Overview

In this course, we’re going to go over analytical solutions to common healthcare problems. I will review these business problems and you’ll build out various data structures to organize your data. We’ll then explore ways to group data and categorize medical codes into analytical categories. You will then be able to extract, transform, and load data into data structures required for solving medical problems and be able to also harmonize data from multiple sources. Finally, you will create a data dictionary to communicate the source and value of data. Creating these artifacts of data processes is a key skill when working with healthcare data.

Syllabus

Solving the Business Problems

In this module, you will explain why comparing healthcare providers with respect to quality can be beneficial, and what types of metrics and reporting mechanisms can drive quality improvement. You'll recognize the importance of making quality comparisons fairer with risk adjustment and be able to defend this methodology to healthcare providers by stating the importance of clinical and non-clinical adjustment variables, and the importance of high-quality data. You will distinguish the important conceptual steps of performing risk-adjustment; and be able to express the serious nature of medical errors within the US healthcare system, and communicate to stakeholders that reliable performance measures and associated interventions are available to help solve this tremendous problem. You will distinguish the traits that help categorize people into the small group of super-utilizers and summarize how this population can be identified and evaluated. You'll inform healthcare managers how healthcare fraud differs from other types of fraud by illustrating various schemes that fraudsters use to expropriate resources. You will discuss analytical methods that can be applied to healthcare data systems to identify potential fraud schemes.

Algorithms and "Groupers"

In this module, you will define clinical identification algorithms, identify how data are transformed by algorithm rules, and articulate why some data types are more or less reliable than others when constructing the algorithms. You will also review some quality measures that have NQF endorsement and that are commonly used among health care organizations. You will discuss how groupers can help you analyze a large sample of claims or clinical data. You'll access open source groupers online, and prepare an analytical plan to map codes to more general and usable diagnosis and procedure categories. You will also prepare an analytical plan to map codes to more general and usable analytical categories as well as prepare a value statement for various commercial groupers to inform analytic teams what benefits they can gain from these commercial tools in comparison to the licensing and implementation costs.

ETL (Extract, Transform, and Load)

In this module, you will describe logical processes used by database and statistical programmers to extract, transform, and load (ETL) data into data structures required for solving medical problems. You will also harmonize data from multiple sources and prepare integrated data files for analysis.

From Data to Knowledge

In this module, you will describe to an analytical team how risk stratification can categorize patients who might have specific needs or problems. You'll list and explain the meaning of the steps when performing risk stratification. You will apply some analytical concepts such as groupers to large samples of Medicare data, also use the data dictionaries and codebooks to demonstrate why understanding the source and purpose of data is so critical. You will articulate what is meant by the general phase -- “Context matters when analyzing and interpreting healthcare data.” You will also communicate specific questions and ideas that will help you and others on your analytical team understand the meaning of your data.