Informs Annual Meeting 2017

TB05

INFORMS Houston – 2017

3 - Fairness in Pollution Regulation Krishnan S. Anand, University of Utah, David Eccles School of Business, Ois Department, University Of Utah, Salt Lake City, UT, 84112, United States, anand@eccles.utah.edu, Francois C. Giraud-Carrier Fairness in pollution regulation is an important and much-debated question, especially given multiple stake-holders with conflicting objectives, including polluting firms, industry lobby groups, consumers, environmentalists, doctors and the government. While there is a vast literature on the notion of fairness in general, we use the term precisely to mean that the polluter pays for the entire pollution damage it causes, and no more— the so-called `polluter pays’ principle. We develop a novel benchmark-mechanism to implement fairness, and assess the fairness of Cap-and-Trade and Taxes against this benchmark. 320B Beyond the Unit - Holistic Approach to Demand- Based Staffing in Hospital Systems Sponsored: Health Applications Sponsored Session Chair: Hari Balasubramanian, University of Massachusetts-Amherst, Amherst, MA, 01003, United States, hbalasubraman@ecs.umass.edu Co-Chair: Ekin Koker, University of Massachusetts, Amherst, MA, 01003, United States, ekoker@umass.edu 1 - Forecasting Demand and Staffing Fluctuations in Hospital Operations Jeffrey Jung, Navigant Consulting, Inc., jeffrey.jung@navigant.com A critical component of labor optimization is the ability to predict short and long term forecasts of demand variations as well as the supply /staffing variations. By projecting fluctuations in demand, hospital leaders are equipped to accurately reflect proportional changes to staffing needs. A variety of different algorithms (e.g. vector auto-regression, ARIMA), provide different approaches to minimize the forecasting error. These predictions lead to hiring decisions in anticipation high volume times and flight risk. The forecasted demand complements the optimal staffing plans to enable our clients to plan their future staffing in a dynamic environment with a higher level of precision. 2 - Creating Optimized Hospital Workforce Solutions Jeffrey Jung, Senior Consultant, Navigant Consulting, Inc., New York, NY, 60606, United States, Jeffrey.jung@navigant.com, Hannan Abdi, Yaqiong Li, Asli Kayaalp Hospitals are faced with the challenge of staffing to demand; however, the demand is sometimes hard to predict until just before a shift starts. Managers are therefore inclined to staff to capacity, or use expensive resources to fill gaps. These methods are inefficient and costly. Using mixed integer programming, we model the needs of a group of like-units and determine the optimal numbers of core and flexible resources (float pool vs. per diem, overtime, etc.) required to satisfy demand while minimizing cost. The results allow managers to allocate the best mix of resources (core vs. flexible, part-time vs. full-time) and plan schedules to meet the varying demand throughout the day/week. 3 - Sustaining Performance using Workforce Management Tools Asli Kayaalp, PhD, Navigant Consulting, Inc., Washington, DC, To achieve and sustain the value of labor optimization, it is critical for hospital operational leaders to use tools and technologies that allows them to measure and track their performance. These tools include a well-defined position control process with mechanisms to track the impact of decision-making, predictive hiring plans and schedule planners all informed by the optimal staffing models. These tools are complemented by standardized productivity reporting analytics across functional areas. We work with the client to assess the optimal utilization of existing management tools and technologies and identify additional tools to supplement those currently available. 320C Healthcare Operations Sponsored: Health Applications Sponsored Session Chair: Hessam Bavafa, Wisconsin School of Business, Madison, WI, 53706, United States, hbavafa@bus.wisc.edu 1 - The Impact of E-visits on Visit Frequencies and Patient Health: Evidence from Primary Care 20036, United States, asli.kayaalp@navigant.com, Vamshi Gunukula, Hannan Abdi, Jeffrey Jung TB06 TB05

Hessam Bavafa, Wisconsin School of Business, 4284C Grainger Hall, 975 University Avenue, Madison, WI, 53706, United States, hbavafa@bus.wisc.edu Secure messaging, or “e-visits,” between patients and providers has increased sharply in recent years, and many hope they will help improve healthcare quality while increasing provider capacity. Using a panel dataset from a large healthcare system in the United States, we find that e-visits trigger about 6% additional office visits, with mixed results on phone visits and patient health. These additional visits come at the sacrifice of new patients: physicians accept 15% fewer new patients each month following e-visit adoption. Our data on nearly 100,000 patients spans from 2008 to 2013, which includes the rollout and diffusion of e-visits in the health system we study. 2 - Black Swans in Patient Costs: Evidence and Implications for Patient Allocation and Reimbursement Michael Freeman, INSEAD, 1 Ayer Rajah Avenue, 138676, Singapore, mef35@cam.ac.uk, Stefan Scholtes Using a unique data set of patient-level clinical information and costs, we identify “black swan” events — i.e. the admission to hospital of patients who cost significantly more than expected — and study their distribution across over 50 hospitals in England. We show that while emergency black swans appear to be allocated at random across hospitals, there is evidence that for planned patients the black swans are concentrated in certain hospitals. We estimate the quality and productivity implications of a black swan patient beeing seen at a hospital that rarely treats such patients, and recommend how these patients might be better Mazhar Arikan, University of Kansas, 931 Drum Dr, Lawrence, KS, 66049, United States, mazhar@ku.edu, Baris Ata, Rodney P.Parker The Centers for Medicare and Medicaid Services’ (CMS) Conditions of Participation requirements evaluate transplant centers based on one-year risk- adjusted patient and organ survival rates post transplantation. Using actual transplant data, we empirically analyze some potential unintended consequences of these regulations such that more risk averse centers choose healthier patients and higher quality organs to transplant. 4 - Personalized Health Care Outcome Analysis of Cardiovascular Surgical Procedures Guihua Wang, Ross School of Business, University of Michigan, 1209 McIntyre Drive, Ann Arbor, MI, 48105, United States, guihuaw@umich.edu, Jun Li, Wallace J. Hopp Using patient-level data from thirty-five hospitals for six cardiovascular surgeries in New York state, we identify patient groups that exhibit significant differences in outcomes with a recently developed instrumental variable tree approach. We find that outcome differences between hospitals are heterogenous across patients. By quantifying these differences, we demonstrate that a large majority of patients can achieve better expected outcomes by selecting providers based on personalized outcome information. We also show how personalized outcome information can help providers to improve their processes and payers to design effective pay-for-performance programs. 322A Operations Research and Analytics for Healthcare Sponsored: Health Applications Sponsored Session Chair: Sharan Srinivas, Penn State University, State College, PA, 16803, United States, sus412@psu.edu 1 - Smart Appointment Scheduling Rules for Outpatient Clinics Sharan Srinivas, Penn State University, 265 Blue Course Dr, Apt 3E, State College, PA, 16803, United States, sus412@psu.edu, Arunachalam Ravindran In this research, a new framework that uses a combination of predictive analytics and simulation is proposed for scheduling patients dynamically. Further, new appointment scheduling rules that leverage patient specific no-show values are proposed to improve patient satisfaction and resource utilization of outpatient clinics. A case study with real data from a Family Medicine Clinic is used to compare the effectiveness of the proposed rules with current scheduling practices. The results indicate that the proposed scheduling rules significantly outperform the current practice for all the clinic settings tested. allocated to hospitals and how to reimburse for them appropriately. 3 - Cherrypicking Kidneys and Patients: Incentives in Transplant Centers TB07

304