Biomedical Informatics and Data Access

Access to rich clinical data environment

• See Data Sources section below

Design of data repositories

• Enterprise-wide clinical data repositories (CDRs)
• Disease or service-specific clinical data marts and registries

Dataset preparation

• Data transformation
• Summarization
• Codeset Mapping (ICD9/ICD10, NDC, RxNorm, etc.)

Data de-identification using validated tools

• Fully de-identified datasets
• Limited datasets

Natural Language Processing

• NLP-PIER information extraction platform for clinical notes

Healthcare Application Development 

• User Interface Design
• Mobile App development

Data Analytics

• Tier 1 Analytics: Dashboards and Self-Service Business Intelligence
• Tier 2 Analytics: Collaboration with expert data scientists and faculties

Secure, Compliant and robust Data Infrastructure

• PHI-compliant storage
• Secure data collection, integration & management
• User access management
• Validated data enrichment tools & algorithms

Informatics Resources

• Informatics Consultants
• Data Analysts
• Database Engineers
• Developers

Assistance with submissions and documentation

• Grant proposals
• IRB submissions (IRB ancillary reviews)
• Manuscripts

Academic Health Center Information Exchange (AHC-IE)

• Clinical Data Repository (CDR): EHR records for more than 4.5 million Fairview and University of Minnesota Physicians' patients
• Genomics data:  Tumor (Somatic) Data and Germline Data
• Fairview Tumor Registry: Minnesota Cancer Surveillance System (NAACCR) 
• CVIS: Discrete clinical data from Fairview Cath Labs
• Optima/PaceArt: Cardiac medical devices interrogation data
• Xcelera/Echo: Discrete data from the cardiology image management system
• UMP Flowcast: Professional billing data (e.g., diagnoses, procedures, departments) for UMP providers

COVID-19 Registry

De-identified electronic medical records of COVID-19 patients as well as those with related conditions and symptoms. COVID-19 registry inclusion criteria.

Masonic Cancer Center Data Registries and Repositories

• Bone Marrow Transplant Research database (BMT): Captures post-transplant outcomes for BMT patients
• Laboratory information management system (LIMS): Subjects’ samples data (number of aliquots, analyte, etc.)
• FreezerWorks: Research samples inventory (freezer name, shelf, rows, box etc.)

BigMouth Network

Oral health records of more than 4.3 million consenting patients from institutions around the country.

OnCore^® enterprise CTMS

Clinical Trial Data

Dental Records

AxiUm dental records from the University of Minnesota School of Dentistry

Minnesota Death Index

Minnesota Department of Health death records issued from 2011 to present for deceased individuals who were born in Minnesota, have died in Minnesota, or have ever had a permanent address in the state.

IBM MarketScan^®

The University of Minnesota has obtained an institutional license to access IBM MarketScan® Commercial Claims and Encounters and Medicare Supplemental databases for years 2013 to 2017. The current license enables access to these data for approved researchers until May 2022.

MarketScan data provide claims (medical, prescription drug purchases, physician encounters, and hospital stays for the population as well as insurance enrollment) for a large population of commercially insured patients and Medicare beneficiaries with supplemental health coverage.

For more details about access to this data, fees, and restrictions, please refer to the Data Use Agreement Document.

Spatial Data: 

• Patient records are geocoded and enhanced with latitude / longitude / block data elements
• Census data at the block group level from the 2010 Census. 
• The Clinical EMR data is geocoded and enhanced latitude / longitude / block data elements

UMLS Data

Standard Terminology and code sets including ICD-9/10, LOINC, NDF-RT, RxNorm, CPT, HCPCS

Other

Datasets obtained by Researchers (e.g., OptumLabs, AHRQ/HCUP, CMS, VA, etc.) 

The CTSI Informatics core participates in several data sharing networks, the goal of which is to facilitate and streamline multi-site clinical research.

Submit a BPIC consultation request to learn how data sharing networks can help your research.

Accrual to Clinical Trials (ACT)

• Supporting cohort identification across 21 sites (approximately 40M patients)
• Contact us to query the NCATS-ACT Network

MARCH

• Offers investigators a streamlined infrastructure for undertaking studies that are better served at multiple sites. 

TriNetX

• Provides anonymized cohort search over UMN data to major Pharmas so that new trials are efficiently initiated.

i2b2

• Self-service cohort discovery tool primarily used for simple feasibility queries
• Clinical Data loaded from Data Environment
• Sign up on the site with your x500 | Login

SHRINE

• Self-service cohort discovery tool used to query sites participating in the ACT Network
• Contact [email protected] for access

TriNetX

• Self-service cohort discovery tool primarily used for simple feasibility queries
• Clinical Data loaded from Data Environment (see Data Sources section above)
• Contact [email protected] for access
• Pharma has access to run feasibility queries on our data environment to identify potential clinical trial collaboration opportunities
• Queries run faster than i2b2
• Clinical trial opportunities flow through TriNetX from Pharmaceutical companies to University of Minnesota researchers

OptumLabs Natural History of Disease (NHD)

• Used to query OptumLabs Data Warehouse on behalf of UMN investigators who are interested in partnering with OptumLabs on a specific research proposal
   

Natural Language Processing – Patient Information Extraction for Research (NLP-Pier)

• Information extraction platform for clinical notes
• Can only be accessed within the data shelter
• Filter and search documents using HL7-LOINC Document Ontology features, keyword searches, UMLS concept searches, Boolean queries, and by UMLS semantic type

Data Extraction Policies

• AHC-IE Extraction of Identifiable Data
• AHC-IE Extraction of De-identifiable Data

Contact [email protected] to set up a training session or to learn more about our services, data and tools.

• Initial Consultation: Scope and Feasibility – Free of charge
• Grant proposal preparation – Free of charge
• Other services: Directly supported by the appropriate funding sources
• Costs for services are directly supported by the appropriate funding sources:
  • Established agreements/collaboration models
  • % effort of personnel
  • Grants
  • Internal hourly rate: $120/hr; $95 subsidized rate for Cancer Center members
• For non-U of M clients, rates are negotiated separately

The clinical data repository supports research in the following ways:

• Study feasibility
• Recruitment
• Retrospective and observational studies
• Operational or quality improvement projects (non-research)
• Data feeds
• Education and fundraising

User authorization

Each partner manages its own process for authorizing its users, but the requirements generally include that you are an employee in good standing, have completed HIPAA training, and have signed an attestation form in which you agree to terms for use of the data (e.g. you will only use the data for the purposes specified in the original request form). If you are affiliated with more than one of the partners, your authorization will be via the affiliation most relevant to your data request. It is not your responsibility to ensure your authorization before submitting a request; we will verify your authorization once your request has been submitted.

External collaborators

External collaborators cannot request data themselves, but qualified University-affiliated individuals or departments can be granted access in some cases. To do so, a University member must secure a sponsored guest x500 account for their external collaborator. This is a free service that can be requested via the Office of Information Technology. 

More information for students and employees of UMN

Addressing the following requirements as soon as possible will allow us to deliver data to you more quickly:

HIPAA training

• The University's HIPAA Privacy and Data Security Training program consists of four separate online courses that are part of the "Public Jobs: Private Data" training program.
• If you've completed the training, you do not need to re-take it. A record of all training completed while at the University is available online through the Training History link at the MyU website. You also can contact the Health Information Privacy and Security Office at [email protected].

Attestation form

• You must provide attestation to indicate that you comply with the terms of AHC Information Exchange (IE) data use. To attest, complete this short online form, which will require that you sign in with your university ID (x500). Your attestation will be valid for one year. 

Research

IRB review and approval is required for research falling under the federal definition of human subjects research. IRB approval is required for data requests that require identifiable private information and involve a systematic investigation designed to contribute to generalizable knowledge. Visit the U of M's IRB website for more guidance (e.g., Does My Research Need IRB Review?) and up-to-date forms, including the Change in Protocol form and Change in Personnel forms.

Activities preparatory to research

Data requests to determine study feasibility or plans for participant recruitment do NOT require IRB review. Participant recruitment itself, however, does need IRB approval. For more information on the disclosure of PHI for activities preparatory to research, see the related Administrative Policy. 

Non-research activities

If you need data in support of clinical operations or quality improvements, you do not need IRB approval however, approval by a Data Steward is required. 

Phase 1: Initial Consultation and Feasibility Analysis

Free consultation during which a BPIC Informatics Consultant meets with the researcher(s) to discuss the proposed project and assess feasibility. This rapid appraisal reduces time delays, expands the scope of technical approaches that can benefit complex research, and reduces errors in the early stages of project design.

Phase 2: Detailed Planning

Detailed planning is a short and quick phase that outlines a technical plan, budget, timeline in a refined consultation report. 

Phase 3: Execution/Implementation

Project execution begins after BPIC and the researcher have an agreement on the scope of work. This is when the informatics work is done, including identifying exact data elements needed, informatics methods design, dataset creation, etc.

Accessing your data

Once your data request has been fulfilled, you'll be able to view and analyze your datasets through a secure desktop environment called the "Data Shelter." Instructions for logging into the Data Shelter. This work environment protects private patient data, provides data backup, includes free, powerful analysis tools, and enables collaboration with colleagues. 

Moving files in and out of the Data Shelter

To import and export files to and from the Data Shelter use our file transfer application. You can import documents from your personal computer using the Data Shelter File Transfer site .  To retrieve uploaded documents or export documents from the shelter, you will access the same website from a browser within the shelter. Please note that all exports will be reviewed by an analyst and approved if they meet our data extraction policies.

Accessing Epic within the Data Shelter

For security purposes, web traffic is disabled within the Data Shelter. However, there are approved sites that you can access with the Firefox browser. For example, you can access the Fairview Secure Gateway, which will allow you to access Epic and other Fairview systems from within the shelter. If you have a site you would like added, please contact us at [email protected]. 

Analyzing data

Any analysis you perform on your secure datasets must be done using the software provided in the shelter. The following software is available for use, free of charge, within the Data Shelter:

• Epi Info
• JMP Pro
• MatLab
• Microsoft Excel
• Microsoft Access
• MS SQL Server
• SAS
• SQL Developer
• Stata
• R, R Studio
• Weka
• NLP-PIER: An Information Extraction (IE) platform that provides direct access to patient data stored in free text of clinical notes. The underlying framework of PIER uses Elasticsearch technology and features the University of Minnesota Clinical NLP/IE program's open source Natural Language Processing (NLP) application, BioMedICUS (BioMedical Information Collection and Understanding System). 

If the tool you need is absent, contact the Informatics Consulting Service at [email protected] and we will do our best to accommodate your need. 

If you would like to send recruitment letters to the patients you have identified with the assistance of  BPIC, we will work with you and Fairview Research Administration (FRA) to coordinate such a mailing.

• When you submit your data request to BPIC, let us know that you plan to send out recruitment letters via Fairview Research Administration.
• Fairview will send out recruitment letters on your behalf provided that your study will occur at Fairview or the University of Minnesota and that you have UMN IRB approval for your study. Read the Fairview Research Recruitment Mailing Guidelines.
• Contact FRA at [email protected] as early as possible to alert them to your plans for a recruitment mailing.

In addition to providing data, experts from BPIC also can help you work with that data. Contact us at [email protected] if you want to:

• Get help managing the data you received through our data request process.
• Get support for the informatics aspects of your grant applications.
• Learn about informatics experts beyond CTSI.

Consultations are available at no charge to researchers at the University of Minnesota, Fairview, and University of Minnesota Physicians. Request a consultation via our website or by submitting a request form directly.

General questions and assistance?

Please contact [email protected]

Forgot your password?

Password assistance for University of Minnesota students and staff, visit [email protected]

x500:
Use your University of Minnesota x500 username and password to access: i2b2, SHRINE, Duo, Box and vpn
AD Account:
Access the Data Shelter via remote desktop using your AD username and password. Instructions for logging into the Data Shelter can be found here.