CORR Insights®: High-volume Arthroplasty Centers Are Associated With Lower Hospital Costs When Performing Primary THA and TKA: A Database Study of 288,909 Medicare Claims for Procedures Performed in 2019

Where Are We Now?

In the context of ever-declining reimbursement rates for orthopaedic procedures [11], the Centers for Medicare and Medicaid Services developed and instituted various bundled payment programs to further decrease expenditures specifically associated with primary hip and knee arthroplasties, given their associated costs and drastically increasing volume [7]. These developments have strained the financial sustainability of inpatient arthroplasty services in health systems; efforts are needed to allow hospitals to adequately adapt to these changes, including those related to the availability of data on, for example, hospital-specific cost drivers. Indeed, when surveyed about preparedness for bundled payments in joint arthroplasty, more than 50% of hospitals reported being unprepared [4]. This highlights the importance of a thorough understanding of cost drivers, because identifying modifiable cost drivers may aid hospitals in their strategies to successfully participate in bundled payment programs.

Although there are many studies on modifiable and nonmodifiable cost drivers [9], one factor has been relatively understudied: increasing hospital arthroplasty volume and the potential for hospital cost savings. In this issue of Clinical Orthopaedics and Related Research®, Blackburn et al. [2] addressed this evidence gap by using Medicare 2019 claims data on 288,909 inpatient THAs and TKAs. Blackburn et al.’s [2] study is particularly meaningful because it used recent data, aimed to estimate the true cost and not Medicare payments, and adjusted for a variety of potential confounders, including standard patient demographics, other hospital variables (including hospital geographic location, setting, teaching status, and resident-to-bed ratio), and patient comorbidity burden. The authors point out that various mechanisms may underlie cost savings in higher-volume centers, including streamlined care pathways, leading to more efficient care processes; lower cost as a function of fewer complications (and thus shorter length of stay) because of a more favorable volume–outcome relationship; or hospitals’ greater bargaining power in negotiating more favorable implant pricing with vendors, because implant costs represent a significant portion of arthroplasty total costs [3].

The study by Blackburn et al. [2] provides some important initial insights into economies of scale in relation to arthroplasty volume and cost. Based on these findings, hospitals or health systems—especially those at risk of financial losses because of the aforementioned developments—might think of strategies to increase arthroplasty volume through various mechanisms. These could include consolidation or restructuring through, for example, shifting patients to appropriate settings in a healthcare system based on patent complexity and appropriateness for outpatient surgery, as recently described in a health system with an academic medical center, an associated community-based hospital, and several ambulatory care centers [5]. In this example, after a case stratification strategy was implemented, increases in arthroplasty volume coincided with greater efficiency and improved net margins. In addition to these important practical implications, Blackburn et al.’s study [2] also raises important questions and exciting opportunities for future research that may further elucidate volume–cost mechanisms in this context.

Where Do We Need To Go?

First, more information is needed on mechanisms driving volume–cost relationships in joint arthroplasty. Although various mechanisms have been proposed, it is unclear which ones primarily drive cost savings. Data are also lacking on the role of surgeon (in addition to hospital) volume; Blackburn et al. [2] merely focused on hospital arthroplasty volume irrespective of the number of individual surgeons responsible for performing these arthroplasties. One could argue that in the presence of a true volume–cost relationship, it would matter whether these procedures were performed by mainly high-volume or low-volume surgeons [10]. Second, because Blackburn et al. [2] focused on only inpatients with Medicare, additional studies are needed that include arthroplasties performed in the outpatient setting and patients with commercial insurance, both of which are rapidly growing subgroups. Because not all patients will be candidates for outpatient arthroplasty, a shift toward the latter will result in ongoing differences between inpatient and outpatient populations, likely impacting volume–cost relationships. Likewise, commercially insured patients are generally younger than those with Medicare, which may impact implant choice. Importantly, arthroplasty care does not happen in a vacuum; surgeons treat a mix of patients with Medicare and those without, and may perform both inpatient and outpatient surgery. Failure to sufficiently account for important subgroups—and hypothesize about the magnitude and mechanism of volume–cost relationships in these groups—may yield inaccurate volume–cost estimates. Third, it is prudent to explore the implications of a potentially real volume–cost relationship in arthroplasty care. Blackburn et al. [2] specifically noted that smaller-volume hospitals may benefit from increasing surgical volume, and from a policy perspective, may be disproportionately affected by declining reimbursement and require special considerations in bundled payment designs. Studies focusing on specific volume thresholds—other than numerically appealing cutoffs based on units of tens or hundreds—to thoughtfully increase arthroplasty volume, and studies weighing the implications of increased volume in terms of access to care (with, for example, consolidation that could result in longer travel time for patients) against lower arthroplasty costs will be important to further refine this ongoing discussion. Fourth, it is currently unclear to what extent participation in bundled payment programs could impact arthroplasty volume–cost relationships. A focus is warranted on the role of volume in discharge destination because savings related to bundled payments have been mainly attributed to reductions in postacute care use [1]; moreover, Blackburn et al. [2] describe home discharge rates almost three times as high in the highest-volume as in the lowest-volume hospitals.

How Do We Get There?

Although a variety of study designs may be required to address the knowledge gaps I mentioned, the primary focus should be the availability of all-payer data and more-detailed and reliable cost data to understand the mechanisms driving volume–cost relationships in joint arthroplasty. Medicare claims data, as used by Blackburn et al. [2], are among the most commonly used for health services research; however, all-payer longitudinal datasets do exist [6] and could provide further insights into volume–cost relationships for arthroplasty, in terms of not only hospital costs, but also 90-day episode costs [3]. Strategies to mitigate the prohibitive costs generally associated with these datasets include interinstitutional collaboration, given that insights on modifiable cost drivers could benefit a wide variety of stakeholders. Given the importance of implant pricing on arthroplasty costs [3], it is disheartening to observe barriers—including those resulting from confidentiality agreements between hospitals and manufacturers—to implant pricing disclosures and subsequent availability of large-scale cost data [8]. Because regulatory pressure is the most likely remedy, these barriers to transparency about true implant costs are not likely to disappear anytime soon. In the meantime, single-institution or multi-institution studies [1] are likely to provide some details on the role of implant pricing in arthroplasty volume–cost relationships. In addition to quantitative data from existing data sources, studies focusing on collecting mixed quantitative and qualitative data are an important opportunity to elucidate volume–cost mechanisms. Detailed knowledge about care processes and protocols in a set of high-volume with low-cost, high-volume with high-cost, low-volume with low-cost, and low-volume with high-cost hospitals could be compared to determine what exactly distinguishes centers from one another in terms of successful cost reduction related to arthroplasty volume. Finally, regarding an assessment of the implications of identifying true volume–cost relationships, it will be prudent to involve all relevant stakeholders including patients, surgeons, physician extenders, hospital administrators, payers, policymakers, and researchers. Each will bring unique perspectives to strategies to increase surgical volume and bundled payment designs.