The first three chapters of this feature focused on the evolution of personal health records (PHRs) and their potential to transform health and health care. However, the simple vision of having meaningful access to one’s complete medical record belies the more complex realities of a fragmented health care system. The capacity to store information electronically and exchange it with other institutions is an essential step toward realizing the benefits of PHRs described in pervious chapters. Yet, health information technology (HIT) struggles to keep pace with the demand for more efficient and more effective use of health data. This chapter examines the challenges of working with health information and aims to illuminate PHRs’ place within the larger ecosystem of HIT infrastructure. This context is crucial to understanding the barriers to HIT adoption and the policy issues confronting PHRs, which are discussed in subsequent chapters.
As we discussed in Chapter 1, one of the primary limitations of first-generation PHRs was their lack of connectivity with medical providers and the health care system. First-generation PHRs could store information, but their usefulness was limited—either because the physician did not employ an electronic heath record or because the patient’s providers could not access the information. HIT offers a promising avenue to combat these systemic shortcomings. The transformative potential of PHRs relies on HIT in two important ways. First, networked PHRs require physicians, hospitals, labs, pharmacies, and other medical care providers to adopt and use health information systems like electronic medical records (EMRs) and electronic health records (EHRs). Second, they require those information systems to share information across institutions and among providers. Or, put another way, PHRs require widespread adoption of records systems that can communicate across health care entities.
The difficulty of implementing this vision for PHRs has revealed several HIT-related challenges facing the current health care system. In Achieving Electronic Connectivity in Healthcare, a report by Connecting for Health and the Markle Foundation, the authors note that it is difficult to create a free- flowing web of information in what is widely acknowledged to be a tremendously diverse and highly fragmented health care system. There are myriad participants in the system—ranging from large hospital networks, to individual providers and physicians, to laboratories, pharmacies and other freestanding units—and patients’ medical records are scattered across these various entities. Some of these participants use EHR software and computer applications to organize patient data, but many do not. Some record systems are “interoperable” (able to communicate with one another), but many are not. Very few providers currently use EHRs—approximately 13 percent reported using a basic EHR and only 4 percent reported using a fully functional EHR, according to recent survey analysis funded by the Robert Wood Johnson Foundation. The rest still rely on paper records, which, of course, are not interoperable. Taken together, these realities point to a health care system in which there is a high degree of fragmentation and a limited degree of interoperability. This hinders the kind of meaningful information sharing that underlies the transformative potential of PHRs.
The complex and (at times) haphazard health care system makes effective information sharing a challenge. But the information stored in health records is also exceptionally diverse, which creates unique obstacles to sharing data effectively. Health information is rich, heterogeneous and rapidly evolving. Clinical records may contain data about medications, dosages, and diagnoses, physician notes, test results, billing codes and images. Sophisticated records systems may also store data captured by biomedical devices (e.g., glucometers) and even some multimedia data like voice or video. The expert panel from Part 3 of the podcast series, “Personal Health Records in a Digital Age,” observes that health information is “not like your bank account, [where] how you describe one dollar or $10,000 is very well defined.” According to the panel, there are many different ways to represent an image, a lab value, or a blood pressure reading, and the fact that different providers record and represent health information differently makes it difficult to share those data among information management systems.
Data standards are an essential piece of the HIT-puzzle, but there are no clear-cut solutions there, either. For example, there is a trade-off between a minutely structured and coded approach to data (which works well for computers), and a more person-friendly approach that allows information to be represented as simple text. Highly structured standards are necessary to cover the gamut of health-related information: symptoms, medications, insurance claims and doctors’ orders. Highly specified data standards are also a requirement for computer-based decision support systems and other aggregate data analyses. Yet, in order for the information to be useful to clinicians and patients, people have to be able to understand it. If a patient needs to “increase dosage to 3mgs daily,” then the computer must be able to translate the reference code supporting that recommendation into English and present it to the patient in context.
A further challenge for creating adequate data standards is finding the right approach to defining and transmitting patient-sourced data like observations of daily living (ODLs). In order for PHRs to effectively communicate with the care system (e.g., EHRs, pharmacies, imaging centers), the data need to be standardized and coded. But the PHR must also have a human vocabulary and be able to communicate with the patient using everyday language. Other HIT challenges include ensuring data quality, providing secure information exchange, and integrating data from multiple sources. Each of these issues presents a challenge for HIT developers trying to keep pace with the rapidly evolving needs of the health care system. (Many of these challenges overlap with policy and workflow questions and we address them in subsequent chapters.)
Recognizing the complexity of these and other HIT challenges, the Pioneer Portfolio funded Connecting for Health and the Markle Foundation’s work to develop the Common Framework. The Common Framework is a suite of standards, policies and methodologies that support “secure connectivity, reliable authentication, [and]…work together to support information exchange.” The set of seven technical guides for medical professionals describes best practices and standards for medication history, lab results and data quality, all of which address aspects of the HIT challenges outlined above.
The Markle Foundation and Connecting for Health have made tremendous strides forward for HIT, and yet the field is still a long way from realizing the vision of an integrated network of seamlessly connected records systems that can interface with PHRs. Partial credit for the slow progress can be attributed to low rates of EHR adoption. (See Chapter 5 of this feature for more on adoption.) On the bright side, the American Recovery and Reinvestment Act (ARRA)—the 2009 stimulus legislation—designates more than $19 billion for advancing HIT and includes cash incentives for physicians and hospitals to adopt EHRs. Of course, only those providers that meet certain requirements will qualify for the ARRA incentives, which has made the “meaningful use” standard a topic of heated debate. The president has set a goal of 100 percent EHR adoption by 2014. While many remain skeptical that everyone will have access to an EHR in the near future, ARRA does stipulate that physicians will be penalized for not implementing an EHR beginning in 2015. What remains to be seen is whether the ARRA incentives and penalties will have a transformative effect on EHR adoption rates.
Further complicating the path forward for HIT is a small but vocal contingent that questions the whole EHR paradigm. According to Pioneer grantee Gordon Moore, M.D., M.P.H., a professor at Harvard Pilgrim Healthcare, Inc., EHRs are overly structured and offer too little benefit to providers to justify their cost. In a white paper supported by the Pioneer Portfolio, Moore argues that HIT is too focused on creating a digitized version of the flawed paper records system. Moore advocates deploying HIT to improve clinical decision support and workflow processes, such as ordering a lab test or transferring responsibility for a patient to a specialist. Moore and others, including the National Research Council, have suggested that EHRs do not supply doctors with the kind of cognitive support (i.e., diagnostic algorithms, clinical guidelines) that would prove truly revolutionary. In its report, Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions, the National Research Council claims that, even when they have fully functional EHRs, clinicians are forced to spend precious time and cognitive energy “sifting through raw data and trying to integrate these data with their general medical knowledge.” Others, like Kenneth Mandl, M.D., M.P.H., and Isaac Kohane, M.D., Ph.D, both professors at Harvard Medical School, have argued that the EHR structure is not flexible enough or modular enough. In their article in the New England Journal of Medicine, “No Small Change for the Health Information Economy,” Kohane and Mandl recommend separating the health record platform from the functional health management applications. Read more about the “platform/app” model in Chapter 7 of this Web feature.
Impressive challenges and opportunities lie ahead for personal health records. Health information management systems, electronic health records, and other HIT tools must continue to evolve in order to transcend the technical challenges discussed in this chapter. Widespread adoption of interoperable systems is a crucial step toward realizing this vision for PHRs and HIT. Imperfect though the current tools may be, the Pioneer Portfolio believes that path toward improving those systems is through adoption and feedback from physicians, hospitals and patients. Chapter 5 and Chapter 6 discuss some of the challenges ahead for HIT adoption. They examine the incentives driving (and hindering) adoption and delve into the prevalent policy issues—like privacy, liability and reimbursement—shaping those incentive structures. Chapter 7 examines the business case for PHRs and looks to recent entrants to the realm of health and health care, like Google and Microsoft, to uncover the potential of the open platform for stimulating adoption and innovation in HIT.