Chart reviews are critical to answering many clinical research questions. At Stanford, there is a robust system that aids researchers in conducting chart reviews called the Stanford Medicine Research Data Repository or STARR. This article highlights ways to utilize STARR to conduct a chart review.
How to conduct a Research Chart Review
During patient care, we are often faced with difficult situations and challenges that spur us into thinking about larger questions.
I wonder how many patients have this side-effect? What are the likely outcomes of patients like mine?
The best approach when attempting to answer these types of questions is to review the available literature for related data. If insufficient data exists, you may want to add to the body of knowledge that exists around this condition by undertaking a chart review of Stanford patients. The temptation may be to acquire a list of patients and dive into EPIC to find the answer to your query, however accessing EPIC for this purpose requires additional permissions from the Institutional Review Board (IRB).
If you are sure the answer to the question can be found in the medical record (EPIC), the first step is to obtain permission for Chart Review. This involves creating a short application to the Stanford Institutional Review Board (IRB) https://eprotocol.stanford.edu/.
Information about the source of the data is required. If the information comes from EPIC, the preferred search mechanism is the Stanford Medicine Research Data Repository (STARR). This is a clinical database that includes adult (SHC) and children (LPCH) electronic data as far back as 1998. STARR covers the multiple electronic health record systems that both hospitals have used over the years.
More information about STARR can be found at: https://med.stanford.edu/researchit/infrastructure/clinical-data-warehouse.html
The Research Information Center (RIC) approves access to STARR and requires all users to complete a Data Privacy Attestation that is linked to the specific IRB application. There are different attestations depending on whether the data has personal health information (PHI) or is deidentified. As the capabilities of data acquisition from STARR are rapidly evolving, please check the link to RIC. http://med.stanford.edu/ric/resources/som-compliance-processes.html
The STARR database does not list the information in the same format as EPIC as there are no flowsheets. The information that was entered into each EPIC cell is available for searching, however, the data is listed in a format more similar to an excel spreadsheet.
For example: once a patient record is selected, the researcher might want to know the urine output. To find out if the patient had a urinary catheter is a separate query. The color of the urine is a separate query. Or a broad category of “urine” might be used. This is the reason it is so important to have the test question and the search criteria identified before starting the study. It may be helpful to make a worksheet to guide the search. For help with a chart review study, contact the Office of Research Patient Care Services (ORPCS) at: firstname.lastname@example.org
Article By: Mary E. Lough