Electronic prescribing done using the Food and Drug Administration-administered National Drug Code identifiers sometimes misrepresent the prescriber's intent through over specification, says the Rand Corp. in a study posted online July 20.

The report, commissioned by the Centers for Medicare and Medicaid Services, says a newer drug nomenclature developed by the National Library of Medicine called RxNorm would do a better job than NDC--and that, based on laboratory and live pilot testing, RxNorm appears ready for use in prescriptions from primary care physicians.

Current e-prescribing systems tend to force selection of a medicine matched to an individual NDC rather than allowing a description of intent,  for example by requiring the e-prescription to specify medicine in capsule form when a tablet would be equally appropriate, the Rand report says.

"Forcing the selection of NDCs make prescriptions over specified, and this in some cases limits the pharmacist's ability to select the optimal manufacturer and package size for the patient" the report adds. Over specification causes confusion and consumes time when pharmacies have to contact physician offices to get clarification, it says.

The report also criticizes the NDC directory, stating that commercial prescription drugs sometimes lack a code and that existing codes in the directory can be wrong. Since each drug packager generates its own NDC and is allowed to reuse them, NDCs also have the potential to change meaning over time, the report adds.

RxNorm differs from NDC in that it's a nomenclature of clinically distinct drugs "as opposed to differences that matter in production and distribution." A laboratory evaluation of RxNorm suggests that it includes nearly all prescribable drug concepts necessary for ambulatory primary care settings, the report says. Diabetes supplies should be the top priority for expanding the scope of RxNorm, since they constitute about 90 percent of out-of-scope ambulatory prescriptions, it adds.

During pilot testing, pharmacy technicians said RxNorm generated a high rate of false alerts when e-prescribing system used it to check for accuracy, but report authors say the majority of the alerts were due to the technician interchanging a brand name or dose form different from the prescription at too early a stage. "These are arguably not false alerts but rather alerts indicating that the user has essentially taken a shortcut," the report says.

A usability challenge that would need fixing is the existence of prescribable concepts that have more than one clinically different drug, the report says. For example, if a prescription is for an extended release table at a certain dosage, the effect is different with 12-hour and 24-hour forms of the medication.

For more:
- download the Rand report, "Evaluation of RxNorm in Ambulatory Electronic Prescribing" (.pdf)

Related Articles:
Associations endorse delay in stage 2 meaningful use attestation 
NASCIO preps state CIOs for health benefit exchange IT challenges 
HHS proposes on-demand EHR 3rd party transaction disclosure rule