Patients with cancer have equal utilization of care at community hospitals for lower cost

Insurers shelled out more money for surgical procedures performed at National Cancer Institute (NCI)-designated cancer centers compared to procedures performed at community hospitals, but that extra spending did not lead to greater care utilization, according to findings from an analysis of privately insured patients undergoing cancer surgery.

NCI-designated cancer centers are academic hospitals that have historically delivered consistently excellent patient outcomes, attracting top clinicians and educators and demonstrating clinical expertise in cancer care, Samuel U. Takvorian, MD, MSHP, of the Perelman Center for Advanced Medicine in Philadelphia, and colleagues explained in JAMA Network Open. However, while some research shows that treatment at NCI centers may be linked to improved outcomes, particularly among those with more advanced cancers, it may also be linked to higher spending for surgery and cancer care post-discharge—spending that may not be representative of the quality of care delivered.

Takvorian and colleagues conducted a retrospective, cross-sectional study to determine the association of cancer treatment in an NCI center or community hospital with insurer spending and care utilization during common cancer surgery; they focused their analysis on patients with private insurance, “for whom health care prices are negotiated between insurers and clinicians and price transparency is lacking.”

The study authors found that “surgery-specific and 90-day postdischarge spending were higher at NCI centers than community hospitals and in an intermediate range at non-NCI academic hospitals without differences in acute care utilization for patients with private insurance undergoing surgery for breast, colon, or lung cancer. Facility rather than physician payments accounted for most of the differences in spending outcomes, consistent with national trends showing that hospital payments occupy a disproportionate and growing share of overall health care spending. These results support our hypothesis that insurer spending would be higher at NCI centers than community hospitals, possibly due to their size, market share, and prestige, affording leverage in negotiations with private payers.”

However, Takvorian and colleagues added that another aspect of their hypothesis—that patients at NCI centers would have decreased acute care utilization post-discharge—did not bear out. Instead, acute care utilization rates were comparable across hospital types, “suggesting that negotiated transaction prices rather than utilization may be driving site-level differences in spending.”

“A 2021 study by Wolfson et al found similarly higher health plan spending at NCI-designated cancer centers versus other hospitals for privately insured young adults with acute lymphoblastic leukemia,” Nancy L. Keating, MD, MPH, of Harvard Medical School in Boston, wrote in an accompanying editorial. “Such price differences are becoming increasingly evident as commercial insurance data become more available to researchers.”

Unlike with Medicare, Keating explained, commercial prices are negotiated by hospitals and physician groups for each commercial payer, and the prices are substantially higher—a cost difference that likely explains why federal health insurance plans often omit NCI centers, she noted.

Given that there is some evidence suggesting care at NCI centers is associated with better outcomes, it is “likely that some patients benefit from the highly specialized care available at NCI-designated cancer centers, particularly patients with rarer or complex clinical conditions, patients requiring complex procedures, or those for whom clinical trials may offer promising treatment options,” she wrote. “But it is also likely that many other patients will do equally well regardless of where they receive their care.”

The analysis by Takvorian and colleagues included adult patients with an incident diagnosis of breast, colon, or lung cancer who received cancer-directed surgery from 2011-2014. Hospitals were categorized into three mutually exclusive categories: NCI centers, non-NCI academic hospitals, and community hospitals. Data was pulled from the Health Care Cost Institute’s national multipayer commercial claims data set, which includes claims data from Aetna, Humana, and UnitedHealthcare.

The primary spending outcomes were surgery-specific insurer price paid and 90-day post-discharge payments, the latter of which was calculated “by aggregating payments on inpatient, outpatient, physician, and pharmacy claims during the 90-day period after discharge,” they explained. Primary utilization outcomes included hospital length of stay (LOS), emergency department (ED) use, and hospital readmissions within 90 days of discharge.

The final analysis included 66,878 patients (51,569 [77.1%] women; 31,585 [47.2%] age ≥65 years), of whom incident breast (35,788 (53.5%]), colon (21,378 [32.0%]), or lung (9,712 [14.5%]) cancer. Surgeries were conducted across 2,995 hospitals: 5,522 (8.3%) at NCI centers, 10,917 (16.3%) at non-NCI academic hospitals, and 50,439 (75.4%) at community hospitals.

The findings showed substantial differences in spending outcomes across hospital types:

  • “Treatment at NCI centers was associated with higher surgery-specific insurer prices paid compared with community hospitals ($18,526 [95% CI, $16,650 to $20,403] versus $14,772 [95% CI, $14,339-$15,204]; difference, $3,755 [95% CI, $1,661-$5,849]; P<0.001), driven predominantly by differences in facility payments ($17,704 [95% CI, $15,845-$19,563] versus $14,120 [95% CI, $13,691-$14,549]; difference, $3,584 [95% CI, $1,525-$5,643]; P<0.001).”
  • Ninety day post-discharge payments were also higher at NCI centers versus community hospitals ($47,035 [95% CI, $43,289 to $50,781] versus $41,291 [95% CI, $40,350-$42,231]; difference, $5,744 [95% CI, $1,659-$9,829]; P=0.006).
  • Non-NCI academic hospitals had numerically higher surgery-specific and 90-day post-discharge payments compared to community hospitals, but the difference was not statistically significant.

Meanwhile, there were no significant differences between hospital types in LOS, ED utilization, or hospital readmission within 90 days:

  • “Mean LOS was comparable at NCI centers and community hospitals (5.1 [95% CI,4.8-5.4] days versus 5.1 [95% CI, 5.1-5.2] days, P=0.73).
  • “The probability of ED utilization (13.1% [95% CI,11.9%-14.3%] versus 13.2% [95% CI, 12.8%-13.5%]; P=0.93) or hospital readmission (10.4% [95% CI,9.2%-11.5%] versus 10.8% [95% CI, 10.5%-11.1%]; P=0.48) within 90 days was also similar between NCI centers and community hospitals.”

Given the lack of differences in short-term post-surgical outcomes by hospital type, the study authors concluded that “Further research examining hospital-level differences in long-term post-surgical outcomes, such as mortality, paired with spending outcomes, is necessary to judge whether and under what circumstances the premium price of NCI centers is justified.”

Keating concurred, adding that such data “could also be used by payers considering tiered networks and by physician organizations participating in risk contracts for decisions about where to refer patients with cancer for treatment. In the search for high-value, patient-centered, and equitable care, it is critical to identify strategies to better allocate health care dollars in ways that achieve the best possible outcomes across populations of patients.”

Study limitations included limiting the analysis to patients undergoing surgery for breast, colon, or lung cancer, which might limit generalizability to other cancer types or those receiving non-surgical cancer care; a lack of analysis into patient out-of-pocket spending; the claims-based analysis did not allow the researchers to adjust for clinical factors such as stage at diagnosis, surgical complexity, and pathologic status; and due to the study’s observational nature, observed differences by hospital type may be attributable to unmeasured factors.

  1. Surgery-specific and 90-day post-discharge spending were higher at National Cancer Institute (NCI)-designated centers than community hospitals without differences in acute care utilization for patients with private insurance undergoing surgery for breast, colon, or lung cancer.

  2. Comparable rates of acute care utilization across hospital types suggests that negotiated transaction prices, rather than utilization, may be driving site-level differences in spending, and a better understanding of the drivers behind prices and spending at NCI centers is needed.

John McKenna, Associate Editor, BreakingMED™

Coauthor Bekelman reported grants from Pfizer, UnitedHealth Group, Embedded Healthcare, and Blue Cross Blue Shield of North Carolina and personal fees from UnitedHealthcare, the Centers for Medicare & Medicaid Services, the National Comprehensive Cancer Network, and Optum outside the submitted work.

Keating had no relevant relationships to disclose.

Cat ID: 22

Topic ID: 78,22,728,791,730,22,23,24,192,925