The authors report the experience of one of the first Southern US communities to develop a comprehensive health care data repository for tracking processes and outcomes of care and identifying areas of greatest need.
ABSTRACTObjectives: To describe an innovative health information technology (HIT) model for supporting community-wide health improvement through multiprovider collaboration in a regional population health registry and practice-based research network (PBRN).
Study Design: Case study.
Methods: We describe the HIT data structure and governance of the Diabetes Wellness and Prevention Coalition (DWPC) Registry and PBRN based in Memphis, Tennessee. The population served and their characteristics were assessed for all adult patients with at least 1 encounter in a participating health care delivery system from January 1, 2013, to March 31, 2019. Disparities in access and health care utilization were assessed by residential zip code.
Results: The DWPC Registry is a chronic disease and population health data warehouse designed to facilitate chronic disease surveillance and tracking of processes and outcomes of care in medically underserved areas of the mid-South. The Registry primarily focuses on obesity-associated chronic conditions such as diabetes, hypertension, hyperlipidemia, and chronic kidney disease. It combines patient data from 7 regional health systems, which include 6 adult hospitals and more than 50 outpatient practices, covering 462,223 adults with 2,032,425 clinic visits and 602,679 hospitalizations and/or emergency department visits from January 1, 2013, to March 31, 2019. The most prevalent chronic conditions include obesity (37.2%), hypertension (34.4%), overweight (26.4%), hyperlipidemia (18.0%), and type 2 diabetes (14.0%). The Registry provides quarterly practice improvement reports to participating clinics, facilitates surveillance of and outreach to patients with unmet health needs, and supports a pragmatic clinical trial and multiple cohort studies.
Conclusions: Regional registries and PBRNs are powerful tools that can support real-world quality improvement and population health efforts to reduce disparities and improve equity in chronic disease care in medically underserved communities across the United States.
Am J Manag Care. 2020;26(7):e211-e218. https://doi.org/10.37765/ajmc.2020.43764
The Diabetes Wellness and Prevention Coalition Registry and Practice-Based Research Network based in Memphis, Tennessee, demonstrate how regionwide population-based chronic disease registries can serve communities and improve equity in chronic disease care and outcomes by:
The prevalence of obesity is rising across all demographics in all regions of the United States.1 Nationally, 39.6% of US adults have obesity.2 Compared with individuals with normal weight, those with obesity are at increased risk for multiple chronic conditions, such as hypertension, diabetes, high cholesterol, and coronary artery disease.3 Chronic diseases cluster in individuals, particularly in those with obesity.4 In the mid-South and Mississippi Delta region, these conditions are at epidemic levels, especially for low-income minority populations with the poorest access to primary and preventive care.5-7 African American patients in medically underserved areas are less likely to get recommended care for these conditions and as a result experience increased adverse outcomes, complications, suffering, and premature death.6,7
Many authors have suggested that health information technology (HIT) has the potential to help mitigate disparities using high-tech features of electronic health records (EHRs), such as patient portals, appointment reminders, automated scheduling, and advanced registry functions, to enhance patient outreach and care coordination.8-12 However, the power and reach of modern EHR systems are often lost or wasted upon vulnerable populations due to discontinuity of care, lack of primary care, and noninteroperability of various EHR systems.13-15 There is a nationally recognized need to address the lack of semantic and syntactic interoperability among EHR systems, but to date, no single proven method addresses all problems. Indeed, health care systems commonly use EHRs for competitive advantage and resist participation in electronic health information sharing initiatives to track and improve health care access, continuity of care, and outcomes across health care systems and provider groups.16-18
Patient-data registries and practice-based research networks (PBRNs) have been proposed to foster collaboration among providers in sharing data for “common good” purposes: namely, regional quality improvement (QI), practice-based QI, and pragmatic clinical research. Other industrialized countries routinely use city-, county-, state-, or regionwide data systems to serve as both registries and PBRNs to support QI and pragmatic clinical research, respectively.19 Yet regional community-wide registries and PBRNs have seldom been implemented on a large scale in the United States. Most previous US efforts have been limited in scope to studies of a single disease, health care delivery system, or provider group10,20-23 rather than the community at large.24,25
This study describes the Diabetes Wellness and Prevention Coalition (DWPC) Registry and PBRN, an innovative US model for combining EHR data from multiple health delivery systems and providers for regional QI and practiced-based research initiatives. The DWPC Registry was created with the aspirational goal of providing a reliable and trusted data backbone to support care management for vulnerable and underserved populations in the mid-South.
We performed a descriptive case study of the DWPC Registry and PBRN based in Memphis, Tennessee. We reviewed the DWPC Registry purpose, governance, and registry use in QI, population health, and research activities using internal and publicly available records and key informant interviews of program leaders. We reviewed the HIT data structure, governance, and policies and procedures in the same manner. Our review assessed characteristics for the population participating in the Registry, including all patients with at least 1 encounter in a participating health system from January 1, 2013, to March 31, 2019. Geographic and racial disparities in access to outpatient care and inpatient health care utilization (HCU) were assessed for the year 2018 through geographic information system mapping by zip code of residence.
The DWPC is a patient, provider, and research partnership initiated in Memphis, Tennessee, in 2009. The DWPC is supported by the University of Tennessee Health Science Center (UTHSC) Center for Health System Improvement in collaboration with the UTHSC Center for Biomedical Informatics (CBMI); the Joint Institute for Computational Science of the University of Tennessee and Oak Ridge National Labs; and 7 regional health systems.26 These regional health systems partner with the DWPC and provide regular data feeds for 6 adult hospitals and more than 50 primary care and specialty clinics within these systems (Figure 1). The majority of the patients served by these health care delivery systems come from West Tennessee, Mississippi, and Arkansas.
Data Collection, Quality Assurance, and Storage
Providers, insurers, and payers participate in the DWPC Registry and PBRN by completing data use agreements with UTHSC and establishing a secure data feed of EHR or billing data with the Registry. The DWPC Registry and PBRN are fully Health Insurance Portability and Accountability Act (HIPAA) compliant and patient privacy is meticulously protected. Patient- and encounter-level data from hospitals and clinics are sent to the Registry database on monthly or quarterly schedules, with each clinic or group of clinics setting a schedule based on resources and capacity. Initially, data feeds are uploaded to an intermediary secure file transfer protocol (SFTP) server at UTHSC (Figure 2). The SFTP server runs an extract/transform/load script to immediately transfer the data to a secure landing zone inside the UTHSC firewall and deletes the memory on the SFTP server. Thus, the SFTP “bastion host” acts as a safe layer between the entity sending files and UTHSC. Once in the landing zone, the data are reviewed through a quality assurance (QA) process that ensures data completeness and uniformity of format using a data quality checklist. Once QA processes have been completed, data are uploaded and stored in a dedicated Oracle database at UTHSC.
Data interoperability issues are inherent to inter-EHR data sharing among independent sites. Even for individual EHR vendors, no 2 EHR instances are exactly alike due to site-specific specifications during installation and subsequent customizations and modifications to meet specific provider needs. Therefore, CBMI technical staff map clinic and hospital EHR feeds to a Common Data Model (CDM) to harmonize them with the larger registry. We are using PCORnet CDM version 5.1 for organization and representation of the data.27 Each patient in the CDM is assigned a unique Registry identifier. Mapping is automated using Talend software.
Following data QA and harmonization, quarterly data quality audits are conducted to identify discrepancies in the Registry data. Specifically, we quantify the accuracy and completeness of data fields within the Oracle database by selecting 50-patient random samples from participating clinics and comparing key variables for harmonized Registry data with the raw data received from data providers. The full HIT infrastructure, data transfer, and storage process is detailed in Figure 2.
Linkage With Administrative Claims Data
The DWPC Registry receives data on a quarterly basis from Tennessee Medicaid (TennCare) for beneficiaries who are served by Registry-participating clinics and/or hospitals. TennCare beneficiaries within the Registry are identified and listed using their name, Social Security number, and date of birth. TennCare uses these variables to match patients with their medical and pharmacy claims. TennCare claims data add information on medication prescriptions filled and provide more comprehensive HCU data for TennCare beneficiaries.
The DWPC is a patient, provider, and research partnership whose purpose is to transform the paradigm of care for obesity-associated chronic conditions in the mid-South from a focus on reactive, rescue care to a focus on high-value, patient-centered care that mobilizes and engages the entire community.26 The Registry initially started as a diabetes registry but is currently being used more broadly to address obesity and obesity-associated conditions including hypertension, hyperlipidemia, and chronic kidney disease. DWPC partners support, contribute to, and help oversee the DWPC Registry and PBRN to track practice-level diabetes care performance, improve care, and participate in practice-based research to improve health care delivery, patient outcomes, and population health for people in the mid-South.26 University and health system partners worked together to launch the DWPC Registry and PBRN in 2014 (Figure 1). Since then, the registry has continued to scale up, adding more community health systems in order to provide an increasingly comprehensive view of health and HCU among the mid-South’s most vulnerable citizens, to better address health and health care disparities, and to improve population health.
The DWPC Registry Data Governance Board is responsible for developing and finalizing standard policies and procedures for using the DWPC Registry data for approved health services research and pragmatic clinical trials. The board is composed of representatives from participating health systems with expertise in health informatics, pragmatic health care research, QI, and human subject protections. A separate DWPC Steering Committee, composed of representatives from each clinic and hospital data provider, is responsible for Coalition strategic planning, oversight and direction, support of initiatives and activities, and final approval of policies and procedures recommended by the Data Governance Board. Finally, a DWPC Registry Data Management Committee is responsible for day-to-day operations, data management, and Registry administration. This committee is composed of UTHSC faculty and staff and is responsible for reviewing researcher requests for data.
The DWPC Registry contains demographic, health plan, diagnosis, vital signs, and selected laboratory data for all individuals served by health system partners. Administrative claims data are also available for patients who are insured by TennCare.
The Table shows the characteristics of the Registry population. For the period from January 1, 2013, through March 31, 2019, the DWPC Registry contains records of more than 550,000 individuals, the majority of whom are adults 18 years and older (84.0%), identify as black or African American (55.0%), and reside in health professional shortage areas (HPSAs) or low-income areas (73.0%). The Registry includes more than 2 million outpatient clinic encounters and 600,000 hospitalizations and/or emergency department (ED) visits. The prevalence of obesity is 37.2% and of type 2 diabetes is 14.0% among adult Registry participants. Other most prevalent chronic conditions include hypertension (34.4%), overweight (26.4%), hyperlipidemia (18.0%), and chronic kidney disease (8.4%).
The DWPC Registry helps Registry-participating providers to track processes and outcomes of care and to improve care for their patients. The DWPC Registry originally qualified as a “specialized registry” for the purposes of achieving “meaningful use” of EHRs. Now, the Registry helps hospitals meet CMS Promoting Interoperability Program requirements by serving as a centralized repository to support hospital interoperability efforts and data exchange with EHRs.28 The Registry also assists participating clinicians in addressing interoperability requirements for meaningful use of certified EHR technology under the Promoting Interoperability category of the Merit-based Incentive Payment System (MIPS) of the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA).29
The DWPC Registry supports QI initiatives of the participating clinics through provision of quarterly practice improvement reports. The reports include detailed demographics and clinical information for patients with and at risk for diabetes and those at risk for diabetes; standard quality-of-care metrics, with measures based on the Healthcare Effectiveness Data and Information Set standards; and information on HCU (ED visits, hospitalizations, and readmissions). The reports compare patient characteristics and quality of diabetes care for each individual practice with those for the entire registry population, thus providing each clinic with a deidentified standard comparison group for benchmarking. An example practice improvement report is included in the eAppendix (available at ajmc.com). The report format is periodically revised in collaboration with participating practices to meet practice needs. Additionally, through the DWPC provider learning collaborative (PLC), we review practice improvement reports with the Registry-participating providers and share best practices. In periodic PLC meetings, practice representatives share best practices that they think account for variations in performance.
The DWPC PBRN is being used to conduct pilot and feasibility studies, health services research, and pragmatic clinical trials using both deidentified and identified data. The DWPC PBRN was recognized by the Agency for Health Research and Quality (AHRQ) in 2019 and is registered with the AHRQ PBRN Resource Center.30 Each participating health system completes a Business Associate Agreement to govern the provision of QI services by the University to the health system and a data use agreement to govern the use of Registry data to support pragmatic trials and health services research according to HIPAA and institutional review board (IRB) requirements. Researchers seeking to use data must receive approval from both the Data Governance Board and the IRB.
Pragmatic Clinical Trials
The DWPC Registry is being used to support the Patient-Centered Outcomes Research Institute—funded Management of Diabetes in Everyday Life (MODEL) study,31 a pragmatic clinical trial focusing on African American patients with uncontrolled diabetes. The DWPC Registry is used to identify potentially eligible patients, confirm eligibility, and track clinical outcomes. The Registry data for MODEL study participants is linked to a REDCap database,32 which is used for collecting survey data on the study participants. A Tableau33 data visualization layer is used for generating reports. Preliminary data show that of 541 MODEL study participants recruited through July 31, 2019, 62.6% were recruited using the Registry. Additionally, the DWPC Registry has also been used for identifying and recruiting patients for 2 pilot/feasibility focus group studies that provided preliminary data for grant funding opportunities.
Furthermore, the Registry has been used to conduct retrospective cohort studies.31,34 These studies have focused on patients with obesity, diabetes, and other obesity-associated chronic conditions. The Registry is currently being used for a UTHSC-funded retrospective cohort study in collaboration with University of Mississippi Medical Center, Tulane University, and Ochsner Health System. The aim of this study is to assess whether continuity of care protects patients living with obesity-associated chronic conditions and type 2 diabetes from potentially preventable ED and hospital utilization.35 In another study, we used Registry data to examine the association of access to primary care with overall ED and hospital utilization for patients with diabetes. We found that primary care access was associated with fewer overall ED visits and hospitalizations among medically underserved patients with diabetes.34 Because the Registry is primarily derived from EHRs, it includes data for all payers (eg, Medicare, Medicaid, private insurance, self-pay/uninsured), increasing database utility for assessing patient socioeconomic status and access to care. And for the TennCare population, the Registry provides more comprehensive information on medication and HCU data of interest to researchers. Additionally, the data can be merged with Census data using patient residence zip code to further evaluate social determinants of health.
Representativeness and Population Health
Figure 3 shows the representativeness of the population included in the Registry. The percentage of overall patients represented in the DWPC Registry and PBRN ranges from 6% to 75% per zip code, with consistent rates more than 50% for the majority of urban core zip codes. Representation of African American patients is even higher, ranging from 3% to 96% per zip code, with consistent rates more than 68% for the majority of urban core zip codes. Figure 4 shows that zip codes with higher proportions of African American patients experience higher numbers of hospitalizations and ED visits and fewer physician office visits. As the DWPC Registry has become more comprehensive, it is being used to help support regional health system improvement efforts to address disparities and improve access to primary and preventive care. The Everyone Has a Provider Initiative was launched by the UTHSC College of Medicine in 2019 to help address these disparities in access to care.
Our study describes the development of an EHR-based chronic disease registry focused on obesity-associated chronic conditions that combines demographic, clinical, laboratory, and vital data of individuals from multiple health systems and providers in the mid-South region. The DWPC Registry and PBRN demonstrate how regionwide population-based chronic disease registries can serve communities and improve equity in chronic disease management and outcomes by (1) identifying health disparities and areas of high health care need and (2) supporting patient-centered outcomes research to support health care innovation and culture change.
The DWPC Registry is a unique data source with a high representation of African American patients (55%) and those residing in HPSAs and low-income areas (73%), thus serving as a resource to answer critical questions regarding health disparities in Memphis and surrounding areas. Our data show that zip codes with higher proportions of African American patients experience higher numbers of hospitalizations and ED visits and fewer physician office visits. The Registry data provide insights on the prevalence of obesity (37.2%) and ambulatory care—sensitive conditions including hypertension (34.4%), hyperlipidemia (18.0%), type 2 diabetes (14.0%), and chronic kidney disease (8.4%). Thus, it serves as a tool for multifaceted interventional strategies targeting multiple chronic conditions rather than focusing on a single condition.
This type of registry can effectively serve as a centralized repository to help participating clinicians and hospitals address interoperability and meaningful use requirements under MIPS and MACRA. Through quarterly practice improvement reports, Registry-participating providers get an opportunity to track and compare their performance on key quality-of-care measures with that of other health systems and providers serving similar patient populations. By including data from multiple health systems and providers, the Registry provides a model for collaborative QI efforts. Most other registries have been limited in scope because they include only data for a single health system, for practices that share a common EHR, or for a single disease.21,23 For example, some of these efforts have focused on disease-specific registries to foster improvements in care for a single disease such as hypertension,20,22 diabetes,36 or cancer.10 Alternatively, many communities have developed PBRNs that have served admirably to support observational and interventional studies of chronic disease care for conditions such as hyperlipidemia.23
The DWPC PBRN demonstrates how regional PBRNs targeting medically underserved populations can support pragmatic research with real-world implications. The DWPC PBRN has been successfully used to conduct pragmatic clinical trials and feasibility/pilot studies, health services research, and QI projects. Our data provide real-world patient demographic and clinical information by including individuals who are less likely to enroll in clinical trials, such as minority patients and those living in low-income areas or HPSAs. In collaboration with other PBRNs in the South, we can address disparities and inequities in chronic disease care and outcomes across the Southern United States.
Our future efforts will focus on adding more clinics and hospitals to increase Registry representativeness and comprehensiveness for medically underserved zip codes in the mid-South. We expect to progressively expand our support for health systems, community health centers, and independent practices through provision of practice-level QI feedback reports and facilitation of surveillance of and outreach to patients with unmet health needs. We also plan to expand the Registry statewide by including data for all TennCare beneficiaries and by collaborating with academic medical centers and federally qualified health centers across Tennessee. Specifically, we are interested in expanding the DWPC Registry statewide to enable participating providers to identify and reach out to patients with uncontrolled diabetes and prediabetes for diabetes self-management education services and participation in Diabetes Prevention Program services, respectively.
Working with multiple health systems presents significant challenges because of differences in EHRs, staff capability, and willingness to share data. We addressed these obstacles through careful and diligent “boots on the ground” approaches, meeting and discussing our goals and challenges with our community partners in an environment of openness and transparency. In collaboration with participating health systems, we developed templates for data files and reassurance of safe data transfer, offering technical support as needed. Major technical, political, funding, and staffing challenges also have potential to interfere with our efforts to achieve data comprehensiveness. For example, data are received at varied frequencies, risking data lag. We plan to overcome this challenge by moving over time to real-time data feeds for all data providers. The Registry also faces similar challenges in achieving full community representation. The DWPC Registry does not yet include all health systems in the Memphis area and therefore lacks information on care provided outside of participating health systems.
In addition, we have struggled to make the DWPC Registry practice improvement reports actionable by the health systems we serve. Each of our participating health systems also creates internal quality and revenue reports that typically serve as their major tools for tracking processes and outcomes of care and their highest-priority QI initiatives. However, through the DWPC PLC, practice representatives share best practices that they think account for variations in performance. Such forums for discussing practice-based QI initiatives in the context of community-wide population health goals are ultimately needed to make practice- and community-level quality reporting initiatives relevant and actionable.
The greatest long-term challenge for the DWPC Registry, however, is to maintain financial viability. Currently, we have sufficient funding from the University; however, as we expand the Registry by including additional health systems, it will be challenging to sustain it long-term. We are seeking philanthropic and government funds to use the Registry in community-wide QI initiatives. The National Patient-Centered Clinical Research Network has also provided funding for many similar registries.
Regional registries have potential to assist local health systems in quality measurement that involves coordination of care across providers. For example, they can enable outpatient providers to get feedback on the inpatient utilization of their patient panels. They can also assist with reporting requirements mandated by payers and can enable integrated delivery networks across independent providers and accountable care organizations. In many communities with a number of independent providers or competing health care delivery systems, regional HIT solutions such as registries and PBRNs have potential to support a more nimble health care delivery system that helps ensure coordinated and equitable care across an entire community.
The authors thank Emanuel Villa, Mark Sakauye, Jennifer Featherstone, Josh Callaway, Ming Chen, and Patti Smith. They also gratefully acknowledge the health systems and their leadership that have made their work possible, including Methodist Le Bonheur Health Care, Regional One Health, Christ Community Health Services, Church Health, Memphis Health Center, University Clinical Health, and West Tennessee Health Care.Author Affiliations: Center for Health System Improvement, College of Medicine, University of Tennessee Health Science Center (SS, SAS, CL, JEB), Memphis, TN; Anne Arundel Medical Center (IMB), Annapolis, MD; Center for Biomedical Informatics, College of Medicine, University of Tennessee Health Science Center (PZ, RLD), Memphis, TN; Department of Preventive Medicine, College of Medicine, University of Tennessee Health Science Center (EAT, JEB), Memphis, TN; Mississippi State University (REC), Starkville, MS; College of Medicine, University of Tennessee Health Science Center (AJS), Memphis, TN; Department of Medicine, College of Medicine, University of Tennessee Health Science Center (JEB), Memphis, TN.
Source of Funding: Intramural funding by The University of Tennessee Health Science Center College of Medicine.
Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (SS, IMB, EAT, JEB); acquisition of data (SS, SAS, REC, RLD, PZ, JEB); analysis and interpretation of data (SS, SAS, PZ, REC, JEB); drafting of the manuscript (SS, IMB, SAS, EAT, JEB); critical revision of the manuscript for important intellectual content (SS, IMB, SAS, PZ, EAT, CL, AJS, JEB); statistical analysis (SS, JEB); provision of patients or study materials (IMB, JEB); obtaining funding (JEB); administrative, technical, or logistic support (IMB, CL, RLD, AJS, JEB); supervision (CL, RLD, AJS, JEB); and figures/maps (REC).
Address Correspondence to: Satya Surbhi, PhD, Center for Health System Improvement, College of Medicine, University of Tennessee Health Science Center, 956 Court Ave, Coleman D224A, Memphis, TN 38163. Email: firstname.lastname@example.org.REFERENCES
1. Overweight & obesity. CDC. Accessed January 2, 2020. https://www.cdc.gov/obesity/index.html
2. Hales CM, Fryar CD, Carroll MD, Freedman DS, Ogden CL. Trends in obesity and severe obesity prevalence in US youth and adults by sex and age, 2007-2008 to 2015-2016. JAMA. 2018;319(16):1723-1725. doi:10.1001/jama.2018.3060
3. The health effects of overweight and obesity. CDC. Accessed January 2, 2020. https://www.cdc.gov/healthyweight/effects/index.html
4. Agborsangaya CB, Ngwakongnwi E, Lahtinen M, Cooke T, Johnson JA. Multimorbidity prevalence in the general population: the role of obesity in chronic disease clustering. BMC Public Health. 2013;13:1161. doi:10.1186/1471-2458-13-1161
5. Ward BW, Black LI. State and regional prevalence of diagnosed multiple chronic conditions among adults aged ≥18 years — United States, 2014. MMWR Morb Mortal Wkly Rep. 2016;65(29):735-738. doi:10.15585/mmwr.mm6529a3
6. Allen NB, Diez-Roux A, Liu K, Bertoni AG, Szklo M, Daviglus M. Association of health professional shortage areas and cardiovascular risk factor prevalence, awareness, and control in the Multi-Ethnic Study of Atherosclerosis (MESA). Circ Cardiovasc Qual Outcomes. 2011;4(5):565-572. doi:10.1161/CIRCOUTCOMES.111.960922
7. Durant RW, Parmar G, Shuaib F, et al. Awareness and management of chronic disease, insurance status, and health professional shortage areas in the REasons for Geographic And Racial Differences in Stroke (REGARDS): a cross-sectional study. BMC Health Serv Res. 2012;12:208. doi:10.1186/1472-6963-12-208
8. Gatwood J, Hohmeier KC, Brooks IM. Beyond the reminder: the next steps in pharmacist-driven, mHealth patient engagement. J Am Pharm Assoc (2003). 2019;59(2S):S21-S24. doi:10.1016/j.japh.2018.10.003
9. Samal L, Lipsitz SR, Hicks LS. Impact of electronic health records on racial and ethnic disparities in blood pressure control at US primary care visits. Arch Intern Med. 2012;172(1):75-76. doi:10.1001/archinternmed.2011.604
10. Levy DE, Munshi VN, Ashburner JM, Zai AH, Grant RW, Atlas SJ. Health IT—assisted population-based preventive cancer screening: a cost analysis. Am J Manag Care. 2015;21(12):885-891.
11. Neuner J, Fedders M, Caravella M, Bradford L, Schapira M. Meaningful use and the patient portal: patient enrollment, use, and satisfaction with patient portals at a later-adopting center. Am J Med Qual. 2015;30(2):105-113. doi:10.1177/1062860614523488
12. Antonio MG, Petrovskaya O, Lau F. Is research on patient portals attuned to health equity? a scoping review. J Am Med Inform Assoc. 2019;26(8-9):871-883. doi:10.1093/jamia/ocz054
13. Sequist TD. Health information technology and disparities in quality of care. J Gen Intern Med. 2011;26(10):1084-1085. doi:10.1007/s11606-011-1812-8
14. Mack D, Zhang S, Douglas M, Sow C, Strothers H, Rust G. Disparities in primary care EHR adoption rates. J Health Care Poor Underserved. 2016;27(1):327-338. doi:10.1353/hpu.2016.0016
15. Reisman M. EHRs: the challenge of making electronic data usable and interoperable. P T. 2017;42(9):572-575.
16. Thune J, Alexander L, Roberts P, Burr R, Enzi M. Where is HITECH’s $35 billion dollar investment going? Health Affairs. March 4, 2015. Accessed June 8, 2020. https://www.healthaffairs.org/do/10.1377/hblog20150304.045199/full
17. Information blocking. HealthIT.gov. Accessed January 2, 2020. https://www.healthit.gov/topic/information-blocking
18. Office of the National Coordinator for Health Information Technology. Report on health information blocking. HealthIT.gov. April 2015. Accessed January 2, 2020. https://www.healthit.gov/sites/default/files/reports/info_blocking_040915.pdf
19. Tan WS, Ding YY, Xia WC, Heng BH. Effects of a population-based diabetes management program in Singapore. Am J Manag Care. 2014;20(9):e388-e398.
20. Jaffe MG, Young JD. The Kaiser Permanente Northern California story: improving hypertension control from 44% to 90% in 13 years (2000 to 2013). J Clin Hypertens (Greenwich). 2016;18(4):260-261. doi:10.1111/jch.12803
21. Ray GT, Bahorik AL, VanVeldhuisen PC, Weisner CM, Rubinstein AL, Campbell CI. Prescription opioid registry protocol in an integrated health system. Am J Manag Care. 2017;23(5):e146-e155.
22. Bisognano JD, Speranza PS, Becker LM, et al. Creating community collaboration to improve the care of patients with high blood pressure: lessons from Rochester, New York. J Clin Hypertens (Greenwich). 2012;14(3):178-183. doi:10.1111/j.1751-7176.2011.00579.x
23. Fuke D, Hunt J, Siemienczuk J, et al. Cholesterol management of patients with diabetes in a primary care practice—based research network. Am J Manag Care. 2004;10(2 pt 2):130-136.
24. Egan BM, Li J, Fleming DO, et al. Impact of implementing the 2013 ACC/AHA cholesterol guidelines on vascular events in a statewide community-based practice registry. J Clin Hypertens (Greenwich). 2016;18(7):663-671. doi:10.1111/jch.12727
25. Egan BM, Zhao Y, Li J, et al. Prevalence of optimal treatment regimens in patients with apparent treatment-resistant hypertension based on office blood pressure in a community-based practice network. Hypertension. 2013;62(4):691-697. doi:10.1161/HYPERTENSIONAHA.113.01448
26. About the DWPC. The University of Tennessee Health Science Center. October 22, 2019. Accessed January 2, 2020. https://www.uthsc.edu/chsi/dwpc/index.php
27. Common Data Model (CDM) specification, version 5.1. PCORnet. Accessed January 2, 2020. https://pcornet.org/wp-content/uploads/2019/09/PCORnet-Common-Data-Model-v51-2019_09_12.pdf
28. Promoting interoperability programs. CMS. Updated May 12, 2020. Accessed June 8, 2020. https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/index?redirect=/ehrincentiveprograms/
29. Promoting interoperability. HealthIT.gov. Accessed January 2, 2020. https://www.healthit.gov/topic/meaningful-use-and-macra/promoting-interoperability
30. Practice-based research networks. Agency for Healthcare Research and Quality. Accessed January 2, 2020. https://pbrn.ahrq.gov
31. Jackson BM, Gutierrez ML, Relyea GE, et al. Unmet primary care needs in diabetic patients with multimorbidity in a medically underserved area. Health Serv Res Manag Epidemiol. 2017;4:2333392817702760. doi:10.1177/2333392817702760
32. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-381. doi:10.1016/j.jbi.2008.08.010
33. Tableau. Accessed January 2, 2020. https://www.tableau.com
34. Shuvo SA, Surbhi S, Dashputre A, Bailey JE. Association of access to primary care with healthcare utilization among medically underserved patients with diabetes. Abstract presented at: AcademyHealth Annual Research Meeting; June 24-26, 2018; Seattle, WA. Accessed September 13, 2019. https://academyhealth.confex.com/academyhealth/2018arm/meetingapp.cgi/Paper/26435
35. Delta Clinical and Translational Health Disparities Conference. University of Tennessee Health Science Center. Accessed June 8, 2020. https://uthsc.edu/research/development/intramural-funding/uthsc-tu-ummc-health-disparities-conference.php
36. Karter AJ, Moffet HH, Liu J, et al. Achieving good glycemic control: initiation of new antihyperglycemic therapies in patients with type 2 diabetes from the Kaiser Permanente Northern California Diabetes Registry. Am J Manag Care. 2005;11(4):262-270.