Cochran, J., Baus, A., Jarrett, T. & Plaugher, C. (Summer 2017). Making the data talk. Online Journal of Nursing Informatics (OJNI), 21 (2), Available at http://www.himss.org/ojni
Background: The gap between theory and practice has not been well bridged in rural, non-academic clinics. Electronic health records (EHR) provide nurse practitioners a venue to support clinical questions and answers. However, it is challenging to use data outside the realms of required reporting.
Purpose: This paper discusses the process of developing networks and collaboration for data acquisition and analysis in rural primary care. The use of informatics to describe populations in a primary care center helps inform interventions. Nurse practitioners can use EHR data to evaluate and improve patient care.
Methodology: West Virginia received funding to form the West Virginia Clinical and Translational Science Institute (WVCTSI), which supports the development of practice-based research. Primary care centers gained access to academic institutions and state agencies for data analysis. Clinicians used the process to ask and answer clinically driven questions.
Results: Clinicians used the data and process to answers questions about specific patient populations. Clinicians can use EHR data to describe patients and to inform interventions.
Introduction and Problem Statement
Rural nurse practitioners are rarely principal investigators in research projects. Further, tightly controlled clinical research trials are the means by which evidence-based best practices are ultimately introduced to primary care. Theory and practice are poorly integrated into rural, non-academic clinics. Clinical evidence is used to maintain the standard of care but is not usually generated at the primary care exam table (Hayes & Burge, 2012). Fortunately, nurse practitioners have a collective knowledge base of practice and experience. Equally important is the applied experience of practicing clinical providers who understand the cultural, familial and social context in which patients live and make decisions (Green, 2008). Better documentation of this collective knowledge stands to benefit quality of care improvement and practice-based research (Parchman, Zeber & Palmer, 2010). Nurse practitioners need tools to examine their daily practice, transform those observations into generalizable research questions, and ultimately disseminate those findings among their peers.
For clinically relevant research to occur, applied clinical knowledge and research skills must be integrated to develop evidence-based best practices. This integration must include practice-based and practice-led research methods (Rolfe, 1998). Out of 101 very promising claims of new discoveries made in major basic science journals between 1979 and 1983, five resulted in interventions with licensed clinical use by 2003, and only one resulted in extensive clinical use (Contopoulos-Ioannidis, Ntzani, & Ioannidis, 2003). A relatively recent movement to support practice-based research networks – multi-clinic collaborations designed to leverage research for the improvement of patient care and population health (Agency for Healthcare Research and Quality, 2012) – allowed for an expansion of applied research within clinical settings and the ability to respond more rapidly to emerging questions and health priority areas (Westfall, Mold, & Fagnan, 2007). However, opportunities for practice-led research from the clinical level are limited.
The introduction and widespread use of electronic health records (EHRs) provide a venue to support the practice-based research process, the ability for nurse practitioners to explore their own data, and a foundation for acting on data for improvements in patient care and ultimately population health. As EHR data are transformed into query-friendly tools, the research interests of nurse practitioners can be better addressed, ideas more carefully articulated and actualized, and team-based collaboration better facilitated. While reporting regulations such as meaningful use and value-based care compel nurse practitioners to view their practices through outcomes and data (Ritchie, Marbury, Verdon, Mazzolini, & Boyles, 2017), the information used in meaningful use reports and value-based reimbursement does not give the level of assessment necessary for population-based care and intervention designs addressing rural health disparities. These required data do not answer clinical-based questions and concerns about patients, or groups of patients, with specific diseases or patterns of health problems. The purpose of this paper is to describe how a small rural clinic collaborated with an academic center to leverage routinely collected clinical data from the EHR to answer important clinical questions and initiate practice-based research.
Developing Academic Institutional Capacity for Practice-Led Research
Primary care is one of the major providers of accessible health care services. Health care needs of patients and communities are integrated and sustained by this partnership (Institute of Medicine (US) Committee on the Future of Primary Care, 1996). The Department of Health and Human Services designated the Agency for Healthcare Research and Quality (AHRQ) as the principal source of funding for primary care research (Agency for Healthcare Research and Quality, 2016). AHRQ recognized primary care as a clinical laboratory and facilitated networks among the centers. The AHRQ described PBRNs as:
Primary care practice-based research networks (PBRNs) involve practicing clinicians in asking and answering clinical and organizational questions central to primary health care. The AHRQ developed its PBRN initiative in recognition of this work, its ability to improve the health of all Americans, and the potential of these networks to engage clinicians in quality improvement activities (Agency for Healthcare Research and Quality, 2001).
In 2011, West Virginia received a National Institute of General Medicine Sciences Institutional Development Award (U54GM104942), with an aim to use innovative recruitment, training, and mentoring strategies to develop clinical and translational opportunities across the state. From this initiative, the West Virginia Clinical and Translational Science Institute (WVCTSI) was formed. A primary component of the WVCTSI commitment was to provide opportunities for quality translational research, enhance understanding of rural primary care, and inform population health through the development of a statewide practice-based research network (PBRN). There were no PBRNs within the state of West Virginia prior to the overarching structure of the WVCTSI.
In 2012, the West Virginia Practice Based Research Network (WVPBRN) officially formed. It registered with the AHRQ in 2013 and currently includes 72 primary care practices. By June of 2014, partnerships with state local health alliances and multiple academic institutions emerged. These alliances foster peer-reviewed publications and presentations, the development of research questions, and an increased capacity of academic institutions and clinical sites to conduct practice-based research. The AHRQ recognizes the importance of PBRN’s role to translate research findings into practice (Agency for Healthcare Research and Quality, 2016).
Many of the practice-led research methods currently employed by WVPBRN clinical partners grew from years of collaboration with the West Virginia University Office of Health Services Research (OHSR), a key partner in the WVCTSI structure. For more than three decades, OHSR, part of the university’s School of Public Health, has provided quality improvement assistance and best practice recommendations, coaching in implementation of clinical information systems such as EHRs and patient registries, data quality reviews, and health analytics support for practice transformation and practice-based research. OHSR leadership provided technical and clinical/community engagement expertise to the newly formed WVPBRN, and staff acted as the inaugural network coordinators.
Clinical Level Methods for Practice-Led Research
To develop practice-driven research questions, clinicians must first reflect on what they would like to know about their patients, treatments, and outcomes based on their day-to-day experiences. Then, they must obtain the data necessary to answer their questions. EHRs were developed to create a consistent record of patient care in order to make clinical decisions at the point of care. However, these systems can also help clinicians answer the clinical implications of courses of care more broadly. Although many of these questions manifest in quality improvement opportunities, a broad picture of population health can emerge through research methods.
The use of EHR data enables the transition from quality of care improvement to practice-based research. This process begins by extracting targeted data (such as patient demographics, visit histories, vitals, health conditions, medications, services provided, and laboratory values) from the EHR using practice analytics. Practice analytics is reporting software linked to the EHR allowing for some initial queries of the data and exporting of these data to external systems for supplemental analyses. Exported data are securely stored on a server at the health clinic. The data extraction occurs in-house via health informatics staff. Using tools built by OHSR, these data are imported into a patient registry, also stored on the same secure server, for analysis. This larger data set is then queried per research project to produce the necessary data sets for analysis. The data are partially de-identified when first pulled from the EHR and then completely de-identified (e.g., limiting zip codes to three digits and truncating dates to years only) as data sets are created for analysis based on the research protocols. A signed business associate agreement between the clinic and OHSR enables this work. The team applies biostatistics principles to identify statistical tests appropriate for each clinical question. Initial data analyses are conducted and clinicians review results for internal validity. Clinicians use a team-based approach to evaluate results of the analysis. The clinical team uses knowledge gained from the process to inform interventions (Figure 1). All research undergoes Institutional Review Board approval procedures as needed.
Using informatics and the methodology described, multiple practice-based and practice-led projects emerged to answer clinically driven questions for improved patient care. Most are basic quality improvement/quality assessment efforts, such as:
- Clinicians noted increased pediatric obesity trends and difficulties with intervention planning but were not sure of locations or age ranges to target in our pediatric practice. The team performed a demographic analysis to view the population as a whole. This project gave valuable insight into the ages, gender, and geographic location of obese pediatric patients (Cochran & Baus, 2015). As a result, providers identified community resources in areas with increased obese children and included this information in patient counseling.
- Pediatric providers were concerned about missing hypertension in pediatric patients or hypertension being improperly identified due to diagnostic criteria. Clinicians extracted data using diagnosis and multiple in-office blood pressure readings. Clinicians validated that hypertension was correctly identified in the majority of patients with the diagnosis (Thorpe, Cochran, & Bridges, 2015). The validated results were well received by the providers and no changes were made in procedures.
- Clinical staff noticed problems with transition of care in the pediatric special needs population. Clinicians identified issues transferring this population to an adult provider who would assume their care. More needs to be known about the special needs group to plan adult transitional care. Our clinic is affiliated with a family practice residency program, and the residents took on the project (Bailey et al., 2015). As a result of the study, transitional patients were identified by a template change that allowed the special needs patients to be identified in a data pull.
- Obesity in the clinic’s population predisposed children to metabolic syndrome. The literature suggested multiple diagnostic criteria. Clinicians investigated how we could assess our obese children with labs that were available for metabolic syndrome. The HDL/triglyceride ratio was used (Bridges, Jarrett, Thorpe, Baus, & Cochran, 2016). There were no change in procedures, but providers continued evaluating methods to identify metabolic syndrome.
- Diabetes is highly prevalent in West Virginia. Meaningful use/value-based care views diabetic control or A1C values for outcome criterion. The clinical staff looked at how our clinic compared to the rest of the state and what proportion of our population was uncontrolled versus controlled. The project also studied how to better assess the population and determine resources/interventions that are appropriate for this population (Bannister, Baus, & Jarrett, 2015). The results suggested that our population of uncontrolled diabetic patients was less than projected, but probably more complex and time consuming for the providers.
AHRQ supports projects that use EHR data to identify gaps in care and needed preventive health screenings such as mammography and colonoscopy (Johnson & Mardon, 2013). EHR data describes structure, process and outcomes within organizations and evaluates delivery of care. International and community-based interventions are initiated and evaluated using population data from EHR systems (National Learning Consortium, 2013).
The projects presented here are similar to previously conducted quality improvement/quality assurance efforts. However, in these instances, providers initiated and performed the research as opposed to participating in research initiated outside of their health care facility. While specific to one clinic, taken as a collective these projects provide assessments of clinical problems from a different perspective – creating solutions not otherwise considered. As existing research efforts evolve and new ones begin, the collaborative partnership has grown in tandem. Clinicians now report findings at regular staff meetings. These sessions initiate dialogue about each project with suggestions for improvement.
Limitations and Challenges
This work takes place within the context of some overarching limitations. First, quality improvement/quality assurance projects are not generally published in mainstream, clinically preferred literature. Further, while major organizations, such as AHRQ and the National Institutes of Health, offer information relevant to daily practice, clinicians may not recognize the value of these agencies. Second, given research that suggests physicians lack acceptance of EHR implementation and challenges to the incorporation of these systems into primary care, providers may not accept EHR data as accurate (Boonstra & Broekhuis, 2010; Emani et al., 2014). The team must extract and prepare data for statistical analysis apart from meaningful use programming. This is outside of the required reporting, uses more resources than are available for most clinics, and is often cost prohibitive for smaller practices (Fryer, Doty, & Audet, 2011). Third, the distinction between quality of care improvement efforts and those deemed specifically to be practice-based research efforts is often challenging to determine. Our clinical teams concentrate efforts on understanding important patterns and trends in patient care and health outcomes. However, research efforts using de-identified data require that we transform our data to meet certain specifications such as 3-digit zip code truncations, year-only truncations for date, and age categorizations rather than date of birth. These informatics steps, while important, have the potential to limit our ability to apply the data in terms that are more granular. Answering specific clinical questions has not provided a simple solution to high-risk, high-need patient problems. It has helped, however, to assess patient populations in an efficient, resource sensitive manner through primary care/public health partnerships that were not necessarily available in the past.
Notable outcomes of clinic-based analyses for nurse practitioners have helped to:
- Understand methods of basic population/demographic assessment within the clinic population and use the information to build interventions and collect resources tailored to the population.
- Further evaluate observational knowledge and confirm it with data analysis.
- Participate in describing similar predictors of poor outcomes for specific disease groups.
As previously described, rural nurse practitioners are rarely in the position of being principal investigators in practice-based research efforts. However, nurse practitioners can not only be principal investigators, but also catalysts for action-oriented research incorporated into daily practice. Nurse practitioners are in a singularly important position of being at the forefront of patient care while concurrently having an in-depth understanding of and insight into outcomes data due to their interactions with administration, providers and patients.
As a practical application of this approach, quarterly research luncheons are held at the clinic during which health care providers, medical staff, administration and practice-based research partners regularly gather to engage and discuss current and planned research efforts. At these forums, data from current efforts are presented and displayed for group discussion, feedback and interpretation. The group vets ideas for future efforts, aiding in design and overall buy-in for the efforts. Results are action-oriented in that they help to inform practice policy development, modification and, ultimately, the patient care experience.
As translational research emerges into the clinical forefront, nurse practitioners will now have opportunities to formulate research questions stemming from quality improvement efforts to guide their practice. The richness of the research lies deep in the everyday occurrences of the clinical setting. Nurse practitioners need to have tools to adequately investigate day-to-day occurrences in the form of basic research questions. Pushing that knowledge to peer-reviewed sources is imperative for health care providers, especially nurse practitioners, in rural areas.
Clinical faculty practicing in academic settings are role models for bringing clinical translational research to the table. This not only fosters excellence in practice within the academic medical community but also models translational research to health care students for use in future practice. As EHR data become universally accepted, nurse practitioners must evaluate the validity and use of their data. The future outcomes of patient care and population health depend on ethical, accurate use of EHR data.
Future Directions/Implications for Practice-Based Research
The efforts of the collaborative academic and clinical research team to build practice-based research infrastructure have, by necessity, required a reexamination of the role of the nurse practitioner in the overall health systems process. Institutional-wide buy-in is essential for a research idea to be transformed into a practical, sustainable project that can both inform and improve the patient care process. Research efforts may or may not be funded initiatives. Regardless, demonstration of the return-on-investment for the work and being able to do so beyond mere dollars and cents is essential. Slight changes to patient care, the delivery of health education, and even the ways in which patients are engaged in their own health can have a significant, clinical impact. Repurposing EHR data to identify, evaluate and better understand changes in patient-level and population-level outcomes over time is a prerequisite to making informed decisions for the sake of patient care.
The future will require that providers have the skills needed to evaluate EHR data. Nurse practitioners are provided the opportunity to view their patient populations through an epidemiological lens. Data gained from their patient population can provide information about patterns of disease process, comorbidities, and geographical pockets of similar health-related issues. Data analysis at the clinical level provides a bridge between theory and practice and opens the door for translational research. The results can inform clinical practice, influence policies, impact reimbursement, and ultimately improve patient care.
Agency for Healthcare Research and Quality. (2016). Practice-based research networks: research in everyday practice. Retrieved from https://pbrn.ahrq.gov/
Agency for Healthcare Research and Quality. (2012). Practice-based research networks: Research in everyday practice. Retrieved from https://pbrn.ahrq.gov/
Agency for Healthcare Research and Quality. (2001). Primary care practice-based research networks. Retrieved from https://www.ahrq.gov/research/findings/factsheets/primary/pbrn/index.html
Bailey, J., Cochran, J., Dilly, J., Lee, J., Hampton, J., Yi, J., & Chung, M. (2015). A demographic analysis of the special needs population in a rural health clinic. WVSOM Mid-Winter Conference, 2015.
Bannister, J. C., J., Baus, A., & Jarrett, T. (2015). Demographic and clinical aspects of diabetics in a rural primary care center. West Virginia Rural Health Conference, 2016.
Boonstra, A., & Broekhuis, M. (2010). Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions. BMC Health Services Research, 10, 231. doi:10.1186/1472-6963-10-231
Bridges, K. G., Jarrett, T., Thorpe, A., Baus, A., & Cochran, J. (2016). Use of the triglyceride to HDL cholesterol ratio for assessing insulin sensitivity in overweight and obese children in rural Appalachia. Journal of Pediatric Endocrinology & Metabolism: JPEM, 29(2), 153-156. doi:10.1515/jpem-2015-0158 [doi]
Cochran, J., & Baus, A. (2015). Developing interventions for overweight and obese children using electronic health records data. Online Journal of Nursing Informatics, 19(1), //www.himss.org/ResourceLibrary/GenResourceDetail.aspx?ItemNumber=39758. Epub 2015 Feb 2. doi:http://www.himss.org/ResourceLibrary/GenResourceDetail.aspx?ItemNumber=39758 [pii]
Contopoulos-Ioannidis, D. G., Ntzani, E., & Ioannidis, J. P. (2003). Translation of highly promising basic science research into clinical applications. The American Journal of Medicine, 114(6), 477-484.
Emani, S., Ting, D. Y., Healey, M., Lipsitz, S. R., Karson, A. S., Einbinder, J. S., & Bates, D. W. (2014). Physician beliefs about the impact of meaningful use of the EHR. Applied Clinical Informatics, 5(3), 789-801. doi:10.4338/ACI-2014-05-RA-0050
Fryer, A., Doty, M. M. & Anne-Marie J. Audet, M D. (2011). Sharing resources: Opportunities for smaller primary care practices to increase their capacity for patient care. Retrieved from http://www.commonwealthfund.org/publications/issue-briefs/2011/mar/sharing-resources
Green, L. W. (2008). Making research relevant: If it is an evidence-based practice, where's the practice-based evidence? Family Practice, 25 Suppl 1, i20-4. doi:10.1093/fampra/cmn055
Hayes, H., & Burge, S. (2012). Creating a practice-based research network from scratch: Where do I begin? Progress in Community Health Partnerships: Research, Education, and Action, 6(3), 369-380. doi:10.1353/cpr.2012.0033
Institute of Medicine (US) (1996). Committee on the Future of Primary Care.
Johnson, M., Mardon, R. (2013). Potential measures for clinical-community relationships. Retrieved from /professionals/prevention-chronic-care/resources/ccrm-atlas-suppl/index.html
National Learning Consortium. (2013). Continuous quality improvement (CQI) strategies to optimize your practice. (No. 2017).HealthIT.gov. Retrieved from https://www.healthit.gov/sites/default/files/tools/nlc_continuousqualityimprovementprimer.pdf. (cqi)
Parchman, M. L., Zeber, J.E., & Palmer, R. F. (2010). Participatory decision-making, patient activation, medication adherence, and intermediate clinical outcomes in type 2 diabetes: A STARNet study, Annuals of Family Medicine, 8(5):410-7.
Ritchie, A., Marbury, D., Verdon, D., Mazzolini, C., & Boyles, S. (2017). Shifting reimbursement models: The risks and rewards for primary care. Medical Economics, (04/08/2014), 02/14/2017. Retrieved from http://medicaleconomics.modernmedicine.com/medical-economics/content/tags/aca/shifting-reimbursement-models-risks-and-rewards-primary-care?page=full
Rolfe, G. (1998). The theory-practice gap in nursing: From research-based practice to practitioner-based research. Journal of Advanced Nursing, 28(3), 672-679.
Thorpe, A., Cochran, J., & Bridges, K. (2015). Hypertension in a rural West Virginia pediatric population. Ohio Osteopathic Symposium Poster Competition.
Westfall, J. M., Mold, J., & Fagnan, L. (2007). Practice-based Research—“Blue highways” on the NIH roadmap. JAMA, 297(4), 403. doi:10.1001/jama.297.4.403
Processes discussed in this publication were supported by the National Institute of General Medical Sciences of the National Institutes of Health under Award Number U54GM104942. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Adam Baus, PhD, MA, MPH, is a research assistant professor in the Department of Social and Behavioral Sciences, assistant director of the Office of Health Services Research - School of Public Health, and the associate director for research and quality improvement in the West Virginia Practice-Based Research Network. His work focuses on health informatics in primary care, helping clinic and community partners better leverage clinical data for improved patient care, and population health. He also works with the West Virginia University Health Science Center’s associate vice president’s office to build health analytics collaborations with and to increase capacity within the West Virginia Department of Health and Human Resources. He has a doctoral degree in public health, a master's of public health, and a master's in applied social research.
Jill Cochran, PhD, APRN, C-FNP is an associate professor, West Virginia School of Osteopathic Medicine. Jill Cochran has been involved in rural health for 39 years. After obtaining her master’s degree in nursing from West Virginia University, she practiced in small rural clinics in Meadow Bridge, Rainelle, Rupert and later in Lewisburg, West Virginia. She completed her doctoral studies in 2011 at West Virginia University and conducted research in children’s obesity. She completed a fellowship with the National Rural Health Association in 2009 and has continued as an advocate for rural health. She has been a nurse practitioner in pediatrics at Robert C. Byrd Clinic for more than 15 years.
Traci Jarrett, PhD. Dr. Jarrett is a research assistant professor in the WVU School of Public Health Department of Social and Behavioral Sciences. She is a faculty member within the West Virginia Prevention Research Center and the West Virginia Office of Health Services Research, and is a member of the WV Clinical Translational Science Institute Community Engagement and Outreach Core. She is housed at the University of Kentucky in Lexington, and serves as a liaison with the University of Kentucky Center for Clinical and Translational Sciences. In addition, her research interests include evaluation, community development and social action, community-engaged research, social capital development and access among young adults, health and social disparities in rural populations, systems and organizational development, social networking, and improvement science as it relates to health access and outcomes in primary care clinical settings.
Christine Plaugher, MS, currently serves as the Statewide Campus (SWC) and Mountain State Osteopathic Postdoctoral Training Institutions, Inc., (MSOPTI) clinical research coordinator at the West Virginia School of Osteopathic Medicine in the Center for Rural & Community Health. In this position, she is working to promote clinical and translation research within the SWC and MSOPTI by offering assistance with all aspects of research and clinical improvement projects. The majority of her career has been spent as a small business owner contracting with Dow AgroSciences, a division of Dow Chemical, in which she processed and evaluated pesticide research data for all product trial phases on multiple types of urban and large and small crop trials. Tina attended West Virginia University where she earned a bachelor’s degree in environmental science and a master’s degree in entomology.