The high cost of health care in the United States in part relates to how clinicians are paid. Performing expensive and often risky procedures simply pays better than engaging patients in a shared decision-making conversation. Shared decision-making (SDM) is a process where clinicians and patients educate each other about treatment options, risks and benefits, and preferences, and then they decide on the best approach together. Patient decision aids can assist in this process to make these interactions practical in the time-starved world of clinical practice. Despite evidence supporting SDM and decision aids, from a recent Cochrane review and numerous randomized trials, SDM and decision aids are underused, and the misalignment of incentives deserves some of the blame.
Recently, the Centers for Medicare & Medicaid Services (CMS) announced a new payment policy for lung cancer screening. Specifically, the new policy provides payment for an SDM visit for patients considering low-dose computed tomography (LDCT) to screen for lung cancer. The SDM visit includes the use of one or more patient decision aids and would help determine clinical eligibility for lung cancer screening, as well as establish “preference eligibility” (the desire of the patient to proceed with screening once fully informed).
Results of the 2012 National Lung Screening Trial (NLST) were key to the creation of this payment policy. More than 50,000 individuals, ages 55 to 74, were enrolled in the NLST. These individuals had a smoking history of at least 30 pack-years and remained smokers or had quit in the past 15 years. Participants were randomized into three rounds of annual screening with either LDCT or chest radiography (the study’s placebo intervention, with no mortality benefit seen in previous randomized trials). After a 6.5-year median follow-up, the absolute risk of dying from lung cancer decreased from 1.7 percent with chest radiography to 1.4 percent with LDCT, a statistically significant reduction. On the other hand, the cumulative risk of a false-positive result raising a “red flag” about possible lung cancer was relatively high at 35 to 40 percent, with a positive predictive value of only 4 percent. The LDCT group had more invasive procedures and more major complications, though the great majority of false positives could be sorted out with serial CT scans … although with further radiation exposure and no doubt some number of sleepless nights for participants with false positive results. Due to the mixed positive and negative results from the NLST, the decision about LDCT screening doesn’t appear to be a “slam dunk.”
Two articles in the September 24, 2014, issue of JAMA, one by Bob Volk and colleagues and the other by Harold Sox, noted the close call between benefits and risks. The articles recommended an SDM approach, as was ultimately reflected in the coverage decision. Another proposal from these articles echoed in the decision memo is for the creation of a registry of people who undergo LDCT, allowing for further study of outcomes in the “real world,” outside of a randomized trial. The registry could be extended to people who are candidates for LDCT but decide not to have it. The registry-generated data could help sharpen future recommendations regarding lung cancer screening.
An implication of this proposed policy requiring the use of patient decision aids as part of an SDM encounter is that some mechanism is needed to ensure that the aids used are accurate and unbiased. A national certification program for patient decision aids, modeled on the International Patient Decision Aid Standards and supported by CMS, would help assure that decision aids are trustworthy.
Payment for physicians who spend time sharing fateful decisions with patients is long overdue. The payment policy proposed by CMS in this decision memo is a very good start. Over time, the same approach should be taken for other tough health care decisions. It’s the ethical and cost-effective thing to do.
Michael J. Barry is president, Informed Medical Decisions Foundation.