Completing the Records Review Section for Category 4 Exemption Applications
How to complete the records review section of the ARROW application
What is the purpose of the Record Review section?
The Record Review section of the initial review application asks study teams to describe the data source(s) being used for research purposes such as:
- electronic medical records
- paper charts
- clinical or research databases
- publically available datasets
- datasets from previously completed research
When does the Record Review section display?
The Record Review section will display when Review of Records/Data/Images is selected on the Special Procedures page for category 4 exemption applications. Please note that this option should be selected for all category 4 exemption applications that involve obtaining or accessing some kind of data source to collect data for analysis. The only time this should not be checked for category 4 exemptions is when the study team will obtain samples for analysis that are not accompanied by data, or are accompanied by very limited data directly on the sample, e.g. diagnosis and gender of subject. Note: If the sample includes date of collection, the study team must clarify whether this date constitutes a limited identifier under HIPAA, e.g. the date of collection is the date of a surgical procedure during which the sample was collected.
How should study teams answer question 1.1 in the Record Review section?
This response should always be “no” for category 4 exemptions.
How should study teams answer question 1.2 in the Record Review section?
Each data source should be checked according to the following guidance:
|| All data from WISCR has been incorporated into HealthLink; should not be checked
|| Electronicmedical records available in and accessed from HealthLink
| VA Electronic Records System
|| Electronic medical records of VA patients available in and accessed from the VA CPRS
| Clinic or departmental paper records
|| Paper medical chart review
| Departmental database
|| Clinical database, such as a QI or outcomes database for a specific department; could also be a departmental research database
| Other database or source (including external sources)
|| Any data source that does not fit into one of the other categories. This could be an external database (e.g. a national research registry), a dataset from past research studies, or a publically available dataset (e.g census datasets)
How should study teams answer question 1.2.1 in the Record Review section?
Any data source checked above other than HealthLink or the VA Electronic Records System must be identified and briefly explained in this response. If a departmental database and/or paper records is selected in 1.2, the study team is asked to clarify what the database is and/or the clinic/departmental source for paper records. An example of a departmental database is the one maintained by the transplant division within surgery.
How should study teams answer question 1.2.2 in the Record Review section?
If a database is identified as a data source in 1.2, please indicate whether it is a clinical database or one that is used primarily for research purposes. This question only needs to be answered if “Departmental database” and/or “Other database or source” is selected in question 1.2. Of note, departmental databases generally are created primarily for clinical rather than research purposes even if used for research studies.
How should study teams answer question 184.108.40.206 in the Record Review section?
If you checked “no” in 1.2.2, please provide the IRB protocol number for the research database. All research databases must have IRB approval; clinical databases only require registration with the HIPAA Privacy Officer.
How should study teams answer question 1.3 in the Record Review section?
This question is asking how it will be determined which patient or research subject records are eligible for inclusion in the study. For example, records with a specific diagnosis or billing code may be flagged for inclusion. This answer should also describe the data source that will be used to identify potential subject records and describe who will conduct this activity (e.g. UWHC ITS query, study team member, PI, etc.).
How should study teams answer question 1.4 and 1.4.1 in the Record Review section?
If the study team must access medical records in order to collect data for the study, someone with valid access to the records must do this. Please note that in 1.4, “valid clinical access” means that the key personnel who will access medical records have a clinical role that affords them access to the records outside the research context. If this is the case, click on “yes” for 1.4. If key personnel who will access medical records do not have a clinical role, click on “no” to 1.4 and proceed to 1.4.1. In order to access medical records without a clinical role, key personnel must have received authorized access to medical records for research purposes. In that case, click on “yes” for 1.4.1. Instructions for obtaining valid access to records for research purposes are provided along with the question for 1.4.1.
How should study teams answer question 2.1 in the Record Review: Continued section?
There are specific regulations about certain vulnerable populations that must be addressed if the subject population includes them. Please indicate whether your subject population includes children/minors, pregnant women/fetuses, or prisoners. If the study team will not know if members of these groups are included, please choose “none of the above.” Of note, federally-funded studies that target prisoners cannot qualify for exemption.
How should study teams answer question 2.3 in the Record Review: Continued section?
The date range of data collection should indicate the dates as of which the data are available for collection, not necessarily the dates during which the data will be collected. For example, a date range may indicate data from patients who received lung transplants and follow-up care between 1/1/2000 and 12/31/2014, but the data could be collected at any time after that. Full dates (mm/dd/yyyy) should be provided. As noted in the instructions for this item, applications for category 4 exemption must show a date range that does not exceed the date of IRB submission. Stated another way, if the end of the date range for data collection is later than the date of IRB submission, the application cannot qualify for category 4 exemption and must be changed to a different application type.
How should study teams address question 2.4 in the Record Review: Continued section?
The study team must upload one or more documents that state what data elements will be collected from the data source(s) listed in 1.2-220.127.116.11. These documents may take the form of a Word document listing individual data elements, a data collection form that will be filled out from the data source(s), or a spreadsheet with column and/or row headings that indicate specific data elements. Please note that, for category 4 exemptions, direct identifiers may not be included in this document or linked to the document via a study code. If the study team intends to collect limited identifiers, they must be included in this document and be documented in the HIPAA section.
How should study teams address question 2.5 in the Record Review: Continued section?
The study team should indicate the number of records that will be accessed to collect the data, such as how many records will UWHC IS pull for the study team OR how many subjects will be enrolled in the study. The number of records accessed may be more than the number of records ultimately used for the research because the study team may need to review records to determine if the desired data are in the record. If the study team will receive data from outside UW-Madison, the response to 2.5 should be the number of records in the data set.
How should study teams address question 2.6 in the Record Review: Continued section?
The study team should explain why the number provided in 2.5 corresponds to their study, such as it is the number of records needed for a certain analysis or it is the number of participants being enrolled in the study. A common response to this question could be that the number of records to be used for the study represents the number of potential eligible patients during the timeframe that was chosen for review.
How should study teams answer question 2.7 and 2.7.1 in the Record Review: Continued section?
These should be checked “yes” if the study team will retain data from this study for future research purposes. Please note that there are similar questions in the Privacy/Confidentiality section (questions 1.4 and 1.4.1), and the answers in both sections must match.