes. This was a 42-item questionnaire that allowed patients to self-report outcome measures in the five domains of pain, symptoms, activities of daily living, function for sports and recreation, and hip-related quality of life. The maximum score a patient could achieve was 100, indicating no hip problems. The minimum score was zero, indicating severe hip problems. The HOOS score has documented use for a variety of hip pathologies and has demonstrated use in patients of various ages (Hung et al., 2018; Nilsdotter et al., 2003).
Statistical Methods and Analysis
All routine descriptive statistical analysis such as mean and median for demographic characteristics and perioperative, postoperative, and outpatient follow-up outcome data were performed with software (Excel; Microsoft Corp). For demographic characteristics, mean was used for age, body mass index (BMI), and distance traveled in miles, whereas median was used for the American Society of Anesthesiologists physical status (ASA score; see Table 2). Mean was used for the preoperative and postoperative outcome scores of HHS, HOOS, KSS (including KSS function), and KOOS (see Tables 3 and 4).
Results
DCOE Patients
A total of 213 patients underwent initial intake screening for participation in the DCOE program, and 100 (47%) were approved for and completed surgery (67 TKAs and 33 THAs; see Table 2). The three foremost reasons for exclusion from the program were BMI greater than 40 kg/m2, lack of exhausting all conservative treatment options, and socioeconomic factors as not being able to afford time off work or not having a caregiver to travel with them. The mean age at surgery was 58 years (range, 37-77 years), 66% of participants were women, and the mean BMI was 33.3 (range, 20.4-44.5). On average, patients traveled 316 miles (range, 4-2,750 miles) for participation in the DCOE program. Further loosening the concept of medical tourism, patients who lived within 50 miles of the medical campus were given the option to stay in their homes rather than the on-campus hotel.
Perioperative Trajectory
The mean length of stay was 2 days (range, 1-4 days) and 1.7 days (range, 1-4 days) after TKA and THA, respectively. Two participants developed major complications during the acute perioperative period that required medical intervention during the DCOE program stay. Of these two patients, one developed deep vein thrombosis and the other had a stroke on postoperative Day 1. The 90-day readmission rate for all patients was 1%, and this was a planned readmission for observation after a closed manipulation under anesthesia (MUA). In total, five patients (7.5%) required MUA for stiffness after TKA, given their limited progress in postoperative rehabilitation. One reoperation after TKA was performed at 9 months postoperatively for nerve pain: a neurectomy of the infrapatellar branch of the saphenous nerve. One revision after THA was required at 5 months postoperatively because the acetabular component failed ingrowth and loosened, with no further sequelae after the revision procedure.
Clinical Outcomes
All patients reported that postoperative outcome measures significantly improved (p < .01) for THAs and TKAs (see Tables 3 and 4). These outcomes were measured by routine survey collection as part of standard department practice. For patients who underwent THA, the final mean HHS and HOOS were 94.24 (range, 86-100) and 93.92 (range, 56.42-100). For patients who underwent TKA, the final mean KSS and KOOS were 91.5 (range, 68-100) and 94.35 (range, 0-58). All final scores represent an "excellent" outcome, according to each grading system.
Discussion
Few studies have reported outcomes of medical tourism (travel specifically for the purpose of healthcare) or surgical DCOE programs (Nwachukwu et al., 2015). Given the evolution of healthcare delivery and payer systems, evaluation is important for the viability and outcomes of such programs (Gogineni et al., 2019).
Our institution (Mayo Clinic in Jacksonville, FL) recently implemented a DCOE program for total joint replacement through the work multidisciplinary care teams. According to the results from early experience, the program has been successful, resulting in an optimized preoperative process, limited postoperative complications, and excellent improvement in clinical outcome scores.
Perhaps the most important concern related to a surgical DCOE implementation was overcoming the distanced patient evaluation and postoperative care for a major surgical experience. Nwachukwu et al. (2015) evaluated this challenge in their review of 38,887 TJAs performed at DCOE hospitals and showed no clinically significant association between complications and travel distance. Similarly in our study, although a smaller patient population, the patients traveled on average 316 miles to participate in the DCOE program and had no major medical complications. We believe that mitigation of adverse events may be attributed to the additional staff hired to streamline the intake process and sustain continuity throughout the care episode. In addition, communication and integrated facilitation pathways were established across all phases of care for efficient patient flow through the process. Unique barriers will arise with the development of any DCOE program. However, in our experience, the multidisciplinary team of stakeholders found innovative and holistic approaches that improved the entire TJA surgical practice. These approaches include the creation of the screening tool for eligibility, increased use of the online patient portal for communication, and increased use of telemedicine for postoperative follow-up visits.
Despite efficient patient throughput, the DCOE TJA patients were not without surgical reoperations. The TKA patients had a higher rate of reoperation, predominantly related to knee stiffness that required MUA in 7.5% of patients. Although stiffness after TKA is common, our MUA rate was slightly higher than the estimated rates of 1.3% and 5.8% (Zachwieja et al., 2018). The cause of this increased rate is likely multifactorial. Yet, of the MUAs, one patient had a previous MUA for arthrofibrosis after a patellar fracture and three patients presented with a preoperative knee flexion less than 95[degrees]. A low threshold was used to have a DCOE patient return for in-person range of motion evaluation with a preplanned MUA, and this may have contributed to surgeon bias in performing the procedure. Only one patient (1%) in the entire DCOE TJA cohort required revision arthroplasty after having pain and limited function from failed ingrowth of the acetabular component. This case could be considered an outlier because survival rates of uncemented acetabular cups have been as high as 98.8% at 10-year follow-up (Karuppal, 2016). Moreover, our results appear to be comparable with those of Mehrotra et al. (2013), who demonstrated equivalent risk of complication in TKA but a statistically significant lower risk of complication for hip replacements performed at hospitals with a designated joint center of excellence.
Value was exhibited with the outcomes deemed important by all stakeholders: patients, surgical team, hospitals, and payers. The needs of the patient and their health outcomes remain our institution's primary value. Implementation of the TJA DCOE program delivered on that value with a significant objective increase in the clinical outcome measures and metrics. Subjectively, overwhelming patient satisfaction was expressed to our RN and clinical navigator through positive comments about patients' medical destination experience, pain reduction, and return of function and quality of life. The phenomenon called distance bias may be argued where patients who travel a considerable distance to receive care experience better outcomes (Munoz & Samet, 2004). Etzioni et al. (2013) examined this potential outcome bias and reported an opposite finding: Locally treated patients had less risk of severe complication and better outcomes than predicted. For purposes of our program, patients who lived within 50 miles of the medical campus were given the option to stay in their homes rather than the on-campus hotel. This did not appear to impact their experiences.
Healthcare value may also be defined as health outcomes accrued to the patient per dollar spent or, more simply, the patient's condition level over the entire care cycle (Hamid et al., 2014). No cost analysis is reported here because reimbursement is confidential. However, our program generally showed financial value and savings through practice standardization, minimization of complications or readmissions, and excellence in patient outcomes. With the exponential increase in TJA procedures nationally, innovative TJA programs such as ours need continuous improvement to provide the highest value care (Etzioni et al., 2013; Inacio et al., 2017).
Limitations
Limitations inherent to the present study are its small number of patients, short-term follow-up, and retrospective design. Meticulous documentation of each patient communication was performed in real time and reviewed systematically as part of the research. External electronic health records were reviewed when connected and available. Still, postoperative complication and outcome data were obtained from the patient and could be subjectively misinterpreted or not captured if the patient did not report a serious event. Individual medical and surgical decision-making bias may exist despite a homogeneous group of surgeons and stakeholders. In addition, it should be noted that some of the success of the DCOE program might be attributed to the stringent selection criteria that screened out high-risk patients. If the patients who were excluded had been included, then the patient outcome measures and use of resources may have been different. It was not investigated whether the creation of the DCOE provided better outcomes for the entire practice population.
Implications for Orthopaedic Nursing
The orthopaedic nurse plays a vital role in the development and implementation of a successful DCOE program. Both inpatient and outpatient nurses collaborated with surgeons and clinicians for the design of the DCOE. Outpatient nursing staff focused on patient education, medication reconciliation, and ensuring adequate home support and patient flow through the perioperative process. Inpatient nursing staff focused on delivering standardized care, medication regimens, and physical therapy. Both outpatient and inpatient nursing staff were instrumental representatives of the multidisciplinary team for the development and implementation of the DCOE program. It is our recommendation that orthopaedic nurses and advanced practice nurses leading orthopaedic programs collaborate with the multidisciplinary healthcare team to integrate evidence-based practices into the development of specialty programs such as the DCOE program.
Conclusion
In summary, development of a DCOE TJA program at our institution required vision from a multidisciplinary care team. The keys to this process were dedicated outpatient clinical resources, optimized patient referral intake, standardized practices, and the use of telemedicine. The early experiences of the DCOE program have been favorable, with successful patient completion through all phases of care, low complication rates, and excellent patient-reported satisfaction. As the program continues to grow and evolve, longitudinal evaluation will be ongoing. Antidotal clinical findings and patient outcomes are not drastically different from the findings of this study. Therefore, plans for further research will be based on findings from ongoing reevaluation.
References