Peripheral Vascular Interventions in Office-Based Laboratories: Good News for Disparities or Profit Margins?

The personal and societal burden of peripheral arterial disease (PAD) and lower extremity amputations cannot be overstated. Annual Medicare expenditures for PAD exceed $84 billion per year, and following an amputation, most patients are permanently disabled and do not survive 5 years.¹ However, PAD’s burden is not uniformly distributed; significant disparities run along racial and socioeconomic lines. For example, Black and low socioeconomic status (SES) patients are more likely to undergo surgical amputation, and Black patients are less likely than Whites to have an attempt at limb salvage prior to amputation.^2,3 Fortuitously, revascularization options for patients with PAD have advanced and expanded substantially over the past few decades. Improvements in the safety and efficacy of endovascular peripheral vascular intervention (PVI), and its minimally invasive nature, have rendered outpatient treatment feasible and resulted in this modality to widely surpass surgery as the predominant revascularization procedure performed in the United States.⁴ In 2008, to improve efficiency and reduce overall revascularization costs while accounting for higher overhead, the Centers for Medicare and Medicaid Services (CMS) increased reimbursement for PVIs performed in outpatient office-based laboratories (OBLs).⁵ Although OBL procedural reimbursement remained lower than that of hospital-based procedures, this boost in reimbursement and the potential for added patient convenience, resulted in a dramatic shift from hospital-based PVI procedures to privately owned OBLs.^6,7 With peripheral arterial atherectomy reimbursed at a significantly higher rate than other OBL procedures, not surprisingly, atherectomy has become the most commonly performed PVI procedure performed in OBLs.⁶

It has been shown that minority patients, patients of low SES, and patients in disadvantaged communities often seek care in community-based office-based practices rather than academic hospitals.⁸ Thus, greater access to community-based physician practices could potentially improve health care access and expand the safety net for vulnerable populations.⁸ This may, at least in part, be due to a perception of large academic medical centers as being more impersonal and requiring more complicated navigation. It is natural for minority patients to seek care from physicians who speak their preferred language and provide culturally concordant care in office settings better designed to make them feel welcomed and comfortable while removing some of the barriers associated with large hospital environments. So, what impact, if any, has the rapid rise in OBL PVI procedures had on clinical outcomes and health care disparities for underserved minorities and low SES PAD patients? Until recently, published data on the demographics, clinical characteristics, and long-term outcomes of patients treated in OBLs versus hospital-based settings has been lacking. In this edition of Circulation: Cardiovascular Quality and Outcomes, Raja et al⁹ shine some much needed light on the question of whether OBLs produce long-term PAD outcomes that match those of hospital-based care. The results are germane as they have potential implications for both health care policy and health care disparities. The authors studied the demographics and clinical characteristics of 134 869 Medicare patients aged ≥66 years identified by CPT and ICD-10 codes as having undergone femoropopliteal endovascular PVI from 2015 to 2017. They compared risk-adjusted outcomes between patients treated in OBLs (29.9%; n=40 345) versus those treated in hospital-based centers (70.1%; n=94 524). The study revealed that OBLs were more often located in the South Atlantic, Pacific, and Southwest Central regions of the United States and OBL-treated patients were more often Black (16.9% versus 11.9%), dually enrolled in Medicare and Medicaid (26.3% versus 19.6%), and residents of lower resourced regions (32.6% versus 25.6%). Revascularization procedures performed in OBLs were much more likely to employ atherectomy (81.7% versus 37.2%). Through a median follow-up of 800 days, OBL-treated patients experienced a lower adjusted risk of major amputation or death compared with those treated in hospital-based centers (HR, 0.92 [95% CI, 0.89–0.95]), and lower adjusted risks of all-cause mortality (HR, 0.93 [95% CI, 0.90–0.96]), leg amputation (HR, 0.84 [95% CI, 0.79–0.89]), and all-cause hospitalization (HR, 0.86 [95% CI, 0.84–0.88]). As expected, more minority and low SES patients were treated in OBLs, as compared with the hospital-based setting, but the favorable outcome findings were consistent across all subgroups and persisted after stratification by critical limb ischemia, race, dual enrollment, regional SES, and operator characteristics in both clinical settings. The authors concluded that, despite treating a racially more diverse and socioeconomically disadvantaged population than hospital-based clinics, OBL-based PVIs were associated with favorable long-term outcomes.

The major strengths of this study include its large sample size, sound study design and statistical analysis, multivariable risk adjustment and stratification of outcomes across race and SES, the latter of which informs outcomes in underserved subgroups. However, as the authors correctly point out, there are major limitations to this analysis. First, although race/ethnicity was listed as a baseline characteristic, only race (Black, White, and Other) was reported, with no information on Hispanic ethnicity. This is an important omission, since both Black and Hispanic non-White patients have been shown to fare worse than Whites with comparable stages of PAD.¹⁰ Second, drawing from a claims-based, retrospective observational registry, the issue of confounding looms large and precludes one from asserting a definitive causal relationship between OBLs and improved outcomes. Claims data do not allow for investigation of the anatomical and angiographic characteristics of patients treated with PVIs. It could be that these office-based practices were generally getting healthier lower risk patients with more straightforward anatomy that was prone to better outcomes. It is not difficult to imagine how operators would be financially incentivized to direct more technically straightforward lower risk patients to privately owned OBLs to undergo a higher reimbursing atherectomy procedure, while relegating more complex patients in need of other non-atherectomy and higher risk procedures to the hospital. However, the baseline patient characteristics do not support this notion, as OBL patients in this study were older and had more comorbidities, including higher rates of diabetes, hyperlipidemia, hypertension, and prior stroke than hospital-based patients.

On the surface, one might celebrate the favorable outcomes noted for minorities treated in OBLs as welcome news. It was certainly reassuring that the degree of benefit associated with OBLs observed in the entire population extended to underserved groups (Black and lower SES patients). This is good news, not only for minority and low SES patients, but for all patients undergoing revascularization in OBLs, especially those who prefer outpatient treatment. However, the study’s findings also point to an economic undercurrent. The higher rate of atherectomy reimbursement compared with other OBL procedures likely explains why these procedures dominated OBLs. Financial incentives may also explain why repeat procedures were more commonly performed in OBL-treated patients compared with those treated in hospitals, a phenomenon observed by others.¹¹ On the other hand, it is also possible that office-based care was associated with more effective revascularization and/or adjunctive care, thereby driving lower rates of major amputation and other favorable outcomes. If true, this would be highly relevant for minority, low SES patients, for whom marked PAD disparities exist. Unfortunately, this is impossible to discern from this study, since many of the key determinants of long-term limb preservation were not available, including PAD lesion length and complexity, below knee vascular runoff, adjunctive preventive care and medications, the management of diabetes (ie, pre- and post-procedure HGA1c levels), diabetic foot care. and other factors.¹

Despite some shortcomings, Raja et al are to be commended for providing important insights into the associations between clinical setting, race, sociodemographic factors and the long-term outcomes of endovascular femoropopliteal revascularization procedures. Their results suggest that Black and lower SES patients are more likely to be treated in OBLs, and within this context, fare at least as well as if treated in a hospital setting. The study results highlight the need for further cardiovascular outcome research studies in office-based settings. Although one might be tempted to interpret these results as a “stamp of approval” for OBLs, we would caution against this, since an independent causal relationship between OBLs and favorable outcomes cannot be established from this work. Instead, the major contribution of this study is that it provides a unique window into the health care delivery and treatment patterns resulting from the rapid growth of OBLs, including the relationship of OBLs to geographic location, race, SES, and endarterectomy. Further investigation is needed to define the relative merits versus faults of our current CMS payment system that incentivizes volume shifts to OBLs. For that, angiographic lesion characteristics, technical success, short- and long-term clinical outcomes, cost-effectiveness, and patient preference/satisfaction must all be considered. Finally, for those truly interested in health care disparities, we also need to better understand how the rapid progression to OBLs has impacted access to revascularization, ancillary preventive care, and overall health-related outcomes for the underserved. It may be that shifts to OBLs have improved access to PVI care for previously underserved patients, which if true would be an important phenomenon. Although this work by Raja et al⁹ takes us one step closer, the answers to these questions remain elusive, but highly relevant, since many lives and limbs lie in the balance.