The Veterans Health Administrations (VHA) has been a leader in health informatics for over 30 years. The VHA helped lead the early development of health information systems in the 1980s and led the first wide-scale deployment of an electronic health record (EHR)—the VHA’s Computerized Patient Record System (CPRS)—in the late 1990s. The clinician-designed CPRS played an integral role in system-wide improvements in VHA, through ensuring access to clinically relevant information and tools, as well as facilitating better use of population-based information by individual practices and regional and national managers.1 With wider deployment of commercial EHR systems accelerating, aided by recent Federal investments and incentives, the audience for research on how to use health information technology (HIT) most effectively has widened.2 As the articles in this special issue illustrate, the VHA continues to develop and deploy important innovations in HIT and conduct critical research on HIT.
It is hard, however, for research to keep ahead of the pace of innovation in HIT. To map out a meaningful research agenda, we must think forward 5–10 years. The HIT landscape is currently undergoing rapid change as a result of increased computing power, changing HIT platforms, and the changing expectations of providers, patients, and payers. The research community needs to identify which questions most need to be answered to make the best use of new HIT tools and approaches in that new landscape. Our research infrastructure is critical to supporting the type of HIT innovation that can support changes in the way we organize and deliver care.3
The HIT landscape of the future will be marked by the following broad trends, many of which are already underway:
1. Connected health will be a central part of the care-delivery system. Secure messaging between providers and consumers will be a central part of shared decision making. Use of Web-based tools, clinical videoconferencing, interactive voice response, home telehealth, and old-fashioned telephone encounters with nurse-care managers will supplant a growing proportion of face-to-face visits.4
2. Patients will control their own data and who has access to that data. There is an inexorable movement toward making all the medical information in the EHR transparent to the patients. The recent publication of a research study on “Open Notes” indicates that, despite clinicians’ reservations, patients strongly support the ability to access their entire medical record, including clinicians’ note.5 This will erase the current division between the EHR (what the clinician sees and controls) and the personal health record (what the patient sees and can modify).
3. An increasing amount of the data available will be provided by the patient rather than the clinician. Data may be provided automatically, as through home-telehealth tools that transmit patient data, such as blood pressure, or manually by the patient through information they enter into the health-risk assessment, as addendums to their record, or through routine capture of established measures of patient-reported outcomes.6, 7
4. HIT tools will be a key element of ongoing efforts to promote and assist behavior change in patients. Patients, their caregivers, and their clinicians will have access to a wide variety of applications to help develop goals for healthier behavior and to provide support to achieve them. Patients wishing to lose weight already can use smart phone applications that measure and chart the amount of exercise one is getting and Web tools that make recommendations and track progress on dietary goals.8, 9
5. Even as more and more data is centrally aggregated, the presentation of data will be increasingly customized. The VHA and many other systems are migrating toward Web-based medical records where clinical information is standardized and centralized, but where local applications can customize what information is retrieved and how it is displayed for patients, individual clinicians, and system managers.10 This will allow the benefits of rapidly changing platforms that are responsive to local needs, while ensuring standards-based data collection.
6. Standardization of data will allow system performance to be assessed in real time for the entire population rather than through retrospective chart-reviews of a small sample. Robert Jesse, MD, the VA’s Principal Deputy Undersecretary for Health, has coined the term “transactional quality improvement” to describe the need to systematically collect the data necessary to improve critical clinical “transactions.” Systems such as the VHA’s Cardiovascular Assessment, Reporting and Tracking system, the automated information system implemented in every VHA cardiac catheterization lab, collects data on every procedure in a standardized manner, allowing more consistent and timely reports, real-time monitoring for safety events, and better ability to monitor variation in system performance.11
7. Providers will be able to understand their patient within the context of the population. Patient data will be aggregated within the context of provider and patient-defined populations, ensuring that the provision of care for the individual is based upon clinical information reflected in the population. An expanded set of clinically relevant data will also include nontraditional determinants of health (eg, information related to community, family, and work).
8. The ability to query large databases in “real time” will give rise to new tools to help with individual diagnostic and therapeutic decisions. “Big data” techniques will allow researchers and clinicians to develop and test new clinical decision algorithms and tools for improving clinical care. Predictive tools such as the Care Assessment Need report system are already being rolled out in VHA, whereas others are using established networks to answer questions about uncommon clinical scenarios.12
9. Researchers will have improved access to and documentation about robust national datasets, including information extracted from text data. The Veterans Informatics Computing Infrastructure has begun the process of standardizing the creation of research datasets from electronic health record text data and allows cumulative improvements based on the experience of researchers working with the data.13,14 The VA Information Resource Center has improved data-linkage processes between VA, Medicare, and Medicaid clinical and claims data for research.15
The range of tools and technologies that fall under the loose terminology of “health information technology” is proliferating. The unifying element is not that they involve technology but that they introduce novel ways to collect and analyze health data and to extract knowledge from the data to inform decisions. How information influences decisions of 3 important groups—patients, health care teams, and managers—raise distinct sets of questions that are ripe for research.
Until recently, many attempts to use HIT to deliver information to patients have emphasized simple binary decisions—for example, an email reminder to a woman that she is due for a mammogram. In the future, patients will struggle with assimilating a growing volume of their own health data (eg, lab and radiology reports) while also being asked to provide more of their own data (eg, behavioral risk factors and preferences). Research can help answer the following questions to help improve the value of HIT in specific contexts:
* How do we improve the ability of patients of varying age, culture, education, and health literacy to report health information reliably and easily?
* How do we incorporate the “narrative” of the patient into structured decision making?
* How do patients understand and use various types of clinical data? How do we customize what data is presented to patients, and how it is presented to their individual needs and abilities?
* What ways of presenting and communicating data and information are most helpful for motivating behavior change?
* How are patients using secure messaging and how can they be educated to get the best use out of electronic communications with their health care team and be reassured about privacy of communications and data?16
HEALTH CARE TEAMS
At present, clinicians are experiencing the double-edged sword of the EHR. Clinicians may be spared thumbing through a multivolume paper record but they now spend a major portion of their clinical encounters typing in data and resolving alerts and reminders. Innovation in usability as well as new data display methods are needed to meet the particular preferences of clinicians. Research is limited on how to customize information to individual clinician preferences, expertise, and practice patterns.
* How do health care teams communicate to improve care? What is the best way to transfer patient care among team members to improve outcomes?
* What presentation modes are most effective for tracking the course of a chronic disease and identifying when treatment needs to be modified?
* How can reminders or other EHR tools intended to guide practice be tailored to best target potentially inappropriate use? For example, can we improve outcomes and efficiency if we target different reminders to specialists versus primary care practitioners? To high users versus low users of a given treatment?
* Where providers-practice patterns are outside the norm, can HIT serve to educate and modify behavior?
* How can secure messaging be managed effectively to improve patient interactions, reduce need for in-person visits, and identify important changes in patient condition?
The biggest challenge for managers in this new world is learning how to use effectively the growing array of data at their disposal. The aim of effective management in health systems is to identify and reduce unwanted variation in practice (ie, raising the floor) by improving practices of low performers while identifying ways to continuously improve performance (raising the ceiling). The article by Trafton et al17 outlines a number of the challenges in analyzing and presenting performance data for managers:
* How can we determine what level of variation truly indicates a problem in performance?
* How can related sets of data be combined into more meaningful clusters to inform actions in a timelier manner—for example, grouping measures related to substance abuse treatment?
* How should we present data so that managers focus on most critical and achievable areas for improvement?
Answering these questions will require a program of qualitative and quantitative research. It will require attention to the cognitive science of decision making and to the science of behavior change. The means by which we collect and deliver information will continue to evolve, just as laptops are giving way to tablet devices and smart phones. What won’t change is the challenge of understanding how to extract knowledge from data and how to use that knowledge to improve decisions and manage systems. This is the challenge that research must help us meet.
1. Brown SH, Lincoln MJ, Groen PJ, et al. VistA-US Department of Veterans Affairs national-scale HIS. Int J Med Inf. 2003;69:135–156
2. Blumenthal D. Launching HITECH. N Engl J Med. 2010;362:382–385
3. Hynes D, Whittier E, Owens A. Health information technology and implementation science: partners in progress in the VHA. Med Care. 2013 51(suppl 13):6–12
4. Dixon R. Enhancing primary care through online communication. Health Aff. 2010;29:1364–1369
5. Delbanco T, Walker J, Bell S, et al. Inviting patients to read their doctors’ notes: a quasi-experimental study and a look ahead. Ann Intern Med. 2012;157:461–470
6. Cella D, Yount S, Rothrock N, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS): progress of an NIH Roadmap Cooperative Group during its first two years. Med Care. 2007;45:S3–S11
7. Cohen A, Chinman M, Hamilton A, et al. Using patient-facing kiosks to support quality improvement at mental health clinics. Med Care. 2013 51(suppl 13):13–20
9. Demiris G, Afrin LB, Speedie S, et al. Patient-centered applications: use of information technology to promote disease management and wellness. A white paper by the AMIA knowledge in motion working group. J Am Med Inform Assoc. 2008;15:8–13
10. Fraser H, Thomas D, Tomaylla J, et al. Adaptation of a web-based, open source electronic medical record system platform to support a large study of tuberculosis epidemiology. BMC Med Inform Decis Making. 2012;12:125
11. Tsai T, Box T, Gethoffer H, et al. Feasibility of Proactive Medical Device Surveillance: The VA Clinical Assessment Reporting and Tracking (CART) Program. Med Care. 2013 51(suppl 13):57–61
12. Wang L, Porter B, Maynard C, et al. Predicting risk of hospitalization or death among patients receiving primary care in the Veterans Health Administration. Med Care. 2012 Dec 23. [Epub ahead of print]
13. Zeng Q, Redd D, Divita G, et al. Characterizing clinical text and sublanguage: a case study of the VA clinical notes. J Health Med Inform. 2011;S3:2
14. Chapman WW, Nadkarni PM, Hirschman L, et al. Overcoming barriers to NLP for clinical text: the role of shared tasks and the need for additional creative solutions. J Am Med Inform Assoc. 2011;18:540–543
15. Humensky J, Carretta H, de Groot K, et al. Service utilization of Veterans dually eligible for VA and Medicare fee-for-service: 1999–2004. Medicare Medicaid Res Rev. 2012;2:E1–E22
16. Shimada S, Hogan T, Rao S, et al. Patient-provider secure messaging in VA: variations in adoption and association with urgent care utilization. Med Care. 2013 51(suppl 13):21–28
17. Trafton J, Greenberg G, Harris A, et al. VHA Mental Health Information System: applying health information technology to monitor and facilitate implementation of VHA uniform mental health services handbook requirements. Med Care. 2013 51(suppl 13):29–36
© 2013 Lippincott Williams & Wilkins, Inc.