Credit: Aetna

Cancer care is expensive. In the US today, tens of billions of dollars are spent each year on direct medical expenses related to the disease, and yet whether Americans are getting value for this massive expenditure depends on whom you ask. If you read the newspaper or watch television, you'd think medicine is moving forward at a rapid pace and, thus, patients are getting good value for their money. But the truth is that survival for the most common adult malignancies has not improved much. For many treatments, the medical enterprise lacks precise measures of benefit that accurately capture the essence of true value.

As practicing physicians and as the medical leadership for Aetna, a large U.S. health insurance carrier, we see this evidence gap influence decisions every day—as do healthcare administrators and physicians across the country. Patients, their families, their communities and their employers (who are often paying for health care) are forced to make individual decisions about treatment and associated costs while simultaneously struggling with life-threatening illness. Society asks them to be consumers without providing them with the data they need to be informed consumers. Today, they must use tools that are rooted in the history of cancer care, in government regulatory standards, in unjustified promises of effectiveness and often in provider bias.

The US Food and Drug Administration (FDA), when it evaluates a drug, does not so much define value as judge safety and efficacy. Health economists argue in favor of metrics such as QALYs, quality-adjusted life years gained, which, although transparent and reproducible, are arbitrary and mostly meaningless to patients and their doctors. And government bodies such as the UK's NICE, or National Institute for Health and Care Excellence, which determines clinical policies in England and Wales, are not felt to be very nice. That is because, to this point, no one has really proposed a way to capture the less tangible benefits that matter most to patients, their doctors and society—such as symptom burden, financial costs, family disruption and the ability to work or serve as caregivers.

Clinical trials do not currently provide much insight into value. Survival is the gold standard, and progression-free survival is often used as a surrogate. But what are these things really measuring? Patients may live longer or better, but then again, these are not really metrics of overall health. What, after all, constitutes 'better'? This has rarely been addressed in the FDA approval process, and, as FDA approval has almost uniformly translated into coverage by the US Centers for Medicare and Medicaid Services and other third-party payers, it remains unanswered after approval.

With an increasing number of therapeutic options with comparable clinical efficacies, toxicities and costs (irrespective of mechanism of action), there is a growing need for a new way to measure value. A new measure could breathe life into comparative effectiveness research and spawn a new field of health outcomes research. Although this research is unlikely to affect the role of the FDA or other regulatory agencies, the results could immediately have an impact on coverage policy for commercial payers. Such health outcomes research could revolutionize the dialogue between payers and employers into one focused on benefits design to promote 'wellness' in the cancer population. This line of investigation is especially needed now with the rise of accountable care organizations, or ACOs—integrated healthcare providers that aim to replace traditional fee-for-service reimbursements with systems that tie costs to quality of care. Increasingly, ACOs will bear significant financial risk and become influential judges of value in US health care going forward.

To develop surrogates for real-world benefit, we need to begin to define what is really important to patients and caregivers, and we need to develop and validate instruments where none exist to facilitate this data acquisition and analysis. These should operate on a standardized platform and should be consensus driven and transparent. Data collection should capitalize on advances in information technology and the explosion of electronic connectedness, allowing the patient's voice to be heard. This would occur in parallel to collection of conventional measures of clinical benefit.

The economic impact of treatment also needs to be redefined. At this point, when considering cost, we focus almost exclusively on the sticker price. We need to ask whether a drug keeps our patients out of emergency rooms and hospitals. We need to ask whether treatment decisions have significant impact on downstream responses, health resource consumption and development of other medical problems. We need to ask whether the drug allows people to continue to work, care for their children or elderly parents and depend less on other caregivers. We need to ask whether the drug allows patients to maintain health longer.

In order to successfully transform the value universe, we will be forced to integrate large amounts of data. These data sets either do not currently exist or are not currently coordinated. The ability to collate, sort and interpret big data will be crucial to success.

For the pharmaceutical industry and other innovators who fear that healthcare reform will destroy their business models, this evolution allows an opportunity to not only position a new drug in what may become a very regulated formulary but also redefine the value afforded by old drugs post-marketing. This could dramatically expand opportunities for health outcomes and comparative effectiveness research with actionable results. It would also allow actualization of the long-desired relevance of patient-reported outcomes. It is time to move beyond the debate of whether six months of progression-free survival justifies arbitrarily defined prices. In the absence of a cure, the research effort needs to focus on what is really important.

figure b

Aetna

figure c

Aetna