Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Mar 15;29(4):660-670.
doi: 10.1093/jamia/ocab269.

Enhancing PCORnet Clinical Research Network data completeness by integrating multistate insurance claims with electronic health records in a cloud environment aligned with CMS security and privacy requirements

Lemuel R Waitman  1 , Xing Song  1 , Dammika Lakmal Walpitage  2 , Daniel C Connolly  3 , Lav P Patel  3 , Mei Liu  3 , Mary C Schroeder  4 , Jeffrey J VanWormer  5 , Abu Saleh Mosa  1 , Ernest T Anye  6 , Ann M Davis  7   8
Affiliations

Enhancing PCORnet Clinical Research Network data completeness by integrating multistate insurance claims with electronic health records in a cloud environment aligned with CMS security and privacy requirements

Lemuel R Waitman et al. J Am Med Inform Assoc. .

Abstract

Objective: The Greater Plains Collaborative (GPC) and other PCORnet Clinical Data Research Networks capture healthcare utilization within their health systems. Here, we describe a reusable environment (GPC Reusable Observable Unified Study Environment [GROUSE]) that integrates hospital and electronic health records (EHRs) data with state-wide Medicare and Medicaid claims and assess how claims and clinical data complement each other to identify obesity and related comorbidities in a patient sample.

Materials and methods: EHR, billing, and tumor registry data from 7 healthcare systems were integrated with Center for Medicare (2011-2016) and Medicaid (2011-2012) services insurance claims to create deidentified databases in Informatics for Integrating Biology & the Bedside and PCORnet Common Data Model formats. We describe technical details of how this federally compliant, cloud-based data environment was built. As a use case, trends in obesity rates for different age groups are reported, along with the relative contribution of claims and EHR data-to-data completeness and detecting common comorbidities.

Results: GROUSE contained 73 billion observations from 24 million unique patients (12.9 million Medicare; 13.9 million Medicaid; 6.6 million GPC patients) with 1 674 134 patients crosswalked and 983 450 patients with body mass index (BMI) linked to claims. Diagnosis codes from EHR and claims sources underreport obesity by 2.56 times compared with body mass index measures. However, common comorbidities such as diabetes and sleep apnea diagnoses were more often available from claims diagnoses codes (1.6 and 1.4 times, respectively).

Conclusion: GROUSE provides a unified EHR-claims environment to address health system and federal privacy concerns, which enables investigators to generalize analyses across health systems integrated with multistate insurance claims.

Keywords: Amazon Web Services private cloud; Centers for Medicare and Medicaid Services; PCORnet; Patient-Centered Outcomes Research Institute; cloud computing; electronic health records; obesity.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Privacy-Preserving Data Linkage between Center for Medicare and Medicaid Services (CMS) claims and EHR data. (1) Each participating Greater Plains Collaborative Greater Plains Collaborative (GPC) site uses its EHR data, to define patients for linkage to CMS data. (2) GPC sites generate a unique hashed ID for each patient. (3) Each GPC site sends “finder files” combining multiple primary and secondary identifiers and hashed IDs to NewWave-GDIT/Chronic Condition Data Warehouse (CCW) following a well-established encryption procedure. (4) NewWave-GDIT/CCW uses the set of identifiers from each of the GPC sites to generate a cross walk file that maps between the hashed IDs and the GPC Reusable Observable Unified Study Environment-specific BENE_ID. (5) NewWave-GDIT/CCW creates an extract of CMS data specific to the states encompassing the GPC sites. The resulting files are sent by NewWave-GDIT/CCW to the GPC CC via encrypted external media (6) GPC Coordinating Center (CC) receives Limited Data Sets containing EHR data from each of the GPC sites along with the hashed IDs sent to NewWave-GDIT/CCW. (7) GPC CC will then use the hashed IDs to link the patient records received from NewWave-GDIT/CCW with the Limited Data Sets received from each site. (8) Each merged data set is deidentified by GPC CC via dynamic views and made available to the collaborating investigators that are listed within the protocol. (9) No identifiers are retained by GPC CC after creation of the deidentified data set. The GPC site data may be refreshed over time upon agreement across sites. CMS data may be refreshed when new data becomes available. For this data refresh, individual sites will either use the same hashed IDs previously used for its patients so that they are linked automatically over time, or if a site chooses to use a different hashed ID for the refresh, then they will provide GPC CC a mapping between the previous hashed ID and the new hashed ID.
Figure 2.
Figure 2.
Multi-Stakeholder Data Access Governance Model. A data access request starts from researcher submitting an Access Request Intake Form and trigger “Study Scope Review,” which is performed by designated stakeholders who determine: (1) whether the study can be covered by the scope of Greater Plains Collaborative Reusable Observable Unified Study Environment institutional review board (IRB) or a new reuse IRB is needed; (2) the appropriate group/role of the requester. Upon the study scope approval, researchers will submit a GPC Data Request Oversight Committee request to approval from participating GPC sites. Then, a compliance review with requirements for CITI Human Subject Research and NIH security and privacy awareness training are checked and collected by administrators, as well as signing the data use agreement to affirm agreement to GPC terms and conditions. Finally, an AWS research user account will be provisioned with self-serviced tools and applications enabled. Each step is to support for annual review and periodic auditing.
Figure 3.
Figure 3.
Infrastructure as a Service (IaaS) versus Platform as a Service (PaaS) versus Software as a Service (SaaS) cloud service models. The modules highlighted in yellow are consumer’s responsibility, while the white modules are cloud provider’s responsibility. SaaS enables the customers to use the cloud provider’s applications/software that are running on a provider’s infrastructure, whereas PaaS enables consumers to create or acquire applications/software and tools and to deploy them on the cloud provider’s infrastructure. IaaS enables a consumer to provision processing, storage, networks and other fundamental computing resources.
Figure 4.
Figure 4.
Data that flow from multiple sources, including (1) NewWave-GDIT physical media and (2) other Greater Plains Collaborative sites will be load into secured S3 bucket via Secure File Transfer Protocol or using AWS S3 management console (TLS 1.2). (3) Raw files are externally staged in S3 buckets and then loaded into Snowflake data warehouse via (4) the Snowpipe automated pipeline (a Snowflake functionality). (5) Data in source Center for Medicare and Medicaid Services (CMS) research identifiable file or site Common Data Model (CDM) schema are first extracted as they are in 1 database. (6) CMS data will then be transformed into PCORnet CDM and integrated with electronic health record data using the finder file provided by CMS. (7) The integrated CDM will be deidentified using the built-in dynamic view functionality provided by Snowflake. (8) Both the limited and deidentified view can be accessed via ODBC or JDBC connector with researchers’ service workbench workspaces. (9) Service workbench provides templated and reusable workspaces (AWS EC2 instances) with various computing power, operating systems and prepackaged software that can satisfy most of the research needs. (10) Approved researchers can deploy the self-serviced applications to perform either advanced analysis using the service workbench or simply discover study cohort using an integrated Informatics for Integrating Biology & the Bedside query tool. Various underlying Amazon Web Services are marked at each step described above as well as at the bottom of the figure.
Figure 5.
Figure 5.
Electronic health record and claims based obesity rates for different age groups. (A1) and (A2) plot Greater Plains Collaborative Weight Cohort and Crosswalk Weight Cohort sizes stratified by 3 age groups (2–19, 20–64, 65, and older) by calendar year. (B1) and (B2) plot obesity rates where the solid lines correspond to body mass index-defined obesity, whereas the dotted line correspond Code-defined obesity.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Fleurence RL, Curtis LH, Califf RM, Platt R, Selby JV, Brown JS.. Launching PCORnet, a national patient-centered clinical research network. J Am Med Inform Assoc 2014; 21 (4): 578–82. - PMC - PubMed
    1. The Greater Plains Collaborative (GPC). Secondary The Greater Plains Collaborative. 2015. http://www.gpcnetwork.org/ Accessed July 1, 2021.
    1. Waitman LR, Aaronson LS, Nadkarni PM, Connolly DW, Campbell JR.. The Greater Plains Collaborative: a PCORnet Clinical Research Data Network. J Am Med Inform Assoc 2014; 21 (4): 637–41. - PMC - PubMed
    1. NIH. National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Science Award (CTSA) Program. Secondary National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Science Award (CTSA) Program. 2013. https://ncats.nih.gov/ctsa.
    1. PCORI. Patient-Centered Outcomes Research Institute Cooperative Agreement Funding Announcement: Improving Infrastructure for Conducting Patient-Centered Outcomes Research. The National Patient-Centered Clinical Research Network: Clinical Data Research Networks (CDRN)—Phase One. Secondary Patient-Centered Outcomes Research Institute Cooperative Agreement Funding Announcement: Improving Infrastructure for Conducting Patient-Centered Outcomes Research. The National Patient-Centered Clinical Research Network: Clinical Data Research Networks (CDRN)—Phase One. 2013. https://www.pcori.org/sites/default/files/PCORI-PFA-CDRN-071713.pdf.

Publication types