Healthcare Quality Improvement Analytics: An Example Using Computerized Provider Order Entry.

Evaluation of sustainability after quality improvement (QI) projects in healthcare settings is an essential part of monitoring and future QI planning. With limitations in adopting quasi-experimental study design in real-world practice, healthcare professionals find it challenging to present the sustained effect of QI changes effectively. To provide quantitative methodological approaches for demonstrating the sustainability of QI projects for healthcare professionals, we conducted data analyses based on a QI project to improve the computerized provider order entry system to reduce patients' dosing frequencies in Korea. Data were collected for 5 years: 24-month pre-intervention, 12-month intervention, and 24-month post-intervention. Then, analytic approaches including control chart, Analysis of Variance (ANOVA), and segmented regression were performed. The control chart intuitively displayed how the outcomes changed over the entire period, and ANOVA was used to test whether the outcomes differed between groups. Last, segmented regression analysis was conducted to evaluate longitudinal effects of interventions over time. We found that the impact of QI projects in healthcare settings should be initiated following the Plan-Do-Study-Act cycle and evaluated long-term effects while widening the scope of QI evaluation with sustainability. This study can serve as a guide for healthcare professionals to use a number of statistical methodologies in their QI evaluations.

The Effect of Regimen Frequency Simplification on Provider Order Generation: A Quasi-Experimental Study in a Korean Hospital.

The multiplicity of dosing frequencies that are attached to medication orders poses a challenge to patients regarding adhering to their medication regimens and healthcare professionals in maximizing the efficiencies of health care service delivery. A multidisciplinary team project was performed to simplify medication regimens to improve the computerized physician order entry (CPOE) system to reduce the dosing frequencies for patients who were discharged from the hospital. A 36-month pre-test-post-test study was performed, including 12-month pre-intervention, 12-month intervention, and 12-month post-intervention periods. Two-pronged strategies, including regimen standardization and prioritization, were devised to evaluate the dosing frequencies and prescribing efficiency. The results showed that the standardized menu reduced the dosing frequencies from 4.3 ± 2.2 per day in the pre-intervention period to 3.5 ± 1.8 per day in the post-intervention period (p < 0.001). In addition, the proportion of patients taking medications five or more times per day decreased from 40.8% to 20.7% (p < 0.001). After prioritizing the CPOE dosing regimen, the number of pull-down options that were available reflected an improvement in the prescribing efficiency. Our findings indicate that concerted efforts in improving even a simple change on the CPOE screen via standardization and prioritization simplified the dosing frequencies for patients and improved the physicians' prescribing process.

Design and Analysis of Cost-Efficient Sensor Deployment for Tracking Small UAS with Agent-Based Modeling.

Recently, commercial unmanned aerial systems (UAS) have gained popularity. However, these UAS are potential threats to people in terms of safety in public places, such as public parks or stadiums. To reduce such threats, we consider a design, modeling, and evaluation of a cost-efficient sensor system that detects and tracks small UAS. In this research, we focus on discovering the best sensor deployments by simulating different types and numbers of sensors in a designated area, which provide reasonable detection rates at low costs. Also, the system should cover the crowded areas more thoroughly than vacant areas to reduce direct threats to people underneath. This research study utilized the Agent-Based Modeling (ABM) technique to model a system consisting of independent and heterogeneous agents that interact with each other. Our previous work presented the ability to apply ABM to analyze the sensor configurations with two types of radars in terms of cost-efficiency. The results from the ABM simulation provide a list of candidate configurations and deployments that can be referred to for applications in the real world environment.