Digital Musculoskeletal Program Is Associated With Decreased Joint Replacement Rates

ABSTRACT

Objectives: To compare 12-month total knee arthroplasty (TKA) and total hip arthroplasty (THA) rates for digital musculoskeletal (MSK) program members vs patients who received traditional care for knee or hip osteoarthritis (OA).

Study Design: Retrospective, longitudinal study with propensity score–matched comparison group that used commercial medical claims data representing more than 100 million commercially insured lives.

Methods: Study participants with hip OA (M16.x) or knee OA (M17.x) International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes were identified in the medical claims database. Digital MSK program members were identified using record linkage tokens. The comparison group had hip- or knee-related physical therapy identified via ICD-10-CM and Current Procedural Terminology codes. Respectively in each knee and hip OA group, digital members were matched to control group patients with similar demographics, comorbidities, and baseline MSK-related medical care use. TKA and THA at 12 months post participation were compared.

Results: In the knee OA group, 739 of 56,634 control group patients were matched to 739 digital members. At 12 months, 3.79% of digital members and 14.21% of control group patients had TKA (difference, 10.42%; P < .001). In the hip OA group, 141 of 20,819 control group patients were matched to 141 digital members. At 12 months, 16.31% of digital members and 32.62% of control group patients had THA (difference, 16.31%; P = .001).

Conclusions: These findings suggest that patients who participated in a digital MSK program to manage OA have lower rates of total joint arthroplasty in the 12 months after enrollment.

Am J Manag Care. 2024;30(4):e103-e108. https://doi.org/10.37765/ajmc.2024.89463

Takeaway Points

• Clinical guidelines recommend conservative care as a first-line therapy for symptomatic osteoarthritis.
• In this study of commercial health plan members with knee or hip osteoarthritis, fewer members who participated in a digital musculoskeletal (MSK) program had total knee arthroplasty or total hip arthroplasty at 12 months compared with those who utilized traditional care when matched on demographics, comorbidities, and MSK-related medical use.
• We also examined inpatient and outpatient total knee arthroplasty and total hip arthroplasty rates at 12 months and found lower rates of utilization for both settings among digital MSK program members vs matched control group patients.

Osteoarthritis (OA) affects 1 in 7 US adults, half of whom are working-aged adults aged 18 to 64 years.¹ A person’s lifetime risk of developing knee OA is estimated to be 44.7%, and 1 in 4 (25.3%) may also develop hip OA.^2,3 OA prevalence is increasing because of rising trends in associated risk factors such as age, obesity, physical inactivity, and joint injuries.⁴ OA imposes an economic burden on patients, payers, and society. Patients with OA bear higher direct health care costs and indirect costs such as loss of work productivity compared with those without OA.^5,6 OA-associated hospital stays were ranked as the most expensive condition among private payers in 2017, accounting for 5.9% ($6.96 billion) of total hospital costs.⁷

Because OA develops slowly over decades, it is commonly viewed as a chronic condition with opportunities to slow the progression of symptoms.⁸ Exercise therapy, education, and weight management are so effective at improving OA-related pain and function that they are included in evidence-based clinical guidelines as the first line of care for OA.^9-14 By reducing pain and increasing function, conservative care can help patients with OA postpone or even avoid total knee arthroplasty (TKA) and total hip arthroplasty (THA). However, limited access or adherence to conservative care may result in overutilization of TKA and THA.¹⁵ Better OA management before progression to TKA or THA is a growing priority for payers given that 850,000 THA surgeries and 1.9 million TKA surgeries are projected to be performed annually in the United States by 2030.¹⁶

Increasingly, conservative care approaches are delivered digitally. Evidence suggests that digital musculoskeletal (MSK) care programs (hereafter, digital MSK programs) are as effective at improving MSK-related pain, function, mental health, and surgery intention outcomes as in-person, traditional care.^17-22 It is unknown whether digital MSK programs also decrease the actual incidence of TKA and THA among individuals with OA.

To address this evidence gap, this study’s primary objective was to examine, in a commercially insured population, whether digital MSK program members with OA who also received traditional MSK care have lower rates of TKA or THA compared with nonparticipants with OA receiving only traditional MSK care. Because arthroplasty surgeries in the outpatient setting have become more common,^23,24 the study’s secondary objective was to examine whether the use of a digital MSK program vs traditional care alone is associated with the site of surgical service (ie, inpatient and outpatient facilities).

STUDY DESIGN AND METHODS

Study Design

We compared TKA and THA rates among patients with OA who were digital MSK program members (hereafter, digital members) with those of a propensity score–matched control group of patients with OA who received traditional care (hereafter, control group patients).

Methods

Digital MSK program description. The digital MSK program was a health benefit for employees and dependents offered through 66 employers. The goal of the digital MSK program was to help participants manage chronic knee or hip pain by offering exercise therapy, education, and personal health coaching. The program provided members with tablet computers with a program app. The program app used playlists to present 3 to 8 different stretching, strengthening, balance, and mobility exercises via animations and videos. The program also provided wearable motion sensors (InvenSense MPU-6050; TDK Corporation) that gave feedback through the app about range of movement and repetitions. After exercises, members received educational resources and support from certified health coaches. Program details were described previously.²⁵

Study participants. Digital members who started the digital MSK program between January 2020 and October 2020 were identified in the medical claims database using privacy-preserving record linkage tokens provided by Datavant. Previous research has shown the validity of this identification approach.^26-30 Digital hip members were coded with a hip OA diagnosis (ie, International Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] code M16.x) in their medical claims between January 2016 and September 2021. Digital knee members were coded with a knee OA diagnosis (ie, ICD-10-CM code M17.x) in medical claims data between January 2016 and September 2021. To include all patients with these chronic conditions, we included those with an OA diagnosis before or in the 12 months after starting the digital MSK program.

The hip control group had a hip OA diagnosis in medical claims data between January 2016 and September 2021 and a hip pain–related physical therapy (PT) visit between January 2020 and October 2020 (hereafter, index event). The knee control group had a knee OA diagnosis in medical claims data between January 2016 and September 2021 and a knee pain–related index event. To rule out acute pain and ensure only chronic pain management, we excluded those who had any hip pain– or knee pain–related physician office visits in the same month as the PT visit (eAppendix [available at ajmc.com]).

Additional inclusion criteria were being aged 18 to 64 years and continuously enrolled in a health plan at least 12 months before (ie, baseline period) and 12 months after (ie, post period) starting the digital MSK program or having their index event. Exclusion criteria were TKA or THA in the baseline period and cancer, pregnancy, childbirth, or outlier total medical cost (99th percentile threshold to represent extreme comorbidity) in the baseline or post period.

Outcome variables.The primary outcome of this study was whether a study participant had a TKA or THA in the post period (ie, during the 12 months after starting the digital MSK program or having their index event). We used the Current Procedural Terminology (CPT)/Healthcare Common Procedure Coding System (HCPCS), diagnosis-related group (DRG), or ICD-10 procedure codes in the medical claims to identify whether a study participant had a TKA or THA (eAppendix).

A secondary outcome was surgery setting. We determined whether the TKA or THA occurred in inpatient (no/yes) or outpatient (no/yes) settings. The surgery was categorized as inpatient if any part of the TKA or THA medical claim came from an inpatient facility. Otherwise, we categorized the surgery as outpatient.

Confounding variables.The data set included the following demographic information: (1) age groups (18-29, 30-39, 40-49, or 50-64 years at the time of the data extraction), (2) sex (male, female), and (3) 9 Census divisions (New England, Middle Atlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain, Pacific).

In addition to the knee and hip OA, some study participants had other MSK conditions in the back, knee, shoulder, hip, and neck regions that existed at the same time as the key index event. Therefore, we identified these concurrent chronic MSK conditions from medical claims in the 3 months before the program start month or index event month (eAppendix).

To determine the presence of comorbidities, we applied the Agency for Healthcare Research and Quality (AHRQ) Clinical Classifications Software Refined (CCSR) taxonomy to primary ICD-10-CM diagnosis codes in the claims data. Specifically, we included 10 comorbidities: hypertension, heart disease, diabetes, obesity, mental health, substance use, autoimmune disorders, neurological disorders, respiratory disorders, and HIV (eAppendix). We also calculated a weighted Charlson Comorbidity Index (CCI) score for each study participant using the medical claims in the baseline period (ie, the 12 months before starting the digital MSK program or the index event).

We identified MSK-related health care use during the baseline period. We first identified services with MSK-related primary ICD-10-CM diagnosis codes. Then we categorized CPT and HCPCS codes into service categories using the Restructured Berenson-Eggers Type of Service Classification System.³¹ Next, ICD-10, Procedure Coding System (ICD-10-PCS) codes were categorized using AHRQ’s CCSR taxonomy for ICD-10-PCS. Finally, we categorized DRG codes based on DRG definitions (eAppendix).

We included the number of orthopedic surgeon office visits and PT visits at baseline. We also included presence/absence of specific MSK service categories, including arthroscopy, joint injection, lumbar laminotomy/laminectomy, nerve block injection, other MSK-related chronic treatments (including injection administration fee, provider-administered medication, acupuncture), MSK-related emergency department visits, MSK evaluation and management services in offices and other settings, chiropractor visits, imaging, anesthesia, durable medical equipment, testing (eg, laboratory), and other invasive services (eg, percutaneous vertebroplasty, osteotomy procedures).

Data sources. We used Health Insurance Portability and Accountability Act–compliant, deidentified medical claims data sourced from a claims database that comprised more than 100 million commercially insured members from January 1, 2016, through September 30, 2021, across all US states and territories. Data with enrollment dates and dates of service between January 2016 and September 2021 were included in this study.

Statistical methods.To address confounding variables, we matched digital members to similar control group patients separately for the knee and hip programs using the following propensity score–based matching steps. First, we calculated a propensity score for each individual using a logit model with the following covariates: demographics (age group, gender, Census division), weighted CCI score at baseline, concurrent MSK conditions (back, knee, shoulder, hip, neck) in the 3 months before the index month, baseline MSK-related health care use (variables described above), and a squared term for number of orthopedic surgeon office visits and PT visits at baseline due to nonlinear relationships with propensity of participating in the digital MSK program.

Next, we matched control group patients to the digital members based on the calculated propensity score using full Mahalanobis matching with 1:1 nearest neighbor without replacement. Matching was conducted separately for each pain region (ie, knee and hip). The final analytic sample included 739 matched pairs for the knee program and 141 matched pairs for the hip program.

To describe study participants and check covariate balance, we generated descriptive statistics for the matched sample for baseline factors. We applied χ² tests for categorical variables and t tests for continuous variables to evaluate differences between groups.

For our main finding, we calculated descriptive statistics for primary and secondary outcomes for the matched sample. We used a χ² test to evaluate postperiod differences between groups. For each outcome, we present the difference as outcome for the control group patients minus outcome for the digital members.

As a robustness check, we employed a multivariate linear regression model to adjust for observable differences in the matched sample to mitigate the effects of potential confounding. Models included all confounding variables (described earlier).

We used Stata 17.0 (StataCorp LLC) to conduct the analyses. Data analysis was performed in January 2023.

The study was reviewed and deemed exempt from institutional review board oversight by WIRB-Copernicus Group Institutional Review Board (IRB) (Office for Human Research Protections/FDA IRB registration number IRB00000533) at WIRB-Copernicus Group.

RESULTS

Descriptive Results

After matching, we detected no significant differences on any of the baseline characteristics between the control group patients and digital members (Table 1 [part A and part B]). More than half of matched study participants were female, and more than 80% were older than 50 years. Most resided in the East North Central, South Atlantic, and Pacific regions. Between 1% and 4% had other concurrent MSK conditions in the 3 months prior to the index event. Hypertension, mental health diagnoses, and diabetes were the 3 most prevalent comorbidities. Weighted CCI scores ranged from 0.11 to 0.22.

Study participants used a range of health care services in the baseline period. For example, joint injection was the most common, with 30% of knee participants having joint injections and 16% to 23% of hip participants having joint injections. Between 53% and 59% of study participants had MSK-related evaluation and management services (eg, office doctor visit), and 47% to 50% had imaging.

Main Finding

Table 2 illustrates TKA and THA surgery utilization in the 12-month post period for digital members vs control group patients. For TKA, 3.79% of digital members had surgery in the post period vs 14.21% of matched control group patients (difference, 10.42%; P < .0001). For THA, 16.31% of digital members had surgery in the post period vs 32.62% of matched control group patients (difference, 16.31%; P = .001).

We observed similar patterns across inpatient and outpatient settings. For TKA, 1.35% of digital members had surgery in inpatient settings vs 4.33% of matched control group patients (difference, 2.98%; P = .0006) and 2.44% of digital members had surgery in outpatient settings vs 9.88% of matched control group patients (difference, 7.44%; P < .0001). For THA, 3.55% of digital members had surgery in an inpatient setting vs 9.93% of matched control group patients (difference, 6.38%; P = .03) and 12.77% of digital members had surgery in an outpatient setting vs 22.7% of matched control group patients (difference, 9.93%; P = .03).

Robustness Check

In the regression-adjusted model, digital MSK program membership was associated with 10.49% (95% CI, 7.73%-13.25%) fewer individuals with TKA and 17.56% (95% CI, 7.75%-27.37%) fewer individuals with THA vs the control group in the post period (eAppendix).

DISCUSSION

Among those with OA, this claims analysis showed lower TKA and THA rates among digital members vs control group patients who received traditional care. Specifically, 73% fewer digital members underwent TKA and 50% fewer digital members underwent THA than control group patients at 12 months. These findings contribute to the evidence base about the role of conservative exercise and therapy for managing the progression of symptoms of knee and hip OA. One possible explanation is that the digital MSK program improved the knee and hip pain of digital members more than traditional care improved pain for the control group, which decreased the need for TKA and THA.

This study suggests that conservative care delivered digitally is associated with significantly lower surgery rates than traditional care. A possible reason is that adherence to traditional care may be lower than that for digital or telerehabilitation programs,^32,33 although the COVID-19 pandemic significantly increased telehealth use and thus adherence.³⁴ Indeed, past randomized controlled trials with high PT adherence have reported that in-person therapy and exercise led to 1-year TKA rates between 5% and 6%,^35,36 which is similar to the 3.79% TKA rate in our observational study.

A study strength is that we examined surgery rates by settings. We wanted to do so because projections show that joint replacement surgeries are on the rise and that the surgeries are shifting from inpatient to outpatient settings.^16,37 We showed that the digital MSK program was associated with fewer surgeries in both settings and that the magnitude of the decrease was similar for both settings. For example, we observed 69% fewer inpatient TKA surgeries and 75% fewer outpatient TKA surgeries among digital members vs control group patients.

Digital MSK programs may be a useful approach for controlling increasing TKA and THA costs. Based on Medicare severity DRG 470, an inpatient hip/knee joint replacement procedure without complications cost on average $17,564 in 2020 in the United States.³⁸ We estimated that outpatient procedures cost approximately $12,295—or 30% less than inpatient procedures.³⁹ For every 100 patients with hip OA and 100 patients with knee OA (ie, a total of 200 patients with OA), the digital MSK program was associated with a mean of 9 fewer inpatient surgeries and 17 fewer outpatient surgeries in the year after the participation. Therefore, a digital MSK program may be associated with gross savings of $158,000 in avoided inpatient costs and $209,000 in avoided outpatient costs of these surgeries.

Limitations

The study had the following limitations. First, this claims-based observational study does not demonstrate the causal effect of the digital MSK program on surgery prevention. Although we applied a quasi-experimental design and accounted for an extensive list of demographics, comorbidities, and health care use at the baseline, the study is still subject to potential selection and omitted variable bias because it does not include all key confounding or surgery prediction variables (eg, pain severity, functional status, race/ethnicity).⁴⁰ At the patient level, it is possible that patients with worse pain and lower function may be less likely to engage with digital health platforms. Chronic MSK pain also disproportionately affects certain race/ethnic groups and thus surgery use.⁴¹ Findings from past studies have shown that expected pain and functional improvement are associated with decisions to have joint replacement surgery.⁴² In addition, some health plans may require patients to fail to respond to conservative care before they are approved for a THA or TKA. Therefore, the control group patients may still differ from the digital members in terms of their MSK severity because we do not observe the specific carrier’s coverage policy. Second, we are unable to test whether higher engagement is associated with lower surgery use because the deidentified claims data cannot be linked to member engagement data (eg, counts of exercise sessions, PT video visits). Third, we were unable to determine whether the digital MSK program completely prevented surgeries or simply delayed them beyond 12 months of follow-up. Fourth, this study was conducted among a commercially insured population, so the results may not be generalizable to other insurance populations (eg, Medicare, Medicaid).

Finally, it is plausible that COVID-19 had an effect on the findings because the study period (2019-2021) coincided with pandemic shutdowns. For example, orthopedic surgeon or PT accessibility and backlogs due to COVID-19 may have affected the likelihood of surgery. Providers in one state may have resumed surgeries sooner than in other states. In addition, although we included Census regions in models, we did not include state because state and zip code information was not available in our deidentified medical claims data. Another bias may come from how PT was delivered after COVID-19, because many PT practices started offering telehealth options. It is possible that many patients in the control group also had PT telehealth visits in addition to traditional in-person care, especially for those with multiple comorbidities.³⁴ As we move out of the COVID-19 period, newer claims with standardized telehealth coding will be helpful to mitigate these biases.

To address these limitations, future research could include prospectively designed randomized controlled trials to address causation and to further prove the efficacy of a digital MSK program in decreasing TKA and THA rates. Studies could use newer data and combine clinical data, patient-reported outcomes, and claims data to better explore and control for clinical and health care utilization factors associated with surgeries. A longer-term study could also examine prevention vs delay of surgeries.

CONCLUSIONS

Using a digital MSK conservative care program in OA management demonstrates potential to avoid joint arthroplasty in the first 12 months after enrollment. As a result, a digital MSK program may decrease the economic burden of OA for patients and payers.

Author Affiliations: Hinge Health, Inc (LL, JHL), San Francisco, CA; Department of Radiology, University of Washington (LSG), Seattle, WA; Department of Surgery and Perioperative Care, Dell Medical School, The University of Texas at Austin (KMK), Austin, TX; Weitzman Institute (GW), Middletown, CT.

Source of Funding: Hinge Health, Inc provided the digital musculoskeletal program to participants and funded this research study.

Author Disclosures: Drs Lu and Lee are employed by Hinge Health. Drs Lu, Wang, and Lee have equity interest in Hinge Health, which produces the digital musculoskeletal program evaluated in this article. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (LL, GW); acquisition of data (LL); analysis and interpretation of data (LL, LSG, KMK, JHL, GW); drafting of the manuscript (LL, LSG, KMK, GW); critical revision of the manuscript for important intellectual content (LL, LSG, JHL, GW); statistical analysis (LL); administrative, technical, or logistic support (LL, JHL, GW); and supervision (KMK, GW).

Address Correspondence to: Louie Lu, PhD, Hinge Health, Inc, 455 Market St, Suite 700, San Francisco, CA 94105. Email: louie.lu@hingehealth.com.

REFERENCES

1. Weinstein SI, Yelin EH, Watkins-Castillo SI. The Burden of Musculoskeletal Diseases in the United States. 4th ed. United States Bone and Joint Initiative; 2020. Accessed December 8, 2022. https://www.boneandjointburden.org/fourth-edition

2. Qin J, Barbour KE, Murphy LB, et al. Lifetime risk of symptomatic hand osteoarthritis: the Johnston County Osteoarthritis Project. Arthritis Rheumatol. 2017;69(6):1204-1212. doi:10.1002/art.40097

3. Murphy LB, Helmick CG, Schwartz TA, et al. One in four people may develop symptomatic hip osteoarthritis in his or her lifetime. Osteoarthritis Cartilage. 2010;18(11):1372-1379. doi:10.1016/j.joca.2010.08.005

4. Neogi T. The epidemiology and impact of pain in osteoarthritis. Osteoarthritis Cartilage. 2013;21(9):1145-1153. doi:10.1016/j.joca.2013.03.018

5. Hawker GA. Osteoarthritis is a serious disease. Clin Exp Rheumatol. 2019;37(suppl 120)(5):3-6.

6. Zhao X, Shah D, Gandhi K, et al. Clinical, humanistic, and economic burden of osteoarthritis among noninstitutionalized adults in the United States. Osteoarthritis Cartilage. 2019;27(11):1618-1626. doi:10.1016/j.joca.2019.07.002

7. Liang L, Moore B, Soni A. National Inpatient Hospital Costs: The Most Expensive Conditions by Payer, 2017. Agency for Healthcare Research and Quality statistical brief 261. July 2020. Accessed December 7, 2022. https://www.hcup-us.ahrq.gov/reports/statbriefs/sb261-Most-Expensive-Hospital-Conditions-2017.jsp

8. Roos EM, Arden NK. Strategies for the prevention of knee osteoarthritis. Nat Rev Rheumatol. 2016;12(2):92-101. doi:10.1038/nrrheum.2015.135

9. Henriksen M, Hansen JB, Klokker L, Bliddal H, Christensen R. Comparable effects of exercise and analgesics for pain secondary to knee osteoarthritis: a meta-analysis of trials included in Cochrane systematic reviews. J Comp Eff Res. 2016;5(4):417-431. doi:10.2217/cer-2016-0007

10. Kelley GA, Kelley KS, Hootman JM, Jones DL. Effects of community-deliverable exercise on pain and physical function in adults with arthritis and other rheumatic diseases: a meta-analysis. Arthritis Care Res (Hoboken). 2011;63(1):79-93. doi:10.1002/acr.20347

11. Penninx BW, Messier SP, Rejeski WJ, et al. Physical exercise and the prevention of disability in activities of daily living in older persons with osteoarthritis. Arch Intern Med. 2001;161(19):2309-2316. doi:10.1001/archinte.161.19.2309

12. Mazzei DR, Ademola A, Abbott JH, Sajobi T, Hildebrand K, Marshall DA. Are education, exercise and diet interventions a cost-effective treatment to manage hip and knee osteoarthritis? a systematic review. Osteoarthritis Cartilage. 2021;29(4):456-470. doi:10.1016/j.joca.2020.10.002

13. Bannuru RR, Osani MC, Vaysbrot EE, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019;27(11):1578-1589. doi:10.1016/j.joca.2019.06.011

14. Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 American College of Rheumatology/Arthritis Foundation guideline for the management of osteoarthritis of the hand, hip, and knee. Arthritis Care Res (Hoboken). 2020;72(2):149-162. doi:10.1002/acr.24131

15. Ward MM, Dasgupta A. Regional variation in rates of total knee arthroplasty among Medicare beneficiaries. JAMA Netw Open. 2020;3(4):e203717. doi:10.1001/jamanetworkopen.2020.3717

16. Singh JA, Yu S, Chen L, Cleveland JD. Rates of total joint replacement in the United States: future projections to 2020-2040 using the National Inpatient Sample. J Rheumatol. 2019;46(9):1134-1140. doi:10.3899/jrheum.170990

17. Valentijn PP, Tymchenko L, Jacobson T, et al. Digital health interventions for musculoskeletal pain conditions: systematic review and meta-analysis of randomized controlled trials. J Med Internet Res. 2022;24(9):e37869. doi:10.2196/37869

18. Latif-Zade T, Tucci B, Verbovetskaya D, et al. Systematic review shows tele-rehabilitation might achieve comparable results to office-based rehabilitation for decreasing pain in patients with knee osteoarthritis. Medicina (Kaunas). 2021;57(8):764. doi:10.3390/medicina57080764

19. Seron P, Oliveros MJ, Gutierrez-Arias R, et al. Effectiveness of telerehabilitation in physical therapy: a rapid overview. Phys Ther. 2021;101(6):pzab053. doi:10.1093/ptj/pzab053

20. Xie SH, Wang Q, Wang LQ, Wang L, Song KP, He CQ. Effect of internet-based rehabilitation programs on improvement of pain and physical function in patients with knee osteoarthritis: systematic review and meta-analysis of randomized controlled trials. J Med Internet Res. 2021;23(1):e21542. doi:10.2196/21542

21. Chen T, Or CK, Chen J. Effects of technology-supported exercise programs on the knee pain, physical function, and quality of life of individuals with knee osteoarthritis and/or chronic knee pain: a systematic review and meta-analysis of randomized controlled trials. J Am Med Inform Assoc. 2021;28(2):414-423. doi:10.1093/jamia/ocaa282

22. Safari R, Jackson J, Sheffield D. Digital self-management interventions for people with osteoarthritis: systematic review with meta-analysis. J Med Internet Res. 2020;22(7):e15365. doi:10.2196/15365

23. Arshi A, Leong NL, Wang C, Buser Z, Wang JC, SooHoo NF. Outpatient total hip arthroplasty in the United States: a population-based comparative analysis of complication rates. J Am Acad Orthop Surg. 2019;27(2):61-67. doi:10.5435/JAAOS-D-17-00210

24. Haas DA, Zhang X, Barnes CL, Iorio RR. The national trend in arthroplasty surgery location and the economic impact on surgeons, hospitals and ASCs. J Arthroplasty. 2022;37(8):1448-1451. doi:10.1016/j.arth.2022.03.036

25. Wang G, Yang M, Hong M, Krauss J, Bailey JF. Clinical outcomes one year after a digital musculoskeletal (MSK) program: an observational, longitudinal study with nonparticipant comparison group. BMC Musculoskelet Disord. 2022;23(1):237. doi:10.1186/s12891-022-05188-x

26. Kiernan D, Carton T, Toh S, et al. Establishing a framework for privacy-preserving record linkage among electronic health record and administrative claims databases within PCORnet, the National Patient-Centered Clinical Research Network. BMC Res Notes. 2022;15(1):337. doi:10.1186/s13104-022-06243-5

27. Mirel LB, Resnick DM, Aram J, Cox CS. A methodological assessment of privacy preserving record linkage using survey and administrative data. Stat J IAOS. 2022;38(2):413-421. doi:10.3233/sji-210891

28. Ziedan E, Simon KI, Wing C. Mortality Effects of Healthcare Supply Shocks: Evidence Using Linked Deaths and Electronic Health Records. National Bureau of Economic Research working paper 30553. October 2022. Accessed February 6, 2023. http://www.nber.org/papers/w30553.pdf

29. Wong RJ, Zhang Y, Thamer M. Chronic liver disease and cirrhosis are associated with worse outcomes following SARS-CoV-2 infection. J Clin Exp Hepatol. 2023;13(4):592-600. doi:10.1016/j.jceh.2023.01.014

30. Curtis JR, Fox KM, Xie F, et al. The economic benefit of remission for patients with rheumatoid arthritis. Rheumatol Ther. 2022;9(5):1329-1345. doi:10.1007/s40744-022-00473-6

31. Restructured BETOS classification system. CMS. Updated October 20, 2022. Accessed November 29, 2022. https://data.cms.gov/provider-summary-by-type-of-service/provider-service-classifications/restructured-betos-classification-system

32. Bennell KL, Marshall CJ, Dobson F, Kasza J, Lonsdale C, Hinman RS. Does a web-based exercise programming system improve home exercise adherence for people with musculoskeletal conditions?: a randomized controlled trial. Am J Phys Med Rehabil. 2019;98(10):850-858. doi:10.1097/PHM.0000000000001204

33. Lambert TE, Harvey LA, Avdalis C, et al. An app with remote support achieves better adherence to home exercise programs than paper handouts in people with musculoskeletal conditions: a randomised trial. J Physiother. 2017;63(3):161-167. doi:10.1016/j.jphys.2017.05.015

34. Miller MJ, Pak SS, Keller DR, Gustavson AM, Barnes DE. Physical therapist telehealth delivery at 1 year into COVID-19. Phys Ther. 2022;102(11):pzac121. doi:10.1093/ptj/pzac121

35. Skou ST, Rasmussen S, Laursen MB, et al. The efficacy of 12 weeks non-surgical treatment for patients not eligible for total knee replacement: a randomized controlled trial with 1-year follow-up. Osteoarthritis Cartilage. 2015;23(9):1465-1475. doi:10.1016/j.joca.2015.04.021

36. Deyle GD, Henderson NE, Matekel RL, Ryder MG, Garber MB, Allison SC. Effectiveness of manual physical therapy and exercise in osteoarthritis of the knee: a randomized, controlled trial. Ann Intern Med. 2000;132(3):173-181. doi:10.7326/0003-4819-132-3-200002010-00002

37. Richards MR, Seward JA, Whaley C. Removing Medicare’s outpatient ban and Medicare and private surgical trends. Am J Manag Care. 2021;27(3):104-108. doi:10.37765/ajmc.2021.88598

38. Healthcare Cost and Utilization Project (HCUPnet). Agency for Healthcare Research and Quality. Accessed January 9, 2023. https://datatools.ahrq.gov/hcupnet

39. Yian EH, Schmiesing AM, Kwong BD, Prentice HA, Patel SP. Procedure cost comparison of outpatient and inpatient shoulder arthroplasty and lower-extremity arthroplasty within a managed-care organization. Perm J. 2022;26(4):6-13. doi:10.7812/TPP/22.069

40. Liu Q, Chu H, LaValley MP, et al. Prediction models for the risk of total knee replacement: development and validation using data from multicentre cohort studies. Lancet Rheumatol. 2022;4(2):e125-e134. doi:10.1016/s2665-9913(21)00324-6

41. Patel M, Johnson AJ, Booker SQ, et al. Applying the NIA Health Disparities Research Framework to identify needs and opportunities in chronic musculoskeletal pain research. J Pain. 2022;23(1):25-44. doi:10.1016/j.jpain.2021.06.015

42. Salimy MS, Humphrey TJ, Katakam A, Melnic CM, Heng M, Bedair HS. Which factors are considered by patients when considering total joint arthroplasty? a discrete-choice experiment. Clin Orthop Relat Res. 2023;481(3):427-437. doi:10.1097/CORR.0000000000002358