Test gets almost 1 in 5 syphilis cases wrong

[February 11, 2011]  ATLANTA, Ga. (AP) -- Hundreds of people may have been told they tested positive for syphilis when they didn't actually have the disease, health officials say.

A study of five U.S. labs shows about 18 percent of the positive results from a test method used since the 1980s were actually negative, the Centers for Disease Control and Prevention said Thursday.

Experts previously thought the statistic was much lower -- under 7 percent.

The CDC recommends additional testing when this particular test gives a positive result. But even if most doctors retest, the new research suggests that some people have unnecessarily worried they were infected with an infamous sexually transmitted disease.

It also may mean some patients have been given unnecessary treatment -- probably penicillin, which can have side effects.

Syphilis testing is recommended for all pregnant women and for people considered to be at high risk for sexually transmitted diseases.

But CDC officials said it's not clear how widely used the problematic testing method is versus another, more traditional one.

The problematic test, which is cheaper than the alternative, does have one advantage: "It doesn't miss people who are infected," said Dr. Karen Hoover, a CDC epidemiologist who co-authored the study.

But it isn't perfect at distinguishing syphilis antibodies from other proteins in the blood, meaning it sometimes gives a positive result for something other than syphilis, CDC scientists said.

Only about 14,000 Americans have the most contagious forms of syphilis, but the number has been rising in recent years. The 14,000 cases are those with symptoms that are carefully confirmed, so the new study's finding does not affect that count.

Hoover's research is published this week in the CDC's Morbidity and Mortality Weekly Report.

___

Online:

The CDC report: http://www.cdc.gov/mmwr/

[Associated Press; By MIKE STOBBE]

Copyright 2011 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.