Characterization of bariatric surgery and outcomes using administrative claims data in the research network of a nationwide commercial health plan.

BACKGROUND: The supplementation of electronic health records data with administrative claims data may be used to capture outcome events more comprehensively in longitudinal observational studies. This study investigated the utility of administrative claims data to identify outcomes across health systems using a comparative effectiveness study of different types of bariatric surgery as a model. METHODS: This observational cohort study identified patients who had bariatric surgery between 2007 and 2015 within the HealthCore Anthem Research Network (HCARN) database in the National Patient-Centered Clinical Research Network (PCORnet) common data model. Patients whose procedures were performed in a member facility affiliated with PCORnet Clinical Research Networks (CRNs) were selected. The outcomes included a 30-day composite adverse event (including venous thromboembolism, percutaneous/operative intervention, failure to discharge and death), and all-cause hospitalization, abdominal operation or intervention, and in-hospital death up to 5 years after the procedure. Outcomes were classified as occurring within or outside PCORnet CRN health systems using facility identifiers. RESULTS: We identified 4899 patients who had bariatric surgery in one of the PCORnet CRN health systems. For 30-day composite adverse event, the inclusion of HCARN multi-site claims data marginally increased the incidence rate based only on HCARN single-site claims data for PCORnet CRNs from 3.9 to 4.2%. During the 5-year follow-up period, 56.8% of all-cause hospitalizations, 31.2% abdominal operations or interventions, and 32.3% of in-hospital deaths occurred outside PCORnet CRNs. Incidence rates (events per 100 patient-years) were significantly lower when based on claims from a single PCORnet CRN only compared to using claims from all health systems in the HCARN: all-cause hospitalization, 11.0 (95% Confidence Internal [CI]: 10.4, 11.6) to 25.3 (95% CI: 24.4, 26.3); abdominal operations or interventions, 4.2 (95% CI: 3.9, 4.6) to 6.1 (95% CI: 5.7, 6.6); in-hospital death, 0.2 (95% CI: 0.11, 0.27) to 0.3 (95% CI: 0.19, 0.38). CONCLUSIONS: Short-term inclusion of multi-site claims data only marginally increased the incidence rate computed from single-site claims data alone. Longer-term follow up captured a notable number of events outside of PCORnet CRNs. The findings suggest that supplementing claims data improves the outcome ascertainment in longitudinal observational comparative effectiveness studies.

Administrative claims data to support pragmatic clinical trial outcome ascertainment on cardiovascular health.

BACKGROUND/AIMS: Health plan administrative claims data present a cost-effective complement to traditional trial-specific ascertainment of clinical events typically conducted through patient report or a single health system electronic health record. We aim to demonstrate the value of health plan claims data in improving the capture of endpoints in longitudinal pragmatic clinical trials. METHODS: This retrospective cohort study paralleled the design of the ADAPTABLE (Aspirin Dosing: A Patient-centric Trial Assessing Benefits and Long-Term Effectiveness) trial designed to compare the effectiveness of two doses of aspirin. We applied the ADAPTABLE identification query in claims data from Anthem, an American health insurance company, and identified health plan members who met the ADAPTABLE trial criteria. Among the ADAPTABLE eligible members, we selected overlapping members with PCORnet Clinical Data Research Networks in the 2 years prior to the index date (1 April 2014). PCORnet Clinical Data Research Networks consist of network partners (or healthcare systems) that store their electronic health record data in the same format to support multi-institutional research. ADAPTABLE outcome events-cardiovascular hospitalizations including admissions for myocardial infarction, stroke, or cardiac procedures; hospitalizations for major bleeding; and in-hospital deaths-were evaluated for a 2-year follow-up period. Events were classified as within or outside PCORnet Clinical Data Research Networks using facility identifiers affiliated with each hospital stay. Patient characteristics were examined with descriptive statistics, and incidence rates were reported for available Clinical Data Research Networks and claims data. RESULTS: Among 884,311 ADAPTABLE eligible health plan members, 11,101 patients overlapped with PCORnet Clinical Data Research Networks. Average age was 70 years, 71% were male, and average follow-up was 20.7 months. Patients had 1521 cardiovascular hospitalizations (571 (37.5%) occurred outside PCORnet Clinical Data Research Networks), 710 for major bleeding (296 (41.7%) outside PCORnet Clinical Data Research Networks), and 196 in-hospital deaths (67 (34.2%) outside PCORnet Clinical Data Research Networks). Incidence rates (events per1000 patient-months) differed between available network partners and claims data: cardiovascular hospitalizations, 4.1 (95% confidence interval: 3.9, 4.4) versus 6.6 (95% confidence interval: 6.3, 7.0), major bleeding, 1.8 (95% confidence interval: 1.6, 2.0) versus 3.1 (95% confidence interval: 2.9, 3.3), and in-hospital death, 0.56 (95% confidence interval: 0.47, 0.67) versus 0.85 (95% confidence interval: 0.74, 0.98), respectively. CONCLUSION: This study demonstrated the value of supplementing longitudinal site-based clinical studies with administrative claims data. Our results suggest that claims data together with network partner electronic health record data constitute an effective vehicle to capture patient outcomes since >30% of patients have non-fatal and fatal events outside of enrolling sites.

The Community and Patient Partnered Research Network (CPPRN): Application of Patient-Centered Outcomes Research to Promote Behavioral Health Equity.

Objective: We describe the rationale, development, and progress on the Community and Patient Partnered Research Network (CPPRN). The CPPRN builds on more than a decade of partnered work and is designed to promote health equity by developing partnered research on behavioral health and social risk factors in Los Angeles and New Orleans. Setting: A community-academic partnership across Los Angeles County and New Orleans. Methods: Review of rationale, history, structure, activities and progress in applying community partnered participatory research (CPPR) to CPPRN. Findings: Patient and community stakeholders participated in all phases of development, including local and national activities. Key developments include partnered planning efforts, progress on aggregating a large, de-identified dataset across county agencies, and development of an information technology-supported screening approach for behavioral and social determinants in health care, social, and community-based settings. Conclusion: The CPPRN represents a promising approach for research data networks, balancing the potential benefit of information technology and data analytic approaches while addressing potential risks and priorities of relevant stakeholders.

Towards a privacy preserving cohort discovery framework for clinical research networks.

BACKGROUND: The last few years have witnessed an increasing number of clinical research networks (CRNs) focused on building large collections of data from electronic health records (EHRs), claims, and patient-reported outcomes (PROs). Many of these CRNs provide a service for the discovery of research cohorts with various health conditions, which is especially useful for rare diseases. Supporting patient privacy can enhance the scalability and efficiency of such processes; however, current practice mainly relies on policy, such as guidelines defined in the Health Insurance Portability and Accountability Act (HIPAA), which are insufficient for CRNs (e.g., HIPAA does not require encryption of data - which can mitigate insider threats). By combining policy with privacy enhancing technologies we can enhance the trustworthiness of CRNs. The goal of this research is to determine if searchable encryption can instill privacy in CRNs without sacrificing their usability. METHODS: We developed a technique, implemented in working software to enable privacy-preserving cohort discovery (PPCD) services in large distributed CRNs based on elliptic curve cryptography (ECC). This technique also incorporates a block indexing strategy to improve the performance (in terms of computational running time) of PPCD. We evaluated the PPCD service with three real cohort definitions: (1) elderly cervical cancer patients who underwent radical hysterectomy, (2) oropharyngeal and tongue cancer patients who underwent robotic transoral surgery, and (3) female breast cancer patients who underwent mastectomy) with varied query complexity. These definitions were tested in an encrypted database of 7.1 million records derived from the publically available Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS). We assessed the performance of the PPCD service in terms of (1) accuracy in cohort discovery, (2) computational running time, and (3) privacy afforded to the underlying records during PPCD. RESULTS: The empirical results indicate that the proposed PPCD can execute cohort discovery queries in a reasonable amount of time, with query runtime in the range of 165-262s for the 3 use cases, with zero compromise in accuracy. We further show that the search performance is practical because it supports a highly parallelized design for secure evaluation over encrypted records. Additionally, our security analysis shows that the proposed construction is resilient to standard adversaries. CONCLUSIONS: PPCD services can be designed for clinical research networks. The security construction presented in this work specifically achieves high privacy guarantees by preventing both threats originating from within and beyond the network.

Mis-mappings between a producer's quantitative test codes and LOINC codes and an algorithm for correcting them.

OBJECTIVES: To access the accuracy of the Logical Observation Identifiers Names and Codes (LOINC) mapping to local laboratory test codes that is crucial to data integration across time and healthcare systems. MATERIALS AND METHODS: We used software tools and manual reviews to estimate the rate of LOINC mapping errors among 179 million mapped test results from 2 DataMarts in PCORnet. We separately reported unweighted and weighted mapping error rates, overall and by parts of the LOINC term. RESULTS: Of included 179 537 986 mapped results for 3029 quantitative tests, 95.4% were mapped correctly implying an 4.6% mapping error rate. Error rates were less than 5% for the more common tests with at least 100 000 mapped test results. Mapping errors varied across different LOINC classes. Error rates in chemistry and hematology classes, which together accounted for 92.0% of the mapped test results, were 0.4% and 7.5%, respectively. About 50% of mapping errors were due to errors in the property part of the LOINC name. DISCUSSIONS: Mapping errors could be detected automatically through inconsistencies in (1) qualifiers of the analyte, (2) specimen type, (3) property, and (4) method. Among quantitative test results, which are the large majority of reported tests, application of automatic error detection and correction algorithm could reduce the mapping errors further. CONCLUSIONS: Overall, the mapping error rate within the PCORnet data was 4.6%. This is nontrivial but less than other published error rates of 20%-40%. Such error rate decreased substantially to 0.1% after the application of automatic detection and correction algorithm.

Electronic health record data quality variability across a multistate clinical research network.

Background: Electronic health record (EHR) data have many quality problems that may affect the outcome of research results and decision support systems. Many methods have been used to evaluate EHR data quality. However, there has yet to be a consensus on the best practice. We used a rule-based approach to assess the variability of EHR data quality across multiple healthcare systems. Methods: To quantify data quality concerns across healthcare systems in a PCORnet Clinical Research Network, we used a previously tested rule-based framework tailored to the PCORnet Common Data Model to perform data quality assessment at 13 clinical sites across eight states. Results were compared with the current PCORnet data curation process to explore the differences between both methods. Additional analyses of testosterone therapy prescribing were used to explore clinical care variability and quality. Results: The framework detected discrepancies across sites, revealing evident data quality variability between sites. The detailed requirements encoded the rules captured additional data errors with a specificity that aids in remediation of technical errors compared to the current PCORnet data curation process. Other rules designed to detect logical and clinical inconsistencies may also support clinical care variability and quality programs. Conclusion: Rule-based EHR data quality methods quantify significant discrepancies across all sites. Medication and laboratory sources are causes of data errors.

Angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and COVID-19-related outcomes: A patient-level analysis of the PCORnet blood pressure control lab.

SARS-CoV-2 accesses host cells via angiotensin-converting enzyme-2, which is also affected by commonly used angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs), raising concerns that ACEI or ARB exposure may portend differential COVID-19 outcomes. In parallel cohort studies of outpatient and inpatient COVID-19-diagnosed adults with hypertension, we assessed associations between antihypertensive exposure (ACEI/ARB vs. non-ACEI/ARB antihypertensives, as well as between ACEI- vs. ARB) at the time of COVID-19 diagnosis, using electronic health record data from PCORnet health systems. The primary outcomes were all-cause hospitalization or death (outpatient cohort) or all-cause death (inpatient), analyzed via Cox regression weighted by inverse probability of treatment weights. From February 2020 through December 9, 2020, 11,246 patients (3477 person-years) and 2200 patients (777 person-years) were included from 17 health systems in outpatient and inpatient cohorts, respectively. There were 1015 all-cause hospitalization or deaths in the outpatient cohort (incidence, 29.2 events per 100 person-years), with no significant difference by ACEI/ARB use (adjusted HR 1.01; 95% CI 0.88, 1.15). In the inpatient cohort, there were 218 all-cause deaths (incidence, 28.1 per 100 person-years) and ACEI/ARB exposure was associated with reduced death (adjusted HR, 0.76; 95% CI, 0.57, 0.99). ACEI, versus ARB exposure, was associated with higher risk of hospitalization in the outpatient cohort, but no difference in all-cause death in either cohort. There was no evidence of effect modification across pre-specified baseline characteristics. Our results suggest ACEI and ARB exposure have no detrimental effect on hospitalizations and may reduce death among hypertensive patients diagnosed with COVID-19.

Cleaning of anthropometric data from PCORnet electronic health records using automated algorithms.

Objective: To demonstrate the utility of growthcleanr, an anthropometric data cleaning method designed for electronic health records (EHR). Materials and Methods: We used all available pediatric and adult height and weight data from an ongoing observational study that includes EHR data from 15 healthcare systems and applied growthcleanr to identify outliers and errors and compared its performance in pediatric data with 2 other pediatric data cleaning methods: (1) conditional percentile (cp) and (2) PaEdiatric ANthropometric measurement Outlier Flagging pipeline (peanof). Results: 687 226 children (<20 years) and 3 267 293 adults contributed 71 246 369 weight and 51 525 487 height measurements. growthcleanr flagged 18% of pediatric and 12% of adult measurements for exclusion, mostly as carried-forward measures for pediatric data and duplicates for adult and pediatric data. After removing the flagged measurements, 0.5% and 0.6% of the pediatric heights and weights and 0.3% and 1.4% of the adult heights and weights, respectively, were biologically implausible according to the CDC and other established cut points. Compared with other pediatric cleaning methods, growthcleanr flagged the most measurements for exclusion; however, it did not flag some more extreme measurements. The prevalence of severe pediatric obesity was 9.0%, 9.2%, and 8.0% after cleaning by growthcleanr, cp, and peanof, respectively. Conclusion: growthcleanr is useful for cleaning pediatric and adult height and weight data. It is the only method with the ability to clean adult data and identify carried-forward and duplicates, which are prevalent in EHR. Findings of this study can be used to improve the growthcleanr algorithm.

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.

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.

Electronic Health Record Network Research in Infectious Diseases.

With the marked increases in electronic health record (EHR) use for providing clinical care, there have been parallel efforts to leverage EHR data for research. EHR repositories offer the promise of vast amounts of clinical data not easily captured with traditional research methods and facilitate clinical epidemiology and comparative effectiveness research, including analyses to identify patients at higher risk for complications or who are better candidates for treatment. These types of studies have been relatively slow to penetrate the field of infectious diseases, but the need for rapid turnaround during the COVID-19 global pandemic has accelerated the uptake. This review discusses the rationale for her network projects, opportunities and challenges that such networks present, and some prior studies within the field of infectious diseases.

Infrastructuring an organizational node for a federated research and data network: A case study from a sociotechnical perspective.

BACKGROUND: Local nodes on federated research and data networks (FR&DNs) provide enabling infrastructure for collaborative clinical and translational research. Studies in other fields note that infrastructuring, that is, work to identify and negotiate relationships among people, technologies, and organizations, is invisible, unplanned, and undervalued. This may explain the limited literature on nodes in FR&DNs in health care. METHODS: A retrospective case study of one PCORnet® node explored 3 questions: (1) how were components of infrastructure assembled; (2) what specific work was required; and (3) what theoretically grounded, pragmatic questions should be considered when infrastructuring a node for sustainability. Artifacts, work efforts, and interviews generated during node development and implementation were reviewed. A sociotechnical lens was applied to the analysis. Validity was established with internal and external partners. RESULTS: Resources, services, and expertise needed to establish the node existed within the organization, but were scattered across work units. Aligning, mediating, and institutionalizing for sustainability among network and organizational teams, governance, and priorities consumed more work efforts than deploying technical aspects of the node. A theoretically based set of questions relevant to infrastructuring a node was developed and organized within a framework of infrastructuring emphasizing enacting technology, organizing work, and institutionalizing; validity was established with internal and external partners. CONCLUSIONS: FR&DNs are expanding; we provide a sociotechnical perspective on infrastructuring a node. Future research should evaluate the applicability of the framework and questions to other node and network configurations, and more broadly the infrastructuring required to enable and support federated clinical and translational science.

Promoting health equity for deaf patients through the electronic health record.

Language status can be conceptualized as an equity-relevant variable, particularly for non-English-speaking populations. Deaf and hard-of-hearing (DHH) individuals who use American Sign Language (ASL) to communicate comprise one such group and are understudied in health services research. DHH individuals are at high-risk of receiving lower-quality care due to ineffective patient-provider communication. This perspective outlines barriers to health equity research serving DHH ASL-users due to systems developed by large-scale informatics networks (eg, the Patient-Centered Clinical Outcomes Research Network), and institutional policies on self-serve cohort discovery tools. We list potential to help adequate capture of language status of DHH ASL-users to promote health equity for this population.

Automated production of research data marts from a canonical fast healthcare interoperability resource data repository: applications to COVID-19 research.

OBJECTIVE: The rapidly evolving COVID-19 pandemic has created a need for timely data from the healthcare systems for research. To meet this need, several large new data consortia have been developed that require frequent updating and sharing of electronic health record (EHR) data in different common data models (CDMs) to create multi-institutional databases for research. Traditionally, each CDM has had a custom pipeline for extract, transform, and load operations for production and incremental updates of data feeds to the networks from raw EHR data. However, the demands of COVID-19 research for timely data are far higher, and the requirements for updating faster than previous collaborative research using national data networks have increased. New approaches need to be developed to address these demands. METHODS: In this article, we describe the use of the Fast Healthcare Interoperability Resource (FHIR) data model as a canonical data model and the automated transformation of clinical data to the Patient-Centered Outcomes Research Network (PCORnet) and Observational Medical Outcomes Partnership (OMOP) CDMs for data sharing and research collaboration on COVID-19. RESULTS: FHIR data resources could be transformed to operational PCORnet and OMOP CDMs with minimal production delays through a combination of real-time and postprocessing steps, leveraging the FHIR data subscription feature. CONCLUSIONS: The approach leverages evolving standards for the availability of EHR data developed to facilitate data exchange under the 21st Century Cures Act and could greatly enhance the availability of standardized datasets for research.

PCORnet® 2020: current state, accomplishments, and future directions.

OBJECTIVE: To describe PCORnet, a clinical research network developed for patient-centered outcomes research on a national scale. STUDY DESIGN AND SETTING: Descriptive study of the current state and future directions for PCORnet. We conducted cross-sectional analyses of the health systems and patient populations of the 9 Clinical Research Networks and 2 Health Plan Research Networks that are part of PCORnet. RESULTS: Within the Clinical Research Networks, electronic health data are currently collected from 337 hospitals, 169,695 physicians, 3,564 primary care practices, 338 emergency departments, and 1,024 community clinics. Patients can be recruited for prospective studies from any of these clinical sites. The Clinical Research Networks have accumulated data from 80 million patients with at least one visit from 2009 to 2018. The PCORnet Health Plan Research Network population of individuals with a valid enrollment segment from 2009 to 2019 exceeds 60 million individuals, who on average have 2.63 years of follow-up. CONCLUSION: PCORnet's infrastructure comprises clinical data from a diverse cohort of patients and has the capacity to rapidly access these patient populations for pragmatic clinical trials, epidemiological research, and patient-centered research on rare diseases.

Development of an electronic health records datamart to support clinical and population health research.

INTRODUCTION: Electronic health record (EHR) data have emerged as an important resource for population health and clinical research. There have been significant efforts to leverage EHR data for research; however, given data security concerns and the complexity of the data, EHR data are frequently difficult to access and use for clinical studies. We describe the development of a Clinical Research Datamart (CRDM) that was developed to provide well-curated and easily accessible EHR data to Duke University investigators. METHODS: The CRDM was designed to (1) contain most of the patient-level data elements needed for research studies; (2) be directly accessible by individuals conducting statistical analyses (including Biostatistics, Epidemiology, and Research Design (BERD) core members); (3) be queried via a code-based system to promote reproducibility and consistency across studies; and (4) utilize a secure protected analytic workspace in which sensitive EHR data can be stored and analyzed. The CRDM utilizes data transformed for the PCORnet data network, and was augmented with additional data tables containing site-specific data elements to provide additional contextual information. RESULTS: We provide descriptions of ideal use cases and discuss dissemination and evaluation methods, including future work to expand the user base and track the use and impact of this data resource. CONCLUSIONS: The CRDM utilizes resources developed as part of the Clinical and Translational Science Awards (CTSAs) program and could be replicated by other institutions with CTSAs.

Assessing the practice of data quality evaluation in a national clinical data research network through a systematic scoping review in the era of real-world data.

OBJECTIVE: To synthesize data quality (DQ) dimensions and assessment methods of real-world data, especially electronic health records, through a systematic scoping review and to assess the practice of DQ assessment in the national Patient-centered Clinical Research Network (PCORnet). MATERIALS AND METHODS: We started with 3 widely cited DQ literature-2 reviews from Chan et al (2010) and Weiskopf et al (2013a) and 1 DQ framework from Kahn et al (2016)-and expanded our review systematically to cover relevant articles published up to February 2020. We extracted DQ dimensions and assessment methods from these studies, mapped their relationships, and organized a synthesized summarization of existing DQ dimensions and assessment methods. We reviewed the data checks employed by the PCORnet and mapped them to the synthesized DQ dimensions and methods. RESULTS: We analyzed a total of 3 reviews, 20 DQ frameworks, and 226 DQ studies and extracted 14 DQ dimensions and 10 assessment methods. We found that completeness, concordance, and correctness/accuracy were commonly assessed. Element presence, validity check, and conformance were commonly used DQ assessment methods and were the main focuses of the PCORnet data checks. DISCUSSION: Definitions of DQ dimensions and methods were not consistent in the literature, and the DQ assessment practice was not evenly distributed (eg, usability and ease-of-use were rarely discussed). Challenges in DQ assessments, given the complex and heterogeneous nature of real-world data, exist. CONCLUSION: The practice of DQ assessment is still limited in scope. Future work is warranted to generate understandable, executable, and reusable DQ measures.

Patient involvement in medical research: what patients and physicians learn from each other.

BACKGROUND: There is increasing interest in actively involving patients in the process of medical research to help ensure research is relevant and important to both researchers and people affected by the disease under study. This project examined the recently formed Vasculitis Patient-Powered Research Network (VPPRN), a rare disease research network, to better understand what investigators and patients learned from working on research teams together. METHODS: Qualitative interviews were conducted by phone with patients, physician/PhD-investigators, and study managers/staff who participated in the network. The question guiding the interviews and observational analysis was: "What have investigators and patients learned about working together while working on VPPRN teams?" Interview transcripts were analyzed in combination with observations from multiple in-person and telephone meetings. RESULTS: Transcripts and notes were reviewed and coded from 22 interviews conducted among 13 patient-partners, 5 study managers/staff, and 4 MD or PhD-investigators, and 6 in-person and 42 telephone/web-conference meetings. Patient-partners and investigators characterized their working relationships with one another, what they learned from their collaborations, and provided recommendations for future teams of patient-partners and investigators. Major themes included the great benefits of communicating about activities, being open to listening to each group member, and the importance of setting reasonable expectations. CONCLUSIONS: Direct engagement in research design and development by patient-partners and co-learning between investigators and patient-partners can result in a positive and productive working relationship for all members of a medical research team. This bi-directional engagement directly benefits and impacts research design, participant recruitment to studies, and study subject retention.

Implementing a hash-based privacy-preserving record linkage tool in the OneFlorida clinical research network.

OBJECTIVE: To implement an open-source tool that performs deterministic privacy-preserving record linkage (RL) in a real-world setting within a large research network. MATERIALS AND METHODS: We learned 2 efficient deterministic linkage rules using publicly available voter registration data. We then validated the 2 rules' performance with 2 manually curated gold-standard datasets linking electronic health records and claims data from 2 sources. We developed an open-source Python-based tool-OneFL Deduper-that (1) creates seeded hash codes of combinations of patients' quasi-identifiers using a cryptographic one-way hash function to achieve privacy protection and (2) links and deduplicates patient records using a central broker through matching of hash codes with a high precision and reasonable recall. RESULTS: We deployed the OneFl Deduper (https://github.com/ufbmi/onefl-deduper) in the OneFlorida, a state-based clinical research network as part of the national Patient-Centered Clinical Research Network (PCORnet). Using the gold-standard datasets, we achieved a precision of 97.25∼99.7% and a recall of 75.5%. With the tool, we deduplicated ∼3.5 million (out of ∼15 million) records down to 1.7 million unique patients across 6 health care partners and the Florida Medicaid program. We demonstrated the benefits of RL through examining different disease profiles of the linked cohorts. CONCLUSIONS: Many factors including privacy risk considerations, policies and regulations, data availability and quality, and computing resources, can impact how a RL solution is constructed in a real-world setting. Nevertheless, RL is a significant task in improving the data quality in a network so that we can draw reliable scientific discoveries from these massive data resources.

Learning to Share Health Care Data: A Brief Timeline of Influential Common Data Models and Distributed Health Data Networks in U.S. Health Care Research.

The last twenty years of health care research has seen a steady stream of common health care data models implemented for multi-organization research. Each model offers a uniform interface on data from the diverse organizations that implement them, enabling the sharing of research tools and data. While the groups designing the models have had various needs and aims, and the data available has changed significantly in this time, there are nevertheless striking similarities between them. This paper traces the evolution of common data models, describing their similarities and points of departure. We believe the history of this work should be understood and preserved. The work has empowered collaborative research across competing organizations and brought together researchers from clinical practice, universities and research institutes around the planet. Understanding the eco-system of data models designed for collaborative research allows readers to evaluate where we have been, where we are going as a field, and to evaluate the utility of different models to their own work.

PCORnet's Collaborative Research Groups.

The Patient-Centered Outcomes Research Institute (PCORI) launched a multi-institutional "network of networks" in 2013 - Patient-Centered Clinical Research Network (PCORnet) - that is designed to conduct clinical research that is faster, less expensive, and more responsive to the information needs of patients and clinicians. To enhance cross-network and cross-institutional collaboration and catalyze the use of PCORnet, PCORI has supported formation of 11 Collaborative Research Groups focusing on specific disease types (e.g., cardiovascular health and cancer) or particular patient populations (e.g., pediatrics and health disparities). PCORnet's Collaborative Research Groups are establishing research priorities within these focus areas, establishing relationships with potential funders, and supporting development of specific research projects that will use PCORnet resources. PCORnet remains a complex, multilevel, and heterogeneous network that is still maturing and building a diverse portfolio of observational and interventional people-centered research; engaging with PCORnet can be daunting, particularly for outside investigators. We believe the Collaborative Research Groups are stimulating interest and helping investigators navigate the complexity, but only time will tell if these efforts will bear fruit in terms of funded multicenter PCORnet projects.