Healthcare Analytics & Operations Workshop 2022
Welcome to Healthcare Analytics & Operations Workshop
Welcome to Healthcare Analytics & Operations Workshop
The Operations Management group at Lee Kong Chian School of Business (LKCSB), Singapore Management University (SMU) proudly presents the Healthcare Analytics & Operations Workshop on 19th December 2022. By bringing together scholars, expert practitioners, and policymakers in Singapore and worldwide, the workshop aims to share the latest development in healthcare research and practice.
The workshop will be fully face-to-face, and registration is free. You are kindly invited to join us. Please register for this event by clicking on the registration button below.
Speakers
Speakers
Northeastern University
Northeastern University
COE Distinguished Professor
Department Editor for M&SOM
Area Editor for OR
University of Pennsylvania
University of Pennsylvania
Assistant Professor of Operations, Information and Decisions
Assistant Professor of Health Care Management
Associate Editor for MS
Purdue University
Purdue University
Gerald Lyles Rising Star Professor of Management
Associate Editor for MS
Associate Editor for M&SOM
Senior Editor for POM
Johns Hopkins University
Johns Hopkins University
Professor of Operations Management & Business Analytics
Associate Editor for MS
Associate Editor for M&SOM
Senior Editor for POM
University of Texas at Dallas
University of Texas at Dallas
Sydney Smith Hicks Faculty Fellow
Assistant Professor at the University of Texas at Dallas
Singapore Management University
Singapore Management University
Associate Professor of Operations Management
Area Coordinator of Operations Management
Practitioner Panelists
Practitioner Panelists
Ministry of Health Singapore
Ministry of Health Singapore
Director (Future Systems Office)
Adjunct Associate Professor at Saw Swee Hock School of Public Health, National University Singapore
Adjunct Associate Professor at Centre of Regulatory Excellence, Duke-NUS Medical School
KK Women’s and Children’s Hospital
KK Women’s and Children’s Hospital
Senior Consultant, Neurology Service, Department of Paediatrics
Chief Medical Informatics Officer
Adjunct Associate Professor at Lee Kong Chian School of Medicine, Nanyang Technological University
Programme
Programme
8:55am - 9:00am Opening Speech
8:55am - 9:00am Opening Speech
9:00am - 10:00am Keynote Speech: Özlem Ergun
9:00am - 10:00am Keynote Speech: Özlem Ergun
Examining Resiliency in Pharmaceutical Drug Supply Chains Incorporating Stakeholder Behaviors
Examining Resiliency in Pharmaceutical Drug Supply Chains Incorporating Stakeholder Behaviors
Abstract: The growing epidemic of drug shortages in the United States causes challenges for providers all across the critical health care infrastructure and demonstrates the lack of resiliency within drug delivery supply chains. With many of these drugs having no acceptable substitute, drug shortages directly translate to a public health and safety risk. Between 2008 and 2014, there was a 393% increase in shortages for direct lifesaving emergency medicines. More recently, there have been multiple drug shortages for sterile injectable products that endangered the lives of patients across the country, leading directly to cancelled heart surgeries and chemotherapy treatments. The underlying causes of some shortages include a wide variety of disruptions from manufacturing suspensions due to production challenges to the physical impact of hurricanes on critical manufacturing sites. While some shortages have been linked to large events, this is not the case for all shortages. One of the understudied elements driving this crisis is the role of stakeholder behavior and decision-making practices across the supply chain echelons. We present an integrated simulation framework, which allows for instantiating, testing, and improving supply chains when accounting for behavioral components of the system. We examine how manufacturing disruptions in a pharmaceutical supply chain impact evolving trust dynamics among the stakeholders during and after a disruption and study the implications of these dynamics for the supply chain performance.
Abstract: The growing epidemic of drug shortages in the United States causes challenges for providers all across the critical health care infrastructure and demonstrates the lack of resiliency within drug delivery supply chains. With many of these drugs having no acceptable substitute, drug shortages directly translate to a public health and safety risk. Between 2008 and 2014, there was a 393% increase in shortages for direct lifesaving emergency medicines. More recently, there have been multiple drug shortages for sterile injectable products that endangered the lives of patients across the country, leading directly to cancelled heart surgeries and chemotherapy treatments. The underlying causes of some shortages include a wide variety of disruptions from manufacturing suspensions due to production challenges to the physical impact of hurricanes on critical manufacturing sites. While some shortages have been linked to large events, this is not the case for all shortages. One of the understudied elements driving this crisis is the role of stakeholder behavior and decision-making practices across the supply chain echelons. We present an integrated simulation framework, which allows for instantiating, testing, and improving supply chains when accounting for behavioral components of the system. We examine how manufacturing disruptions in a pharmaceutical supply chain impact evolving trust dynamics among the stakeholders during and after a disruption and study the implications of these dynamics for the supply chain performance.
10:00am - 10:45am Talk: Hummy Song
10:00am - 10:45am Talk: Hummy Song
Leapfrogging for Last-mile Delivery in Health Care
Leapfrogging for Last-mile Delivery in Health Care
Abstract: Radical technological innovations may allow leapfrogging over traditional solutions to improve access to quality medical care, especially in hard-to-reach areas. Using data from Rwandan public hospitals, we examine the impact of using drones for the delivery of blood products on inventory management and health outcomes. We find that adopting drone delivery leads to a 62% reduction in on-hand inventory of blood products, 42% reduction in their wastage, and 88% decrease in inpatient mortality from postpartum hemorrhage (PPH). Hospitals that experienced road infrastructure improvements prior to adopting drone delivery see a quarter of the decline in PPH mortality compared to facilities that only adopted drone delivery, suggesting a leapfrogging effect.
Abstract: Radical technological innovations may allow leapfrogging over traditional solutions to improve access to quality medical care, especially in hard-to-reach areas. Using data from Rwandan public hospitals, we examine the impact of using drones for the delivery of blood products on inventory management and health outcomes. We find that adopting drone delivery leads to a 62% reduction in on-hand inventory of blood products, 42% reduction in their wastage, and 88% decrease in inpatient mortality from postpartum hemorrhage (PPH). Hospitals that experienced road infrastructure improvements prior to adopting drone delivery see a quarter of the decline in PPH mortality compared to facilities that only adopted drone delivery, suggesting a leapfrogging effect.
10:45am - 11:00am Coffee/tea break
10:45am - 11:00am Coffee/tea break
11:00am - 11:45am Talk: Susan Lu
11:00am - 11:45am Talk: Susan Lu
What Can Personal Statements Tell Us? Insights about Physicians’ Personality Traits and Their Medical Performance
What Can Personal Statements Tell Us? Insights about Physicians’ Personality Traits and Their Medical Performance
Abstract: In this study, we investigate the impact of physicians’ personality traits on their clinical performance and find leveraging personality traits provide a promising pathway toward understanding physicians’ cognitive behavior and their clinical performance. Specifically, we use physicians’ personal statements posted on an online physician review platform and extract 2,073 physicians’ personality traits from the unstructured physician personal statement following the Big Five model of personality. To address non-random matching between patients and physicians, we adopt a quasi-random setting to study patients who arrived at emergency departments (EDs) with accidental injuries in Florida. Our analyses show that being treated by physicians with higher openness scores is associated with lower in-hospital mortality rates, lower lab test costs, and shorter length of stay taking advantage of new treatment techniques. In contrast, physicians with higher conscientiousness scores tend to incur more lab test costs and make patients wait longer to undergo principle procedures. Moreover, agreeable physicians are more likely to help patients save lab test costs because of empathy, which is manifested more pronouncedly for low-income patients. These findings deepen the understanding of the Big Five personality theory in healthcare ED context, reveal the hidden value of supply-side user-generated content on predicting service quality, and provide managerial implications regarding the future potential of leveraging physician personality in healthcare operations management. The physicians’ latent personality traits could be considered in healthcare management such as online review platform, physicians recommendation, ED scheduling, among others.
Abstract: In this study, we investigate the impact of physicians’ personality traits on their clinical performance and find leveraging personality traits provide a promising pathway toward understanding physicians’ cognitive behavior and their clinical performance. Specifically, we use physicians’ personal statements posted on an online physician review platform and extract 2,073 physicians’ personality traits from the unstructured physician personal statement following the Big Five model of personality. To address non-random matching between patients and physicians, we adopt a quasi-random setting to study patients who arrived at emergency departments (EDs) with accidental injuries in Florida. Our analyses show that being treated by physicians with higher openness scores is associated with lower in-hospital mortality rates, lower lab test costs, and shorter length of stay taking advantage of new treatment techniques. In contrast, physicians with higher conscientiousness scores tend to incur more lab test costs and make patients wait longer to undergo principle procedures. Moreover, agreeable physicians are more likely to help patients save lab test costs because of empathy, which is manifested more pronouncedly for low-income patients. These findings deepen the understanding of the Big Five personality theory in healthcare ED context, reveal the hidden value of supply-side user-generated content on predicting service quality, and provide managerial implications regarding the future potential of leveraging physician personality in healthcare operations management. The physicians’ latent personality traits could be considered in healthcare management such as online review platform, physicians recommendation, ED scheduling, among others.
11:45am - 12:30pm Talk: Sarah Yini Gao
11:45am - 12:30pm Talk: Sarah Yini Gao
Optimizing Initial Screening for Colorectal Cancer Detection with Adherence Behavior
Optimizing Initial Screening for Colorectal Cancer Detection with Adherence Behavior
Abstract: Two-stage screening programs are widely adopted for early colorectal cancer (CRC) detection. Individuals receiving positive outcomes in the first-stage (initial) test are recommended to undergo a second-stage test (colonoscopy) for further diagnosis. We study the initial test design—i.e., selecting cutoffs to report test outcomes—to balance the trade-off between screening effectiveness (i.e., cancer detection) and efficiency (i.e., colonoscopy costs), considering that not all individuals adhere to the guidelines to follow up with a colonoscopy after receiving positive outcomes. We integrate the Bayesian persuasion framework with information avoidance to model the problem. We show that under certain conditions, using a single cutoff in the initial test is optimal for follow-up maximization, and a continuous test (i.e., showing exact readings of the biomarker) is optimal for effectiveness maximization. We apply the framework to Singapore's CRC screening design and calibrate the model using various sources of data, including a nationwide survey in Singapore. Our results suggest that compared with the current practice, increasing the cutoff to the level that maximizes expected follow-ups by cancer patients can detect 969 more CRC incidences and prevent 37,820 colonoscopies. The current practice of using lower cutoffs to achieve high sensitivity can backfire and lead to excessive unnecessary colonoscopies and low adherence. Leveraging the interpretable clustering technique, we find that using a lower cutoff for males older than 60 and females older than 70 (high-risk and high-adherence groups) and a higher cutoff for the remaining screening population (low-risk and low-adherence groups) can further improve screening effectiveness and efficiency.
Abstract: Two-stage screening programs are widely adopted for early colorectal cancer (CRC) detection. Individuals receiving positive outcomes in the first-stage (initial) test are recommended to undergo a second-stage test (colonoscopy) for further diagnosis. We study the initial test design—i.e., selecting cutoffs to report test outcomes—to balance the trade-off between screening effectiveness (i.e., cancer detection) and efficiency (i.e., colonoscopy costs), considering that not all individuals adhere to the guidelines to follow up with a colonoscopy after receiving positive outcomes. We integrate the Bayesian persuasion framework with information avoidance to model the problem. We show that under certain conditions, using a single cutoff in the initial test is optimal for follow-up maximization, and a continuous test (i.e., showing exact readings of the biomarker) is optimal for effectiveness maximization. We apply the framework to Singapore's CRC screening design and calibrate the model using various sources of data, including a nationwide survey in Singapore. Our results suggest that compared with the current practice, increasing the cutoff to the level that maximizes expected follow-ups by cancer patients can detect 969 more CRC incidences and prevent 37,820 colonoscopies. The current practice of using lower cutoffs to achieve high sensitivity can backfire and lead to excessive unnecessary colonoscopies and low adherence. Leveraging the interpretable clustering technique, we find that using a lower cutoff for males older than 60 and females older than 70 (high-risk and high-adherence groups) and a higher cutoff for the remaining screening population (low-risk and low-adherence groups) can further improve screening effectiveness and efficiency.
12:30pm - 1:30pm Lunch
12:30pm - 1:30pm Lunch
1:30pm - 2:30pm Keynote Speech: Tinglong Dai
1:30pm - 2:30pm Keynote Speech: Tinglong Dai
Purposeful Design for AI-Augmented Healthcare: Harnessing Physician-in-the-Loop Systems to Improve the Patient Journey
Purposeful Design for AI-Augmented Healthcare: Harnessing Physician-in-the-Loop Systems to Improve the Patient Journey
Abstract: Artificial intelligence (AI) has become an integral part of healthcare delivery, as evidenced by the FDA's approval of more than 500 medical AI systems by October 2022. Many people now believe AI will not replace physicians, but that physicians who use AI will replace those who do not. However, incorporating AI into health care delivery requires a rethinking and redesign of day-to-day workflows, which most health care systems are not ready for. In this talk, I discuss several ongoing projects aimed at understanding and overcoming barriers to the appropriate use of AI in clinical practice. I will also discuss the feasibility of using AI to improve clinician productivity, expand access to care, and reduce disparities, drawing on a series of studies conducted in the U.S. and the developing world. I propose a research agenda for business scholars to consider as our community continues to discover principles of purposeful design for AI-augmented healthcare delivery systems.
Abstract: Artificial intelligence (AI) has become an integral part of healthcare delivery, as evidenced by the FDA's approval of more than 500 medical AI systems by October 2022. Many people now believe AI will not replace physicians, but that physicians who use AI will replace those who do not. However, incorporating AI into health care delivery requires a rethinking and redesign of day-to-day workflows, which most health care systems are not ready for. In this talk, I discuss several ongoing projects aimed at understanding and overcoming barriers to the appropriate use of AI in clinical practice. I will also discuss the feasibility of using AI to improve clinician productivity, expand access to care, and reduce disparities, drawing on a series of studies conducted in the U.S. and the developing world. I propose a research agenda for business scholars to consider as our community continues to discover principles of purposeful design for AI-augmented healthcare delivery systems.
2:30pm - 3:15pm Talk: Guihua Wang
2:30pm - 3:15pm Talk: Guihua Wang
The Spillover Effect of Suspending Non-essential Surgery: Evidence from Kidney Transplantation
The Spillover Effect of Suspending Non-essential Surgery: Evidence from Kidney Transplantation
Abstract: The COVID-19 pandemic has posed an epic challenge to the U.S. healthcare industry. Between March and April 2020, multiple state governors issued orders to temporarily suspend non-essential surgical procedures. The suspensions caused the healthcare industry to shed millions of jobs, raising concerns about the availability of essential procedures. In this paper, we estimate the potential spillover effect of suspending non-essential surgery on patient access to essential health services, using deceased-donor kidney transplantation as the clinical setting. Through analyzing a dataset of all U.S. kidney transplantation procedures, we observe a steep reduction in the volume of deceased-donor kidney transplantation across nearly all states amid the initial months of the pandemic. However, states that suspended non-essential surgery experienced far steeper reductions than those without. Using a difference-in-differences approach, we estimate a state-level suspension of non-essential surgery led to a 23.6% reduction in the transplant volume. Our study reveals the spillover effect of state-level health policies on patient access to essential services such as deceased-donor kidney transplantation. Our mediation analysis shows 38.7% of the spillover effect can be attributable to the change in healthcare employment, indicating these suspensions caused hospitals to reduce the size of their workforces required for all procedures, which ultimately had a negative impact on access to essential procedures. Instead of suspending all non-essential surgery in the event of a future pandemic, policymakers should consider more granular approaches to safeguarding the healthcare workforce critical to supporting essential services.
Abstract: The COVID-19 pandemic has posed an epic challenge to the U.S. healthcare industry. Between March and April 2020, multiple state governors issued orders to temporarily suspend non-essential surgical procedures. The suspensions caused the healthcare industry to shed millions of jobs, raising concerns about the availability of essential procedures. In this paper, we estimate the potential spillover effect of suspending non-essential surgery on patient access to essential health services, using deceased-donor kidney transplantation as the clinical setting. Through analyzing a dataset of all U.S. kidney transplantation procedures, we observe a steep reduction in the volume of deceased-donor kidney transplantation across nearly all states amid the initial months of the pandemic. However, states that suspended non-essential surgery experienced far steeper reductions than those without. Using a difference-in-differences approach, we estimate a state-level suspension of non-essential surgery led to a 23.6% reduction in the transplant volume. Our study reveals the spillover effect of state-level health policies on patient access to essential services such as deceased-donor kidney transplantation. Our mediation analysis shows 38.7% of the spillover effect can be attributable to the change in healthcare employment, indicating these suspensions caused hospitals to reduce the size of their workforces required for all procedures, which ultimately had a negative impact on access to essential procedures. Instead of suspending all non-essential surgery in the event of a future pandemic, policymakers should consider more granular approaches to safeguarding the healthcare workforce critical to supporting essential services.
3:15pm - 3:30pm Coffee/tea break
3:15pm - 3:30pm Coffee/tea break
3:30pm - 4:15pm Talk: Daniel Zhichao Zheng
3:30pm - 4:15pm Talk: Daniel Zhichao Zheng
Estimating Patient Health Transition From Data Censored By Treatment-Effect-Based Policies
Estimating Patient Health Transition From Data Censored By Treatment-Effect-Based Policies
Abstract: Treatment-effect-based decision policies are increasingly used in healthcare problems. Such policies leverage predictive information on patient health transitions and treatment outcomes for treatment recommendations. However, these policies can significantly censor the observation of patients’ health transitions and distort the estimation of transition probability matrices (TPMs). We propose a structural model to recover the underlying true TPMs from censored transition observations. We show that the estimated TPM from the structural model is consistent and asymptotically normally distributed and also maximizes the log-likelihood of observing the data. Using hypothesized data with known ground-truth TPMs, we demonstrate the advantages of our model against benchmark estimation methods that ignore the censoring mechanism. We further implement our model to estimate patient health transitions using observed data for the extubation problems in an intensive care unit (ICU). Formulating the extubation problem as a classical optimal stopping Markov Decision Process model, we show that the proposed method, with more accurate estimated TPMs considering treatment-effect-based policy censoring, can reduce patients’ length of stay in the ICU compared to benchmark methods.
Abstract: Treatment-effect-based decision policies are increasingly used in healthcare problems. Such policies leverage predictive information on patient health transitions and treatment outcomes for treatment recommendations. However, these policies can significantly censor the observation of patients’ health transitions and distort the estimation of transition probability matrices (TPMs). We propose a structural model to recover the underlying true TPMs from censored transition observations. We show that the estimated TPM from the structural model is consistent and asymptotically normally distributed and also maximizes the log-likelihood of observing the data. Using hypothesized data with known ground-truth TPMs, we demonstrate the advantages of our model against benchmark estimation methods that ignore the censoring mechanism. We further implement our model to estimate patient health transitions using observed data for the extubation problems in an intensive care unit (ICU). Formulating the extubation problem as a classical optimal stopping Markov Decision Process model, we show that the proposed method, with more accurate estimated TPMs considering treatment-effect-based policy censoring, can reduce patients’ length of stay in the ICU compared to benchmark methods.
4:15pm - 4:30pm Coffee/tea break
4:15pm - 4:30pm Coffee/tea break
4:30pm - 6:00pm Panel Discussion
4:30pm - 6:00pm Panel Discussion
Panelists:
Panelists:
Kelvin Bryan Tan - Director (Future Systems Office) at Ministry of Health Singapore
Terrence Thomas - Senior Consultant of Neurology Service and Chief Medical Informatics Officer at KK Women’s and Children’s Hospital
Bee Keow Goh - Assistant Director, Data Analytics Office at KK Women’s and Children’s Hospital
Özlem Ergun - COE Distinguished Professor at Northeastern University
Tinglong Dai - Professor of Operations Management & Business Analytics at the Johns Hopkins University
Susan Lu - Gerald Lyles Rising Star Professor of Management at Purdue University
Moderator:
Moderator:
Leon Liang Xu - Assistant Professor of Operations Management at Singapore Management University
6:00pm - 6:05pm Concluding Remarks
6:00pm - 6:05pm Concluding Remarks
Organising Committee
Organising Committee
Faculty, SMU
Faculty, SMU
Faculty, SMU
Faculty, SMU
PhD Student, SMU
Questions?
Questions?
Email us at: Healthcare.Workshop.SMU2022@gmail.com