Can the Patient-Centered Outcomes Research Institute Become Relevant to Controlling Medical Costs and Improving Value?

One result of health care reform legislation was establishment of the Patient-Centered Outcomes Research Institute (PCORI).¹ This relatively generously funded institute is intended to provide support for research that “helps people make informed healthcare decisions and allows their voice to be heard in assessing the value of healthcare options.”² This kind of research “assesses the benefits and harms of preventive, diagnostic, therapeutic, or healthcare delivery system interventions to inform decision-making, highlighting comparisons and outcomes that matter to people.”²

Research that the PCORI funds will not find the vaccine to cure cancer, the medication to eliminate obesity, or the treatment to cure Alzheimer disease. But the PCORI's research is important. This research can help physicians and patients decide which of multiple tests or therapies is slightly better or no different in terms of its effects on patients with specific symptoms or diseases. PCORI research can help to determine how the delivery of health care can be improved, perhaps by altering the way that teams of health professionals are assembled, or the ways that e-mail or telephone communication are used to deliver care.

Collecting detailed data about the cost of interventions being compared is not within the PCORI's scope, even though value, which most individuals believe includes cost, is central to the institute's mission. An example from past research brings the effect of this decision into sharper focus. More than 40 years ago, the US government funded the RAND Health Insurance Experiment (HIE).³ Its purpose was to test whether patient cost sharing affected health care use, quality, health status, and cost of care. The HIE was so large that both the Secretary of Health, Education, and Welfare and the president had to authorize its funding. Indeed, the HIE was a page in the president's budget book. To replicate the HIE today might cost $1 billion—an amount similar to the PCORI's budget. Among the findings of the HIE that have been adopted is the use of patient cost sharing to incentivize patient and physician behavior.

What if the president, instead of signing off on the experiment, had said that the US government would pay for understanding how cost sharing affects health (which is what patients value) but that no cost data could be reported, and that the HIE's results could not be used to affect policy directly?

Within a few years, hundreds of studies funded by the PCORI will begin to appear in medical journals. Whether the studies focus on a test, a therapy, or a health delivery intervention, the results will be 1 of 2 types: (1) the test, therapy, or interventions being assessed led to small but statistically significant changes in health or another health outcome (some effects are likely to be small for an individual, but important when generalized to a large population)⁴ ; or (2) the alternatives being compared were deemed to be equivalent (all interventions produce different outcomes but small differences are difficult to detect, and powerful designs and large sample sizes are needed to detect them). However, regardless of the results of these studies, cost information almost certainly will not be included.

How might such studies be used by a patient and physician? Fast-forward to a potential conversation. Assume that the results are equivalent and the patient says to the physician: “I heard the distinguished leader of the Medicare program speak about the need for affordable health care. I want to do my part to keep Medicare costs down so give me the less expensive intervention, as long as my health is not impaired as a result.” Other than scratching his or her head, how can the physician determine which of the 2 therapies would turn out to be less expensive either for the patient or for the system? If the results of the interventions showed a difference and the patient, who has a high deductible health plan, wants to know whether it is worth paying for X vs Y, how will the physician respond?

To the patient and the health system, value should mean both health and cost. For example, some employed individuals with chronic disease decreased their use of statins and antihypertensive medications in response to small increases in co-payment. This outcome would not have occurred if the patient's share of the cost of the medication were not important.⁵ But in the absence of data about the cost of the intervention, it is impossible to inform patients about the effects of the trade-off they are making. So what can be done?

A scientific and political case could be made to amend health care legislation to require that cost be included in the studies that the PCORI funds. But in the current political environment, such a move could place the institute at grave risk of being defunded.

The research community could reject the policy of excluding cost by not applying for and not accepting funding under the conditions established by the PCORI. There is precedent for this kind of stance. For example, the federal government funds some important work by contract with the proviso in some contracts that the work cannot be published.⁶ This condition could be considered analogous to excluding cost data in the PCORI studies. To their credit, many researchers and institutions do not apply for, or accept, federal monies under such conditions.

Medical journals choosing to publish the PCORI-funded research that does not include cost data could encourage authors to consider including comments about the potential implications of cost in clinical decision making in the discussion section of the article. The journals could also ensure that an accompanying editorial discusses cost implications.

Another approach might be to establish an institute for value-based health care, governed with complete independence from its sponsors. Private industries and foundations both have interests that might motivate them to fund such an institute. After an article based on the PCORI-funded research was published, investigators and analysts at the institute for value-based health care would conduct a cost analysis, which could be made available in the public domain.

With these approaches, the health care legislation governing the PCORI need not change, and the risk of defunding the PCORI would be averted. The resources for an institute for value-based health care could go farther because it need not fund the effectiveness part of many studies that would be relevant to health care professionals because those assessments will already have been funded by the PCORI. If such an approach proves feasible, perhaps this broader principle could be adopted: regardless of who funds the clinical research or where in the world it is conducted, it would be complemented by a cost analysis. This approach could apply to clinical trials funded by the National Institutes of Health, work performed by the US Preventive Services Task Force, or national coverage decisions made for the Medicare program—all of which ordinarily exclude cost considerations.

Many details would need to be resolved, but the world is truly flat when it comes to research. With cutbacks in health care being contemplated throughout the Western world, physicians and patients should be able to have fact-based conversations about value.⁷ But to make that possible, creative ways must be found to address the limitations placed on work funded by the PCORI. Incorporating the cost and value perspective is essential so that the work of the PCORI can be clinically relevant and optimally used by patients and physicians.