Cost Reduction

Why addressing unexplained variations in care should be a healthcare system priority

October 3, 2022 1:12 pm

Healthcare researchers believe Medicare could save $2 billion by using a cheaper drug for retinal disease, but some physicians don’t want to switch.

Population health experts at CommonSpirit Health have singled out a costly drug with a much cheaper alternative that is not being prescribed by many physicians who could.

If just a portion of those non-prescribing physicians switched to the lower-cost retinal disease drug, researchers at Chicago-based CommonSpirit estimate that it would generate yearly savings for Medicare of $2.3 billion, an amount that exceeds the annual savings from the entire Medicare Shared Savings Program (MSSP).

The situation underscores some major flaws both with CMS’s payment model for the drugs involved and with the structure of Medicare accountable care organizations (ACOs).

But more important, experts say there are other similarly simple ways to save the system money while maintaining the level of healthcare.

“This is the tip of the iceberg — the iceberg is heart failure, diabetes, angina, lung disease,” said David B. Nash, MD, MBA, founding dean emeritus and the Dr. Raymond C. and Doris N. Grandon Professor of Health Policy at the Jefferson College of Population Health.

Nash considers the case of retinal-disease biologics to be one of many examples of unexplained variation in care.

Nash points to a fundamental problem in U.S. healthcare that also was highlighted by the CommonSpirit analysis.

“Who is accountable for tackling unexplained variation, and what’s in their toolkit?” Nash said.

A cheaper alternative

What the CommonSpirit researchers found were drastic cost differences in the treatment of retinal disease. Their findings, published in NEJM Catalyst, describe how three biologics have shown similar outcomes in clinical trials, but with widely varied cost. Two of the agents — ranibizumab (Lucentis) and aflibercept (Eylea) — cost $1,883 and $2,098 per injection, respectively, and are approved by the Food and Drug Administration (FDA).

The third biologic, bevacizumab (Avastin), costs just $90 per injection, but only has FDA approval for the treatment of advanced colon cancer and is prescribed off-label for retinal disease.

The highest-cost option, aflibercept, is the most common choice for ophthalmologists, garnering 41% of prescriptions in 2019, according to CommonSpirit Health’s analysis. Low-cost bevacizumab came in second with 38%, followed by ranibizumab with 21%.

The CommonSpirit team found no obvious pattern in prescribing behavior. Of the 3,017 ophthalmologists who wrote at least 100 prescriptions in 2019, 9% used bevacizumab exclusively, 7% never chose it, and the vast majority used it as well as one or both of the other options.

If 75% of first-line prescriptions for retinal disease were for the low-cost option, the government would save $2.3 billion a year in the Medicare fee-for-service program alone, the CommonSpirit research found. By comparison, the 513 ACOs that participated in MSSP in 2020 generated about $1.86 billion in net savings for CMS, compared with benchmarks.

As a result, aflibercept is the top Part B expense for Medicare, and ranibizumab weighs in at No. 6, according to an analysis of 2019 data by Kaiser Family Foundation.

Contrast in cost


Medicare cost

Estimated physician markup

Percentage of prescriptions













Three injectable drugs — aflibercept, bevacizumab and ranibizumab — have similar efficacy but very different reimbursement to physicians for administering the agents for retinal disease.
Source: CommonSpirit Health research, HFMA estimates

No easy fix

Landing on a solution will require work from more than a single, albeit huge, health system.

“This isn’t something we can figure out how to do alone,” said Nicholas Stine, MD, system senior vice president for population health for CommonSpirit Health. “This is something that we need everybody at the table if it’s actually going to change.”

Stine and his colleagues published their analysis of the savings opportunity, in part, to demonstrate the challenges to capturing those types of savings in care provided by ACOs run by CommonSpirit Health.

ACOs’ traditional strategies for reducing total cost of care have been focused on improving care coordination and improving outcomes in vulnerable populations. The savings that come from those labor-intensive strategies pale in comparison to the savings opportunities associated with high-cost drugs, Stine said.

He and the co-authors are calling for “a broader sense of stakeholder responsibility, attention and prioritization” of savings opportunities like the retinal-treatment example. They predict that as high-cost Part B drugs with similar efficacy to existing agents continue to emerge, they will become more common.

“Part B drugs are one of the fastest growing areas of cost in medicine, and it’s sneaking up on people,” Stine said. “The way ACOs are set up, we have limited sticks and limited control over formulary. We need a conversation about how ACOs and other healthcare stakeholders collaborate to manage these types of costs.”

The inability of ACOs to influence specialty spending reflects a flaw in the MSSP design, according to Robert Berenson, MD, a fellow at the Urban Institute and a former member of the Medicare Payment Advisory Commission.

Primary care services account for between 2.1% and 4.9% of overall Medicare spending, according to a 2019 analysis by RAND Corp.

“Spending on specialists is substantially more than that,” Berenson said.

Because ACOs are designed to focus on primary care — and because physicians are paid under the Medicare physician fee schedule — the ACOs are inherently limited in their ability to control costs. ACOs have the incentive, but counting on non-
employed specialists who are paid via fee-for-service to rein in costs is wishful thinking, Nash said.

By contrast, provider organizations that employ both primary care and specialty physicians are positioned to influence physician behavior. Nash cited Optum, which employs more than 50,000 physicians, as an example.

“This is their secret weapon,” he said. “If you own the providers and you have the data (about physician performance), you’re off to the races.”

Comparing physician performance allows identification of best practices, which can be mandated and monitored via technology.

“Eventually, they can get to a point of saying, ‘This is the Optum way. If you don’t do it our way, see you later,’” he said.

Drugmakers’ role

But ophthalmologist David Glasser, MD, secretary of federal affairs for the American Academy of Ophthalmology, said that holding physicians accountable for drug expenditures ignores the root cause of the problem.

“The problem doesn’t lie with the physician,” Glasser said. “The problem lies with who is making the decisions on how much drugs cost.”

Genentech sells both ranibizumab and bevacizumab in the United States. It has not sought FDA approval for the latter, even though CMS, at the Academy’s urging, funded a trial that showed bevacizumab is not inferior to ranibizumab, Glasser said.

“It would have cost them a lot of money to get it approved for that indication, and it may have cannibalized their more expensive drug, Lucentis,” he said.

Genentech declined comment.

A counterintuitive fix

Glasser, who has been tracking the retinal- disease treatments for years, pointed to a solution that aims to reduce the disincentive to prescribe the cheaper drugs.

“We found that if Avastin [bevacizumab] use increased by 5% to 10%, Medicare could save hundreds of millions of dollars a year,” Glasser said. “So, we agree with the need for increased Avastin use.”a

In the past seven years, however, the use of bevacizumab has significantly decreased, according to a study published in Ophthalmology Retina. Glasser and his co-authors suggest that a reimbursement rate that disincentivizes the use of bevacizumab is the cause.b

Glasser and his colleagues believe that bevacizumab use would rise modestly — by 9% to 10% — if physicians were reimbursed for it at a higher rate. At the current rate of $90 per injection, the margin over a practice’s acquisition cost is too low for many practices to cover the storage and handling costs for the biologic.

Raising the reimbursement rate to about $126 per injection would cover those expenses and still save money by reducing use of the higher-priced agents.

“It’s not the case that physicians are primarily interested in just making more money off the increased overhead payment they get with the more expensive drug,” he said. “We see (nearly) 40% are using Avastin already and that number would be greater except for the physicians who can’t use it because they’re losing money on it.” 

Why some ophthalmologists choose a $1,900 injection over a $90 injection

CommonSpirit Health, a 21-state health system that operates 14 accountable care organizations (ACOs) serving 334,000 beneficiaries in the Medicare Shared Savings Program (MSSP), has had limited success convincing independent ophthalmologists to increase their use of bevacizumab within its sponsored ACOs. CommonSpirit population health leaders identified several possible reasons why.

MSSP ACOs have little leverage over independent specialists to influence their prescribing habits.

Specialists compensated via Medicare’s fee-for-service system often have no incentive to curtail spending. Even if an ophthalmologist is in-network as part of an ACO, the incentive pay for working to lower costs would be a small fraction of their fee-for-service compensation.

Under the buy-and-bill model for Part B medications, providers purchase drugs themselves and bill Medicare for the price plus an additional 6% as a handling fee. This financially incentivizes providers to opt for higher-priced medications, which generate higher handling fees.

Physicians often have entrenched prescribing habits. One CommonSpirit Health ACO saw significantly increased bevacizumab prescriptions after a local respected retina specialist championed its use, suggesting that physicians are influenced by peers.


a. Glasser, D.B., Parikh, R., Lum, F., and Williams, G,A., “Intravitreal anti-vascular endothelial growth factor cost savings achievable with increased bevacizumab reimbursement and use,Ophthalmology, December 2020.
b. Repka, M.X., Glasser, D.B., Cohen, C.G., et al., “Use of bevacizumab injections in Medicare fee for service in the IRIS Registry,” Ophthalmology Retina, September 2021.


googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text1' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text2' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text3' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text4' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text5' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text6' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text7' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-leaderboard' ); } );