Wednesday, October 19, 2011

Screening tests that do more harm than good

The U.S. Preventive Services Task Force may have been slow to finalize new recommendations recently, but the same can't be said for its main Evidence-Based Practice Center at Oregon Health and Science University, which has been producing systematic reviews at a furious pace. On the heels of a news-making evidence update that I co-authored on screening for prostate cancer, this Monday another USPSTF review team published an update of screening for cervical cancer with liquid-based cytology and human papillomavirus (HPV) testing.

This review led the USPSTF to the preliminary conclusion that HPV-enhanced screening, though widely utilized in the U.S., does more harm than good in women younger than age 30. Furthermore, the Task Force concluded that there is insufficient evidence to support HPV screening in women age 30 or older. HPV, it seems, is yet another example of a test that, despite having genuine value in diagnostic situations (for example, evaluation of abnormal cervical cytology or histology), has flunked as a screening test.

Being carried away by premature enthusiasm to provide a proven test in unproven situations is nothing new; doctors have been doing it for years. As I wrote in an editorial in American Family Physician in 2007:

As family physicians, we often face difficult decisions about ordering tests for the early diagnosis or prevention of disease in healthy-appearing persons. It is hard to convince many patients to think about prevention, and those who come in for health maintenance visits often expect to undergo tests that they have heard about from advertising on the Internet, radio, or television, or in popular magazines. For example, a colleague recently saw a healthy, asymptomatic woman who scheduled an appointment to receive the results of an ultrasound examination that had been ordered by another physician to screen for abdominal aortic aneurysm (AAA). The results were normal, but because the test was not indicated by generally accepted standards, our colleague was perplexed at what reassurance to provide the patient, if any.

Although the U.S. Preventive Services Task Force (USPSTF) recommends against performing AAA screening in asymptomatic women of any age, the existence of these and other evidence-based guidelines have not prevented direct-to-consumer marketing of costly screening tests of uncertain value. There is a striking contrast between widespread public enthusiasm for technology (e.g., whole-body computed tomography [CT], coronary calcium scans) and the paucity of evidence that performing these tests improves outcomes for patients.

“Big-ticket” tests are easy targets for those seeking to reduce waste in health care. But what about the seemingly innocuous practice of performing routine tests such as a complete blood count (CBC) or urinalysis? Both are far less expensive than CT scans and can often be performed in the office at the time of the visit. More than one third of family physicians in the United States think that CBC and urinalysis should be offered routinely at health maintenance examinations, and these tests are ordered for 25 to 37 percent of patients who present for such visits.

Read the rest of "Are Some Screening Tests Doing More Harm Than Good?" here.