Master of Science in Healthcare Analytics

Foundations of Data Management provides students exposure to fundamental data management skills used in modern information systems that support various operational and functional areas within a business organization. Topics covered in the course emphasize how data are identified, organized, processed and managed, in a manner warranting being considered one of the most valuable organizational assets. The course emphasizes a combination of theoretical ‘why’ and experiential ‘how’ within the confines of relational data modeling and querying. It is intended to help students develop fundamental data management skills that are essential to succeeding in subsequent courses.

This course provides students with foundational knowledge of descriptive and inferential statistics. The scope of this course is limited to univariate and bivariate statistics that are commonly used to conduct basic analyses of data, including reading-in data, reviewing individual variables and their properties, conducting any requisite data corrections and enhancements, developing an overall descriptive baseline, and taking initial steps of discovering and testing possible relationships. Also included in the course scope is an introductory overview of the basics of probability and sampling theories. Topic-wise, students will develop an understanding of numerous core statistical notions, including variable types and the underlying measurement scales, variable distributions, sampling distributions, the Central Limit Theorem, Bayes Theorem, measures of central tendency and variance, statistical significance, hypothesis testing, tests of difference, as well as correlation and cross-tabulation.

This course equips students with foundational concepts and techniques required for telling compelling stories with large complex data sets, using static and interactive graphics to depict outcomes, networks, time and maps. The importance of visualizing information for many analysts is often overlooked or downgraded, but if the visualization is ineffective, the decision-making processes and knowledge discovery will be compromised. This is a project-based course that begins with reviewing concepts of human perception and cognition and perceptual accuracy and preferences, followed by exploration of the basics of graphic design and making a “good” graph. Students explore why some data visualizations present information effectively and others do not, all while learning to appreciate visualization as a key component of Data Scientists’ and Business Analysts’ skill set.

This course introduces students to the US healthcare, a dynamic and complex system in which data analytics plays an increasingly greater role, influencing the ever-greater array of its many aspects. Students will develop a foundational knowledge of the mechanics of the US healthcare system, and will gain insights into how data-driven decisions influence clinical and business outcomes, with particular emphasis on value-based metric framework commonly used to manage outcomes. In a more operational sense, students will be introduced to roles and responsibilities of public, private and regulatory healthcare professionals in the context of their organizational roles, capabilities, and manners in which they make use of diverse sources and types of healthcare-related data.

Introduces the key concepts, common data collection approaches, sources and methods of epidemiology and population-based health research. For all methods, issues associated with healthcare data quality control, validity and reliability will be covered. Medical insurance, vital records, epidemiological research data and other major sources of public health data will be discussed. This course will introduce students to common and newly emerging data collection sources and methods used in epidemiological research. It will also introduce students to the challenges corporate America is undergoing due to the increase in the prevalence of non-communicable diseases in the workplace, and what population health strategy solutions are implemented to mitigate cost and facilitate optimal population health (for self-insured and fully insured companies). The course will also prepare students to manage the collection and storage of biological specimens, understand the fundamental concepts and methods of biostatistics as applied predominantly to clinical research and public health analyses. Clinical trials will also be addressed.

The goal of this course is to introduce students to the dual role and the dual benefit of dependence-focused multivariate statistical modeling techniques, commonly used in estimating the likelihood of outcomes of interest, and delineation of the key drivers of those outcomes. The former, often referred to as ‘predictive analytics,’ is focused on making forward-looking predictions, while the latter, commonly referred to as ‘prescriptive analytics,’ is geared at delineating and describing factors contributing to those predictions. The course is focused on developing rudimentary understanding of those two broad data analytical dimensions, and on introducing students to some of the more widely used predictive and prescriptive statistical techniques, including linear and logistic regression models, decision trees, as well as some of the more recent, machine learning focused techniques such as random forest.

This course introduces students to the fundamental conceptual, operational and experiential aspects of automated data mining approaches and techniques, focusing primarily on the use of pattern recognition algorithms to address common business problems involving classification and prediction. In this course students develop basic theoretical and experiential knowledge of supervised and unsupervised data mining approaches, framed in the context of the common data mining related challenges including dimensionality, regularization, overfitting and generalization. Various pattern and association rules and pattern discovery modes are also addressed, including compressed, sequential and spatial patterns, as well as multi-level and multi-dimensional associations. Lastly, common data mining applications, including clustering and prediction, are discussed and illustrated.

Students will develop a capstone project that will include collecting, analyzing and developing insights from healthcare data in an aspect of healthcare that is of interest to them. Faculty will work to pair students with partner companies or institutions to supply data and business challenges to use as the foundation of the capstone project or students can develop a research plan individually with their employer or another business, organization or institution. The Capstone experience will also provide to students critical skills needed to succeed in the healthcare analytics workforce today and to ensure they succeed and are valuable in the workforce today and in the future, as the healthcare business evolves, such as understanding of optimal project management skills, process and performance improvement techniques (Lean), and leadership skills.