Price Variation in Outpatient Percutaneous Coronary Intervention.
Oct 26, 2026DOI: 10.1161/circulationaha.126.081539Journal ArticleAHA
Written by: Kriyana P ReddyNavneeth MuraliNicholas K SmithSameed Ahmed M KhatanaPaula ChatterjeeAlexander C FanaroffPeter W GroeneveldJay GiriAshwin S Nathan
Find these authors on Roon
AF
Dr. Alexander Fanaroff
Interventional Cardiology · UPenn
JG
Dr. Jay Giri
Interventional Cardiology · HUP
Methods
This study was exempt from review by the institutional review board at the University of Pennsylvania. Data used for this study are covered by a Data Use Agreement and are not available for distribution by the authors. Price data were obtained from Turquoise Health, which standardizes payer-provided machine-readable files into structured data by harmonizing billing codes and prices across heterogeneous file formats.6 We identified commercial and Medicare prices associated with Current Procedural Terminology code 92928 between April 1, 2025, and July 31, 2025. Medicare prices were derived by Turquoise Health using Centers for Medicare & Medicaid Services Inpatient Prospective Payment System/Outpatient Prospective Payment System hospital prices and ASC fee schedules. Commercial prices may not include device or implant costs because some contracts use device carve-outs not identifiable in the available data. Therefore, direct comparisons between Medicare and commercial prices were not performed.Facilities were classified as HOPDs or ASCs using National Uniform Claim Committee Taxonomy codes. Analyses were limited to facilities with ≥1 commercial and ≥1 Medicare price, excluding facilities unlikely to perform PCI using predefined keyword matching. Prices flagged as outliers by Turquoise Health as well as prices 1.5 interquartile ranges below the first quartile or above the third quartile within each payer class were excluded. Data from Dartmouth Atlas, US Census Bureau, and Health Care Cost Institute were used to define hospital referral region (HRR), Census region, metropolitan status, and market concentration.Prices were summarized using medians (interquartile ranges) and ranges. Within-facility commercial price variation was quantified as the ratio of maximum to minimum commercial price. The primary aim was to compare price differences between HOPDs and ASCs. First, price distributions by facility type were compared using Kruskal-Wallis tests. Second, across-facility variation between HOPD and ASC prices was assessed using ratios of median HOPD-to-ASC prices within each HRR. Third, the association between facility type and median commercial prices was evaluated using a generalized linear mixed-effects model with a gamma distribution, log link, random intercepts for HRR, and adjustment for facility type, Census region, metropolitan status, and market concentration. The gamma-log specification was selected given positive and right-skewed prices and because variance was found to increase approximately with the square of the conditional mean (modified Park test variance of power, 1.68 [95% CI, 1.31–2.04]), consistent with a gamma-type variance function. Previous methodologic work recommends selecting models based on observed mean-variance relationship, and empirical comparisons have demonstrated favorable performance of gamma-log models for skewed cost data.7,8 Analyses were performed using R version 4.4.1 with 2-sided <0.05 considered statistically significant.
P
Results
Across 8486 outpatient facilities, we identified 44 871 prices for Current Procedural Terminology code 92928, including 36 285 commercial and 8586 Medicare prices. Commercial prices varied substantially, with a median (interquartile range) of $4149 ($980–$11 460). Medicare prices were comparatively uniform, with a median (interquartile range) of $10 787 ($7103–$11 382). Median Medicare prices were significantly higher in HOPDs than ASCs ($11 341 versus $6874; P<0.001). Within-facility commercial price variation was greater among HOPDs than ASCs, with median maximum-to-minimum commercial price ratios of 6.27 versus 2.21, respectively (P<0.001; Table). Across HRRs, median HOPD-to-ASC commercial price ratios exceeded 10 in some markets, whereas Medicare price ratios were generally <2 (Figure 1). In the adjusted generalized linear mixed-effects model with HRR-level clustering, ASCs had 26% lower median commercial prices than HOPDs (rate ratio, 0.74 [95% CI, 0.7–0.78]; P<0.001). Facilities in micropolitan areas also demonstrated lower commercial prices than metropolitan facilities (rate ratio, 0.39 [95% CI, 0.17–0.92]; P=0.032).
Variation in Commercial and Medicare Prices for Current Procedural Terminology Code 92928
Across-facility variation in commercial and Medicare prices for Current Procedural Terminology code 92928. Each bar represents 1 hospital referral region.
Discussion
In this national analysis of >8000 outpatient facilities reporting prices for PCI, commercial prices demonstrated substantial variation. Within-facility commercial price variation was greater among HOPDs than ASCs, with median maximum-to-minimum price ratios of 6.27 versus 2.21. After adjustment for geography, metropolitan status, and market concentration, ASCs had 26% lower negotiated commercial prices than HOPDs.
These findings highlight important differences between administratively determined Medicare reimbursement and commercially negotiated pricing. Medicare prices clustered within narrow ranges across facilities, whereas commercial prices varied widely both within and across outpatient settings. HOPDs demonstrated significantly greater within-facility variation than ASCs, which may reflect more complex hospital contracting across multiple payers and products. In contrast, ASCs may use more standardized contracting structures across a narrower portfolio of services and payers. Lower prices among ASCs may additionally reflect lower operational and infrastructure costs relative to HOPDs, which generally maintain greater staffing, regulatory, and hospital-based support requirements and may care for more clinically complex patients.9
Our findings may also have implications for ongoing discussions regarding site-neutral payment reform and expansion of outpatient cardiovascular care. Previous studies have demonstrated safety and feasibility of outpatient PCI in selected patients, and ASCs may represent a lower-cost setting for appropriately selected elective procedures. PCI was not included among the high-volume service categories proposed by MedPAC for site-neutral payments in 2023, yet the substantial ASC–HOPD commercial price differences observed in this study suggest potential opportunities to reduce unwarranted spending variation and generate savings as outpatient cardiovascular spending continues to rise.10 Simultaneously, payment reform should consider differences in patient complexity, procedural risk, and quality outcomes across outpatient settings. State-level regulatory frameworks, including certificate-of-need laws and restrictions on ASC PCI programs, may also limit the extent to which potential savings from site-neutral payment policies can be realized uniformly across markets.
This study has several limitations. Price transparency data may incompletely capture facilities performing PCI because of variable compliance with federal reporting requirements and heterogeneous reporting practices. Some extreme commercial prices may reflect differences in contract structure or reporting rather than true reimbursement differences. In addition, commercial contracts may differ in the degree to which device-related costs are bundled within reported prices, thereby limiting direct comparisons between commercial and Medicare prices. Analyses were limited to Current Procedural Terminology code 92928, which represents standard outpatient stent-based PCI and the largest share of posted PCI prices in the database, but findings may not generalize to other PCI procedures. Furthermore, posted prices do not capture professional fees or full episode-level reimbursement. Professional reimbursement represents a distinct payment structure that may independently influence physician incentives regarding site of care and warrants further investigation. We also could not account for differences in case mix, procedural acuity, or facility PCI volume, which may contribute to price differences between ASCs and HOPDs.
Conclusion
Commercial prices for outpatient PCI vary substantially across HOPDs and ASCs, whereas Medicare prices are relatively uniform. HOPDs demonstrate greater commercial price variation and higher negotiated prices than ASCs. These findings may inform ongoing discussions regarding site-neutral payment reform and efforts to reduce unwarranted variation in outpatient cardiovascular spending.
Article Information
None.
AF
Dr. Alexander Fanaroff
Cardiology · Interventional CardiologyUniversity of PennsylvaniaalexanderPhiladelphia, PA
