Predictive Analytics Workshop

Predictive Analytics for Population Health Management

June 2, 2015
8:00 am – 4:30 pm
Research Triangle Foundation

Click here to view the agenda.

Topics include:

Predicting the Invisible Patient: Using Predictive Analytics to Reduce Suffering, Save Lives, and Optimize Cost of Care

Patient-centered care can be improved exponentially if the caregiver can forecast illness or complications so that preventative measures can be applied. Recognizing this, leadership from Baptist Health collaborated with a leading software developer and clinical experts to implement a predictive solution that analyzed patient phenotypes to predict disease and infection across the system. Their objectives were to reduce suffering and save lives while optimizing cost of care. The speaker will present predictive findings against one use case, CAUTI, including the business case for prevention; models used; and preliminary health and financial outcomes. Shantanu Nigam (Jvion)

Patient-centric Predictive Analytics

The Carolinas Center for Medical Excellence’s ConvergX applies predictive statistical modeling to proactively identify high-risk patients, so ACO care coordinators can deploy targeted resources that address patient needs and prevent unnecessary utilization. In this way, ConvergX enables ACOs to attain their goals of improved population health, reduced cost of care, and realized “shared savings” to providers.  ConvergX enables ACOs to do more than just report data to CMS: it combines analytics with clinical consulting expertise to manage population health, navigate patient-level data, and improve both clinical and health outcomes for providers and patients. Angela Diaz (CCME) & Stephen Nuckolls (Coastal Carolina Health Care)

Real-Time Healthcare with BI and Analytics

Many healthcare organizations have turned to business intelligence (BI) as a primary method to bring information from disparate systems together. Implementing BI is a first step in the right direction; however, BI will not provide the answer to this question: How will Healthcare Service Organizations become more cost-effective in the future? Organizations not only need to know the present conditions and patient’s history, but also should be able to predict future outcomes. BI provides insight into the “now”: predictive analytics will provide insight into the future. Predictive analytics can help a healthcare service organization effectively identify patients who are highly-likely to return to a healthcare provider due to a failure to execute post-visit care instructions, and implement follow-up procedures with the patient. Predictive revenue analytics can help a healthcare organization predict revenue by merging complex patient accounting data, insurance claims revenue statistics, collection payments and historical patient visits. Kelly Crossley & J. C. Layton (InfraScience)

Care Insights: Actionable Information and High-Yield Care Opportunities in Population Health Management

As a 2015 starter in the Medicare Shared Savings Program, the Carolina Medical Home Network ACO seeks to better coordinate care for its safety net Medicare population by improving clinical, operational, and financial quality. Community Care of North Carolina is a key partner in this value-driven endeavor, as they provide the necessary analytics tools to create actionable data on clinical quality, cost, utilization, and patient risk, identifying those patients most likely to benefit from more intensive care management.” Robert Eick (NC Community Health Center Association) and Annette DuBard & Marya Upchurch (CCNC)

Controlling Cost and Quality in Post-Acute Care

In the world of value-based reimbursement, it is essential for provider organizations to manage the quality of care for their patients across the full spectrum of care while also balancing the cost. One area where providers can impact what happens outside their offices is in managing the patient transition from inpatient hospital to post-acute care setting. By using information obtained from claims data (readmission rates, cost of care, etc.) together with publicly available quality data, CHESS has developed a post-acute care strategy based on a scorecard method to identify high quality / low cost post-acute care providers. Armed with information about both the patient and the post-acute facility, CHESS can utilize predictive analytics to assess the “risk of an adverse event for patient x at facility y.” This information helps CHESS develop and manage a narrow referral network of post-acute care providers for their clients in different geographic regions while also being able to predict the best choice of facility for a particular patient. Lisa Shock & Keith Thompson (Cornerstone Health Enablement Strategic Solutions)

Analytics for Population Health Management

This presentation will focus on the use of administrative healthcare data transformed into Episode of Care data and novel statistical methods to investigate different population health management issues such as risk segmentation of patients for targeted chronic care management, cost and quality of care analytics, as well as health care fraud and abuse. We will present a case study featured in the Cloudera Data Science Challenge for detecting costs variation and anomalies in the US Medicare Health Insurance System plus how we used various data wrangling and mining methods in combination with machine learning tools to identify: (a) providers that overcharge for certain procedures and regions where procedures are too expensive; (b) the three providers that are least similar to other providers and the three regions that are least similar to other regions; and (c) 10,000 Medicare patients who are involved in anomalous activities.  We will also discuss a standardized predictive analytical process undertaken to solve the big data problem that focused on identifying anomalous Medicare patients, anomalous procedures, and anomalous providers in the different health care markets across the United States. At the end of the presentation a panel of SAS statisticians will be available for a Q/A session to discuss the analytical strategy, techniques, and big data technologies used to solve each identified problem area. David Olaleye, Brad Klenz, Patrick Hall and Lina Clover (SAS Institute)


Kelly Crossley is the APPI Practice Lead for InfraScience, where his responsibilities include standardizing application and cloud architectures and developing analytics architecture solutions for clients. Prior to joining InfraScience, he was a SharePoint Architect for Intellinet, Matrix Resources and Critigen. He is a Certified Microsoft Solutions Developer.

Angela Diaz, MBA is the Director of Provider Services at CCME. She is responsible for developing, marketing, and operating self-sustaining programs and services that assist physicians and providers improve the clinical effectiveness of health care. Her background includes over 25 years of leadership experience in operations, fiscal management, marketing and business development in various health care settings. She earned an MBA in Executive Management and is a licensed physical therapist in North Carolina.

Annette DuBard, MD, MPH is the SVP for Informatics and Evaluation at Community Care of NC. She has had a central role in developing web-based information systems used statewide to facilitate care coordination and quality improvement activities in practices and communities; to support targeted care management for patients with complex healthcare needs; and to monitor quality, utilization, and cost outcomes. She received her medical degree from the Johns Hopkins School of Medicine and her MPH in  Health Policy and Administration from UNC.

Robert A. Eick, MD, MPH is the Health Informatics Specialist for the North Carolina Community Health Center Association’s Health Center Controlled Network (HCCN). Serving as project lead, he manages partner relationships related to electronic health system infrastructure and technology. Additionally, he facilitates coordination of HCCN staff activity with work done by contracted partners, monitoring quality and effectiveness, expenditures, and progress toward the goals of the HCCN. Robert completed his medical and graduate studies at the Boonshoft School of Medicine at Wright State University.

J. C. Layton is Director of Healthcare and Application Infrastructure at InfraScience, where he develops infrastructure solutions specific to Healthcare and Life Sciences organizations. Previously, he was Sr. Vice President at Co-Designed, a SharePoint solutions provider, where his major projects were focused on the SharePoint foundation for Lakeview Center and Gwinnett Medical. He holds a Business Administration and Management degree from The Citadel.

Shantanu Nigam is the CEO of Jvion, a healthcare technology company that develops software designed to predict and prevent patient-level disease and financial losses. Previously, he spent the majority of his career serving as a leader in Accenture’s healthcare consulting practice as an executive, helping multiple healthcare clients going through compliance exercises by building strategy, roadmaps and leading large teams.

Stephen Nuckolls is the CEO of Coastal Carolina Health Care, PA, a 50 provider multi-specialty medical practice in New Bern, NC. He also serves as CEO of Coastal Carolina Quality Care, an advance payment model ACO, and as Chief Financial Officer for Atlantic Integrated Health. He earned his Master’s degree in Accounting and Finance from the UNC Kenan-Flagler Business School.

Lisa Shock, MHS, PA-C is Director of Care Transformation for CHESS (Cornerstone Health Enablement Strategic Solutions). She is a Duke University Physician Assistant graduate with more than 15 years of clinical experience in a variety of healthcare settings. Her experience includes Internal medicine and hospital specialty based practice as well as teaching geriatric medicine and long-term care.

Keith Thompson, MA, MS is a Data Scientist for CHESS (Cornerstone Health Enablement Strategic Solutions). His current interests include the application of data mining, predictive analytics and operations research to improve decision support for providers transitioning to value-based reimbursement. He holds Master’s degrees in both Applied Mathematics and Systems Science and is currently a PhD candidate in Systems Science at Binghamton University.

Marya Upchurch, MHA is the VP for Solution Development at CCNC Services. She works with stakeholders to design applications and tools for population health improvement. Her work includes population management, quality metric process development, ACO and regulatory reporting, financial modeling, multi-source data integrations and performance improvement projects. She earned her MHA in Healthcare Administration and Master’s in Accounting from UNC-Chapel Hill.

Exhibiting Opportunities

We have a limited number of exhibit tables available. The fee is $500 NCHICA members/$750 non-members and includes an exhibit table, registration for two staff members, breakfast, lunch and breaks. Click here to reserve a table.


The registration fee is $199 NCHICA members/$299 non-members/$50 students and includes continental breakfast and lunch. To register, click here.


Research Triangle Foundation of North Carolina, 12 Davis Drive, Research Triangle Park, NC. For directions, click here.

NCHICA has created a collaborative organization that no other State has achieved. It includes vendors, providers, payers, physicians, nurses, clinicians, attorneys, information technologists and other professionals, working together to improve healthcare within North Carolina in every way, from the use of technology, to public policy considerations, to patient privacy, and more. Belonging to, and participating in, this organization helps you to not only keep up with the changing healthcare environment, but to have some influence on those changes.

— David B. Dillehunt
FirstHealth of the Carolinas
Pinehurst, NC

Join our alliance of more than 300 organizations dedicated to transforming healthcare through information technology, informatics and analytics.

Learn More