Traditional Managed Care Tools Just Don’t Work in Oncology

Peter G. Ellis, MD

May 2010, Vol 1, No 1 - VBCC Perspectives

As payers focus on the rising cancer care spend, they are turning to tools previously applied to other diseases. Each seems to have potential, but upon closer review, these have fundamental shortcomings when used for cancer.

Disease management. Payers pay to have nurses in national call centers reach out telephonically to patients to help avoid hospitalizations. It is a labor-intensive model, often perceived by physicians as redundant at best and conflicting at worst with direct patient care. Return on investment in cancer care has proven elusive due to the high number of confounding variables. Most importantly, reductions in side effects and hospitalizations are better achieved through evidence-based treatments and supportive care combined with accountability and financial incentives to physicians for managing the entire continuum of care.

Utilization management (UM) imposes prior authorization as an external control on patient care. Oncologists have to obtain approval before initiating a treatment plan. With the complexity of cancer diagnoses and potential toxicities, it is difficult to second-guess cancer treatment decisions from outside the process of direct patient care. Oncologists’ therapy decisions are rarely outside national guidelines, and therefore cannot and should not be denied. Thus, UM adds a costly, cumbersome administrative process that is likely to have minimal impact on patient treatment. Moreover, UM imposes compliance costs on practices and offers physicians no incentive to narrow the variability of care or prioritize cost-effectiveness. Far greater savings could be achieved by empowering and incentivizing physicians to adopt evidence-based best practices.

Pharmacy benefit management (PBM), including specialty pharmacy, has long been utilized to drive down pharmaceutical costs by enforcing use of generics over brands through formularies or tiered pricing along with negotiating volume discounts with pharmaceutical companies. This can work for high-volume drugs, such as statins for lowering cholesterol. But it is not applicable in cancer care. For oral oncology drugs, most of the spending is on single-source drugs whose price cannot be negotiated with pharma based on market share or volume targets.

Managing injectable or infusional agents through PBM mail order is problematic due to 3 factors. First, with oncology reimbursement on the basis of Healthcare Common Procedure Coding System codes, all versions of a multisource drug, such as paclitaxel, are reimbursed by insurers at the same rate, whether the drug is the branded or generic version, which leaves no advantage to managing drug selection. Second, the waste associated with drop shipments of injectable or infusional drugs is high due to the frequent changes in cancer treatment plans. Drugs are shipped as patient-specific prescriptions and cannot be used for any other patient or returned to the mail order pharmacy. Finally, until the reimbursement system for oncology changes, the revenues associated with drugs are the vital financial engine to keeping community-based oncology care viable.

Benefit design. Coinsurance can be tiered by drug or increased in general by putting cancer drugs under pharmacy benefit plans instead of treating them as a medical benefit. In addition to possibly being an unfair burden on cancer patients, since the vast majority of cancer therapies are extremely expensive, this tactic is likely to fail because patients lack the expertise to unsubstantiated therapies), in cancer there is often more than one treatment regimen with demonstrated antitumor efficacy for a given tumor type, histology, stage of disease, and treatment intent. How should we prioritize these therapies and select among them? A next challenge to achieve better care at lower cost is to define the components of value (such as potential for prolonging life or cure, side effects [both acute and long-term complications], cost, patient experience, etc) and to devise appropriate weighting of the components. With more than 750 therapeutic agents in the development pipeline for cancer alone, systems-based decision-support at the moment of clinical prescribing that aligns proper prescribing with appropriate reimbursement is now needed to assist therapeutic decision-making in community practice.