ABSTRACT
PURPOSE: The effects of an adverse-drug-event (ADE) alert system on cost and quality outcomes in community hospitals were evaluated. METHODS: This retrospective observational study evaluated the effects of an ADE alert system in seven hospitals in the Trinity Health network. Outcomes for all inpatients admitted to these hospitals after and one year before the deployment of an ADE alert system were evaluated. Inpatients in two network hospitals that lacked any computerized ADE alert system constituted the external control group. Administrative data were gathered for patients from these facilities for the same time frames as for the preimplementation and postimplementation groups. Primary outcomes evaluated included pharmacy department costs, variable drug costs, and mortality rates. Secondary outcomes included total hospitalization costs, length of hospital stay (LOS), rate of readmission, and case-mix index. Mean differences in primary and secondary outcome measures across all four groups were examined using analysis of variance. RESULTS: Significant decreases in mean pharmacy department costs per patient were observed from preimplementation to postimplementation (p < 0.001), while pharmacy department costs increased significantly in the external control group (p = 0.029). Drug costs decreased significantly from baseline (p < 0.001) in the postimplementation group. Drug costs increased significantly in the external control group (p = 0.029). Severity-adjusted mortality rates and LOS decreased significantly in the postimplementation group. Total patient hospitalization costs, both crude and severity adjusted, significantly increased in both groups. CONCLUSION: Implementation of an ADE alert system in seven community hospitals demonstrated significant decreases in pharmacy department costs, variable drug costs, and severity-adjusted mortality rates.
Subject(s)
Adverse Drug Reaction Reporting Systems/organization & administration , Hospitals, Community/organization & administration , Pharmacy Service, Hospital/organization & administration , Adverse Drug Reaction Reporting Systems/economics , Bias , Costs and Cost Analysis , Databases, Factual , Diagnosis-Related Groups , Hospital Mortality , Hospitalization/economics , Hospitals, Community/economics , Humans , Patient Readmission/economics , Pharmacy Service, Hospital/economics , Quality Assurance, Health Care , Retrospective Studies , Treatment Outcome , United StatesABSTRACT
OBJECTIVE: The purpose of this study was to compare the impact of trauma patient outcomes before and after Level II American College of Surgeons (ACS) verification was received in a not-for-profit community hospital. METHODS: This was a retrospective analysis of hospital discharge data for timeframes before and after Level II ACS verification was conducted. Originally, 8,674 patients were identified using the International Classification of Diseases, 9th Revision codes for trauma. These data were parsed to 7,811 patients by using International Classification of Diseases, 9th Revision codes 800 xx through 959.9 x, which signify an admitting diagnosis of trauma; 3,835 of the patients were admitted after the July 28, 1998, verification date. Blunt injuries constituted the vast majority of the patients (n = 7,488). Outcome measures studied included changes in length of stay (LOS), mortality, and total cost. Internal control was coronary artery bypass graft patients at the same hospital, and external control was trauma patients at a non-ACS hospital over the same time period. Data are presented with p values and SE and the ratio of observed/expected values on the basis of the all-payer severity-adjusted diagnosis-related group severity model. RESULTS: The two timeframes exhibited statistically different outcomes in several variables. Adjusting for severity postverification, LOS was 10% less (p < 0.000). Similarly, severity-adjusted mortality observed/expected ratios were significantly different: 0.81 before versus 0.59 after (p < 0.000). The severity-adjusted ratio of costs found that the postverification era was 5% lower (p < 0.000). The contribution margin of the trauma patient population to the hospital well exceeded any postverification costs. Both control groups exhibited no significant changes in their severity-adjusted outcomes, which could have invalidated these results. CONCLUSION: This study suggests that the efforts and resources consumed achieving ACS Level II trauma center verification appear to result in desired outcomes as evidenced by decreased LOS, reduced in-hospital mortality rates, reduced cost, and improved contribution margins.
