Screening for cancer has become a routine part of health care, and because early detection typically increases survival rates, there has been much debate over just how early testing should begin and how often screening should be done.

Less discussion has occurred about when screening should stop.

In a report published online today (Oct. 12) in the Journal of the American Medical Association (JAMA), researchers at the Memorial Sloan-Kettering Cancer Center in New York City write that efforts to "foster adherence to screening have led to deeply ingrained habits," and some patients continue to get tested even though they are not necessarily likely to benefit.

Screening in patients with advanced cancer

The researchers, led by physician and biostatistician Camelia Sima, collected data on almost 90,000 patients, ages 65 and older, who had been living with advanced lung, colorectal, gastroesophageal, breast or pancreatic cancer for at least two months. All patients received health care through Medicare, and were part of the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) cancer registry.

"We chose these cancers because patients typically have less than a 20 percent chance of surviving more than five years after diagnosis," Sima said.

The researchers looked at the percentage of these patients who underwent at least one mammogram, Papanicolaou (Pap) test for cervical cancer, prostate-specific antigen (PSA) test, lower gastrointestinal endoscopy test or cholesterol test. Patients were compared with cancer-free people of the same age, race and sex.

Sima found about half of the patients underwent cancer screening of some type within a year of being diagnosed with advanced cancer. Among women with advanced cancer, about 9 percent had a mammogram, compared with 22 percent of women in the cancer-free group. Fifteen percent of men with advanced cancer took a PSA test, versus 27 percent of the control group.

In all cases, patients who were accustomed to screening before their diagnosis were more likely to continue screening, the researchers said.

In the case of the gastrointestinal endoscopy test, 6 percent of men and women screened before their advanced-cancer diagnosis got the test done afterwards, compared with less than 5 percent of those in the control group.

Sima and her team also found a correlation between income levels and marital status and screening rate. Married women with an advanced cancer were twice as likely to continue Pap tests as unmarried women. The rates "are likely to be even higher among younger patients with advanced cancer and who are commercially insured," they wrote in their report.

Routine screening not necessary for everyone

For many patients diagnosed with an advanced-stage cancer, cancer screenings are not likely to affect health, the researchers said. For example, according to a 2001 JAMA study, researchers at the University of California, San Francisco, predicted women over 70 whose life expectancy is limited by a current disease have less than a 2 percent chance of dying from breast cancer.

"Cancer screening is compelling for most members of the population," Sima said, "but it shouldn't be a routine procedure in patients with an advanced cancer, or any disease that severely limits life expectancy."

To some extent, old age is a factor in cancer-screening guidelines, but there is a lack of consistency among health care organizations.

According to the Agency for Healthcare Research and Quality, patients without a family history of cancer who have received normal results in previous tests can discontinue screening for breast, colorectal or prostate cancers after age 75. The American Cancer Society does not list an upper limit for getting mammograms, but does recommend women over 70 discontinue Pap testing if they've had no abnormal tests for 10 years.

In their report, Sima and colleagues wrote that unnecessary screening is costly and exposes patients to the risks of testing procedures, such as side effects from treatments and anxiety that is perhaps unnecessary.

It is unclear, however, how best to approach the issue of over-testing, the researchers said. Some possibilities include changes in Medicare restrictions, improvements to electronic record-keeping systems that can track the health status of patients, and better communication between physicians and patients.

Still, any attempt to revise screening guidelines to account for individuals who won't benefit, such as those with a terminal illness, is likely to be just as challenging as determining who will benefit. Moreover, individual variations between patients cannot be forgotten.

"A very small proportion of patients with advanced cancers will prove to be exceptions and will live longer than their prognosis," Sima told MyHealthNewsDaily. "In the end, any decisions about screening should be based on a realistic conversation between doctor and patient."

This article was provided by MyHealthNewsDaily, a sister site to LiveScience.