Abstract
This cohort study explores the role of hospital-level factors in the quality and cost of oncologic care.
Over the past decade, health system mergers and expansions, along with evidence supporting the volume-outcome relationship, have contributed to the centralization of oncologic care.1 While regionalization to high-volume hospitals (HVHs) has been associated with improved outcomes,2 hospital consolidation has also been recognized for reducing competition with surrounding institutions. The association of hospital market competition with the quality and cost of oncologic care remains poorly understood.3,4
Methods
This retrospective cohort study identified all patients 18 years or older within the Healthcare Cost and Utilization Project State Inpatient Databases (of Arizona, California, Florida, Maryland, New Jersey, New York, and Washington) who underwent elective resection for nonmetastatic cancer of the colon, esophagus, stomach, liver, lungs, or pancreas from 2016 to 2021. The University of California, Los Angeles, Institutional Review Board deemed this study exempt from review and informed consent because it used deidentified data. We followed the STROBE reporting guideline.
Hospital market competition was computed using the Herfindahl-Hirschman Index (HHI). Competition was defined for each hospital using a variable radius comprising 75% of discharges across all service lines. Patients were stratified into the competitive (HHI <0.15 indicating unconcentrated) or noncompetitive (HHI ≥0.15 indicating moderately to highly concentrated) market cohorts. We identified hospitals as HVHs or non-HVHs based on Leapfrog criteria (≥40 lung, ≥20 pancreatic, ≥20 esophageal cancer resections per year) or annual procedural volume (top decile: ≥4 hepatic, ≥59 colon, ≥10 gastric cancer resections per year).
The primary end point was composite morbidity (in-hospital mortality or any major complication). A secondary consideration was total hospitalization expenditures.
Multivariable logistic and linear regression models were developed. Model covariates were automatically selected using elastic net regularization to optimize fit and reduce bias. Outputs were reported as adjusted odds ratios (AORs) or β coefficients with 95% CIs. Statistical significance was P < .05. Analyses were performed from July 2024 to April 2025 using Stata 18.0 (StataCorp).
Results
Among 131 357 patients (median [IQR] age, 69 [60-76] years; 66 310 males [50%]), 82 080 (62%) were treated at competitive-market hospitals. The competitive and noncompetitive groups were of similar age, sex, and comorbidity burden, but the competitive group more frequently underwent pancreatectomy for pancreatic cancer and received care at HVHs (Table).
Table. Demographic, Clinical, and Hospital Characteristics.
| Patients, No. (%)a | P valueb | ||
|---|---|---|---|
| Competitive market | Noncompetitive market | ||
| All patients | 82 080 (62) | 49 277 (38) | NA |
| All hospitals | 1873 (52) | 1750 (48) | NA |
| Age, median (IQR), y | 68 (60-76) | 69 (60-76) | <.001 |
| Sex | |||
| Female | 40 607 (49) | 24 440 (50) | .66 |
| Male | 41 473 (50) | 24 837 (50) | |
| Elixhauser Index, median (IQR) | 3 (2-4) | 3 (2-4) | <.001 |
| Surgical approach | |||
| Open | 32 139 (39) | 20 819 (42) | <.001 |
| Minimally invasive | 49 941 (61) | 28 458 (58) | |
| Cancer type | |||
| Colon | 37 920 (46) | 26 567 (54) | <.001 |
| Esophageal | 1108 (1) | 551 (1) | |
| Gastric | 6525 (8) | 3143 (6) | |
| Hepatocellular | 3787 (5) | 1373 (3) | |
| Lung | 23 733 (29) | 13 765 (28) | |
| Pancreatic | 9007 (11) | 3878 (8) | |
| Race and ethnicityc | |||
| Asian or Pacific Islander | 7493 (9) | 2002 (4) | <.001 |
| Black | 5703 (7) | 3605 (7) | .01 |
| Hispanic | 14 018 (17) | 5366 (11) | <.001 |
| White | 51 916 (63) | 37 111 (75) | <.001 |
| Otherd | 1927 (4) | 4044 (5) | <.001 |
| Median household income percentile | |||
| 76th-100th | 16 939 (30) | 10 852 (26) | <.001 |
| 51st-75th | 14 122 (25) | 11 070 (26) | |
| 26th-50th | 14 159 (25) | 11 762 (28) | |
| 0-25th | 11 559 (20) | 8288 (20) | |
| Insurance coverage | |||
| Private | 25 920 (32) | 14 961 (30) | <.001 |
| Medicare | 48 233 (59) | 29 275 (59) | .02 |
| Medicaid | 5788 (7) | 3639 (7) | .02 |
| Not insured or other payer | 1610 (2) | 1100 (2) | .001 |
| Treatment history | |||
| Chemoradiation | 6262 (8) | 2884 (6) | <.001 |
| Treated at HVH | 44 397 (54) | 21 572 (44) | <.001 |
| Comorbidities | |||
| CHF | 4076 (5) | 3030 (6) | <.001 |
| Cardiac arrhythmia | 8942 (11) | 5842 (12) | <.001 |
| Diabetes | 18 758 (23) | 11 261 (23) | .99 |
| Hypertension | 46 588 (57) | 29 232 (59) | <.001 |
| Liver disease | 5319 (6) | 2385 (5) | <.001 |
Abbreviations: CHF, congestive heart failure: HVH, high-volume hospital.
Baseline characteristics for patients treated at hospitals in high-competition (competitive) markets and low-competition (noncompetitive) markets.
Statistical significance was set at P < .05.
Race and ethnicity were obtained from the National Cancer Center Database. These data were collected in this study because of the well-described association of race and ethnicity with access to and outcomes following oncologic surgical care.
Other category included American Indian, Aleutian, or Eskimo as well as other (as categorized by the National Cancer Database.
After risk adjustment, treatment at hospitals in competitive markets was associated with an incremental increase in major morbidity risk (AOR, 1.07; 95% CI, 1.02-1.11; P = .002) (Figure). Stratifying by hospital volume, this association remained at non-HVHs (AOR, 1.09; 95% CI, 1.03-1.16; P = .002) but not HVHs (AOR, 1.05; 95% CI, 0.99-1.11; P = .10).
Figure. Association of Market Competition With Clinical and Financial End Points of Elective Cancer Resection.
A, The proportion of patients treated at hospitals in competitive markets increased from 56% in 2016 to 67% in 2018 and then decreased to 62% in 2021 (P for trend < .001). B, Greater market competition was associated with higher morbidity risk among non–high-volume hospitals (HVHs) but with similar morbidity among HVHs. C, Treatment at hospitals in competitive markets was associated with significantly greater risk-adjusted expenditures, particularly for colon, pulmonary, and pancreatic cancer resections. D, Considering the interaction of market competition with hospital volume, treatment in competitive markets remained associated with greater per-patient expenditures among HVHs and non-HVHs.
Moreover, care in competitive markets was associated with a $1172 (95% CI, $802-$1541; P < .001) increase in per-patient expenditures. Increased market competition remained associated with higher expenditures among HVHs and non-HVHs.
Discussion
Treatment in competitive markets was associated with higher morbidity risk and costs following elective cancer resection. Increased morbidity was limited to low-volume hospitals, while higher costs were observed across both non-HVHs and HVHs. These findings underscore the role of hospital-level factors in delivering high-value oncologic care in competitive settings.5,6
Hospital-level differences in outcomes and costs may be associated with variability in available resources, care coordination, and adherence to standardized clinical practices. Non-HVHs in competitive environments may be less well equipped to manage complex cancer cases, leading to worse outcomes.2 Meanwhile, increased costs across all institutions suggest that competition may drive greater resource utilization, such as additional tests, procedures, or services, regardless of clinical benefit. In contrast, centralization may promote standardized care pathways, optimized perioperative and surgical management, and reduced use of unnecessary services.6
Study limitations include the lack of granular radiographic, laboratory, and staging data. Yet, we present a large, all-payer, multistate analysis of hospital market competition and oncologic surgery outcomes. Our findings suggest that, by reducing competition, centralization may increase both quality and value of care. Future work should explore these associations directly and support the dissemination of effective care models across hospital settings.
Data Sharing Statement
References
- 1.Cutler DM, Scott Morton F. Hospitals, market share, and consolidation. JAMA. 2013;310(18):1964-1970. doi: 10.1001/jama.2013.281675 [DOI] [PubMed] [Google Scholar]
- 2.Birkmeyer JD, Sun Y, Wong SL, Stukel TA. Hospital volume and late survival after cancer surgery. Ann Surg. 2007;245(5):777-783. doi: 10.1097/01.sla.0000252402.33814.dd [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Thumma SR, Dualeh SHA, Kunnath NJ, Bonner SN, Ibrahim AM. Outcomes for high-risk surgical procedures across high- and low-competition hospital markets. JAMA Surg. 2023;158(10):1041-1048. doi: 10.1001/jamasurg.2023.3221 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Adler JT, Sethi RKV, Yeh H, Markmann JF, Nguyen LL. Market competition influences renal transplantation risk and outcomes. Ann Surg. 2014;260(3):550-556. doi: 10.1097/SLA.0000000000000896 [DOI] [PubMed] [Google Scholar]
- 5.Cerullo M, Chen SY, Dillhoff M, Schmidt C, Canner JK, Pawlik TM. Association of hospital market concentration with costs of complex hepatopancreaticobiliary surgery. JAMA Surg. 2017;152(9):e172158. doi: 10.1001/jamasurg.2017.2158 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dafny LS, Lee TH. The good merger. N Engl J Med. 2015;372(22):2077-2079. doi: 10.1056/NEJMp1502338 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
