Generic drugs often differ from brand name drugs in their safety labeling despite standards mandated by the US Food and Drug Administration (FDA), according to a new report.

"Physicians should be aware of such differences and regulators should consider new strategies for harmonizing bioequivalent labels," Jon Duke, MD, from the Regenstrief Institute and Indiana University School of Medicine, both in Indianapolis, and colleagues write.

The researchers report their findings in an article published online in Pharmacoepidemiology and Drug Safety.

According to Dr. Duke and colleagues, the FDA requires that manufacturers of generics demonstrate that "their medications are 'bioequivalent' to reference-branded drugs — meaning that they have identical active ingredients, dose forms, and approximate pharmacokinetics." In addition, safety data on the labels of generics should be identical to those found on the brand label.

To investigate whether the safety data are equivalent, the researchers analyzed safety labels across bioequivalent generic medications using an automated system called the Structured Product Label Information Coder and Extractor (SPLICER). The team showed previously that the system has a 92.8% sensitivity and 95.1% positive predictive value.

Dr. Duke and colleagues indexed the adverse drug reactions (ADRs) found in the adverse reactions and postmarketing sections of 9105 structured product labels and calculated the difference between brand name and generic adverse events.

In addition, the researchers conducted a manual review to identify causes of inconsistency for the 25 drugs with the greatest discrepancy in labeled ADRs.

"Of the 1540 drugs, 500 were represented by only a single label and thus, by definition, had no discrepancies. Of the 1040 drugs with more than one manufacturer's label, 68% (n=711) showed some discrepancies within their labeling," the authors write.

In addition, the degree of such discrepancy was highly variable: For a given drug, the mean deviation in number of labeled ADRs was 4.4, and the median was 0.8 (interquartile range, 0 - 3.2). The mean range in number of labeled ADRs was 12 ± 0.9, and the median was 2 (interquartile range, 0 - 9).

Most (77.9%) generic manufacturers produced labels differing from those of the brand. The most common causes of inconsistency included missing tables, obsolete postmarketing information, and problems with formatting.

"The presence of such labeling differences may complicate physician practice, raising the possibility that important safety information may be missed even when a current label is fully reviewed," Dr. Duke and colleagues conclude.

"While the clinical significance of such labeling discrepancies remains unclear, we suggest for now that physicians review branded drug labeling even when a patient is taking a generic version of a medication," they suggest.

Funding was provided by the Regenstrief Institute. Dr. Duke is also supported by a Young Investigator's award from the Indiana Clinical Translational Sciences Institute. The other authors have disclosed no relevant financial relationships.

Pharmacoepidemiol Drug Saf. Published online October 8, 2012. Abstract