in synthesizing this heterogeneous evidence base. Hospital care is the setting with the most extensive research. Studies have estimated the rate of ADEs incurred in hospitals and error rates at each stage of the medication-use process. There is also an extensive literature on errors of omission in prescribing—failure to prescribe medications in appropriate situations.

Other care settings are much less well researched. For nursing home care, there are estimates of the rates of ADEs incurred while in a nursing home, plus a few studies on error rates at various stages of the medication-use process.4 Little attention has been paid to errors of omission in the nursing home population. For ambulatory care, a modest amount of research has been carried out, spread thinly over a large number of topics— ADE and error rates at various stages of the medication-use process, and omissions of effective therapies in specific populations. Similarly for pediatric care, a modest amount of research has been carried out, again thinly spread over a wide range of topics.

For the remaining care settings considered in this report, little or no research has been conducted on ADE and error rates. Of the limited number of studies relating to self-care, most addressed adherence issues. No study was found on medication error rates in the school setting. Just two studies were found on medication error rates in psychiatric care. Finally, there has been hardly any research on medication errors relating to OTC medications, and no study was found on error rates associated with complementary and alternative medications.

The discussion in this section is based on a large number of studies reviewed by the committee. It first addresses the incidence of medication errors in general, and then the incidence of three specific categories of medication errors—preventable ADEs, underutilization of medications, and overutilization of medications.

Incidence of Medication Errors

As noted, hospital care is the most researched setting for medication error incidence rates, although no study was identified that addressed medi-


There have been many studies of inappropriate prescribing for the elderly in nursing homes, ambulatory care, and home health care, based on such criteria as the Beers criteria (Beers et al., 1991) and subsequent updates/extensions (Beers, 1997). The committee did not include these studies in its synthesis since the causal link between inappropriate prescribing and poor health outcomes has not been documented.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement