Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

SC58

n SC58 West Bldg 101C Healthcare Operations: Improving Access to Care and Patient-Welfare Sponsored: Health Applications Sponsored Session Chair: Sila Cetinkaya, Southern Methodist University, Dallas, TX, 75205, United States Co-Chair: Olga Bountali, Southern Methodist University, Dallas, TX, 75219, United States 1 - Analysis of Primary Care Delivery with E-Visits: Improving Access to Care and Profitability Xiang Zhong, University of Florida, Aditya Mahadev Prakash We consider a medical institution who offers both office and e-visit to their panel patients. A key planning problem is to determine the medical resource capacity that improves access to care for patients and ensures profitability of the medical institution. A novel analytical framework for modeling a care delivery system with two horizontally substitutable channels and a heterogeneous patient population is proposed. We identify that when the service system is of high traffic intensity, based on patient segmentation, medical institutions can equip their e- visit services with an optimal amount of resources to ensure an economical outcome while providing care to more patients. 2 - Modelling Enrollment in Clinical Trials Erinc Albey, Ozyegin University, Istanbul, 34794, Turkey Clinical trial planning and execution depends on enrollment, yet few tools exist that allow prediction of subject accrual to inform study design and conduct. We sought to develop a model to estimate the probability that a site will enroll a specified number of subjects in a trial within a given time interval, allowing for better clinical trial planning and execution. We hypothesized that the enrollment process is a finite, first-order, discrete time, homogeneous Markov chain. We based our analyses on data from two large multicenter, completed clinical trials. We modeled the enrollment performance of sites as a discrete set of states (zero/low/medium/high enrollments) over a given time interval and assessed whether the model accurately reflected trial enrollment. We also performed a set of Monte-Carlo simulations based on this Markov model to estimate the time required for the trial to enroll a predetermined number of patients. The enrollment prediction performance of the proposed Markov model is superior to linear and polynomial regression models, especially in the early phases of the trial. These estimates may be used in predicting the overall duration and in managing future similar trials in real-time. 3 - The Impact of Information Sharing on Organ Transplantation: A Simulation Model Michael Hahsler, Southern Methodist University, Bobby B. Lyle School of Engineering, P.O. Box 750123, Dallas, TX, 75275, United States, Zahra Gharibi Recently, several changes to the kidney allocation rules were approved, and modern IT solutions are developed to improve kidney assignment. We present a flexible simulation framework that considers the effect of several important factors for organ transplantation. We use the model to investigate how the optimal individual-level center listing and kidney acceptance decisions are affected by differences in supply due to different allocation rules. At the macro- level, we evaluate the potential effect of information sharing on the social welfare and organ discard rates. 4 - Improving the Uninsured Patients’ Experience: Practice vs. Theory Olga Bountali, Southern Methodist University, 4210 Fairmount Street, Apartment 3063, Dallas, TX, 75219, United States, Sila Cetinkaya, Farnaz Nourbakhsh Under EMTALA, the current practice of care delivery for the uninsured prescribes that patients receive treatment only after being evaluated as in æemergent, life- threatening condition’. This practice often leads to excessive overhead cost, severe treatment delays, and congestion at publicly funded hospitals with a hefty impact on the cost and quality of healthcare for the uninsured. Existing literature in service operations offers abundant ideas for improving both cost and quality of care including but not limited to congestion mitigation strategies via alternative service protocols. We explore and compare the value of patient-centric service protocols relative to the current practice.

n SC59 West Bldg 102A

Joint Session HAS/Practice Curated: Stochastic models of Hospital Admission and Discharge Services Sponsored: Health Applications Sponsored Session Chair: Nan Kong, Purdue University, West Lafayette, IN, 47906-2032, United States Co-Chair: Michelle M. Alvarado, University of Florida, Gainesville, FL, 32611-6595, United States 1 - Stochastic Models for Inpatient Discharge Planning Maryam Khatami, Texas A&M University, College Station, TX, USA; Michelle M. Alvarado, Nan Kong, Pratik J. Parikh, Mark Lawley. The inpatient discharge planning problem requires the efficient assignment and sequencing of ready-for-discharge patients to resources. Delay in discharge processes deteriorates patient satisfaction and increases hospital costs. We model and solve the inpatient discharge planning problem as a two-stage stochastic program with uncertain inpatient discharge processing time and bed request times. The objective is to minimize patient dissatisfaction, discharge lateness, and patient boarding. We derive managerial insights by comparing the results of a two-stage stochastic program, the mean value problem, and two heuristics from current practice using simulation modeling. 2 - Missed Opportunities in Preventing Hospital Readmissions: Redesigning Post-discharge Checkup Policie Xiang Liu, University of Michigan, Ann Arbor, MI, United States, Mariel Lavieri, Jonathan Helm, Ted Skolarus Hospital readmissions affect hundreds of thousands of patients, placing a tremendous burden on the healthcare system. Post-discharge checkup can reduce readmissions through early detection of conditions. Our work develops optimal checkup plans to monitor patients following hospital discharge using methods including phone calls and office visits. By analyzing the structure of optimal policies, we develop checkup schedules that mitigate 32% more readmissions. 3 - Improving Discharge Process at a University Hospital: A System-theoretic Method Xiaolei Xie, Tsinghua University, 614 Shunde Building, Beijing, 100084, China, Nan Chen, Zexian Zeng, Xiang Zhong, Maria Brenny-Fitzpatrick, Barbara A. Liegel, Li Zheng, Jingshan Li This paper introduces a system-theoretic approach to improve inpatient discharge process at a university hospital. The complex hospital discharge process is modeled by a stochastic process with parallel sub-processes, splits, and merges. A system analysis method is introduced to approximate the discharge time and evaluate the mean, variability, and discharge-time performance. It is shown such a method results in a high accuracy in performance evaluation. To improve the discharge process at the university hospital, bottleneck and what-if analyses are carried out and improvement recommendations are discussed. 4 - Admission Planning Problem with Stochastic Length of Stay Jorge Vera, Universidad Catolica de Chile, Dept. Industrial and System Engineering, Santiago, 7820436, Chile, Ana Celeste Batista, David Pozo Effective admission planning process can improve inpatient throughput and waiting times. The uncertain in the patient s length of stay complex the admission and may cause bottlenecks and long waiting times in the patient’s flow. We study the admission planning problem considering uncertain in the length of stay. Classically length-of-stay is modeled as a sum over a time windows constraint. This makes it very complex to consider uncertainty on this variable. In this work, we developed a new formulation in which the length of stay is on the right hand of the constraint by employing a single binary variable.

82