Daily Health Policy Report

Prescription Drugs | Researchers Use Data Mining To Find Possible Rare Side Effects of Prescription Drugs
[Jul 03, 2007]
The Wall Street Journal on Wednesday examined prescription drug data mining, a process by which sophisticated software enables health officials to search through large databases looking for possible drug dangers. Data mining software allows health authorities to identify "rare side effects that didn't show up in clinical trials," the Journal reports. However, "it can also raise false alarms and force regulators to divert time and money from more pressing dangers," according to the Journal.

The Journal profiled the experience of World Health Organization Drug Monitoring Center Director Ralph Edwards. Edwards and his team in the mid-1990s developed software to mine drug data, and national drug agencies, including FDA, in 2002 allowed the center to publish and share data mining findings without permission. Edwards, who receives about 200,000 adverse-event reports and identifies about 60 serious signals annually, last year discovered a possible link between cholesterol-lowering statins and amyotrophic lateral sclerosis, or Lou Gehrig's disease.

However, the analysis based on data mining did not "prove anything," and Edwards was "wary of creating a drug scare and mindful that statins have been shown to reduce heart attacks significantly," the Journal reports. After "months of hesitation," Edwards published his findings in the journal Drug Safety and recommended that patients taking statins consult their physician if they experience any neuromuscular symptoms.

FDA also studied the connection between statins and ALS but determined that "it didn't need to issue any caution" about the drug, the Journal reports. Robert Temple, medical director at the FDA division that evaluates drugs, said, "People reach different judgments on when to shout and when not to shout. It's the hardest single thing -- the value and danger to screaming early."

Edwards said his paper is intended to prompt more research of the possible connection, adding that FDA's clinical trial data might not show the risk for ALS because it is so rare (Johnson, Wall Street Journal, 7/3).



http://www.kaisernetwork.org/daily_r...fm?DR_ID=46001