Older people and those with limited life expectancy – as little as five years -- often get cancer screenings that likely won’t help and could cause harm, despite national guidelines that recommend against them.
That’s the consensus of two studies and a commentary published Monday in the journal JAMA Internal Medicine that add to the ongoing debate about the consequences of overscreening -- testing inappropriately and without evidence for common kinds of cancer.
“It truly will be a new era when providers will be evaluated, in part, by their ability to refrain from ordering cancer screening tests for some of their patients,” writes Dr. Cary P. Gross, a professor with the Cancer Outcomes, Public Policy and Effectiveness Research (COPPER) Center at the Yale University School of Medicine.
In one study, researchers at the University of North Carolina at Chapel Hill found that between 31 percent and 55 percent of study participants with less than nine years to live were still receiving screenings for four common cancers.
“If you have less than 10-year life expectancy, screening is probably going to harm, but not help,” says Dr. Ronald Chen, an assistant professor with UNC’s Lineberger Comprehensive Cancer Center. “The cancer would never have been the cause of their death.”
Getting that message across to doctors and to patients is necessary – but difficult, says Dr. Scott Ramsey, a health economist and director of the Hutchinson Center for Cancer Outcomes Research, or HICOR, at the Fred Hutchinson Cancer Research Center.
“Overscreening is clearly a problem,” Ramsey says. “This is clearly an area where there is little or no benefit, lots of costs, and potential harm.”
But doctors don’t have good tools for estimating life expectancy – a point acknowledged in Chen’s paper – and it can be easier to simply continue screening sick patients than to have the conversation about why it’s time to stop.
“Frankly, a lot of physicians don’t want to go there,” Ramsey says.
In recent years, once-requisite exams such as the prostate-specific antigen, or PSA, test for prostate cancer, and mammograms for breast cancer have spurred fierce debate among advocates for screening and scientists who argue the results aren’t worth the medical and financial costs. Overall, the key is to have good evidence to support practice-based guidelines and decision-making, adds Dr. Ruth Etzioni, a biostatistics expert in Fred Hutch’s Public Health Sciences division.
“Cancer screening should be used smartly and conservatively,” she says.
Chen and his team analyzed data from more than 27,000 participants aged 65 and older in the population-based National Health Interview Survey from 2000 to 2010. They used a validated tool that measures life expectancy based on 11 factors – function, illnesses, behaviors, demographics – that influence mortality.
More than half of men who had a 75 percent chance of dying within nine years were still getting prostate screening tests, and nearly a third of women with similar dire odds were still receiving cervical cancer exams.
Among those expected to live five years or less, 34 percent of women had recent breast cancer screening and nearly 41 percent of all participants had colorectal exams, the study found.
“Screening tests may seem to be pretty easy tests, but what happens ultimately if a screening test is abnormal?” Chen says.
People may not live long enough to benefit from screening, and even those who do survive may find that a questionable result leads to further tests and potentially invasive treatment that can cause overall harm. Even the anxiety about the results can be a problem, experts say.
The overscreening occurred despite clear guidelines that recommend stopping tests based either on age or life expectancy, the study found. The American Society of Clinical Oncology and the American Cancer Society urge that PSA tests be avoided in men with a less than 10-year life expectancy, for instance. The American College of Physicians recommends no colorectal cancer screening in people older than 75 or with a less than 10-year life expectancy. (The highly regarded U.S. Preventive Services Task Force recommends against PSA screening for all men, and against breast cancer screening in women aged 75 or older.)
In a second study published Monday, researchers found that offering colonoscopies to hypothetical cohorts of Medicare patients more frequently than recommended resulted in more harm to the patients -- and higher costs to the health system overall.
Compared with conducting colonoscopies every 10 years, as recommended, performing the tests at five or three years or in patients past age 85 actually increased complications and boosted costs while cutting the number of colorectal cancer deaths only slightly, found Frank van Hees, a researcher with the Department of Public Health at Erasmus University Medical Center in the Netherlands. Conducting the tests at five years required 909 additional colonoscopies and an additional cost of $711,000 per quality-adjusted life-year, or QUALY, gained.
“This study provides strong evidence and a clear rationale for clinicians and policymakers to actively discourage this practice,” van Hees and colleagues write.
Changing screening practices will be a challenge, the Hutch’s Etzioni and Ramsey agree. Doctors may continue screening despite little or no value in order to avoid lawsuits, increase income -- or simply to avoid difficult conversations with patients who insist on the tests.
“Part of it is force of habit,” Ramsey says. “It’s easier to say yes than to say no.”
The real key will be getting the message across to skeptical consumers that more tests aren’t necessarily better, Ramsey adds. “We need to educate the public about when screening stops being useful.”
JoNel Aleccia is a staff writer at Fred Hutchinson Cancer Research Center. From 2008 to 2014, she was a national health reporter for NBC News and msnbc.com. Prior to that she was a reporter, editor and columnist for more than two decades at newspapers in the Northwest. Reach her at firstname.lastname@example.org.