Informs 2015

221 / 552

INFORMS Philadelphia – 2015
219
MC41
41-Room 102A, CC
Joint Session MSOM-Health/HAS/Analytics: Data-
Driven Modeling in Healthcare III
Sponsor: Manufacturing & Service Oper
Mgmt/Healthcare Operations
Sponsored Session
Chair: Mehmet Ayvaci, Asst Professor, University of Texas-Dallas,
School of Management, Richardson, TX, 75080,
United States of America,
1 - Making the Case for Case Management
Margret Bjarnadottir, Assistant Professor of Management Science
and Statistics, Robert H. Smith School of Business, University of
Maryland, 4324 Van Munching Hall, College Park, MD, 20742,
United States of America, ,
David Anderson
Most case management programs target current high-cost patients. However the
real cost savings potential is including lower cost patients at high risk of future
high costs. We demonstrate the potential of association rules for identification of
these high value patients and derive a general upper bound methodology on
classification performance.
2 - Managing Office Revisit Intervals and Patient Panel Sizes in
Primary Care
Hessam Bavafa, Assistant Professor, Wisconsin School of Business,
Madison, WI, United States of America,
In recent years, the drive to contain health care costs in the US has increased
scrutiny of the traditional mode of delivering primary care where a patient is
treated by his primary care physician during a face-to-face visit. In particular, two
approaches, the use of “e-visits” and greater reliance on non-physician providers,
have been suggested as lower-cost alternatives to the traditional set-up. In this
paper, we consider a patient panel and develop a new model of patient health
dynamics.
3 - Outpatient-clinic Capacity Management when Continuity of
Care Matters
Yichuan Ding, UBC, 2053 Main Mall, Sauder School of Business,
Vancouver, BC, V6T1Z2, Canada,
Diwakar Gupta, Xiaoxu Tang
We study how to manage capacity in an outpatient clinic with the goal of
maximizing service volume as well as maintaining high level of continuity of care
(COC). We consider a simple strategy that doctors may use to improve COC —
book a follow-up appointment (FUA) for a patient before she leaves the clinic. In
order to encourage the doctor to use this strategy, the current fee-for-service
mechanism must be revised to compensate doctors for FUAs that are no show or
late cancelled.
4 - The Impact of Health Information Exchanges on Emergency
Department Length of Stay
Jan Vlachy, PhD Student, Georgia Institute of Technology, 755
Ferst Drive, NW, Atlanta, GA, 30332, United States of America,
Turgay Ayer, Mehmet Ayvaci, Zeynal Karaca
Electronic exchange of health information (HIE) is expected to improve
coordination in emergency departments (ED). We empirically study the impact of
HIEs on ED length of stay (LOS) using a large longitudinal dataset comprising
about 5.8 million visits to 63 EDs over three years. Overall, we find that HIE
adoption is associated with substantial reductions in LOS, but this impact depends
on various contextual and situational factors.
MC42
42-Room 102B, CC
Joint Session MSOM-Health/HAS/Analytics:
Healthcare Analytics
Sponsor: Manufacturing & Service Oper
Mgmt/Healthcare Operations
Sponsored Session
Chair: Tinglong Dai, Assistant Professor, Johns Hopkins University, 100
International Dr, Baltimore, MD, 21202, United States of America,
Co-Chair: Song Hee Kim, Assistant Professor, Marshall School of
Business, University of Southern California, Los Angeles, CA,
United States of America,
1 - Efficient Spatial Allocation of Epidemic Intervention Resources
with a Focus on Ebola in West Africa
Eike Nohdurft, Research Assistant, WHU - Otto Beisheim School
of Management, Burgplatz 2, Vallendar, 56179, Germany,
, Elisa Long, Stefan Spinler
The recent Ebola outbreak has shown that containment of an infectious disease
relies on deployment and allocation of intervention resources. A model reducing
the number of infections through improved allocation is proposed. Allocation
decisions are based on a spatial compartmental epidemic model with a novel
factor dynamically incorporating behavioral change in the population. Our
approach could avoid up to 23% of the infections.
2 - Information Aggregation and Classification under Anchoring Bias:
Application to Breast Imaging
Mehmet Eren Ahsen, Researcher, IBM Research, 1101 Route 134
Kitchawan RD #13-146C, Yorktown Heights, NY, 10598, United
States of America, , Mehmet Ayvaci,
Srinivasan Raghunathan
We study optimal aggregation and subsequent classification for the case of two
sources of information where the interpretation of the primary information
(mammography) is biased by the secondary information (risk profile). We
examine the relationship between bias, weights assigned, and the decision
thresholds in the context of optimal utility or the optimal discriminative ability.
3 - Priority and Predictability
Jillian A Berry Jaeker, Assistant Professor, Boston University, 595
Commonwealth Avenue, Boston, MA, 02215, United States of
America,
This study explores how patient admission characteristics (i.e. whether a patient is
scheduled or emergent; medical or surgical) moderate the effects of high
workload and demand. In particular, the probabilities of admission and discharge,
by patient type are analyzed. The results of this study provide an estimation of the
impact of predictability on patient flow.
4 - Decision Ambiguity and Conflicts of Interests in Interventional
Cardiology Decision-Making
Tinglong Dai, Assistant Professor, Johns Hopkins University, 100
International Dr, Baltimore, MD, 21202, United States of
America, , Chao-wei Hwang, Xiaofang Wang
With the rapidly rising cost of health care, there is a renewed urgency for
reducing inappropriate use of percutaneous coronary interventions (PCI). In this
work, we provide a quantitative analytical model of clinical and non-clinical
factors influencing PCI decision-making processes. Our model helps inform
policy-makers designing guidelines to optimize the use of PCI.
MC43
43-Room 103A, CC
Game Theoretic Models in Revenue Management II
Sponsor: Revenue Management and Pricing
Sponsored Session
Chair: Santiago Balseiro, Assistant Professor, Duke University,
100 Fuqua Drive, Durham, NC, 27708, United States of America,
Co-Chair: Ozan Candogan, University of Chicago, Booth School of
Business, Chicago, IL, United States of America,
1 - Learn and Screen: A Strategic Approach to Collaborative
Inventory Management
Bharadwaj Kadiyala, PhD Candidate, The University of Texas at
Dallas, 800 West Campbell Road, 3.218, Dallas, TX, 75080, United
States of America, Ozalp Ozer
We propose a dynamic mechanism for a supplier who periodically replenishes
inventory with partial knowledge of demand distribution. By combining the best
of Bayesian updating and screening mechanism, we show that in addition to
maximizing profit, inventory decisions also serve a strategic purpose in eliciting
demand information from the buyer.
2 - Optimal Contracts for Intermediaries in Online Advertising
Santiago Balseiro, Assistant Professor, Duke University, 100
Fuqua Drive, Durham, NC, 27708, United States of America,
, Ozan Candogan
The prevalent method online advertisers employ to acquire impressions is to
contract with an intermediary. We study the optimal contract offered by the
intermediary when advertisers’ budgets and targeting criteria are private. We
introduce a novel approach to tackle the resulting multi-dimensional dynamic
mechanism design problem, and show that an intermediary can profitably
provide bidding service to a budget-constrained advertiser and at the same time
increase the overall market efficiency.
MC43