Return to 2001 Abstacts List

*179. Hospital Admissions for Medication Adverse Events in Florida: Implications for Chronic Disease Outcomes.

RR Campbell, Patient Safety Center of Inquiry, JAH VAMC; DC Werner, Patient Safety Center of Inquiry, JAH VAMC; NA Coblio, Phamacy QM, JAH VAMC; A Spehar, Patient Safety Center of Inquiry, JAH VAMC; J Wolfson, College of Public Health, USF/PSCI JAH VAMC

Objectives: A study of the Florida state inpatient dataset (SID)was conducted to determine the distribution of medication adverse events by specific drug classes coded using the ICD-9-CM classification system in non-VHA Florida hospitals. The objectives of the study were to determine the types and distribution of medication adverse events coded as the primary diagnosis in Florida hospitals to allow future comparisons with VHA facilities in VISN 8. Particular emphasis was placed on selecting chronic disease conditions, i.e. diabetes, cardiac disease, mental illness, etc. to determine the incidence of hospitalizations due to medication adverse events for these patients.

Methods: A retrospective descriptive analysis of Florida’s state inpatient dataset (SID) was conducted for calendar years 1996 through 1998. Using the CDC’s "Matrix for External Cause of Injury Mortality and Morbidity Data’s" coding scheme for poisoning codes related to medications, an analysis was conducted of the frequency and distribution of medication adverse event admissions to Florida hospitals between 1996 and 1998. Primary and secondary diagnoses coded for "poisoning" (ICD-9-CM codes 960-979) and the external cause of injury codings (E codes 850-8589) were examined. These diagnoses correspond to the AHRQ’s Clinical Classification System codes for diagnosis categories 241 and 242. The distribution of medication errors by classes of drugs were analyzed by sex, age, geographic region, costs, length of stay, and major disease conditions.

Results: Nearly six million admissions occurred in Florida hospitals during the time period under study. Over 30,000 admissions were coded with a primary diagnosis indicating a medication adverse event. The medication types most frequently associated with these admissions were psychotropic drugs, cardiac drugs, and analgesics. Males, 65 and over, had medication adverse events primarily associated with psychotropic and cardiac drug classes.

Conclusions: By studying the civilian hospital population’s admissions for medication adverse events, insight has been gained into the distribution and types of drugs associated with these hospitalizations in Florida. Further study of the subset of older male patients’ admissions provided valuable information and insights into what kinds of medication adverse events may be expected to result in hospitalizations in the VHA enrolled population in VISN 8.

Impact: According to national datasets, over 350,000 hospital admissions occur annually because of medication adverse events. Over 50,000 medication adverse events resulting in hospitalizations nationally occur annually in males 65 and over. Over 1,000 elderly males’ hospitalizations due to medication adverse events were identified and studied as a subset of the 30,000 hospitalizations during the study timeframe in Florida civilian hospitals. Psychotropic and cardiac drugs were found to be the major classes of drugs associated with these admissions. Similar studies of VHA enrolled patients in VISN 8 may provide additional insight into programs to target these potentially preventable medication adverse events in Veterans with chronic disease conditions.