Doing More for More: Unintended Consequences of Financial Incentives for Oncology Specialty Care

Physician responses to Medicare reimbursement rates

This paper investigates how office-based physicians respond to Medicare reimbursement changes. Using variation from an Affordable Care Act policy that increased reimbursements for office-based care in four states, we use a triple difference analysis, comparing physicians with higher and lower reimbursement changes in treated states to similar physicians in untreated states. We find two mechanisms through which physicians respond. First, the reimbursement change affected integration-physicians with larger increases in office-based reimbursement were less likely to vertically integrate with hospitals and more likely to continue providing office-based care than physicians with smaller reimbursement increases. Second, we find some evidence that physicians who continued practicing in an office setting increased the volume of services provided.

Trends in Urethral Suspension With Robotic Prostatectomy Procedures Following Medicare Payment Policy Changes

IMPORTANCE

In 2016, the Centers for Medicare and Medicaid Services cut payments for robotic prostatectomy performed for Medicare beneficiaries. Although regulations mandate that billing for urethral suspension is only acceptable for preexisting urinary incontinence, reductions in reimbursement may incentivize billing for the use of this procedure in other scenarios.

OBJECTIVE

To assess trends and geographic variations in payments for urethral suspension with robotic prostatectomy in the context of Medicare payment policy.

DESIGN, SETTING, AND PARTICIPANTS

This US population-based retrospective cohort study analyzed data from the IBM MarketScan Commercial Claims and Encounters and Medicare Supplemental Database for men with employer-based insurance (primary commercial or Medicare supplemental coverage) who underwent robotic prostatectomy (Current Procedural Terminology [CPT] code 55866) between 2009 and 2019.

EXPOSURES

Time period and metropolitan statistical area of patient residence.

MAIN OUTCOMES AND MEASURES

Payment for urethral suspension (CPT code 51990) with robotic prostatectomy.

RESULTS

We identified 87 774 men with prostate cancer treated with robotic prostatectomy; 3352 (3.8%) had undergone urethral suspension. The mean (SD) patient age was 59.7 (6.5) years; 16 870 patients (19.2%) had Medicare supplemental coverage. From 2015 to 2016, median payments for robotic prostatectomy changed by -$358 (-17.0%) for Medicare beneficiaries vs -$9 (0%) for commercially insured patients. With urethral suspension vs without, median (IQR) episode payments for robotic prostatectomy were higher for commercially insured men ($3678 [$3090-$4503] vs $3322 [$2601-$4306]) and Medicare beneficiaries ($2927 [$2450-$3909] vs $2379 [$2014-$3512]). Compared with men treated between 2013 and 2015, those treated between 2016 and 2017 were twice as likely to undergo urethral suspension (8.5% vs 4.1%; odds ratio, 2.17 [95% CI, 1.96-2.38]). The proportion of patients who underwent urethral suspension was stable for 2018 to 2019 and 2016 to 2017 (8.5% vs 9.0%; odds ratio, 1.06 [95% CI, 0.96-1.18]). From 2015 to 2019, the proportion of patients who underwent urethral suspension was highest in Charleston, South Carolina (92.0%), Knoxville, Tennessee (66.0%), and Columbia, South Carolina (58.0%). These regions neighbored high-volume areas without patients who underwent prostatectomy with urethral suspension (eg, 146 patients in Greenville, South Carolina, and 173 in Nashville, Tennessee).

CONCLUSIONS AND RELEVANCE

In this study, urethral suspension was associated with increased costs for patients with both commercial insurance and Medicare. Patients treated between 2016 and 2017 were more likely than those treated between 2013 and 2015 to undergo this procedure. Geographic variation in use exceeded what was expected for the preexisting condition for which billing is permitted for Medicare beneficiaries. Policy statements from professional societies highlighting appropriate billing for urethral suspension may have tempered, but not reversed, the broad adoption of this procedure.

Medicare fee reductions and the overuse of intensity-modulated radiotherapy

OBJECTIVE

To estimate the impact of a large Medicare fee reduction for intensity-modulated radiation therapy (IMRT) on its use in prostate and breast cancer patients.

DATA SOURCES/STUDY SETTING

SEER-Medicare.

STUDY DESIGN

We compared trends in the use of IMRT between patients treated in practices directly affected by fee reductions (for prostate cancer, men treated in urology practices that own IMRT equipment; for breast cancer, women treated in freestanding radiotherapy clinics) and patients treated in other types of practices.

DATA COLLECTION/EXTRACTION METHODS

We identified breast and prostate cancer patients receiving IMRT using outpatient and physician office claims. We classified urology practices based on whether they billed for IMRT and radiotherapy clinics based on whether they were reimbursed under the Physician Fee Schedule.

PRINCIPAL FINDINGS

Between 2006 and 2015 the payment for IMRT delivered in freestanding clinics and physician offices declined by $367 (-54.7%). However, the use of IMRT increased in physician practices subject to payment cuts, both in absolute terms and relative to use in practices unaffected by the payment cut. Use of IMRT in prostate cancer patients treated at urology practices that own IMRT equipment increased by 9.1 (95% CI: 2.0-16.2) percentage points between 2005 and 2016 relative to use in patients treated at other urology practices. Use of IMRT in breast cancer patients treated at freestanding radiotherapy centers increased by 7.5 (95% CI: -5.1 to 20.1) percentage points relative to use in patients treated at hospital-based centers.

CONCLUSIONS

A steep decline in IMRT fees did not decrease IMRT use over the period from 2006 to 2015, though use has declined since 2010.

Detecting bad actors in value-based payment models

The U.S. federal government is spending billions of dollars to test a multitude of new approaches to pay for healthcare. Unintended consequences are a major consideration in the testing of these value-based payment (VBP) models. Since participation is generally voluntary, any unintended consequences may be magnified as VBP models move beyond the early testing phase. In this paper, we propose a straightforward unsupervised outlier detection approach based on ranked percentage changes to identify participants (e.g., healthcare providers) whose behavior may represent an unintended consequence of a VBP model. The only data requirements are repeated measurements of at least one relevant variable over time. The approach is generalizable to all types of VBP models and participants and can be used to address undesired behavior early in the model and ultimately help avoid undesired behavior in scaled-up programs. We describe our approach, demonstrate how it can be applied with hypothetical data, and simulate how efficiently it detects participants who are truly bad actors. In our hypothetical case study, the approach correctly identifies a bad actor in the first period in 86% of simulations and by the second period in 96% of simulations. The trade-off is that 9% of honest participants are mistakenly identified as bad actors by the second period. We suggest several ways for researchers to mitigate the rate or consequences of these false positives. Researchers and policymakers can customize and use our approach to appropriately guard VBP models against undesired behavior, even if only by one participant.

Combining Drive Time and Urologist Density to Understand Access to Urologic Care

OBJECTIVE

To improve our understanding of timely access to urologic care, we leveraged driving time combined with a measure of urologist density.

MATERIALS AND METHODS

We identified all urologists who billed Medicare using National Provider Identifier in 2015 and geocoded their practice location. We developed drive-time based service areas for each provider using Esri's street network dataset stratified into 30, 60, 90, and 120-minute areas. Population characteristics were aggregated and block groups were assigned to a Hospital Referral Region.

RESULTS

We identified 10,170 urologists that billed Medicare in 2015 in the United States. Compared to the northeast, vast expanses of land across the western United States have drive times to urology care >60 minutes. However, less than 13% of the US population is unable to obtain urologic care within 30 minutes. Likely reflecting rural populations, White and American Indian populations are represented in greater proportion among those requiring a longer drive time to urologic care. Disparities were noted between areas with timely access to a high versus low density of urologists; low density areas have a greater proportion of Black and Asian populations and greater income inequality.

CONCLUSIONS

Drive time to urologists combined with urologist density is a novel approach to investigating urologic care access and a tool for health disparities research. While almost all of the US population lives within 1-hour drive time to a urologist there remains important differences in the population severed by high compared to low provider density.

Mind the Gaps: Adoption and Underutilization of Holmium Laser Enucleation of the Prostate in the United States from 2008 to 2014

Purpose: There is increasing recognition of the advantages of holmium laser enucleation of the prostate (HoLEP) for benign prostatic hypertrophy (BPH) treatment in the United States but relatively little is known about the extent of HoLEP adoption over time. This study aims to assess national HoLEP adoption rates and regional trends from 2008 to 2014. Methods: We retrospectively analyzed a data set of 100% Medicare claims to determine the rate of U.S. HoLEP adoption in 2008, 2011, and 2014. Rates were adjusted by age and race and stratified by hospital referral region (HRR). Linear and logistic regression models were used to assess for trends in HoLEP adoption over time. Results: Total U.S. BPH cases decreased 24% from 2008 to 2014 and HoLEP cases increased significantly from 1086 (2008) to 3368 (2014). Despite this, HoLEP accounted for just 4% of total BPH cases in 2014. In 2008, 28/306 (9%) of HRRs recorded >10 HoLEP cases per year. This increased to 89 HRRs (29%) in 2011 but stabilized at 94 HRRs (31%) in 2014. In 2014, over 50% of states still had only 0 to 1 sites doing 10+ HoLEP cases per year. Conclusions: Based on this 100% sample of Medicare claims from 2008 to 2014, surgical BPH treatment volume has decreased, while HoLEP volume and regional adoption have tripled. However, rates of HoLEP remain extremely low at just 4% of all BPH procedures in 2014, and large regional gaps in care exist. These data indicate that HoLEP remains substantially underutilized and the majority of regions still lack access to centers performing >10 HoLEP cases per year.

Site of service influence on stent use for hemodialysis access interventions

OBJECTIVE

With rising health care spending in the United States, the Centers for Medicare and Medicaid Services (CMS) in recent years attempted to use reimbursement rates to influence use of less expensive care sites for covered patients, such as ambulatory surgery centers (ASCs) and office-based laboratories (OBLs), in lieu of hospital service sites. It has been suggested that cost savings have not been realized because of more procedures being performed by physicians with ownership interests in nonhospital facilities. CMS adopted massive reimbursement changes for 2019 OBL and ASC-based procedures, which reduced dialysis access angioplasty reimbursement in the ASC setting by 50%, whereas facility reimbursement for stenting increased by 33% above prior levels. The clinical utility of adjunctive stenting in treating dialysis access stenosis remains controversial and highly discretionary. As a vascular group performing such procedures in both a hospital and nonhospital facility in which we have equity interest, we reviewed our use of stents in dialysis access procedures both in the hospital and in the ASC/OBL to determine whether site of service affected stent use.

METHODS

A retrospective review of a prospectively maintained database was performed from 2014 to 2018. All patients undergoing dialysis access angiography with angioplasty and adjunctive stent placement at our OBL (later ASC) and our primary hospital were included in the study.

RESULTS

There were 961 angioplasty or stent procedures performed for dialysis accesses between the two sites, 564 (58.7%) in the hospital setting and 397 (41.3%) at the OBL/ASC. There was a significant difference in race and age between the two sites, with younger, minority patients more frequently being treated in the hospital and older, white patients more likely to be treated in the ambulatory setting; 153 (27.1%) underwent adjunctive stent placement in the hospital and 127 (32.0%) in the ambulatory setting (P = .09).

CONCLUSIONS

Whereas financial incentives have not yet had an appreciable influence on stent use for dialysis access within previous reimbursement paradigms, the dramatic changes recently adopted by CMS may well alter this dynamic and could lead to substantially higher overall costs without proven clinical advantage. Interventionalists may be incentivized to add stents when performing balloon angioplasty in ASCs. With high failure and reintervention rates and increasingly expensive adjuncts (drug-coated balloons and stents, covered stents), the cost implications of attempts to incentivize interventionalists toward a specific type of procedure or site of care are substantial, and unintended negative consequences are likely to occur.

Implementing risk-aligned bladder cancer surveillance care

Implementation science is a rapidly developing field dedicated to the scientific investigation of strategies to facilitate improvements in healthcare delivery. These strategies have been shown in several settings to lead to more complete and sustained change. In this essay, we discuss how refined surveillance recommendations for non-muscle-invasive bladder cancer, which involve a complex interplay between providers, healthcare facilities, and patients, could benefit from use of implementation strategies derived from the growing literature of implementation science. These surveillance recommendations are based on international consensus and indicate that the frequency of surveillance cystoscopy should be aligned with each patient's risk for recurrence and progression of disease. Risk-aligned surveillance entails cystoscopy at 3 and 12 months followed by annual surveillance for low-risk cancers, with surveillance every 3 months reserved for high-risk cancers. However, risk-aligned care is not the norm. Implementing risk-aligned surveillance could curtail overuse among low-risk patients, while curbing underuse among high-risk patients. Despite clear direction from respected and readily available clinical guidelines, there are multiple challenges to implementing risk-aligned surveillance in a busy clinical setting. Here, we describe how implementation science methods can be systematically used to understand determinants of care and to develop strategies to improve care. We discuss how the tailored implementation for chronic diseases framework can facilitate systematic assessment and how intervention mapping can be used to develop implementation strategies to improve care. Taken together, these implementation science methods can help facilitate practice transformation to improve risk-aligned surveillance for bladder cancer.

Unintended Consequences in Cancer Care Delivery Created by the Medicare Part B Proposal: Is the Clinical Rationale for the Experiment Flawed?

PURPOSE

Medicare currently enrolls ≥ 45 million adults, and by 2030 this is projected to increase to ≥ 80 million beneficiaries. With this growth, the Centers for Medicare & Medicaid Services (CMS) issued a proposal, the Medicare Part B Drug Payment Model, to shrink drug expenditures, a major contributor to overall health care costs. For this to not adversely affect patient outcomes, lower-cost alternative medications with equivalent efficacy and no increased toxicity must be available. This is often not true in the treatment of cancer. Herein, we examine the flaws in the rationale of the CMS and the potential unintended consequences of this experiment.

METHODS

We identified the top three oncology expenditures (rituximab, bevacizumab, and trastuzumab) and their vetted alternatives (per the National Comprehensive Cancer Network guidelines) to ascertain whether lower-cost equivalent alternatives are available. Drug cost was based on April 2016 average sale price. We explored both efficacy of the agents and, when applicable, toxicity to compare alternatives to these high-dollar medications.

RESULTS

For the largest Medicare oncology drug expenditures, there is not a lower-cost option with equal efficacy for their primary indications. Without lower-cost alternatives, the unintended consequence of this CMS experiment may include curtailing access to care or an increase in patient/program costs.

CONCLUSION

The CMS proposal, by simply lowering reimbursement for drugs, does not acknowledge the value of these agents and could unintentionally reduce quality of care. Alternative approaches to value-based care, such as the Oncology Care Model and similar frameworks, should be explored.