Studies found that centers provided incomplete or unbalanced information, which could lead to unnecessary screenings and health complications in older adults.
Say a postcard arrives in the mail, a reminder to make an appointment for a mammogram. Or a primary care doctor orders a PSA test to screen a man for prostate cancer, or tells him that because of his years of smoking, he should be screened for lung cancer.
These patients, trying to be informed customers, may look online for a cancer center to learn more about screening, when it is recommended and for whom.
It might not be the best move. Medical societies and the independent U.S. Preventive Services Task Force publish guidelines about who should be screened for lung, prostate and breast cancers and how frequently, among many other prevention recommendations. But websites for cancer centers often diverge from those recommendations, according to three studies published recently in JAMA Internal Medicine.
Researchers found that some sites discussed the benefits of screening but said little about the harms and risks. Some offered recommendations about the age at which to start screening but glossed over when to stop — an important piece of information for older adults.
“If we acknowledge that these websites are important sources of information, based on screening according to the guidelines we have room for improvement,” said Dr. Behfar Ehdaie, a urologist at Memorial Sloan Kettering Cancer Center in New York and an author of the study on prostate cancer screening recommendations.
Screening refers to tests for patients with no symptoms or evidence of disease, including prostate-specific antigen tests, mammograms, colonoscopies and CT scans.
The researchers analyzed more than 600 cancer center websites that provided recommendations for prostate screening, and found that more than one-quarter recommended that all men be screened. More than three-quarters did not specify an age at which to stop routine testing.
Yet guidelines from both the Preventive Services Task Force and the American Urological Association state that men over 70 should not be routinely screened, because, according to the Task Force guidelines, “the potential benefits do not outweigh the expected harms.”
For men aged 55 to 69, both groups urge individual decisions after a discussion with a clinician about benefits and harms. Neither group, though, recommends routine screening for younger men at average risk.
Moreover, the study reported, 62 percent of cancer center websites did not include information on the potential harms of screening. Because prostate cancer grows slowly, it often causes no problems. But detection and treatment can lead to complications from surgery or radiation, including lower quality of life from incontinence and sexual dysfunction.
The surveys found similar problems on websites discussing other cancer screenings. In a study of over 600 breast cancer centers, more than 80 percent of those recommending a starting age and intervals for screening mammograms were at odds with guidelines. The study did not address whether the websites included information on when to stop.
The Preventive Services Task Force’s 2016 guidelines, which are currently being updated, recommend screening mammograms every other year for women aged 50 to 74; it found insufficient evidence of benefit and harms for those 75 and older. The American Cancer Society recommends annual or biennial screening for women over 55 at average risk, as long as they have a 10-year life expectancy.
In a study of over 600 breast cancer centers, more than 80 percent of those recommending a starting age for mammograms were at odds with guidelines.Credit…Kimberly P. Mitchell/Detroit Free Press, via Associated Press
Lung cancer screening, however, is recommended only for those at high risk because of smoking history and older age. Here, too, an analysis of 162 cancer center websites showed that about half did not address potential harms.
“We think it’s important to present a balanced account,” said Dr. Daniel Jonas, an internist at Ohio State University College of Medicine and senior author of the study. “It’s fair to say they could do a better job.”
Concerns about overtesting and overtreatment of certain cancers in older adults have persisted for years. “The harms of screening occur early,” said Dr. Mara Schonberg, an internist and health care researcher at Beth Israel Deaconess Medical Center in Boston. But the benefits of screening can accrue years later; older patients with other health problems may not live long enough to experience them.
With mammography, for instance, harms include false positives, leading to repeat mammograms or biopsies, the psychological consequences of which can continue for months, Dr. Schonberg’s research has shown.
And while most breast cancers diagnosed in women over 70 are very low risk and might never progress, “nearly all are treated with surgery,” Dr. Schonberg said, and sometimes thereafter with radiation and endocrine drugs, all of which can have negative side effects.
As for benefits, the data showed that 1,000 women aged 50 to 74 would have to undergo mammography for nearly 11 years to prevent one death from breast cancer.
Why would some cancer center websites omit possibilities like false positives, repeat testing, radiation exposure or the aftereffects of surgery? Why don’t they include information on how many lives screenings actually save at particular ages?
“In the U.S. health care system, the more procedures you do, the more you get paid,” said Dr. Alexander Smith, a palliative medicine specialist and geriatrics researcher at the University of California, San Francisco. Radiology, which is required for both lung and breast screenings, “is one of the biggest moneymakers for health systems,” he noted.
Some websites may have been developed by marketers with little input from health professionals, Dr. Jonas added. Talking about risks could discourage patients from clicking the “Make an Appointment” button.
On the other hand, it can be hard to dissuade older patients from screening, even when research shows little benefit.
Dr. Schonberg has developed and tested decision aids — pamphlets to help women over 75 and their doctors reach evidence-based conclusions about mammograms.
To some extent, they work. Older women who receive the pamphlets are more knowledgeable and more apt to discuss benefits and risks with their doctors; they are less inclined to continue screening. But over 18 months, about half of women who received decision aids got mammograms anyway, as did 60 percent of those who hadn’t.
Dr. Schonberg explained it as habit or “the need for reassurance.” Patients may also overestimate their risk level; the average 75-year-old woman has a 2 percent chance of a breast cancer diagnosis over five years, she pointed out.
Moreover, screening choices involve an issue some older patients (and doctors) prefer to avoid: life expectancy. The American Cancer Society and some medical groups use 10-year life expectancy, rather than age cutoffs, as guidelines for when older patients can stop screening.
“Prognosis is one of the key factors in decision making,” Dr. Smith said. “Are patients going to live long enough to experience the benefits?” That can be an uncomfortable conversation involving age, health and mortality.
How should older adults inform themselves about cancer screenings? In addition to discussing pros and cons with their doctors — Medicare requires such a visit before it will cover a lung cancer screening — patients can go to the U.S. Preventive Services Task Force website for the latest assessments.
They can also use ePrognosis, an online guide that Dr. Schonberg, Dr. Smith and colleagues at U.C.S.F. developed a decade ago. Most visitors are health care professionals, but patients can also use the site’s calculators to determine whether they are likely to benefit from breast and colon cancer screenings. They can use questionnaires that help to determine their probable life expectancy, as well as several decision aids.
Of course, patients can consult cancer center websites, too — but with an eye to what may be missing.