Author Archives: oitadmin

Incentive contract design for food retailers to reduce food deserts in the US

In the US, obesity affects over 37% of the adult and over 16% of the child and adolescent population. Although not-for-profit agencies cannot directly control what a person eats, they can influence the supply side of the obesity epidemic by incentivizing food retailers to open stores in regions of the US where healthy food options do not exist. CHOT researchers present an incentive contract design for food retailers to reduce food deserts in the US. These subsidies are designed to create financially viable conditions for food retailers to offer high quality, healthy food alternatives. The researchers developed optimization models to determine the most effective and equitable resource allocations. The impact of retailer location on obesity rate is based on estimates of marginal effect of incentives on obesity rate. Given an example initiative in metropolitan Atlanta, Georgia and surrounding counties, the overall countywide obesity rate would decrease by 1.17%. This incentive contract design strategy is a positive step toward ensuring that the underserved US population has better access to healthy foods while helping solve the obesity epidemic. To learn more about how to design an incentive program for your community please contact the authors.

Bastian, N. D., Swenson, E. R., Ma, L., Na, H. S., & Griffin, P. M. (2017). Incentive contract design for food retailers to reduce food deserts in the US. Socio-Economic Planning Sciences.

For more information, please contact: Nathanial Bastian, PhD at

Reducing Rate of Hospital Acquired Infections through Active Barrier Apparel

Healthcare professionals are at great risk of contracting healthcare-associated infections, but appropriate workplace apparel technology can provide a protective barrier.  CHOT researchers recently wrote an editorial that outlines guidance from professional societies and legislative actions that have provided positions on active barrier apparel as a best practice for healthcare apparel. This type of technology can replace today’s traditional apparel or uniforms, such as scrubs and white coats worn in healthcare settings throughout the United States. To learn more about the newest textile technology used to keep healthcare workers safe, please contact the authors.

Kash, B. A., & Davis, E. (2016). Active Barrier Apparel: The Simple, Evidence-Based Workplace and Patient Safety Strategy. Occupational Medicine and Health Affairs, 4(234), 2.

For more information, please contact: Bita Kash, PhD at


Comparing success and sustainability in two health systems

A recent CHOT study investigating how to successfully implement an innovative care management model such as Studer Group’s Evidence-based Leadership (EBL) tool found that culture (accountability, buy-in and communication) is most important for sustainability of initiative (long-term) and leadership seems most important initially during kick-off and early implementation stage. This research study analyzed the implementation of EBL in two large, US health systems by comparing and contrasting the factors associated with successful implementation and sustainability of the EBL initiative. Researchers found three themes associated with success and sustainability of EBL: leadership; culture; and organizational processes. This study offers health system leaders practical guidance on how to effectively implement organizational change initiatives by designing the appropriate system and environment conducive to successful and sustainable implementation of key patient care and management change initiatives.

Schuller, K. A., Kash, B. A., & Gamm, L. D. (2015). Comparing success and sustainability in two health systems. Journal of Health Organization and Management, 29(6), 684-700.


Research Contact:
Bita Kash, PhD

Predicating movement disorders using non-verbals

Design team interactions are one of the least understood aspects of the engineering design process. Given the integral role that designers play in the engineering design process, understanding the emotional states of individual design team members will help us quantify interpersonal interactions and how those interactions affect resulting design solutions. The methodology presented in this paper enables automated detection of individual team member’s emotional states using non-wearable sensors. The methodology uses the link between body language and emotions to detect emotional states with accuracies above 98%. The practical implications include that machine learning to detect emotions using non-invasive sensors in design teams and that there was a high accuracy over 90% achieved for detecting many body language poses.

Behoora, I., & Tucker, C. S. (2015). Machine learning classification of design team members’ body language patterns for real time emotional state detection. Design Studies, 39, 100-127.


Research Contact:
Conrad Tucker, PhD

How to increase diabetic patient adherence to self-management programs

Patient adherence is one of the most difficult changes to healthcare providers. This study models human behavior, both intentional and unintentional, relating to error in patient adherence of diabetes treatment. In a recent study, Penn State-CHOT researchers found that patient adherence was primarily driven by skill-based errors and intentional violations which lent itself to several risk mitigation strategies. This research study indicates that error classifications may be helpful in individualizing treatment interventions. By including device design modification, such as: (1) reduce patient inattention, (2) increase motivation to adhere, and (3) reduce pain barriers for glucometer use, this study demonstrated the feasibility of using an error classification approach in the identification and mitigation of the diabetes patients’ non-adherence with self-monitoring of blood glucose (SMBG).

Vaughn-Cooke, M., Nembhard, H. B., Ulbrecht, J., & Gabbay, R. (2015). Informing Patient Self-Management Technology Design Using a Patient Adherence Error Classification. Engineering Management Journal, 27(3), 124-130.

Research Contact:
Harriet Nembhard, PhD

Which one first? Prioritizing Vaccines for effective pandemic response

When limited vaccines are available, prioritized vaccination is considered the best strategy to mitigate the impact of a pandemic. A recent study led by Dr. Eva Lee, Georgia Institute of Technology-CHOT Site Director, found that without delay in vaccination start time, there is a reduction in prevalence of more than twofold of H1N1. Policy makers can use the results from this study to more rapidly evaluate better trade-offs to save more lives and better utilize limited resources during a pandemic event.

This study is believed to be the first mathematical computational model to combine disease propagation, dispensing operations, and optimization capability. It is also the first to define and allow for rapid determination of optimal switch triggers. The CDC confirms that this is the first time an actionable and operation switch trigger has been defined, an advance that is critical and vital to better mitigation of infections and mass casualties.

Lee, E. K., Yuan, F., Pietz, F. H., Benecke, B.A., & Burel, G. (2015). Vaccine Prioritization for Effective Pandemic Response. Interfaces, 45(5), 425-443.


Research Contact:
Eva Lee, PhD

Reinventing emergency department flow via physician-directed queuing system results in decreased wait time

In this CHOT study, a new flow model of emergency care delivery, physician-directed queuing (PDQ), was created after analysis of operational data and staff input of an overcrowded academic health center emergency department (ED) with increasing patient volumes and limited physical space for expansion. After implementing the PDQ model, the researchers observed a 91% decrease in door-to-bed time (211 to 19 minutes), an 83% decrease average waiting time (70 to 12 minutes), and increased patient satisfaction (from 17% to 85%).  In order to design this PDQ, researchers analyzed the operational data and staff input of an overcrowded academic health center emergency department (ED) with increasing patient volumes and limited physical space for expansion. EDs are a critical point of entry into the healthcare system. ED overcrowding create barriers to access and provision of appropriate care. ED crowding is associated with less timely care, decreases in patient satisfaction, and poor outcomes. With the new PDQ model, providers passively evaluate all patients upon arrival, actively manage patients requiring fewer resources, and direct patients requiring complex resources to further evaluation in ED areas. This model of practice can be applied to other patient care settings such as ambulatory care and imaging.

DeFlitch, C., Geeting, G., & Paz, H. L. (2015). Reinventing emergency department flow via healthcare delivery science. HERD : Health Environments Research & Design Journal, 8(3), 105-115.


Research Contact:
Harriet Nembhard, PhD

Significant cost saving attained with the implementation of a new ED decision support system

Improving an emergency department’s (ED) timeliness of care, quality of care, and operational efficiency while reducing avoidable readmissions, is fraught with difficulties, which arise from complexity and uncertainty. A recent CHOT study describes an ED decision support system that allows healthcare administrators to optimize workflow globally, taking into account the uncertainties of incoming patient diseases and associated care, thereby significantly reducing the length of stay (by 33% in the study hospital) and waiting time of patients (by 70%). This system couples machine learning, stimulation, and optimization to address these complex challenges. Using this system offers significant advantages in that it permits a comprehensive analysis of the entire patient flow from registration to discharge, enables a decision maker to understand the complexities and interdependencies of individual steps in the process sequence, and ultimately allows the users to perform system optimization. Overall benefits and impacts include improved timeliness of care, improved efficiency and emergency care, annual financial savings and revenues, encouragement of external sponsorship; health cost reductions, and improved quality of care in other facilities. It has been used successfully since 2010.

The quality of care impacts include 30% reduction of length of stay, and 70% of the associated average waiting time; 28% reduction in ED readmissions; 32% re-direct of non-urgent-care cases; 19% increase in ED throughput; and 30% reduction of patients who left-without-being-seen. The “golden hours of treatment” for trauma patients was reduced by 10%. The ED efficiencies at Grady led to a 10% reduction in ED-to-hospital admissions. 

The financial impacts of the ED efficiencies at Grady include a $7.5 million yearly savings in penalties due to a reduction in revisits. The alternative care facility for non-urgent conditions reduces ED costs by $21.6 million and results in $12.5 million in revenue. Expansion of trauma care and efficiency results in $19.1 million in revenue; the reduction in left-without-being-seen patients leads to $96.6 million in revenue. For a critical safety-net hospital with $1.5 billion of annual economic impact, only 8% of which results from private insurance, these financial gains have a tremendous impact on maintaining financial healthiness.

Lee, E. K., Atallah, H. Y., Wright, M. D., Post, E. T., Thomas IV, C., Wu, D. T., & Haley Jr, L. L. (2015). Transforming Hospital Emergency Department Workflow and Patient Care. Interfaces.


Research Contact:
Eva Lee, PhD

Patient segmentation and prediction of patient behaviors using the EMR and Machine Learning

In a recent CHOT study, researchers found healthcare organizations can effectively target specific segments of joint replacement patients with readily available information from the electronic medical record (EMR). The goal of this study was to apply unsupervised learning (clustering) methods to a large subset of the patient data to identify unique clusters of patients. By using a new algorithm developed by CHOT researchers, clinicians, can adopt specific protocols during the perioperative phases of Total Joint Arthroplasty by patient cluster, health insurers can isolate high costs and high-risk patients and then incentivize them and their primary care providers to reduce health risks, and patients can learn from their segment what actions to take now to optimize outcomes. Health organizations don’t need to purchase expensive market data to fully understand how patients cluster and respond to health promotions. This innovative algorithm presents an opportunity to efficiently and effectively target segments of the market with health promotions tailored specifically to positively impact health outcomes using existing data in the EMR.

Swensn, E. R., Bastian, N. D., & Nembhard, H. B. (2016). Data analytics in health promotion: Health market segmentation and classification of total joint replacement surgery patients. Expert Systems with Applications60, 118-129.


Research Contact:
Eric Swenson

A new and more accurate model for predicting compliance to evidence-based bundles of practice

CHOT researchers used probability generating functions to rapidly predict compliance and illustrate the improved detection of non-compliance over other methods. This new prediction model for monitoring compliance is practical, fast, and accurate compared to more commonly used types of compliance control charts based on binomial or normal distribution of compliance data, which are shown to produce poor detection properties. CHOT researchers also show that by using this correct method, exact results can be obtained fairly easily, in turn accelerating improvement towards greater compliance to evidence-based practice bundles. Compliance to evidence-based practices, individually and in ‘bundles’, remains an important focus of healthcare quality improvement for many clinical conditions.

Chen, B., Matis, T., & Benneyan, J. (2016). Computing exact bundle compliance control charts via probability generating functions. Health Care Management Science, 103-110.


Research Contact:
James Benneyan, PhD