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

Abstract and Introduction


Purpose: Specific patient and clinical characteristics associated with an increased risk of sustaining an adverse event (AE) were identified.
Methods: AE reports for patients in a 658-bed tertiary care medical center between January 1, 2000, and June 30, 2002, were analyzed. The data collected from each report included medical record number, patient sex, patient age, clinical service, date of occurrence, diagnoses, type of error, suspected medication, and severity of the AE. A three-stage logistic regression model with high-risk indicators was used to evaluate key indicators of the most vulnerable patient populations.
Results: The number of control patients and those with AEs totaled 60,206. This population was then randomly split into two equal groups of patients: the training data set (n = 30,103) and the validation data set (n = 30,103). AEs occurred in a higher percentage of patients who were age <1 year, 1-15, 47-59, and ≥60 years than in other groups. A higher percentage of AEs were reported in men than women, but the groups were not significantly different when comparing those with an AE and those without an AE. Asian Indian patients demonstrated a high rate of AEs, but this may be a statistical artifact, reflecting their very small percentage in the study. Evaluation of admission sources revealed that doctors' offices, clinic referrals, and local hospital transfers accounted for higher rates of AEs than other sources.
Conclusion: Certain age groups, diagnoses, admission sources, types of insurance, and the use of specific medications or medication classes were associated with increased AE rates at a tertiary care medical center.


As many as 1.3 million accidental patient injuries are caused each year by medical treatments,1 with mistakes occurring in 19% of all medication doses given in hospitals, usually during drug administration.[2] Patient safety in hospitals has come under intense scrutiny in both public and private sectors. The amount of literature describing the evaluation and prevention of adverse events is large and diverse. While some studies focus on adverse drug events, others encompass all adverse events (AEs), including adverse drug reactions (ADRs) and medication errors.

In a survey conducted by the American Society of Health-System Pharmacists in May 2002, 85% of respondents expressed concerns about at least one medication-related issue, such as receiving interacting medications, having harmful adverse effects from a medication, or receiving the wrong medication.[3] Institutions accredited by the Joint Commission on Accreditation of Healthcare Organizations and nonaccredited institutions exhibited comparable rates of AEs, revealing that no institution is immune to such errors.[2] With an estimated cost of $2000 per adverse drug event (ADE), institutions should not ignore this problem.[4]

Although published literature has isolated some variables associated with an increased risk of ADEs,[1,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34] no study has definitively specified a model predictive of ADEs. Bates et al.[4] identified a number of factors associated with preventable ADEs in hospitalized patients but found no independent indicators after multi-variate analysis. In a descriptive study based on abstracted case reports of ADEs, Kelly[15] identified a variety of predictive variables and determined that 52% of the ADEs reported were preventable. Of those, half could have been averted if better monitoring, computer screening, and assessment of risk by a pharmacist had been applied before drug administration. Attempts have been made to identify predictors of ADEs in a computerized prescriber-order-entry (CPOE) system with limited success.[35]

Controversy exists regarding predictive modeling of ADEs. Early work to develop an ADE predictive model yielded discouraging results.[5] The authors concluded that ADEs could not be predicted by observing the patient characteristics and hospital environment factors used in their study. The authors concluded that each incident could have a unique set of contributing factors. However, several design flaws were acknowledged in their study, including insufficient information.[5] Alternatively, other investigators identified common and repeated patterns of preventable ADEs.[35,36]

In a nested case-control study of the presence, preventability, and severity of ADEs among patients, sicker patients with longer lengths of hospitalization were more prone to ADEs.[11] No other variables were found to be predictive of ADEs. The authors concluded that a systems approach would be a more effective prevention strategy than would patient-risk stratification. Characteristics of preventable ADRs are reported in most hospitals using a voluntary reporting system. Pearson et al.[19] found that most preventable ADRs involved (1) a documented allergy to the medication ordered, (2) anticoagulants or thrombolytics, (3) medication that required monitoring of serum drug concentrations, and (4) renally eliminated drugs for which dosage adjustments were not made in patients with impaired renal function. In a similar study, Seeger et al.[20] found that preventable ADRs were associated with dosing and previous allergic reactions to the drug. In a prospective randomized study, Gholami and Shalviri[21] found that the risk of preventable ADRs increased with patient age and that these ADRs were most commonly hematologic in nature.

Other research has determined that the number of drugs prescribed (i.e., over seven) per hospitalized patient is the most important risk factor for the development of ADEs.[6] Other risk factors included newly prescribed drugs, drugs to treat gastrointestinal or central nervous system disorders, and antibiotics.[6]

The purposes of this study were to (1) identify specific patient and clinical characteristics associated with an increased risk of sustaining an ADR or medication error, which were collectively termed adverse events (AEs) and (2) identify factors or medication-use process steps that require further investigation.


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.