REDCap - Clinical Data Pull from Epic

What is 'Clinical Data Pull' (CDP)?

Clinical Data Pull is a special feature for importing data into REDCap from an EHR (electronic health record system), such as Epic, Cerner, etc. It provides an adjudication process whereby REDCap users can approve all incoming data from the EHR before it is officially saved in their REDCap project. Clinical Data Pull can only be enabled by a REDCap Administrator, so you should contact them if you wish to utilize Clinical Data Pull for this project.

How 'Clinical Data Pull' works

Clinical Data Pull has the ability to fetch data from the EHR system both manually in real time and automatically at a regular interval. From the EHR interface, Clinical Data Pull can create new records in a CDP-enabled REDCap project. Additionally, if a user knows the patient identifier (e.g. medical record number), then they could optionally enter the MRN for a record in a CDP-enabled REDCap project, after which it will then go and immediately retrieve the patient data from the EHR in real time.

Enabling the feature

After a REDCap administrator has enabled the CDP for your REDCap project, the adminstrator can then give any user in the project access to the CDP (assuming the user has been authorized to access source system data), in which the user can be given user privileges to perform the mapping/setup process and/or to adjudicate/import data from the source system. Only an administrator can grant individuals access to begin utilizing the CDP. After being granted access to the CDP, the user will then see the CDP step on the Project Setup page where they can navigate to the CDP mapping/setup page, which allows the user to map fields already created in their REDCap project to fields from the external source system. This mapping allows REDCap to know how where to put the data when importing it from the external system.

The field mapping process

A user on the mapping/setup page can choose any available fields from the external source system to map to one of their REDCap project fields. (Note: This assumes that the user has already created their data collection instruments and fields in their project.) There are two types of fields that may be mapped: 1) one-time data fields, and 2) temporal data fields. One-time data fields are those where data will only be stored one time in the source system, such as demography data. Conversely, temporal data fields are those where data may be collected many times over a period of time (e.g., labs, vitals, any kind of longitudinal data). When mapping temporal fields, you must also specify a REDCap date/datetime field that will be used for determining the window of time in which it should look when fetching data from the source system. It will determine the window of time in conjunction with the day offset value defined on the mapping page. For example, if the value of the mapped REDCap date/datetime field is 2001-11-29 and the day offset is 1 day, it will query the source system and return only data saved for the field from 2001-11-28 until 2001-11-30. Any source data outside of that range will be ignored during the adjudication process. It is important to note that the source data for a temporal field will NOT be fetched from the external source system UNLESS its associated REDCap date/datetime field has a value. Once a value is entered for the associated date/datetime field, source data will then be fetched for that temporal field.

There are a few demography fields (race, sex, and ethnicity) that, if used in a project, are required to have their choices coded a very specific way. Listed below are the codings that should be used for these fields.