Cost-Effectiveness of Computer-Assisted Fluoroscopic Navigation in Primary Total Hip Arthroplasty Compared With Robotic or Manual Techniques Authors: Jenna Bernstein, Jean-Baptiste Trouiller, Mina Kabiri, William G. Hamilton Published: 2026 ABSTRACT Background:Enabling technologies used in primary total hip arthroplasty (THA), including computer-assisted fluoroscopic navigation (cTHA) and robotic-assisted THA (rTHA), aim to optimize the accuracy and precision of surgery to improve patient outcomes. This study evaluates the cost-effectiveness of cTHA compared with manual THA (mTHA) and rTHA for primary THA in the United States, from an “all-payer” perspective.Methods:A Markov state-transition model was developed to estimate costs and quality-adjusted life year gains over a 1-year time horizon. Inputs were primarily derived from published literature that analyzed data from an electronic healthcare database between 2016 and 2021. The model population consisted of patients undergoing primary THA, with a mean age of 66 years. The model included six health states: post-THA, readmission without revision, readmission with revision, post-readmission, post-revision, and death. Sensitivity analyses, including deterministic, probabilistic, and scenario analyses, were conducted to test the robustness of the results.Results:cTHA demonstrated cost savings of $1595 compared with rTHA and $949 compared with mTHA. Savings with cTHA were primarily due to shorter length of stay and fewer readmissions with or without revision. cTHA also showed minor quality-adjusted life year gains of 0.0013 and 0.0009 versus rTHA and mTHA, respectively. These findings establish cTHA as the cost-effective strategy, combining lower costs with slightly improved health outcomes. Results remained robust across all sensitivity analyses.Conclusions:cTHA is the most cost-effective technique for primary THA in the United States, compared with both rTHA and mTHA. Primary total hip arthroplasty (THA) is one of the most common elective orthopaedic procedures performed in the United States.1,2 The annual number of THA procedures has been increasing globally as a result of advancements in surgical technique and implant materials, together with an aging population with higher activity levels in many countries.3,4 Despite the overall success of THA, the rising incidence of complications and revisions poses a notable burden to patients and the healthcare system.5-8 The cost burden of complications after THA are substantial, with dislocations leading to an average increase of nearly $20,000 (USD) per patient within the first year after surgery and costs accumulating beyond the initial year.7 Revision THA procedures are linked to even greater expenses compared with primary THA, with average postoperative outcome costs of $75,037 versus $53,324, respectively.8 Procedural accuracy is imperative in limiting complications and avoiding revision surgery.9-11 Enabling technologies, such as computer-assisted fluoroscopic navigation (cTHA) systems and robotic-assisted systems, are aimed at optimizing implant placement and alignment in THA.12 Robotic-assisted systems use a navigated robotic arm, which aids surgeons in executing optimal implant positioning.13 Computer-assisted fluoroscopic navigation was developed as an alternative to CT-based computer navigation to overcome the time and cost constraints of the CT imaging process.12 Computer-assisted fluoroscopic navigation provides real-time intraoperative feedback and calculations to enhance implant positioning.14 Several published studies assessed the effect of using enabling technologies on patient outcomes and healthcare utilization.15-20 When compared with manual THA (mTHA), cTHA was associated with lower hip-related readmission rates and greater discharge to home.15 While the early use of robotic-assisted THA (rTHA) was associated with technical challenges and complications,12,21 recent studies have reported favorable outcomes with robotic-assisted systems compared with mTHA.16,17 Both enabling technologies are associated with reduced readmission and revision rates compared with mTHA. Inpatient readmissions associated with deep infections (ie, infections involving the implant) are reportedly lower for patients who underwent rTHA compared with cTHA.18 There are some reported differences between cTHA and rTHA in terms of healthcare resource utilization, including operating room time, length of stay, and associated hospital costs between cTHA and rTHA that could affect considerations for use of either technique.15,18-20 With the emergence of value-based care initiatives, improving quality of care and patient outcomes while minimizing costs has become a critical focus for healthcare providers. Enabling technologies for THA are rapidly advancing, and their use has become increasingly popular.22 As such, it is important for physicians and hospitals to select the most optimal technology while considering associated costs. Multiple studies have demonstrated the cost-effectiveness of rTHA compared with mTHA, primarily because of reduced revision risks, improvement in patients' quality of life, and lowering costs (excluding capital equipment costs).23-25 However, the cost-effectiveness of cTHA compared with rTHA or mTHA has not previously been assessed. The purpose of this study was to evaluate the cost-effectiveness of cTHA compared with both rTHA and mTHA in patients undergoing primary THA in the United States. Methods Model Structure and Study Population In this study, a state-transition Markov model was developed in Microsoft Excel 365 (Microsoft) to evaluate the cost-effectiveness of cTHA versus rTHA and cTHA versus mTHA. Markov models use transition probabilities so that patients can move from one discrete health state to another.26 The model population consisted of patients undergoing primary THA, with a mean age of 66 years (SD 10.44) and 57% female. This population was aligned with the two publications used to inform most clinical and economic inputs in the model, which used the Premier Healthcare Database to identify patients between 2016 and 2021.15,18 The analysis included three treatment arms to represent different THA procedures (cTHA, rTHA, and mTHA). The analysis was conducted from the US healthcare payer perspective, as per the guidelines provided by the Institute for Clinical and Economic Review for conducting economic evaluations, including all direct costs relevant to the US healthcare system.27 Indirect costs (eg, productivity loss, absenteeism, presenteeism, and disability payments) were excluded. The Markov model included six health states, defined according to the occurrence of complications leading to readmissions and revisions (Figure 1). All patients started in the “post-THA” health state (ie, they had all undergone a THA procedure) and could either remain in this state (where the patient does not experience any complication) or could experience a complication, which would lead to either a readmission without revision or a readmission with revision.24,25 Readmissions with or without revision could occur because of fracture, dislocation, superficial infection, and hip pain.15,18,24,25 Based on clinical feedback, it was confirmed that general infections in regions of the body excluding the hip are not considered surgery-related, whereas revision surgery is essentially always required for deep infections. Thus, the two complications were not included as states under readmission without revision. Separate health states were not assigned to various reasons for revision because a revision may result from multiple complications. Readmissions and/or revisions were associated with a one-time cost and quality-of-life decrease. After one model cycle, the patient entered the “post-readmission” or “post-revision” states. Given the low number of patients with multiple readmissions and/or revisions, the model assumed that patients could undergo a maximum of one readmission and one revision; this assumption was validated by a clinical expert as multiple revisions are relatively rare and would not affect the model outcomes. Each heath state could lead to death, which was an absorbing state. Model schematic. After THA and based on 3-month transition probabilities, patients could experience readmission for complications (dislocation, fracture, superficial infection, or hip pain) or readmission with revision THA, transitioning to post-readmission or post-revision or transitioning to death. *All health states can lead to death, which is an absorbing state, so patients do not transfer out of it. THA = total hip arthroplasty The Markov model had a cycle length of 3 months to align with data from the literature.15,18 Consistent with good modeling practice guidelines, a half-cycle correction was applied to reflect that transitions between health states occur continuously throughout each model cycle, and not strictly at the beginning or at the end.28 A model time horizon of 1 year was selected for the base-case scenario because of the limited availability of input parameters beyond 1-year follow-up. However, scenario analyses were conducted using 3-month and 5-year time horizons. Costs and outcomes were discounted at 3% per year, per Institute for Clinical Economic Review guidelines.27 Model Assumptions and Input Parameters Model parameters are summarized in Table 1. Clinical and economic inputs were informed by Bernstein et al18 for the cTHA versus rTHA comparison and by Hamilton et al15 for the cTHA versus mTHA comparison. Both publications used the Premier Healthcare Database (2016 to 2021). Baseline covariates were adjusted using fine stratification methods and analyzed using generalized linear models. Table 1Model ParametersItemcTHArTHA18mTHA15cTHA Versus mTHA15 aCohort characteristics Mean age6615 Percentage male43%15Clinical Reasons for readmission Dislocation readmission37%15,1827%52%— Fracture readmission14%15,185%7%— Superficial infection readmission31%15,1825%41%— Hip pain readmission18%15,1843%0%— Readmission rates 3-mo readmission1.26%182.21%—0.69 6-mo readmissionb0.27%180.42%—0.67 9-mo readmissionb0.27%180.42%—0.65 12-mo readmissionb0.27%180.42%—0.63 Long-term readmissionb0.27%180.42%—0.63 Revision rates 3-mo revision0.07%15,180.24%—0.48 6-mo revisionb0.01%180.02%—0.50 9-mo revisionb0.01%180.02%—0.51 12-mo revisionb0.01%180.02%—0.53 Long-term revisionb0.01%180.02%—0.53Utilities Post-THA0.9325 Dislocation0.5425 Superficial infection0.7625 Fracture0.65c Hip pain0.65c Revision0.6425 Post-readmission without revision0.9325 Post-readmission with revision0.7029 dResource use Length of stay (d)1.54181.68—−0.25 Operating room time (min)137.7418156.00—+2.61Costs (inflated to 2024 USD) Length of stay cost per day ($)3,22130 Operating room cost per minute ($)40.7831 Dislocation ($)21,15524 Fracture ($)22,82732 Superficial infection ($)25,83124 Hip pain ($)23,271e Revision ($)45,43625 fcTHA = computer-assisted fluoroscopic navigation, mTHA = manual total hip arthroplasty, rTHA = robotic-assisted total hip arthroplasty, THA = total hip arthroplastyaOdds ratio is reported for all items except for length of stay and operating room time, for which mean differences are reported.bEstimated from reference calibration exercise and assumption.cAssumption − average between infection and dislocation.dSeventy-five percentage of unrevised post-THA.eAverage of all other readmission costs.fAverage of septic and aseptic revision costs. Model Parameters cTHA = computer-assisted fluoroscopic navigation, mTHA = manual total hip arthroplasty, rTHA = robotic-assisted total hip arthroplasty, THA = total hip arthroplasty Odds ratio is reported for all items except for length of stay and operating room time, for which mean differences are reported. Estimated from reference calibration exercise and assumption. Assumption − average between infection and dislocation. Seventy-five percentage of unrevised post-THA. Average of all other readmission costs. Average of septic and aseptic revision costs. Transition Probabilities Readmission types and their occurrence were based on the studies by Bernstein et al and Hamilton et al and reweighted to exclude general infection and deep infection, as per clinical expert opinion.15,18 An average of readmission types from both studies was used for the cTHA arm in each comparison. Readmission rates for cTHA and rTHA were based on the study by Bernstein et al18 and converted into transition probabilities. The 3-month rate from the study by Bernstein et al18 was used for the first model cycle (3-month), while the probabilities for the second (6-month), third (9-month), and fourth (12-month) cycles were estimated assuming a constant evolution, to reach the 1-year rate from the publication (Table S1, https://links.lww.com/JG9/A540). Hamilton et al15 reported comparative data between cTHA and mTHA using odds ratios. The odds ratios from the study by Hamilton et al were applied to the cTHA arm from the study by Bernstein et al to estimate the mTHA readmission risk.15,18 Because Hamilton et al only reported a 3-month and 1-year odds ratio, to estimate the 6-month and 9-month odds ratio, it was assumed that the 3-month odds ratio would gradually converge to the 1-year odds ratio (Table S1, https://links.lww.com/JG9/A540). To estimate readmission risk after 1 year in the scenario analysis, the probabilities and odds ratios at 1 year were used, assuming constant increase in the readmission risk and constant effect across the three arms beyond 1 year (Table S2, https://links.lww.com/JG9/A541). The 3-month revision risk of cTHA and rTHA was sourced from the study by Bernstein et al. Because the 1-year revision risk was lower in the rTHA arm in the study by Bernstein et al than the 3-month risk due to a change in sample size, the 1-year revision risk of rTHA was calculated based on the ratio between the 3-month and 1-year rates for cTHA (Table S3, https://links.lww.com/JG9/A542). The same methodologies and assumptions as for readmission risk were applied to revision risk to estimate the transition probabilities and cumulative probabilities at each time point (Table S4, https://links.lww.com/JG9/A543). Utilities Utility values in the model were dependent on health states and not on the technology used during primary THA because this is not expected to affect a patient's quality of life. Post–primary THA and readmission and revision utility values were sourced from a recent cost-effectiveness analysis conducted in the United States for patients with primary THA.25 Owing to a lack of published data, utility values for fracture and hip pain were calculated as an average quality-of-life decrease between dislocation and infection. The utility value for revision was the average of septic and aseptic revisions reported by Ong et al.25 The model incorporated a decrease in quality of life after readmissions and revisions, for a 3-month period or longer. Patients were assumed to return to their initial post–primary THA quality of life 3 months after a readmission. According to literature and clinical expert opinion, it was expected that revision THA would reduce patients' quality of life and that they would not go back to their initial quality of life. Therefore, utility values after revision were assumed to be 75% of post–primary THA quality of life, based on previous research.29,33 The Ara and Brazier34 regression model was applied to account for age-related quality-of-life decline. Resource Use and Cost Inputs Costs associated with readmissions and revisions were collected from the literature (Table 1). Length of stay and operating room time for cTHA and rTHA were based on the study by Bernstein et al,18 and the mean difference between cTHA and mTHA from the study by Hamilton et al15 was used to estimate the length of stay for mTHA. All costs were converted to 2024 USD using the Consumer Price Index. Analyses The outcomes of the model were expressed in terms of per-patient difference in quality-adjusted life years (QALYs) and difference in costs between treatment arms, which were used to estimate the incremental cost-effectiveness ratio (ie, Δ costs/Δ effectiveness). Sensitivity analyses were conducted to evaluate uncertainty in the model results. The model parameters that were included in the sensitivity analyses included probabilities of the complications and revision, resource use, utilities, and cost data. To determine which parameters had the most notable effect on the costs and QALYs of the base-case scenario, one-way deterministic sensitivity analyses were conducted. Deterministic sensitivity analyses were conducted by varying a single parameter in its 95% confidence interval, if reported, or using baseline values ±10%, while all other parameters remained unchanged. Results of deterministic sensitivity analyses were presented in tornado diagrams, indicating the effect of each model parameter on the cost and QALY difference. Probabilistic sensitivity analyses were conducted by assigning probability distributions to model parameters and repeatedly (ie, 1000 iterations) sampling values from these distributions to estimate the cost-effectiveness ratios. Beta distribution was assigned to probabilities, proportions, and utilities, whereas gamma distribution was assigned to costs and resource use and log-normal distribution was assigned to odds ratios. Results were presented in cost-effectiveness scatter plots. A scenario analysis was conducted to explore different time horizons (3 months and 5 years). Results Base-Case Analyses The base-case analyses found that cTHA reduced costs with a slight increase in QALYs compared with rTHA and mTHA. cTHA demonstrated cost savings of $1595 per patient and minor QALY gains of 0.0013 versus rTHA, and cTHA also demonstrated cost savings of $949 per patient and QALY gains of 0.0009 versus mTHA (Table 2). Savings with cTHA compared with mTHA were attributed to shorter length of stay and fewer readmissions with/without revisions (Table 3). cTHA also demonstrated lower costs compared with rTHA, associated with reduced operating room time, length of stay, and fewer readmissions with/without revisions. Table 2Results of the Base-Case and Scenario AnalysesGroupCosts ($)LYQALYIncremental Costs ($)Incremental LYIncremental QALYICER ($/QALY)Base-case analysis cTHA11,0610.9910.9201———— rTHA12,6570.9910.9188−159500.0013Dominant mTHA12,0110.9910.9192−94900.0009DominantScenario analyses Time horizon: 3 mo cTHA versus rTHA———−1346—0.0004Dominant cTHA versus mTHA———−781—0.0002Dominant Time horizon: 5 yr cTHA versus rTHA———−2125—0.0053Dominant cTHA versus mTHA———−1447—0.0035DominantcTHA = computer-assisted fluoroscopic navigation, ICER = incremental cost-effectiveness ratio, LY = life year, mTHA = manual total hip arthroplasty, QALY = quality-adjusted life year, rTHA = robotic-assisted total hip arthroplasty Results of the Base-Case and Scenario Analyses cTHA = computer-assisted fluoroscopic navigation, ICER = incremental cost-effectiveness ratio, LY = life year, mTHA = manual total hip arthroplasty, QALY = quality-adjusted life year, rTHA = robotic-assisted total hip arthroplasty Table 3Breakdown of Cost Results in the Base-Case AnalysisItemcTHArTHAmTHAIncremental Costs Versus rTHAIncremental Costs Versus mTHALength of stay496054115765−451−805Operating room time561763625511−745106Readmissions with/without revisions484884735−400−251cTHA = computer-assisted fluoroscopic navigation, mTHA = manual total hip arthroplasty, rTHA = robotic-assisted total hip arthroplasty Breakdown of Cost Results in the Base-Case Analysis cTHA = computer-assisted fluoroscopic navigation, mTHA = manual total hip arthroplasty, rTHA = robotic-assisted total hip arthroplasty Sensitivity Analyses Across all deterministic sensitivity analyses, cTHA was associated with cost savings and QALY gains versus rTHA and mTHA (Figure S1 [https://links.lww.com/JG9/A534], Figure S2 [https://links.lww.com/JG9/A535], Figure S3 [https://links.lww.com/JG9/A536], and Figure S4 [https://links.lww.com/JG9/A537]). The deterministic sensitivity analysis demonstrated that the length of stay was the main driver of cost difference between cTHA and both mTHA and rTHA (Figure S1 [https://links.lww.com/JG9/A534] and Figure S2 [https://links.lww.com/JG9/A535]). Varying the length of stay across its 95% confidence interval produced the largest change ($1466 and $1724 savings compared with rTHA and $821 and $1078 of savings compared with mTHA). Importantly, no variation in model input parameters reversed the cost difference (ie, all variations resulted in cost savings for cTHA). This indicated that the cost-effectiveness modeling findings, demonstrating that cTHA remained cost-saving relative to both rTHA and mTHA, were robust across plausible parameter variations. The results from the probabilistic sensitivity analysis (Figure S5, https://links.lww.com/JG9/A538 and https://links.lww.com/JG9/A539) indicated that cTHA achieved cost savings in 100% of 1000 iterations versus both rTHA and mTHA. cTHA was the cost-saving strategy with a slight QALY increase in 93% of simulations versus rTHA and 90% of simulations versus mTHA. Scenario Analyses Scenario analyses were conducted with 3-month and 5-year time horizons, compared with 1 year in the base case. cTHA demonstrated cost savings compared with both rTHA and mTHA across both evaluated scenarios, and the cost-effectiveness of cTHA (cost savings and QALY gain) was maintained (Table 2). As expected, compared with the 1-year base-case results, the changes in costs and QALYs were lower with a 3-month time horizon (−$1346 and 0.0004 for cTHA versus rTHA; −$781 and 0.0002 for cTHA versus rTHA) and higher with a 5-year time horizon (−$2125 and 0.0053; −$1447 and 0.0035) for both cTHA versus rTHA and cTHA versus mTHA, respectively. Discussion cTHA was a cost-effective strategy for primary THA in the United States when compared with both rTHA and mTHA. cTHA was associated with notable cost savings with a minor QALY gain per patient. Cost savings were largely driven by decreased operating room time compared with rTHA and decreased length of stay compared with mTHA. The results were robust to different sensitivity analyses. This study is the first to compare the cost-effectiveness of cTHA with that of rTHA and mTHA. Similar to our results, previous studies have demonstrated the use of enabling technologies (ie, cTHA and rTHA) to be cost neutral or more cost-effective than mTHA, respectively, but seem to have encountered similar challenges in estimating and including capital equipment costs in their analyses.15,24,25 A 5-year cost-effective analysis conducted by Maldonado et al reported lower Medicare and private-payer costs and minor QALY gains for patients undergoing rTHA versus mTHA.24 Similarly, Ong et al demonstrated that rTHA compared with mTHA yielded lower treatment costs and greater utility up to a 5-year time horizon. To note, both studies were conducted at high-volume institutions, and only Ong et al included costs related to the robotic technology (robot implant and robot disposables) in their analysis. Neither analysis included costs associated with acquiring and maintaining the capital equipment for enabling technology. Hamilton et al15 reported total hospital costs for cTHA and mTHA cases to be similar, suggesting that adoption of cTHA may be cost neutral. Both Hamilton et al and Bernstein et al reported that cTHA was associated with lower readmission rates compared with both rTHA and mTHA,15,18 suggesting improvements in clinical outcomes with cTHA and associated costs, despite the upfront cost of implementing the technology. Given the predicted increase in THA procedures in the future,35 saving costs and time will be critical to keeping up with procedure volume and managing surgical workload. Complications arising from THA markedly affect resource use and cost, in particular when hospital readmission or THA revision surgery is needed.5-7 Studies comparing cTHA with mTHA and rTHA have shown that utilization of cTHA lowers hip-related readmission rates and allows patients to be discharged to home sooner,15,18 both of which have important effects on cost. Furthermore, research indicates that although rTHA may offer improved precision compared with manual surgery, it is associated with increased costs due to longer surgical times and the expense of robotic systems.36 This additional financial burden remains a concern, especially when considering the lack of notable improvement in functional outcomes.36 Understanding the comparative cost-effectiveness of cTHA versus mTHA and rTHA is, therefore, important to enable appropriate decision making, balancing outcomes and costs, and considering individual patient factors, surgical expertise, and institutional capabilities. Hospitals and health systems in the United States are more than ever under financial pressure.37 The estimated per-patient cost savings using cTHA in this study are small compared with the total procedure cost, although we anticipate that the total savings in hip arthroplasty become meaningful within a value-based care framework, where financial performance depends on efficiently delivering high-quality outcomes. As an example, beginning in 2026, the Transforming Episode Accountability Model38 will hold hospitals accountable for total 30-day episode spending in lower extremity joint arthroplasty, with reconciliation payments or losses tied to whether actual costs fall below or exceed Centers for Medicare & Medicaid Services' target reimbursement. Because hip arthroplasties are high-volume procedures, modest per-case savings quickly accumulate, reducing the health systems' financial burden within bundled payment models, such as the Transforming Episode Accountability Model. This study has some limitations. The analyses used comparative data derived from the US population and applied appropriate adjustment methods for the analysis. It is important to acknowledge that the model shares inherent limitations of the real-world evidence studies used for inputs. Moreover, it was not possible to capture factors such as procedure complexity, surgeon experience, surgical approach (ie, anterior versus posterior THA), technology learning curves, and regional variations in cost structures in the model. These factors may affect model outcomes. For example, procedures with lower complexity and/or performed by a more experienced surgeon or a higher volume surgeon, or at a higher volume institution, may lead to better outcomes (eg, no complications and reduced pain), contributing to a shorter length of stay and reduced readmission rates, as was found for cTHA in previous studies.15,18 Nonetheless, the model included factors that contribute highly to costs such as readmissions and revisions, ensuring that the most costly complications were considered; thus, we do not anticipate that the factors excluded would greatly affect the results. Owing to limited availability of input parameters beyond 1 year of follow-up, a 1-year time horizon was selected for the base-case scenario. Using the available data, assumptions were made to predict inputs beyond 1 year to enable a 5-year scenario analysis. cTHA continued to demonstrate cost savings compared with both rTHA and mTHA in the 5-year scenario analyses. Another limitation to note was that the model did not include the capital and disposable costs associated with the technologies themselves because these can vary markedly between hospitals and are often not disclosed. To address this, the incremental savings from the use of cTHA in comparison with rTHA and mTHA were presented, allowing providers to assess whether cTHA would yield savings in their specific settings. The cost saving in this context represents the “breakeven cost”: if the additional costs of implementing cTHA are lower than the realized savings compared with the alternative technology ($1595 per patient versus rTHA or $949 versus mTHA), cTHA would provide cost savings. If the additional costs are higher, then cTHA may not be an option that provides cost savings for that center. Conclusion This study demonstrated that cTHA is a cost-effective strategy (delivering cost savings and modest QALY gains) compared with both rTHA and mTHA, with results remaining robust across all sensitivity and scenario analyses. These findings may inform physicians and hospitals to consider the optimal technology for THA.