Document Type

Research Project

Publication Date



The master patient index is one of the most important components within a healthcare system. It ensures that an individual patient is given a unique identifier that is used across the various separate clinical, financial and administrative systems and ensures that all information about that patient is organized and complete. Ensuring the MPI is accurate is critical since errors can have significant costs- both financially and in terms of patient outcomes. Patient registrars are the first line of defense when it comes to correctly identifying incoming patients and is where many errors in the MPI occur. Errors can be simple misspellings in a patient name, unknowingly assigning a patient a second medical record number, or the worst type, registering a patient under a different patient’s MRN leading to intertwined medical histories being viewed by providers. This study looked specifically at a major health system in Richmond, Virginia to determine the existing workflows on how patients were identified in the MPI and how errors were corrected once known. A literature review was performed to determine if any evidence based practices exist for maintaining MPI data integrity and how the focus health system compares. In addition, two other comparable health organizations were surveyed to determine how they compare with the target and with national standards. It was found that the health system of focus did not have a MPI quality program in place at all which explains why it struggles with errors. A plan for a MPI program using criteria in the Improvement Focused Model was created and implemented. This plan focused on better collection of data on errors, more standardized procedures for registrars, and constructive feedback to registrars when errors do occur. Although after a relatively short period of implementation, there was little difference seen in the number of new errors being generated, it did find that there were a relatively few areas that were creating the majority of errors. Focus on these areas will undoubtedly help reduce future errors. Other factors that impact the lack of a decrease of errors is the complexity of how new registrations are created- some are generated by interfaces and have no human intervention at all. The takeaway from this study is that there are many more players with a critical role in data integrity than previously expected and that a quality MPI integrity program needs to be an ongoing program that focuses on continuous education, monitoring, and feedback to those players involved.