The Centers for Medicare and Medicaid Services (CMS) offers public data samples of claims data (diagnosis, procedure, billed amounts, benefit amounts). Our client seeks to model treatment patterns for cancer patients, but first we need to find those treatment patterns. Patterns may include: Evaluation and management aggregate data Sequence of procedures/therapies/visits Cost summaries via histograms Items you may discover along the way Ultimately, you’ll need to return a data set with the oncology patients claims as well as a dashboard summary of the data and analyses you performed. To complete this project, you’ll need to acquire diagnostic codes related to one or more types of cancer (nobody’s expecting medical expertise here), high level knowledge of procedure codes (they’re hierarchical so we can focus on the easiest level to manage), and some data wrangling to get things sorted. Samples can be acquired here: https://www.cms.gov/Research-Statistics-Data-and-Systems/Downloadable-Public-Use-Files/SynPUFs/DE_Syn_PUF