UMB ICTR

• UMMS Epic Data* from 13 facilities (see Date Range of Data tab below for list):
  • Patient demographics: age, gender, race, ethnicity
  • Encounters: Site of Care, Admit date
  • Diagnoses: billed, recorded at encounter
  • Lab Results:  panels, microbiology
  • Medications: ordered, administered, patient-reported
  • Procedures: surgical, imaging, rehabilitative, therapeutic

• TriNetX Global Network Research Data 

*Currently, UMMS genomics data is not available. UMMS data is patient-level only at this time with very little encounter-level data. Also, imaging data results/comments are not reported in the TriNetX output. 

TriNetX is compliant with the Health Insurance Portability and Accountability Act (HIPAA), the US federal law which protects the privacy and security of healthcare data. TriNetX is certified to the ISO 27001:2013 standard and maintains an Information Security Management System (ISMS) to ensure the protection of the healthcare data it has access to and to meet the requirements of the HIPAA Security Rule.

Any data displayed on the TriNetX Platform in aggregate form, or any patient level data provided in a data set generated by the TriNetX Platform, only contains de-identified data as per the de-identification standard defined in Section §164.514(a) of the HIPAA Privacy Rule. The process by which the data is de-identified is attested to through a formal determination by a qualified expert as defined in Section §164.514(b)(1) of the HIPAA Privacy Rule. This formal determination by a qualified expert, refreshed in December 2020, supersedes the need for TriNetX’s previous waiver from the Western Institutional Review Board (IRB). Please find more information about the Expert Determination here.

The TriNetX network contains data provided by participating Healthcare Organizations (HCOs), each of which represents and warrants that it has all necessary rights, consents, approvals and authority to provide the data to TriNetX under a Business Associate Agreement (BAA), so long as their name remains anonymous as a data source and their data are utilized for research purposes. The data shared through the TriNetX Platform are attenuated to ensure that they do not include sufficient information to facilitate the determination of which HCO contributed which specific information about a patient.

For UMMS Epic data, the date range is [date] to present. UMMS consists of 13 facilities with varying Epic "go-live" dates. 

• UMMS Epic data is updated monthly in TriNetX
• TriNetX Global Network Research Data - receives data from multiple centers with varying data refresh dates. 

Data have been cleaned, curated, and more fully mapped to improve data quality and analysis. However, a few things you need to consider:

• How Current are the Data?
  • UMMS pushes updated data to TriNetX monthly. This is typical for the 70+ HCOs contributing data to TriNetX
• Erroneous Data Output:
  • If data pushed to TriNetX from the HCOs has errors - such as extreme high or low values due to sample mislabeling - the TriNetX data will contain those errors as well. TriNetX does not correct erroneous data.. As they do their quality checks, they will notify the institution of errors they discover. It is up to the institution to make corrections, if possible.  Corrections will not be appear until the next data push to TriNetX. Best practice is to enter valid ranges to avoid pulling erroneous data.
• Non-numeric Lab Values:
  • Most institutions, including UMMS, do not report numeric values for some labs that fall under or exceed a value threshold. Instead, the values are reported as text values, such as < 1.0 or >9.0 (contains less than or greater than signs).  Patients with a text value will not appear in your TriNetX query count because the query can only pull numeric values.  This means you may miss patients that meet your criteria.  If you request an UMMS Epic data pull, discuss how this can be managaged. Identifying the records with text values becomes a manual process.
• Patient Counts:
  • UMMS has multiple hospitals. A patient will be included in the count for each, respective institution's count where they have received care. However, for the overall UMMS sum of patients, a patient will be counted only once. 

Researcher Must First Complete - at a Minimum - Basic TriNetX Training

University of Maryland Medical Center (UMMS 13 facilities) - Patient-level Data. 

1. On the right panel of your TriNetX query portal, select the yellow "Request Dataset" button. This will send your saved query to the UMMS/UMB Research Informatics Core (RIC) team. If you haven't already, proceed with submitting your project to the UMB Institutional Review Board (IRB).  Requests cannot be processed without a UMB IRB determination/approval letter specific to the request. 

2. Once you have IRB approval, submit an ICTR Resource Request. The application link can be found in the "How to Apply for ICTR Resources" section HERE at the bottom of the page. In the application, navigate to the Informatics section and select Access to Clinical Data. Your application will be forwarded to the Research Informatics (RIC) team who will review your request and your saved TriNetX query. The data pull will be exported to the University of Maryland Baltimore (UMB) Secure Research Environment (SRE) Virtual Desktop for your use.

Suggested IRB language: 

We will be using the TriNetX Research Network Database, a web-based tool for research population cohort and feasibility queries that provides access to aggregate data of dozens of academic medical centers and health care organizations. The University of Maryland Medical System (UMMS) is part of this network. 

We will perform a search query within the TriNetX portal, designed in consultation with the TriNetX analysts and/or the UMB TriNetX team, using [specify data elements (ICD codes, lab results, etc.)] to identify a cohort of patients meeting the criteria. This output is de-identified aggregate data only. To obtain record level data that may contain identifiers, a data request is sent to the UMMS Research Informatics Core (RIC) through the TriNetX portal. The dataset will be downloaded as an Excel file in coordination with the UMMS RIC team and honest broker. The data is downloaded to the University of Maryland Baltimore (UMB) Secure Research Environment (SRE) in the Azure Virtual Desktop (AVD) for our access. The SRE AVD is the required storage platform for UMMS electronic medical record data for research and TriNetX Research Network downloads. The data cannot be moved from this environment without IRB approval and the required approvals, such as Data Use Agreements from the UMB/UMMS campus and with assistance from the RIC team. 

TriNetX Global Research Network Patients Limited Dataset

1. Select the 'Browse Network' tab, then select the 'Research' option instead of University of Maryland.

2. Create study (query) in TriNetX. We recommend using Summary Statistics and Design Assistance features to analyze criteria and validate your cohort. 

3. Use the yellow 'Request Data' button on far right. TriNetX will contact you to discuss the purpose of the data request and ask about study funding. TriNetX will ask you to include costs for the data in the grant-related requests. Costs vary from free, for most non-grant related use, to high, depending on use and grant funding. The investigator must negotiate the costs with TriNetX.  Retain the email from TriNetX that indicates whether or not your study will need to pay TriNetX. If you will need to pay TriNetX, they should provide you with a TriNetX Dataset Order Form with negotiated cost and terms.

4. Seek IRB approval to receive the desired data UMB Institutional Review Board (IRB). Confirm that your query matches what was approved by the IRB or submit an IRB modification to update your protocol to indicate use of the TriNetX Research Network as the data source. 

Suggested IRB language:

We will be using the TriNetX Global Research Network Database, a web-based tool for research population cohort and feasibility queries that provides access to clinical data of dozens of academic medical centers and health care organizations. The University of Maryland Medical System (UMMS) is part of this network. 

We will perform a search query within the TriNetX portal, designed in consultation with the TriNetX analysts and/or the UMB TriNetX team, using [specify data elements (ICD codes, lab results, etc.)] to identify a cohort of patients meeting the criteria. The output is deidentified aggregate data only. To obtain a limited dataset, we will request the data from TriNetX who will follow up with the PI of this protocol to discuss the purpose of the data request. If approved, the UMMS/UMB Research Informatics Core (RIC) honest broker will work with TriNetX to move the limited dataset from TriNetX to the UMB Secure Research Environment (SRE) Azure Virtual Desktop (AVD) for our access. The SRE AVD is the required storage platform for TriNetX Research Network data downloads. The data cannot be moved from this environment without IRB approval and other required approvals, such as Data Use Agreements from the UMB/UMMS campus, and with assistance from the RIC team. 

5. Once your IRB protocol is approved, submit a request through the ICTR Resource Request form HERE. Scroll to the Informatics section of the form and select 'Access to Clinical Data".  Include the IRB protocol and approval letter for your protocol in the request, and indicate that you are approved to receive data from the TriNetX Global Research Network. The UMMS/UMB Research Informatics Core (RIC) honest broker will schedule a brief consult with the PI, write a specification document, and review the IRB protocol to ensure that the query aligns with IRB approval.

6. The honest broker will submit a download request to TriNetX on behalf of the PI. TriNetX will create the dataset and make the dataset available to the RIC team/honest broker who will download the data and move to the SRE AVD environment.

Analytics available within the TriNetX environment include:

• Number of patients in a cohort
• Exploration of that cohort’s demographics, co-morbidities, procedures, medications, and labs.
• Clinical phenotype criteria analysis.
• Rate of arrival prediction.

Information about TriNetX Access can be found here in the TriNetX Access Management Manual

Who can access TriNetX

TriNetX accounts have been set up for current UMB faculty. Logging in requires a UMID. We are working towards providing access to non-facutly at UMB as well as exploring processes that will allow non-UMB faculty access.  

Requesting a TriNetX Account