Site of Care Potentially Limits Cost Savings From Biosimilars

ABSTRACT

Objectives: The first FDA-approved biosimilar was launched in 2015: filgrastim-sndz (Zarxio), a biosimilar for the reference drug filgrastim (Neupogen). Filgrastim is a granulocyte colony-stimulating factor used to prevent and treat neutropenia. In this study, we examined the association between site of care and drug cost across reference filgrastim, tbo-filgrastim (Granix; a version of filgrastim approved as a biosimilar in Europe and as a new drug in the United States), and biosimilar filgrastim administrations among the commercially insured.

Study Design: Retrospective study using administrative claims data.

Methods: We used OptumLabs Data Warehouse to identify the site of care of each short-acting filgrastim administration among commercial enrollees between January 1, 2014, and December 31, 2019.

Results: For each filgrastim product, model-adjusted median drug costs were higher in the outpatient hospital setting than for the same drug administered in the office setting. Comparing drug costs within the same setting, in the office setting, costs of biosimilar and tbo-filgrastim were $103.61 and $94.07 lower than reference filgrastim, respectively (P < .001 for each comparison). In the outpatient hospital setting, adjusted median costs for tbo-filgrastim were lower than those for reference filgrastim (–$132.90; P < .001), but adjusted median costs of the biosimilar were slightly higher ($20.50; P = .025).

Conclusions: Although previous work has found lower costs for biosimilar filgrastim compared with reference filgrastim, here we found that site of care can change this calculus, reducing savings. After adjusting for patient characteristics and geography, we found that drug cost savings for biosimilar filgrastim were limited to the office setting, with no savings in the outpatient hospital setting.

Am J Manag Care. 2021;27(8):e287-e289. https://doi.org/10.37765/ajmc.2021.88730

Takeaway Points

Biosimilars—drugs that are similar in quality, safety, and efficacy to reference biologic drugs—are intended to increase competition, reduce prices, and improve patient access to important treatments. The first FDA-approved biosimilar was launched in 2015: filgrastim-sndz (Zarxio), a biosimilar for the reference drug filgrastim (Neupogen), a granulocyte colony-stimulating factor used to prevent and treat neutropenia. In this study, we examined the association between site of care and drug cost across reference filgrastim, tbo-filgrastim (Granix), and biosimilar filgrastim administrations among the commercially insured.

• We used claims data to compare drug costs across site of care (outpatient hospital vs office) for short-acting filgrastim administration among commercial enrollees between January 1, 2014, and December 31, 2019.
• In both settings, median drug costs were higher in the outpatient hospital setting than in the office setting.
• After adjusting for patient characteristics and geography, we found that drug cost savings for biosimilar filgrastim were limited to the office setting, with no savings in the outpatient hospital setting.

The 2010 Biologics Price Competition and Innovation Act created a pathway to approve biosimilars—drugs similar in quality, safety, and efficacy to reference biologic drugs—aiming to increase competition, reduce prices, and improve access. In 2015, the first FDA-approved biosimilar was launched: filgrastim-sndz (Zarxio), a biosimilar for filgrastim (Neupogen). Filgrastim is a granulocyte colony-stimulating factor used to prevent and treat neutropenia. A third product—tbo-filgrastim (Granix)—was approved as a new biologic in the United States before there was a biosimilar regulatory pathway but is considered a biosimilar in Europe. Although patients receive identical drug administrations in office or hospital outpatient settings, insurers generally reimburse higher payments to outpatient hospital settings to cover presumed greater overhead costs.

We examine the association between site of care and drug cost across reference filgrastim, tbo-filgrastim, and biosimilar filgrastim administrations among the commercially insured.

METHODS

We used OptumLabs Data Warehouse, which contains longitudinal, deidentified pharmacy and medical claims data for enrollees across the United States.¹ We identified the site of care of each filgrastim administration in commercial enrollees (January 1, 2014, to December 31, 2019) using place-of-service and procedure billing codes. Our unit of observation was an episode of filgrastim use: a period with 1 or more filgrastim administrations and fewer than 7 days between administrations. We required enrollment in medical benefits from 6 months prior to the index date (patient’s first episode in study period) through the episode end. Patient demographic information as of the index date included age, sex, race/ethnicity, age-adjusted Deyo-Charlson Comorbidity Index score, and state of residence. We excluded episodes with multiple filgrastim products or unknown demographic information. To control for geographic differences in drug administration patterns, we estimated median adjusted episode-level drug costs, defined as the sum of out-of-pocket and health plan paid amounts for the drug per date of service in the episode using a quantile regression model with robust standard errors. Drug costs were defined as the amount paid for the claim line with the drug code (J-/Q-code).

RESULTS

The Table presents patient characteristics and unadjusted drug costs. In 65,372 included episodes, patients treated in office settings were slightly older, were more likely to be female, and had modestly higher comorbidity scores.

The Figure presents model-adjusted median costs for each filgrastim product in each setting; the difference in median adjusted costs compared with reference filgrastim in the same setting is also provided for biosimilar and tbo-filgrastim. For each filgrastim product, model-adjusted median drug costs were higher in the outpatient hospital setting than for the same drug administered in the office setting. Comparing drug costs within the same setting, in the office setting, costs of biosimilar and tbo-filgrastim were $103.61 and $94.07 lower than reference filgrastim, respectively (P < .001 for each comparison). In the outpatient hospital setting, adjusted median tbo-filgrastim costs were lower than those for reference filgrastim (–$132.90; P < .001), but adjusted median costs of the biosimilar were slightly higher ($20.50; P = .025).

DISCUSSION

Although previous work has found lower costs for biosimilar filgrastim compared with reference filgrastim,² here we found that site of care can change this calculus, reducing savings. After adjusting for patient characteristics and geography, we found that drug cost savings for biosimilar filgrastim were limited to the office setting, with no savings in the outpatient hospital setting. Tbo-filgrastim was less expensive than reference filgrastim in both settings. Although findings in this study are limited to commercially insured enrollees, these data suggest that savings from biosimilar drugs are context specific. It is unknown whether other biosimilar drugs show similar discrepancies in cost differences by setting. New site-neutrality policies implemented in 2019 reduce CMS payments for off-campus hospital outpatient departments. We do not observe in this analysis whether hospital outpatient departments are on or off campus.

Author Affiliations: OptumLabs Visiting Fellow (JC, MMJ), Eden Prairie, MN; University of Minnesota School of Public Health (JC), Minneapolis, MN; National Bureau of Economic Research (PK-M), Cambridge, MA; University of Minnesota Carlson School of Management (PK-M), Minneapolis, MN; Mayo Clinic (RSG, MMJ), Rochester, MN; University of Minnesota College of Pharmacy (SS), Minneapolis, MN; University of Minnesota Medical School (DW), Minneapolis, MN.

Source of Funding: American Cancer Society (P006396501).

Author Disclosures: In the past 36 months, Dr Karaca-Mandic has provided consulting services to Tactile Medical, Precision Health Economics, and Sempre Health for work unrelated to this paper. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (JC, PK-M, RSG, SS, DW, MMJ); acquisition of data (PK-M, MMJ); analysis and interpretation of data (JC, PK-M, RSG, SS, DW, MMJ); drafting of the manuscript (JC, PK-M, RSG, SS, MMJ); critical revision of the manuscript for important intellectual content (JC, PK-M, RSG, SS, DW, MMJ); statistical analysis (JC, PK-M, MMJ); obtaining funding (PK-M, MMJ); administrative, technical, or logistic support (MMJ); and supervision (PK-M, MMJ).

Address Correspondence to: Molly Moore Jeffery, PhD, MPP, Mayo Clinic, 200 First St SW, Rochester, MN 55905. Email: jeffery.molly@mayo.edu.

REFERENCES

1. Wallace PJ, Shah ND, Dennen T, Bleicher PA, Crown WH. Optum Labs: building a novel node in the learning health care system. Health Aff (Millwood). 2014;33(7):1187-1194. doi:10.1377/hlthaff.2014.0038

2. Karaca-Mandic P, Chang J, Go R, Schondelmeyer S, Weisdorf D, Jeffery MM. Biosimilar filgrastim uptake and costs among commercially insured, Medicare Advantage. Health Aff (Millwood). 2019;38(11):1887-1892. doi:10.1377/hlthaff.2019.00253