OBJECTIVES: To provide a source document in the creation of a dataset for a hospital and health insurer merger. METHODS: A review of the literature was performed to determine best practices in the development of a dataset to link hospital data with health insurance data.  Peer reviewed articles and reports were retrieved from the published literature from 2002 to 2012. RESULTS: Results revealed a total of 30 articles and reports. A summary of this information is provided as a step-by-step guide on how to setup a dataset. The purpose of data collection should be discussed in detail. The data must be de-identified due to HIPAA regulations. Patient data, such as date of birth, hospital admission, discharge date, sex, and ICD-9 code can be combined to form an identifier that is nearly 100% unique and found in both hospital and payer databases.  This is stored as a separate dataset.  Additionally, a dataset from the payer should be extracted and stored.  The resulting datasets can then be sent to a Data Coordinating Center, where data is evaluated for quality assurance and combined into one master dataset.  Prior to combining datasets it is helpful to identify potential research outcomes of interest to ensure that the variables are accurately represented by the data. CONCLUSIONS: Currently, there are few databases that allow researchers to follow a patient from hospital admission to post discharge. Researchers who obtain data offering a cross sectional view of a patient’s health status must find creative ways to link patients over time either making many assumptions or using various simulation methods.  A hospital and health insurer merger offers a unique opportunity to develop a dataset to track patient outcomes.