Hospital Bed Capacity Influences Patient Charges in Outpatient Robotic Total Knee Arthroplasty: A Propensity-Matched Study

Abstract

Background

The rise of robotic-assisted total knee arthroplasty (RA-TKA) surgeries has correlated with increased outpatient procedures aimed at improving efficiency and outcomes. However, patient charges vary widely due to differences in reimbursement structures, resource utilization, and the higher supply costs associated with robotics. Little is known about the impact of institutional characteristics, specifically hospital bed capacity, limiting understanding of charge variability and cost-control strategies. This study aims to examine whether hospital size affects patient charges for RA-TKA, focusing on (1) total charges and (2) operating room (OR) charges.

Methods

Outpatient RA-TKA cases from January 1, 2017, to December 31, 2021, were identified from a national database. Hospitals were categorized by bed capacity: <300 beds and ≥ 300 beds. A 1:1 propensity score match controlled for demographics, clinical characteristics, and insurance status, resulting in 464 patients per group. Outcomes included total patient charges and OR (anesthesia, OR block time, and postanesthesia care) charges. Statistical analyses included Chi-square tests and independent t-test, with P < .05 considered significant.

Results

Hospitals with ≥300 beds showed significantly higher mean total charges ($65,315.6) compared to hospitals with <300 beds ($53,427.5) (P < .001). Mean OR charges were also elevated in larger hospitals ($30,367.2) vs smaller hospitals ($26,000.0) (P < .001).

Conclusions

Larger hospitals were associated with significantly higher total and OR patient charges for robotic RA-THA, even after matching for clinical and insurance-related factors. These findings suggest that hospital size may be associated with variation in charge profiles beyond patient and payer characteristics.

Introduction

The adoption of robotic-assisted total knee arthroplasty (RA-TKA) has been rapidly increasing, with a significant proportion of knee arthroplasty procedures utilizing robotic technology. From 2015 to 2020 alone, RA-TKA use in the United States increased from 0.84% to 5.89%, equating to a 600% increase [1]. This growth coincides with evidence that RA-TKA improves surgical precision, component alignment, and reduces radiological outliers compared to conventional techniques [2,3]. If current trends continue, RA-TKA has the potential to represent nearly half of all knee arthroplasties by 2030 [4]. While outcomes may be optimistic, the adoption of RA-TKA requires significant capital investment. Initial purchase costs for robotic systems range from $700,000 to over $1 million, plus annual maintenance fees and disposable instrument costs [[5], [6], [7]]. There is also a time investment associated with the adoption of RA-TKA systems. While implementation initially increases operative time, efficiency improves with experience and may surpass conventional techniques in some settings [8,9]. Therefore, understanding which institutional factors influence cost variability is critical in determining settings where RA-TKA may achieve true cost-efficiency.

Potential factors contributing to cost variability in the adoption and use of RA-TKA include hospital procedural volume, intraoperative supply costs, length of stay, complication rates, discharge disposition, and post-acute care utilization [6,7,[10], [11], [12]]. The relatively increased index cost of RA-TKA compared to manual techniques has implications for patient charges. With continued adoption, the high upfront expenditures may be increasingly reflected in institutional billing practices. Investigating this relationship provides insight into the downstream financial consequences of adopting new surgical technologies and their role in value-based orthopaedic care.

To better understand the factors contributing to cost variability, our study aimed to investigate the effects of hospital size on (1) total patient charges and (2) total operating room (OR) charges for primary RA-TKA procedures.

Material and methods

Data sourcing

Data were sourced from a national, service-level, all-payer database (Premier Healthcare Database, Premier Inc., Charlotte, North Carolina). The Premier database includes over 1 billion Health Insurance Portability and Accountability-complaint patient encounters across a variety of payer types, including commercial, Medicare, Medicaid, government, and self-pay. International Classification of Diseases, 10th Edition codes, Current Procedural Terminology codes, and patient demographics were used to isolate the population of interest. Institutional review board approval was waived due to the anonymous nature of data sourcing.

Patient selection

Patients who had a diagnosis of osteoarthritis and underwent outpatient RA-TKA between January 1, 2017, and December 31, 2021, were included in the study. The International Procedural Classification of Diseases, 10th Revision and Current Procedural Terminology codes used for RA-TKA identification were International Classification of Diseases, 10th Edition-P-8E0Y0CZ and 27447, respectively. Outpatient RA-TKA was considered any patient with a length of stay of 0 days. Individuals who had a history of rheumatoid arthritis, malignancy, and/or trauma were excluded. Patients were partitioned into 2 groups based on the respective hospital size where the RA-TKA procedure was performed: fewer than 300 beds and 300 beds or more. There were 588 patients in the <300 bed group and 621 patients in the ≥300 bed group.

Matching and data analysis

Patient demographics included age, sex, severe obesity (body mass index ≥40), alcohol abuse, tobacco use, diabetes, TKA fixation type, and insurance type. Significant differences existed between groups (P > .05) (Table 1). Propensity score matching was performed to minimize the risk of confounding bias. A 1:1 nearest-neighbor matching algorithm was used to create comparable cohorts incorporating demographics and clinical covariates, including age, sex, body mass index ≥40 kg/m², diabetes, tobacco use, alcohol abuse, fixation type, and insurance payer type. Covariate balance after matching was assessed using standardized mean differences, with values <0.10 indicating adequate balance. The final <300 bed and ≥300 bed cohorts were comprised of 464 patients each. We analyzed total patient charges and total OR charges. OR charges included expenses for anesthesia, OR block time, and the postanesthesia care unit; implant-related and disposable supply costs could not be separately identified. Statistical analyses used Chi-square tests for categorical variables and independent t-tests for continuous variables. A P value <.05 was considered statistically significant.

Table 1.

Patient demographics.

Variable	Pre-match			Post-match
	<300 beds (n = 588)	≥300 beds (n = 621)	SMD	<300 beds (n = 464)	≥300 beds (n = 464)	SMD
	N (%)	N (%)		N (%)	N (%)
Age in y (SD)	66 (9.5)	63 (9.0)	0.017	65 (9.4)	65.2 (9.1)	0.010
BMI ≥ 40	54 (9.2)	55 (8.8)	0.014	40 (8.6)	46 (9.9)	0.042
AA	4 (0.7)	1 (0.2)	0.034	1 (0.2)	1 (0.2)	0
TU	167 (28.4)	245 (39.4)	0.225	153 (33.0)	158 (34.1)	0.023
Diabetes (complicated)	40 (6.8)	43 (6.9)	0.008	25 (5.4)	28 (6.0)	0.027
Cemented fixation	380 (64.6)	249 (40.0)	0.501	258 (55.6)	247 (53.2)	0.048
Men	210 (35.7)	274 (44.1)	0.185	190 (40.9)	190 (40.9)	0
Commercial insurance	103 (17.5)	110 (17.7)	0.006	91 (19.6)	88 (19.0)	0.016
Medicaid	152 (25.9)	238 (38.3)	0.279	145 (31.2)	128 (27.6)	0.077
Medicare	321 (54.6)	257 (41.3)	0.269	216 (46.6)	233 (50.2)	0.072
Other insurance	12 (2.0)	17 (2.7)	0.044	12 (2.6)	15 (3.2)	0.04

SD, standard deviation; BMI, body mass index; AA, alcohol abuse; TU, tobacco use; SMD, standard mean difference.

Results

Hospitals with <300 beds had significantly lower total charges ($53,427.5 ± $26,729.4) than hospitals with ≥300 bed capacity ($65,315.6 ± $22,775.1) (Table 2; P < .001). Similarly, hospitals with <300 beds had significantly lower OR charges in comparison to hospitals with ≥300 beds ($26,000.0 ± $14,646.2 vs $30,637.2 ± $13,060.8, respectively; P < .001).

Table 2.

Total and operating room patient charges by hospital size.

Charges	<300 beds (n = 464)	≥300 beds (n = 464)	P value
	x (SD)	x (SD)
Total charges	$53,427.5 (15)	$65,315.6 (22,775.1)	<.001
Operating room charges	$26,000.0 (14,646.2)	$30,637.2 (13,060.8)	<.001

SD, standard deviation.

Bold indicates statistical significance at P < .05.

Discussion

The present study demonstrated that patients who undergo RA-TKA at hospitals with <300-bed capacity were associated with lower total charges and OR charges compared to larger hospitals. These differences persisted after adjustment for patient demographics, TKA fixation method, and insurance payer types. Institutional scale may influence cost variation in RA-TKA, and hospital bed capacity may be associated with variability in patient charges.

There is limited evidence that hospital bed capacity, a surrogate for institutional size, is associated with patient charges for orthopaedic surgical procedures. However, hospital volume has been consistently associated with lower patient charges for total joint arthroplasty (TJA) procedures. In a study of 2702 hospitals and 458,259 primary arthroplasty procedures, Courtney et al.[13] reported that hospitals performing fewer than 100 TJAs had significantly higher mean charges and payments than higher volume hospitals ($60,895 vs $56,323; P < .001). Similarly, Porter et al. [14] demonstrated that primary TJA at high-volume hospitals was associated with a cost decrement of $1900 to $2200 per case relative to low-volume hospitals. It is important to consider that hospital surgical volume does not always correlate with bed capacity. In fact, overnight hospital stays are becoming less necessary as outpatient TJA procedures increase. There has been a dramatic rise in same-day discharge rates for TKA from 2016 to 2020, increasing from 1.2% to 62.4% in the United States [15]. High procedural volume of robotic TKA procedures does not inherently require a large hospital bed capacity. The observed cost differences in the present study may be enhanced in smaller hospital settings (<300 beds). Our findings support a model in which reduced charges are driven not only by procedural volume but also by institutional efficiency and streamlined perioperative pathways, such as same-day discharge.

The effects of hospital size have been investigated in fields outside of orthopaedics. For example, Robinson et al. [16] found that hospitals with greater bed capacity had higher median charges and payments for common acute care diagnoses, including gastrointestinal bleed, gastrointestinal obstruction, and laparoscopic cholecystectomy. This study directly stratified hospitals by bed size and demonstrated a positive association between larger hospital size and higher patient charges [16], which is consistent with our findings for RA-TKA. While not specific to arthroplasty, a retrospective study by Kim et al.[17] of 268,809 patients in Korea concluded that their small hospital cohort incurred fewer inpatient charges, however had longer lengths of stay for treatment of joint disease. These results reinforce the concept that hospital size may be a meaningful factor associated with patient charges and warrants consideration in orthopaedic cost analyses.

Beyond demonstrating an association between hospital bed capacity and patient charges, our findings may have implications for the continued expansion of RA-TKA within value-based care frameworks. Recent systematic reviews and scoping studies confirm that the United States leads global research and implementation of value-based care, with hospitals increasingly adopting strategies to improve cost-efficiency [18,19]. The observed differences in charges by hospital size may reflect underlying structural and organizational differences between institutions, although these specific mechanisms cannot be determined from the present data. Therefore, variation in capital investment structures and fixed costs across hospitals may contribute to differences in patient charges. Of note, larger institutions may face greater fixed costs and administrative complexity [20,21] that translate into higher charges despite comparable procedural delivery. These findings suggest that hospital structural characteristics should be considered in future cost-effectiveness analyses of RA-TKA. Future research should examine specific institutional practices associated with cost variation across hospital sizes to inform scalable cost-containment strategies.

Our study is not exempt from potential limitations. We were unable to control for intraoperative performance, which may influence surgical time and perioperative resource utilization that contribute to overall cost. Additionally, geographic location was not available in the dataset used for this analysis and therefore could not be controlled for. We acknowledge that patient charges may vary by region, and residual confounding related to geographic variation may be present. To provide the best representation of total OR charges, anesthesia, OR block time, and postanesthesia care unit charges were included; however, there may be miscellaneous charges that were not captured in our model. For that reason, the total OR charges reported in the study may be an underrepresentation of actual OR charges for RA-TKA. Similarly, “total charges” for a RA-TKA may be quantified differently across institutions, and itemized billing data were not available within the database utilized for the study. Nevertheless, the use of a large, nationally representative database provides substantial value. The dataset offers broader generalizability and a greater depth of RA-TKA-specific information than could be achieved through single-institution analyses alone.

Conclusions

RA-TKAs performed at hospitals with fewer than 300 beds were associated with lower total charges and total OR charges compared to those performed at larger hospitals. These findings suggest an association between institutional size and cost variability in primary RA-TKA procedures. This information may help inform discussions regarding cost variation and strategies aimed at improving efficiency in elective arthroplasty care.

Conflicts of interest

J. Nace is a paid consultant for Microport Orthopedics; receives research support from Microport Orthopedics, Stryker, and United Orthopedic Corporation; is in the editorial or governing board of Journal of Arthroplasty, Journal of the American Osteopathic Medicine Association, Orthopedic Knowledge, Journal of Knee Surgery, and Knee; and is a board member in Arthritis Foundation. M.A. Mont receives royalties from Medicus Works LLC and Stryker; is a paid consultant for 3M, Exactech, Inc, Johnson & Johnson, Kolon TissueGene, Next Science, Pacira, Smith & Nephew, and Stryker; has stock or stock options in CERAS Health, MirrorAR, Peerwell, and USMI; receives research support from Johnson & Johnson, Medtronic, National Institutes of Health (NIAMS & NICHD), Organogenesis, Orthofix, Inc, Patient-Centered Outcomes Research Institute (PCORI), and Stryker; receives royalties, financial or material support from Medicus Works LLC, Up-to Date, Wolters Kluwer Health - Lippincott Williams & Wilkins; is a member of the medical/orthopaedic publications editorial/governing board at Journal of Arthroplasty; and is a board member/committee appointments for American Association of Hip and Knee Surgeons, Hip Society, and Knee Society. R.E. Delanois receives research support from Johnson & Johnson, Biocomposites, CyMedica Orthopedics, Depuy Synthes Product, Flexion Therapeutics, Microport Orthopedics, Orthofix, Patient-Centered Outcomes Research Institute (PCORI), Smith & Nephew, Stryker, Tissue Gene, and United Orthopedic Corporation; is a member of the medical/orthopaedic publications editorial/governing board at Journal of Knee Surgery; and is a board member/committee appointments for Baltimore City Medical Society; all other authors declare no potential conflicts of interest.

For full disclosure statements refer to https://doi.org/10.1016/j.artd.2026.102091.

CRediT authorship contribution statement

Monica Misch: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Conceptualization. Reza Katanbaf: Writing – review & editing, Visualization, Validation, Methodology, Formal analysis, Data curation, Conceptualization. Chase Smitterberg: Writing – review & editing, Writing – original draft, Visualization, Methodology, Conceptualization. Michael A. Mont: Writing – review & editing, Visualization, Validation, Supervision, Conceptualization. Ronald E. Delanois: Writing – review & editing, Visualization, Validation, Supervision, Investigation, Conceptualization. James Nace: Writing – review & editing, Validation, Supervision, Investigation, Conceptualization.