The Impact of Enhancing Quality of Provider Practices for Older Adults in the Emergency Department (EQUiPPED) on Potentially Inappropriate Medications (PIMs) Open Access

Song, Jiayang (Spring 2019)

Permanent URL:


Background: The elderly play an important role in health care. One of the biggest issues of the health care for the elderly is medical prescriptions. Some medications perform well in the general population while they would cause side effects that outweigh the benefits in the elderly, which are called potentially inappropriate medications (PIMs). PIMs were prescribed very often in the emergency department (ED), where the pace is fast and chaotic. The program “Enhancing Quality of Provider Practices for Older Adults in the Emergency Department (EQUiPPED)” was created with the hope to reduce the use of PIMs. The previous study found it effective in VA settings. We implemented EQUiPPED and analyzed its performance on reducing PIMs count/rate in the first non-VA setting in this paper.

Methods: The intervention include provider education, EHR-based clinical decision tools and provider feedback. The intervention was divided into three periods: Pre-Intervention, Post- Intervention and Intervention (five months). Both weekly- and monthly-based PIMs rates were calculated, and the changing trends were analyzed. Bonferroni adjusted proportion test and exact ratio test were conducted to find the unadjusted effect of EQUiPPED on PIMs rate. Several regression models for count outcomes were fitted and compared. A regression-based test was used to determine the adjusted effect on the final model.

Results: The EQUiPPED program successfully was implemented. The Bonferroni adjusted proportion test indicated significant differences between both Pre-Intervention vs Post-Intervention and Intervention vs Post-Intervention (P-value both < 0.001). The exact rate ratio test gave us similar outcome: with an estimated ratio of 2.51(95% confidence interval 2.30 – 2.73, P-value < 0.001). Negative binomial model was selected given the outcome of Vuong test, dispersion parameter, AIC and BIC. Two models with covariates selected using different criteria were finally fitted. The adjusted effects of EQUiPPED period showed no signs of effectiveness (P-value = 0.40 and 0.44 respectively for these two models) 

Table of Contents

1.     Introduction: 1

2.     Methods: 4

2.1.    Setting and Interventions: 4

2.2.    Provider education: 4

2.2.1.     EHR-based clinical decision support tools: 4

2.2.2.     Provider feedback: 5

2.3.    Data collection and measures: 5

2.4.    Data analysis: 6

2.4.1.     Model selection on different modeling methods: 6

3.     Results: 12

3.1.    Demographic characteristics: 12

3.2.    The implementation of EQUiPPED: 12

3.3.    Unadjusted effectiveness of EQUiPPED: 13

3.4.    Modelling method selection: 13

3.5.    Model selection of negative binomial model and adjusted effectiveness: 15

4.     Discussion: 16

5.     References: 20

6.     Figures and Tables:

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Last modified

Primary PDF

Supplemental Files