Lung Screening Proposal Renews Scrutiny of Preventative Services Task Force

Once fairly obscure, the U.S. Preventive Services Task Force has gotten headlines in recent years by questioning the value of mammograms for women in their 40s and recommending that men not have the prostate-specific antigen (PSA) test unless they have already been diagnosed with prostate cancer.

These decisions drew an uproar from patient advocacy groups and members of Congress, who accuse the task force — an independent board of medical specialists selected by the federal Agency for Healthcare Research and Quality — of trying to take away important cancer-screening tools.

Last month, the task force drew a very different kind of rebuke, and this time, it came from what amounts to a peer group. Medicare directed its own high-level committee to examine the task force’s December 2013 recommendation that many current and former heavy smokers ages 55 to 80 should get high-tech scans annually for lung cancer.

After reviewing the scientific research available to the task force, Medicare’s Evidence Development & Coverage Advisory Committee in effect rejected the recommendation, concluding that such widespread (and potentially costly) screening could lead to more harm than good.

Only one of the Medicare panel’s nine voting members said he was confident that the benefits of these CT scans of the lungs outweighed the risks for the Medicare population. The other eight voted that they had little confidence, giving grades of 1, 2 or 3 on a scale of 1 to 5.

Steven Woolf, a former member of the Preventive Services Task Force and a non-voting participant in Medicare’s review, took his former employer to task at the panel’s April 30 meeting.

“In my day, looking at the evidence that has been presented, this would not have received a B recommendation,” said Woolf, now the director of the Center on Society and Health at Virginia Commonwealth University.

As Woolf sees it, the task force has shifted away from its duty to focus solely on the scientific evidence in weighing the merits of preventive services, largely because of the new clout that the 2010 health care overhaul gave to its recommendations. Under the law, while Medicare isn’t bound by the work of the task force, private insurance plans are required to cover services that get the task force’s top ratings of A or B. Woolf says the task force, stung by the backlash against its mammography and PSA recommendations in 2009 and 2012, respectively, may be tilting toward approving services too readily

“Now that this new law has entered into the picture, I worry that the task force members can’t avoid thinking about economic implications, and the impact that this might have on patients,” Woolf said in an interview.

Michael L. LeFevre, the chairman of the task force, declined to comment on the Medicare panel’s vote. In an interview, he stressed that the mandate for the task force has remained unchanged. Its members deliberate intensely over the available evidence of both benefits and potential harms of a service, he said. While the task force members try to insulate this work from political and financial questions, the landscape in which they work has changed in recent years, he added.

“The scrutiny of the task force’s work has increased. I would be somewhat dishonest to suggest to you that it hasn’t,” he said. “Equally, I would suggest that it would not be realistic to say that we don’t get some pressures from interest groups, for example, that didn’t happen before.”

Medicare has said it expects to unveil by November its proposed decision on covering lung CT scans. The agency, which pays for health care for about 50 million elderly and disabled Americans, has already been asked by medical groups such as the American College of Radiology to cover the scans.

The key study here is the National Lung Screening Trial, a project that involved more than 53,000 participants. Otis Brawley, the chief medical and scientific officer of the American Cancer Society, explained its results in a public posting: “One way of looking at this is: among about 27,000 people screened with a CT scan, 87 lung cancer deaths were prevented, but 356 lung cancer deaths still occurred,” he wrote.

Brawley also noted that the test led to problems for some patients. Abnormal readings on the scans triggered many follow-up procedures, such as needle biopsies and bronchoscopies, where a tube is threaded from the mouth down into the lungs. These can lead to complications and perhaps even prove fatal, Brawley said, noting that some experts have interpreted the NLST results as showing one life lost due to screening and diagnostic procedures for each five or six saved.

With these risks in mind, the American Cancer Society recommends that lung CT scans be done at specialized centers to lower the number of false positive screens and extra diagnostic procedures.

Members of the Medicare panel last month raised concerns about whether more widespread use of lung CT scans could trigger a replay of what happened with PSA screening, in which many men suffered harm as a result of screening. Brawley, who is also a professor at Emory University in Atlanta, earlier had urged the task force to recommend against widespread use of the PSA test. The task force said that 29 of 1,000 men would suffer impotence and 18 of 1,000 would suffer incontinence due to prostate treatments. Among those who have surgery, there is a rate of 1 death for every 3,000 men screened.

After reviewing the evidence available, the task force said in 2012 that “at best, PSA screening may help one man in 1,000 avoid death from prostate cancer after at least 10 years.”

Lawmakers including Sen. Jeff Sessions, R-Ala., have urged the task force to change that view, and Medicare still covers the PSA test, and many states require insurers to do so.

Many people “have a blind faith in early detection of cancer and subsequent aggressive medical intervention whenever cancer is found,” Brawley has said. “There is little appreciation of the harms that screening and medical interventions can cause.”

Topics: