Background
The regionalization of surgical care has been proposed as a strategy to improve clinical outcomes by leveraging the volume–outcome relationship.1,2 Although hospital system mergers and acquisitions are primarily driven by financial and regulatory factors, they are often presented to patients and policymakers as beneficial to patient care.3,4 The proposed benefits include claims of enhanced care coordination among hospitals and the potential for increased efficiency and investment in advanced technology due to economies of scale.5–7 Merging of hospital platforms and health care systems may also enhance the ability to deliver coordinated, interdisciplinary patient care.8 Affiliation with a prominent regional hospital system is often marketed by smaller hospitals as an opportunity for patients to receive the same high standard of care locally as at the system’s flagship hospital.9,10 This “brand sharing” with the regional main tertiary/quaternary hospital provides affiliates with the opportunity to set themselves apart from neighboring hospitals.
The impact of hospital flagship affiliation on health care is complex and not fully understood. Notably, the Federal Trade Commission has expressed concerns that hospital consolidation may drive up prices, primarily due to heightened bargaining power, reduced competition, diminished price transparency, patient redirection toward higher-cost facilities, and barriers to entry for new contenders.11–13 Nevertheless, other research has suggested that hospital consolidation across health care systems may yield cost efficiencies and improve patient outcomes due to streamlining care delivery and promoting continuity of care.8 In fact, there exists a common perception among patients that care provided at hospitals affiliated with a major regional system is comparable to the care received at the system’s main flagship hospital.10,14–16 In turn, patients may seek care at flagship-affiliated hospitals rather than hospitals outside a flagship system.10,14–16
Previous studies have primarily examined the benefits of hospital mergers in terms of outcomes and financial performance before and after affiliation.5,17–19 However, the impact of hospital mergers and system affiliations has not been investigated relative to complex oncologic surgical procedures. These complex elective operations are particularly sensitive to hospital affiliation and mergers due to differences in outcomes, patient demographics, costs, and care quality across hospital markets.20 Furthermore, complication rates for these surgical procedures can differ significantly based on the hospital where they are performed.21 Moreover, the question arises whether hospitals within major regional systems, including those that are not the flagship facility, deliver better surgical outcomes and financial performance than hospitals unaffiliated with these systems. To the best of our knowledge, this is the first work to holistically explore clinical outcomes and financial expenditures associated with complex cancer surgery across hospital flagship systems, flagship hospitals, and flagship affiliates in 35 of the nation’s largest hospital referral regions (HRRs).
Methods
Data Source and Study Population
Data were obtained from 100% Medicare Standard Analytic Files claims by CMS, covering fee-for-service beneficiaries enrolled in both Medicare Parts A and B. Patients aged ≥65 years who underwent complex oncologic procedures—defined as surgical resection for esophageal, lung, gastric, liver, pancreatic, biliary, colon, or rectal malignancies—between 2018 and 2021 were identified using ICD-10 diagnosis codes. Index procedural codes were used to identify patients who underwent surgical resection after diagnosis. Patients were excluded if they were not enrolled in Medicare Parts A and B during the surgical episode month, received additional payments from an HMO, or had missing data. For patients undergoing multiple surgical procedures, the first operation was considered the analytic procedure. The study was approved by the Institutional Review Board of The Ohio State University, and informed consent for limited data was waived.
Hospital characteristics including region, teaching status, cancer program accreditation, and bed size were merged using the American Hospital Association annual survey data.22 Procedure-specific volume for each hospital was calculated and categorized into tertiles (low, moderate, high), as outlined previously.23 Data on social vulnerability index at the county level were sourced from the CDC’s Agency for Toxic Substances and Disease Registry, providing a recognized measure of community susceptibility to external stressors.24 Patient comorbidities were identified using the Elixhauser Comorbidity Software Refined for ICD-10-CM codes.25
Hospital Systems
The Agency for Healthcare Research and Quality (AHRQ) maintains a database of health systems in the United States, defining hospitals as entities comprising “at least one hospital and at least one group of physicians that provides comprehensive care…who are connected with each other and with the hospital through common ownership or joint management.”26 Using the 2018 AHRQ list, “hospital systems” were defined as medical systems with a minimum of 2 acute care hospitals within the same HRR. Among the 306 HRRs in the United States, 35 met the criteria for volume (≥20,000, both medical and surgical) and system classification. Within each HRR, the “flagship hospital” was designated as the largest major Council of Teaching Hospitals and Health Systems (COTH) hospital, characterized by the highest combined medical and surgical patient volume within the same HRR. The “flagship system” was defined as the system encompassing the flagship hospital. Consequently, each HRR was defined as having only one flagship hospital and one flagship system (see the Appendix and Supplementary Tables S9 and S10, available online in the supplementary material). These potential controls could originate from unaffiliated hospitals, hospitals associated with other academic centers within the same HRR, or even other major COTH hospitals within the same HRR but not part of the flagship system. Certain HRRs contained multiple major COTH hospitals. Patients from all such major COTH hospitals not classified as part of the flagship system were included as potential controls, as previously described.27 A stability analysis was performed to evaluate the association between flagship hospital status and outcomes after excluding HRRs with other major COTH (Supplementary Table S7).
Outcomes of Interest
The primary outcome of interest was 30-day mortality. Secondary outcomes included 30-day readmission, perioperative complications, an extended length-of-stay (LoS), as well as index and 90-day postdischarge expenditures. Perioperative complications were determined using previously validated ICD-10-CM and ICD-10-PCS codes.28,29 Complications included minor infections (urinary tract infections, surgical site infections, and Clostridium difficile infections), major infections (sepsis, ventilator-associated pneumonia, and drug-resistant infections), transient ischemic attacks, cerebrovascular accidents, myocardial infarctions, venous thromboembolic events (deep vein thrombosis and pulmonary embolism), and disseminated intravascular coagulation. LoS was defined as time elapsed from the date of admission to the date of discharge, with an extended LoS identified as the index hospitalization with LoS >75th percentile for the respective procedure. Expenditures included Medicare’s portion in US dollars and were adjusted by wage index, indirect medical education, and disproportionate share hospital to allow for equitable comparisons.30
Statistical Analysis
Continuous variables were reported as medians with interquartile ranges (IQRs), and discrete variables were reported as frequencies with percentages. Univariable comparisons were performed using the Wilcoxon rank-sum test for continuous variables, and chi-square test for categorical variables. Balance measurements between patients at flagship system hospitals and other hospitals outside the flagship system were calculated for each covariate using absolute standardized differences (ASDs), with a cutoff of 10% (0.1) representing a meaningful difference.31
A balanced cohort was obtained by pairing exposed and unexposed patients in a 1:1 ratio through propensity score matching, a method commonly used to mitigate bias stemming from observed confounding factors.32 Propensity scores, representing the likelihood of undergoing care at a flagship system hospital, were computed for each patient using multivariable logistic regression analysis after adjusting for all covariates. Subsequently, a greedy nearest-neighbor matching approach was applied, wherein patients at a flagship system hospital (the exposure group) were matched with similar patients from the unexposed group based on their propensity scores after restriction to a caliper distance of 0.2 of the pooled standard deviation of the logit of the propensity score, in accordance with recommended guidelines.33,34 To evaluate balance between the exposed and unexposed groups in this propensity score–matched cohort, the ASD for each covariate was assessed, with values under 0.1 indicative of acceptable balance.31,35 Separate propensity score matching was performed between flagship hospitals and controls, affiliate hospitals and controls, and flagship and affiliate hospitals. Covariates used in propensity score matching are listed in the Supplementary Tables.
Four primary analyses were conducted using matched pairs of patients undergoing identical complex surgical procedures. Initially, all matched pairs were analyzed to determine whether patients treated at any flagship system hospital had a lower risk of mortality compared with those treated at hospitals outside the flagship system. Differences in individual binary outcomes within matched pairs were assessed using the methodology described by Fleiss et al.36
Results were reported as odds ratios (OR) with 95% confidence intervals. All statistical analyses were derived from 2-tailed tests, and a P value of <.05 was considered statistically significant. The analyses were performed using Stata, version 18.0 (StataCorp LLC).
Results
Baseline Characteristics
A total of 159,909 patients underwent a complex oncologic operation (esophagus: n=1,841 [1.2%]; lung: n=45,096 [28.2%]; gastric: n=4,611 [2.9%]; hepatopancreatobiliary: n=15,556 [9.7%]; colorectal: n=92,132 [57.6%]). Differences in patient characteristics in the prematched cohort are noted in Supplementary Table S1. Due to imbalances in several variables with an ASD >0.1, confounder adjustment was performed using propensity score matching. After matching, a 1:1 cohort of 55,335 patients who underwent complex surgical oncologic procedures at a flagship system hospital (median age, 73.0 years [IQR, 69.0–79.0]; n=29,381 [53.1%] female) was paired with 55,335 patients receiving complex surgical care at a hospital outside a flagship system (median age, 73.0 years [IQR, 69.0–79.0]; n=29,274 [52.9%] female) across 35 of the largest HRRs. After matching, the groups were well-balanced with respect to covariates (Supplementary Table S1). Similar matching was performed between flagship hospitals and controls, affiliate hospitals and controls, and flagship and affiliate hospitals to obtain well-balanced cohorts for matched pair analysis (Supplementary Tables S2–S4). Differences in baseline characteristics between HRRs that include flagship hospitals and those do not are highlighted in Supplementary Table S6.
Compared with their matched counterparts, flagship system hospitals had higher procedural volumes, more beds, and a higher likelihood of being a teaching institution, being accredited by the Joint Commission, and offering more nursing resources (Table 1). Of note, these differences in hospital system characteristics were predominantly attributable to the flagship hospitals themselves, whereas affiliate hospitals were more likely to have characteristics similar to the matched unaffiliated hospitals. For instance, flagship hospitals had a mean of 833 beds versus 253 for matched controls, whereas affiliate hospitals had a mean of 191 beds versus 180 for matched controls (P<.05). Interestingly, more flagship and affiliate hospitals were in the Northeast (18.7% vs 11.7%) and Midwest (34.6% vs 28.9%) regions compared with matched controls (P<.05) (Table 1). Details of hospital level services are noted in Supplementary Table S5.
Hospital-Level Characteristics Across Flagship Hospital Systems
| Variables | Total n (%) |
Flagship System | Flagship Hospital | Affiliate Hospital | |||
|---|---|---|---|---|---|---|---|
| No n (%) |
Yes n (%) |
No n (%) |
Yes n (%) |
No n (%) |
Yes n (%) | ||
| Total, N | 3,581 | 2,486 | 1,113 | 1,402 | 33 | 2,424 | 1,083 |
| Hospital region | |||||||
| Midwest | 1,181 (30.7) | 718 (28.9) | 386 (34.6) | 373 (26.6) | 10 (30.3) | 700 (28.9) | 375 (34.6) |
| Northeast | 529 (13.7) | 292 (11.7) | 209 (18.7) | 175 (12.5) | 8 (24.2) | 292 (12.0) | 205 (18.9) |
| South | 1,404 (36.5) | 934 (37.6) | 372 (33.3) | 532 (37.9) | 11 (33.3) | 911 (37.6) | 357 (33.0) |
| West | 738 (19.2) | 542 (21.8) | 150 (13.4) | 322 (23.0) | 4 (12.1) | 521 (21.5) | 146 (13.5) |
| Hospital volumea | |||||||
| Low | 2,275 (63.1) | 1,612 (64.8) | 663 (59.4) | 757 (54.0) | 3 (9.1) | 1,668 (68.8) | 738 (68.1) |
| Moderate | 853 (23.7) | 571 (23.0) | 282 (25.2) | 442 (31.5) | 6 (18.2) | 575 (23.7) | 268 (24.7) |
| High | 475 (13.2) | 303 (12.2) | 172 (15.4) | 203 (14.5) | 24 (72.7) | 181 (7.5) | 77 (7.1) |
| Number of beds, mean [SD] | 187 [210] | 177 [201] | 210 [226] | 253 [230] | 833 [460] | 180 [203] | 191 [184] |
| Teaching hospital | 293 (7.5) | 496 (20.0) | 263 (23.5) | 325 (23.2) | 31 (93.9) | 163 (6.7) | 85 (7.8) |
| Cancer program accreditation by the Joint Commission | 2,574 (67.9) | 1,677 (69.0) | 812 (74.8) | 1,102 (79.9) | 31 (93.9) | 1,658 (69.9) | 780 (74.2) |
| Nurse-to-bed ratio, mean [SD] | 1.7 [1.2] | 1.7 [1.3] | 1.7 [0.9] | 1.8 [1.4] | 2.8 [1.0] | 1.7 [1.3] | 1.7 [0.9] |
Univariable comparisons were performed using the Wilcoxon rank-sum test for continuous variables, and chi-square test of independence for categorical variables.
Procedure-specific volume for each hospital was calculated and categorized into tertiles (low, moderate, and high).
Hospital Flagship Systems Status and Outcomes
Patients who underwent a high-risk surgical procedure at a flagship hospital had lower 30-day mortality (2.76% vs 3.82%; difference, −1.06% [95% CI, −1.62% to −0.50%]; P<.001) and perioperative complications (19.86% vs 21.40%; difference, −1.54% [95% CI, −2.77% to −0.31%]; P=.014) than matched controls treated outside a flagship system after the same surgical procedure. Notably, there were no differences in extended LoS and 30-day readmission among flagship hospitals and matched controls (Table 2, Model 2). Moreover, compared with matched controls, patients who underwent cancer surgery at flagship system hospitals had higher index hospitalization expenditures (difference, +$3,426 [95% CI, $2,792 to $4,060]; P<.001) and similar 90-day postdischarge expenditures (difference, +$1,009 [95% CI, $−9 to $2,027]; P=.051). The association between 30-day mortality and flagship system status relative to comorbidity burden is depicted in Figure 1.
Association Between Flagship Hospital System and Postoperative Outcomes
| Model 1 | Flagship Hospital System | Control | Rate Difference | 95% CI | P Valuea |
|---|---|---|---|---|---|
| 30-day mortality | 4.23% | 4.88% | −0.65 | −0.89 to −0.40 | <.001 |
| 30-day readmission | 13.43% | 13.78% | −0.35 | −0.05 to 0.75 | .089 |
| Perioperative complicationsb | 23.80% | 24.05% | −0.25 | −0.75 to 0.26 | .334 |
| Extended length-of-stay | 21.78% | 21.17% | 0.61 | 0.13 to 1.10 | .013 |
| Index expenditure | $21,011 | $20,016 | $995 | $797 to $1,193 | <.001 |
| 90-day postdischarge expenditure | $8,828 | $8,297 | $532 | $324 to $739 | <.001 |
| Model 2 | Flagship Hospital | Control | Rate Difference | 95% CI | P Valuea |
| 30-day mortality | 2.76% | 3.82% | −1.06 | −1.62 to −0.50 | <.001 |
| 30-day readmission | 13.53% | 13.12% | 0.41 | −0.67 to 1.49 | .454 |
| Perioperative complicationsb | 19.86% | 21.40% | −1.54 | −2.77 to −0.31 | .014 |
| Extended length-of-stay | 20.16% | 19.21% | 0.94 | −0.28 to 2.17 | .131 |
| Index expenditure | $23,488 | $20,062 | $3,426 | $2,792 to $4,060 | <.001 |
| 90-day postdischarge expenditure | $8,986 | $7,976 | $1,009 | −$9 to 2,027 | .051 |
| Model 3 | Affiliate Hospital | Control | Rate Difference | 95% CI | P Valuea |
| 30-day mortality | 4.46% | 4.79% | −0.32 | −0.58 to −0.07 | .013 |
| 30-day readmission | 13.82% | 13.29% | 0.53 | 0.10 to 0.96 | .016 |
| Perioperative complicationsb | 24.42% | 23.64% | 0.78 | 0.28 to 1.29 | .002 |
| Extended length-of-stay | 22.04% | 20.87% | 1.17 | 0.68 to 1.66 | <.001 |
| Index expenditure | $20,880 | $19,932 | $947 | $758 to $1,136 | <.001 |
| 90-day postdischarge expenditure | $8,802 | $8,193 | $609 | $392 to $826 | <.001 |
| Model 4 | Flagship Hospital | Affiliate Hospital | Rate Difference | 95% CI | P Valuea |
| 30-day mortality | 2.76% | 3.29% | −0.53 | −1.06 to −0.00 | .049 |
| 30-day readmission | 13.53% | 13.56% | −0.03 | −1.11 to 1.06 | .962 |
| Perioperative complicationsb | 19.86% | 21.30% | −1.43 | −2.66 to −0.21 | .022 |
| Extended length-of-stay | 20.16% | 18.90% | 1.26 | 0.04 to 2.49 | .044 |
| Index expenditure | $23,488 | $20,925 | $2,562 | $2,027 to $3,098 | <.001 |
| 90-day postdischarge expenditure | $8,986 | $8,720 | $265 | −$334 to $865 | .386 |
P<.05 was considered statistically significant.
Complications included minor infections, major infections, transient ischemic attacks, cerebrovascular accidents, myocardial infarctions, venous thromboembolic events, and disseminated intravascular coagulation.
Association between predicted 30-day mortality, flagship system status, and comorbidity burden among patients undergoing surgical resection for cancer.
Citation: Journal of the National Comprehensive Cancer Network 23, 5; 10.6004/jnccn.2024.7096
Similar to flagship systems as a whole and flagship hospitals specifically, patients receiving complex surgical cancer care at affiliate hospitals within flagship systems experienced a lower risk of 30-day mortality compared with control patients undergoing the same procedure at hospitals outside a flagship system (4.46% vs 4.79%; difference, −0.32% [95% CI, −0.58 to −0.07]; P=.013) (Table 2, Model 3). Unlike flagship systems, affiliate hospitals had a slightly higher incidence of an extended LoS (22.04% vs 20.87%; difference, +1.17% [95% CI, 0.68 to 1.66]; P<.001), perioperative complications (24.42% vs 23.64%; difference, +0.78% [95% CI, 0.28 to 1.29]; P=.002), and 30-day readmission (13.82% vs 13.29%; difference, +0.53% [95% CI, 0.10 to 0.96]; P=.016); however, this difference was negated after multiplicity adjustment (both adjusted P>.05). Compared with matched controls, patients treated at affiliate system hospitals did have higher index hospitalization expenditures (difference, +$947 [95% CI, $758 to $1,136]; P<.001) and 90-day postdischarge expenditures (difference, +$609 [95% CI, $392 to $826]; P<.001). To further compare surgical outcomes at flagship hospitals versus affiliate hospitals, propensity score matching was performed to form flagship–affiliate matched pairs (Table 2, Model 4). Moreover, analysis was conducted to evaluate longer-term mortality at 90 days (Supplementary Table S8) across different hospital systems.
Discussion
Higher surgical volume in hospitals has been linked to better patient outcomes, such as reduced mortality, fewer complications, and improved survival among patients with cancer.1,37 In turn, hospital systems have frequently asserted that mergers and acquisitions, which have notably increased in recent years, are linked to enhanced quality of care due to improved care coordination and economies of scale.3,8,17–19 To date, few, if any, studies have examined whether flagship system affiliation is associated with adverse clinical and financial outcomes for patients and payers. Notably, we identified only one study by Ramadan et al27 that explored mortality differences across flagship hospital system affiliation for inpatient general surgical procedures. This study used a difference-in-difference (DiD) design that does not align with standard DiD analysis for assessing the impact of a specific intervention by comparing exposed and unexposed groups. It also lacked validation through parallel trends analysis and employed matching, which may increase bias in DiD analyses. Moreover, the study by Ramadan et al27 examined a single outcome measure, 30-day mortality, and demonstrated that only flagship hospitals provided a modest mortality benefit. Expanding on this work, our current study is important because it specifically evaluates whether hospital flagship systems, flagship hospitals, and flagship affiliates are associated with surgical quality using multiple clinical outcome measures and cost of care among patients undergoing complex cancer surgery, which are particularly sensitive to hospital mergers and acquisitions due to differences in outcomes, patient demographics, costs, and care quality across different care markets.20 Complication rates also vary widely depending on the hospital performing the procedure.21 Using matched pairs for each individual cohort from a nationally representative dataset and matched pair analysis methodology outlined by Fleiss et al,36 the current study demonstrated that Medicare beneficiaries who underwent complex oncologic surgical procedures at flagship system hospitals—including both flagship and affiliate hospitals—had a lower risk of 30-day mortality compared with their counterparts treated at hospitals outside the flagship system undergoing the same oncologic procedure. However, these differences in mortality were primarily driven by flagship hospitals, specifically the “brand-name” hospital within each flagship system. Collectively, the data suggest that the main flagship hospital, rather than the local affiliate hospital, was the main driver of differences in outcomes among flagship and non–flagship-affiliated health care systems. In addition, these better postoperative outcomes were associated with higher expenditures at the main flagship hospital.
Despite advances in cancer care delivery nationwide, significant differences persist among different populations of society.14–16,38 For instance, Sheetz et al16 reported marked variability in surgical outcomes between hospitals recognized on the US News & World Report Honor Roll and those in the same system that were not designated as honor roll hospitals. In this study, outcomes also varied based on flagship versus affiliate status, suggesting that multihospital networks may not be leveraging all resources across the system to optimize clinical care.16 In the current study, prior to matching, flagship and affiliated hospitals in flagship systems were less likely to care for patients residing in areas with high social vulnerability (Supplementary Table S1). Rather, patients treated in flagship systems were more likely to reside in larger, metropolitan areas with greater access to surgical care. These data highlight that socially vulnerable patients have less access to flagship and flagship-affiliated hospitals, while also suggesting that hospital system mergers tend to occur in more population-dense, affluent urban areas.
Flagship systems offer the promise of specialized services and advanced medical technologies compared with other hospitals, which may help patients benefit from advanced treatments, technology, and expertise.39 In the current study, hospitals within flagship systems were more likely to be teaching institutions, have higher bed capacity, and possess superior nursing resources. Indeed, outcomes following complex surgical procedures at the main flagship hospital were better than those at either affiliate hospitals or nonaffiliated hospitals.
Specifically, the incidence of 30-day mortality was lower at flagship hospitals compared with affiliate hospitals. Perhaps of more interest, although flagship-affiliated hospitals were associated with lower postoperative mortality than nonaffiliated hospitals, the difference was much more modest (difference, −0.32% [95% CI, −0.58 to −0.07]; P=.013). Prior research investigating the provision of high-quality care in relation to local hospital affiliation with a flagship system has yielded conflicting results. Jain et al39 reported an association between affiliate status and reduced short-term mortality among patients admitted for general medical conditions, such as pneumonia. In contrast, Ramadan et al27 failed to find a benefit from treatment at an affiliate hospital versus a hospital outside a flagship system among patients undergoing general surgical procedures. Data from the current study suggest that patients undergoing higher-risk surgical operations for a malignant diagnosis benefit from receiving care within a flagship health care system.
Centralization efforts have predominantly targeted patients receiving complex, high-margin procedures, which has resulted in hospital mergers, alterations in payment structures, and the realignment of care across affiliated hospitals.37,40–42 Moreover, cancer care at designated centers is characterized by its multidisciplinary, evidence-based approach, the integration of advanced technologies, centralized resources, and robust data systems. These factors contribute to a higher degree of standardization and systemization in cancer care at these centers compared with lower-volume hospitals. The rise in centralization may drive up health care expenditures by bolstering market power.12,13,18,43 Indeed, there were increased index hospitalization and 90-day expenditures among patients receiving care at flagship and affiliate hospitals versus nonaffiliate hospitals. Although the incidence of complications among flagship-affiliated versus nonaffiliated hospitals was largely comparable, mortality was different. This may be due to the better resources and more comprehensive postdischarge care at flagship hospitals. Therefore, surgical and postdischarge expenditures at flagship hospitals may have been driven, in part, by increased case complexity, greater resource allocation, advanced medical facilities, and elevated staffing levels.39 Greater market consolidation and hospital acquisition of physician practices may also result in higher spending per beneficiary due to cost sharing of hospital expenses.3,44
Several limitations should be considered when interpreting results from the current study. Use of a large administrative dataset has inherent limitations given the reliance on diagnosis and procedural codes from billing data. Additionally, the Medicare data included only patients aged ≥65 years, limiting generalizability to other patient populations. Moreover, the HRRs included in the study cohort had more beds, higher resident-to-bed ratios, and higher case volumes versus the national average. The Medicare dataset did not allow for assessment of market competition factors such as negotiated reimbursement rates and cost structures. Moreover, the Medicare database lacks detailed information on physicians’ training and cancer severity. Nevertheless, the utilization of Medicare data offered a unique resource of patient- and hospital-level identifiable information, providing a geographically representative sample cohort of hospitals in the United States.
Conclusions
Flagship hospitals were the primary drivers of lower postoperative mortality following complex oncologic surgical procedures performed in flagship hospital systems, although the cost was higher at these flagship hospitals. Smaller hospitals that were affiliated with a flagship system had a more modest difference in 30-day mortality compared with non–flagship-affiliated hospitals. Consolidation of surgical care in flagship hospital systems may help improve care quality for patients with cancer, yet the benefit largely derives from the “hub” hospital.
References
- 1.↑
Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative mortality in the United States. N Engl J Med 2003;349:2117–2127.
- 2.↑
Sahni NR, Dalton M, Cutler DM, et al. Surgeon specialization and operative mortality in United States: retrospective analysis. BMJ 2016;354:i3571.
- 3.↑
MedPAC. Chapter 15: Congressional request on health care provider consolidation (March 2020 report). Accessed March 12, 2024. Available at: https://www.medpac.gov/document/http-www-medpac-gov-docs-default-source-reports-mar20_medpac_ch15_sec-pdf/
- 4.↑
Gaynor M, Ho K, Town RJ. The industrial organization of health-care markets. J Econ Lit 2015;53:235–284.
- 5.↑
Wang E, Arnold S, Jones S, et al. Quality and safety outcomes of a hospital merger following a full integration at a safety net hospital. JAMA Netw Open 2022;5:e2142382.
- 6.↑
Scanlon DP, Harvey JB, Wolf LJ, et al. Are health systems redesigning how health care is delivered? Health Serv Res 2020;55(Suppl 3):1129–1143.
- 7.↑
Heeringa J, Mutti A, Furukawa MF, et al. Horizontal and vertical integration of health care providers: a framework for understanding various provider organizational structures. Int J Integr Care 2020;20:2.
- 9.↑
Reames BN, Anaya DA, Are C. Hospital regional network formation and ‘brand sharing’: appearances may be deceiving. Ann Surg Oncol 2019;26:711–713.
- 10.↑
Chiu AS, Resio B, Hoag JR, et al. Why travel for complex cancer surgery? Americans react to ‘brand-sharing’ between specialty cancer hospitals and their affiliates. Ann Surg Oncol 2019;26:732–738.
- 11.↑
Gale AH. Bigger but not better: hospital mergers increase costs and do not improve quality. Mo Med 2015;112:4–5.
- 12.↑
Glied SA, Altman SH. Beyond antitrust: health care and health insurance market trends and the future of competition. Health Aff (Millwood) 2017;36:1572–1577.
- 13.↑
Cooper Z, Craig SV, Gaynor M, Van Reenen J. The price ain’t right? Hospital prices and health spending on the privately insured. Q J Econ 2019;134:51–107.
- 14.↑
Hoag JR, Resio BJ, Monsalve AF, et al. Differential safety between top-ranked cancer hospitals and their affiliates for complex cancer surgery. JAMA Netw Open 2019;2:e191912.
- 15.↑
Boffa DJ, Mallin K, Herrin J, et al. Survival after cancer treatment at top-ranked US cancer hospitals vs affiliates of top-ranked cancer hospitals. JAMA Netw Open 2020;3:e203942.
- 16.↑
Sheetz KH, Ibrahim AM, Nathan H, Dimick JB. Variation in surgical outcomes across networks of the highest-rated US hospitals. JAMA Surg 2019;154:510–515.
- 17.↑
Beaulieu ND, Dafny LS, Landon BE, et al. Changes in quality of care after hospital mergers and acquisitions. N Engl J Med 2020;382:51–59.
- 19.↑
Mutter RL, Romano PS, Wong HS. The effects of US hospital consolidations on hospital quality. Int J Econ Bus 2011;18:109–126.
- 20.↑
Munir MM, Woldesenbet S, Endo Y, et al. Association of hospital market competition with outcomes of complex cancer surgery. Ann Surg Oncol 2024;31:4371–4380.
- 21.↑
Thumma SR, Dualeh SHA, Kunnath NJ, et al. Outcomes for high-risk surgical procedures across high- and low-competition hospital markets. JAMA Surg 2023;158:1041–1048.
- 22.↑
AHA Data & Insights. AHA Annual Survey Database™. Accessed September 19, 2023. Available at: https://www.ahadata.com/aha-annual-survey-database
- 23.↑
Khan MMM, Munir MM, Woldesenbet S, et al. Association of surgeon–patient sex concordance with postoperative outcomes following complex cancer surgery. J Surg Oncol 2024;129:489–498.
- 24.↑
ATSDR. Social Vulnerability Index. Accessed September 19, 2023. Available at: https://www.atsdr.cdc.gov/place-health/php/svi/index.html
- 25.↑
Healthcare Cost and Utilization Project (HCUP) and Agency for Healthcare Research and Quality. Elixhauser Comorbidity Software Refined for ICD-10-CM. Accessed November 3, 2023. Available at: https://hcup-us.ahrq.gov/toolssoftware/comorbidityicd10/comorbidity_icd10.jsp
- 26.↑
Agency for Healthcare Research and Quality. Compendium of U.S. Health Systems, 2018. Accessed March 12, 2024. Available at: https://www.ahrq.gov/chsp/data-resources/compendium-2018.html
- 27.↑
Ramadan OI, Rosenbaum PR, Reiter JG, et al. Impact of hospital affiliation with a flagship hospital system on surgical outcomes. Ann Surg 2024;279:631–639.
- 28.↑
Abbas A, Dalmacy D, Paro A, Pawlik TM. Impact of neighborhood characteristics on textbook outcome following major surgery. Am J Surg 2022;224:959–964.
- 29.↑
Merath K, Chen Q, Bagante F, et al. Textbook outcomes among Medicare patients undergoing hepatopancreatic surgery. Ann Surg 2020;271:1116–1123.
- 30.↑
Hyer JM, Ejaz A, Diaz A, et al. Characterizing and assessing the impact of surgery on healthcare spending among Medicare enrolled preoperative super-utilizers. Ann Surg 2019;270:554–563.
- 31.↑
Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083–3107.
- 32.↑
Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res 2011;46:399–424.
- 33.↑
Yuan Y, Yung YF, Stokes M. Propensity score methods for causal inference with the PSMATCH procedure. Accessed September 19, 2024. Available at: https://support.sas.com/resources/papers/proceedings17/SAS0332-2017.pdf
- 34.↑
Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat 2011;10:150–161.
- 35.↑
Lenis D, Nguyen TQ, Dong N, Stuart EA. It’s all about balance: propensity score matching in the context of complex survey data. Biostatistics 2019;20:147–163.
- 36.↑
Fleiss JL, Levin B, Paik MC. Statistical Methods for Rates and Proportions. Wiley; 2003. Available at: https://onlinelibrary.wiley.com/doi/book/10.1002/0471445428
- 37.↑
Munir MM, Endo Y, Woldesenbet S, et al. Variations in travel patterns affect regionalization of complex cancer surgery in California. Ann Surg Oncol 2023;30:8044–8053.
- 38.↑
Shaikh CF, Munir MM, Woldesenbet S, et al. Association of persistent poverty and U.S. News and World Report hospital rankings among patients undergoing major surgery. Am J Surg 2024;228:11–19.
- 39.↑
Jain S, Rosenbaum PR, Reiter JG, et al. Mortality among older medical patients at flagship hospitals and their affiliates. J Gen Intern Med 2024;39:902–911.
- 40.↑
Sage WM. Getting the product right: how competition policy can improve health care markets. Health Aff (Millwood) 2014;33:1076–1082.
- 41.↑
Vladeck BC. Paradigm lost: provider concentration and the failure of market theory. Health Aff (Millwood) 2014;33:1083–1087.
- 42.↑
Munir MM, Yutaka E, Alaimo L, et al. Impact of community privilege on access to care among patients following complex cancer surgery. Ann Surg 2023;278:e1250–1258.
- 43.↑
Tsai TC, Jha AK. Hospital consolidation, competition, and quality: is bigger necessarily better? JAMA 2014;312:29–30.
- 44.↑
The Commonwealth Fund. Which parts of the United States have the most consolidated Medicare hospital spending? Accessed March 13, 2024. Available at: https://www.commonwealthfund.org/blog/2023/which-parts-united-states-have-most-consolidated-medicare-hospital-spending
