Assessment of Adverse Drug Events Among Patients in a Tertiary Care Medical Center

Philip E. Johnston; Daniel J. France; Daniel W. Byrne; Harvey J. Murff; Byron Lee; Renee A. Stiles; Theodore Speroff


Am J Health Syst Pharm. 2006;63(22):2218-2227. 

In This Article


Health care organizations, regulatory agencies, consumers, and advisory groups are intent on preventing AEs. Essential to any organization's safety efforts is a reporting system from which to glean information about AEs. Though often used for retrospective study of safety indicators, data on AEs may also have utility in the design of mechanisms that prevent the repetition of errors in the future. Once identified and validated, such a model could be used to guide the development of policies and procedures that guard against AEs in high-risk circumstances. Successful implementation of preventive measures may then yield improvements in patient safety and concomitant reductions in the costs, morbidity, and mortality associated with AEs.

Published case reports of ADEs have identified several potential factors associated with a high rate of ADEs, including female sex, age 40-69 years,[10,11,29,31] renal or liver dysfunction, and use of orally administered medications.[5,10,11,12,13,14,15]

In many studies, the age group identified to be at highest risk of an AE are the 15-60-year-old patients, reflecting the number of patients in the study, the increased number of patients in this age group who experience trauma or surgery, and the increased rate of disease onset during these years. Our data showed a very different pattern, with the very young and the older patient population demonstrating an increased risk of AEs.

Relatively healthy and moderately ill patients tended to exhibit a higher risk of sustaining a significant AE than did sicker patients. The drugs most commonly implicated in AEs in other studies were central nervous system agents, antimicrobials, antineoplastics, and cardiovascular agents, as well as those requiring therapeutic monitoring.[12,13,14,16]

Several studies have identified disparities in the rates of AEs among different units of hospitals. Specifically, AEs have been reported to occur more commonly in intensive care units than in general care units.[1,8] We did not evaluate this in our study.

As is true of other studies of AEs, this study had several limitations, not the least of which were underreporting of AEs and incomplete data regarding reported events. We found the rate of AEs to be less than 2% overall, lower than rates that have been reported in other institutions. However, this low percentage is representative of documented AEs at our institution. The true rate of AEs may be higher.

Our study relied on voluntarily reported AE data, which have numerous limitations, including underreporting, a bias toward nurse reporting, overlapping of diagnostic coding, differences in definitions for universal terms (e.g., ADRs, medication errors), and an imbalance in the reporting of errors of commission more than errors of omission. In addition, specialized quality improvement initiatives, such as a pediatric chemotherapy safety project, may have resulted in increased reporting on certain floors or units. Finally, unit-to-unit variations in reporting culture may have positively or negatively influenced reporting behavior during our study.

We did take several steps to mitigate the effects of this study's limitations. First, the high-risk drugs selected for inclusion reflect both local experience and reports in the literature; however, the preponderance of case reports in the literature suggests that drugs not included in this study may have proven to be good, if not better, indicators of AEs. Second, although the study was performed retrospectively on data already known to be subject to reporting bias, this limitation may have been offset by the large sample size, which lent more power to the analysis than in previous work. Further, retrieving information from institutional administrative systems to supplement data collected via ADE reporting forms yielded a more complete picture of patients with and without AEs than would have been possible by relying solely on data from the ADE records themselves. Although each AE may have arisen from a unique combination of circumstances, we found a measure of predictability in patterns of AEs. Future efforts should include drugs specifically involved in AEs, as well as separate analyses of medication errors and ADRs.

Using the key indicators and factors associated with AEs, we will conduct a detailed analysis of AEs occurring in congestive heart failure patients at our institution. Over three years, an education effort will be launched for health care practitioners, advising them of the factors associated with AEs in this group, and soliciting their assistance to verify these indicators and develop actions to avoid them. Multiple measures will then be implemented to prevent AEs associated with these indicators. Finally, these indicators will be analyzed retrospectively to determine what effect our efforts have had.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.