In response to a request from the National Library of Medicine (NLM), and with support from the Warren Grant Magnuson Clinical Center of the National Institutes of Health and from the Massachusetts Health Data Consortium, the Computer Science and Telecommunications Board (CSTB) initiated a study in October 1995 on maintaining privacy and security in health care applications of the national information infrastructure (NII). As one of the lead agencies within the executive branch for facilitating the development and expansion of health care applications of the NII, NLM identified privacy and security as primary issues that need to be addressed in order to facilitate greater use of information technology within the health care sector.¹ Several reports written over the last two decades note the potential vulnerabilities of health information systems and the potential risks to patient privacy that could result from the unauthorized use of health data.² Although they outline risks and discuss possible corrective measures, these earlier reports do not attempt to evaluate the effectiveness of alternative mechanisms for protecting electronic health information. To remedy this situation, CSTB was asked to investigate the threats to electronic health information and to evaluate alternative technical and nontechnical means of protecting health information that are being used today. A natural outgrowth of this assessment is a judgment about the technical and nontechnical means that can be used to maintain privacy and security in health care information systems, about future research that is needed to develop additional mechanisms, and about the obstacles that stand in the way of future advances.

THE COMMITTEE AND ITS CHARGE

To conduct this study, CSTB formed a committee of 15 members and a special advisor with expertise in computer and information security, medical informatics, health information management, health care privacy, law, medical sociology, and health information systems. Both developers and users of health information systems were included. NLM charged the committee to do the following:

Observe and assess existing technical and nontechnical mechanisms for protecting the privacy and maintaining the security of health care information systems, identify other mechanisms worthy of testing in a health care environment, and outline promising areas for further research.

In carrying out this charge, the committee was asked to address questions in the following areas:

· Threats to health care information:   What problems have health care organizations encountered to date regarding unauthorized access to individually identified patient data? To what extent has the security of health information systems been compromised or threatened by the introduction of electronic medical records and networked information systems? What problems could be encountered in the future related to unauthorized access to individually identifiable patient data? How significant is the threat posed by inferential identification through the linking of databases with unidentifiable information?

· Adequacy of existing privacy and security measures:  What types of policies are in place to provide privacy, security, and confidentiality? How adequate are these policies in practice? What technical features are incorporated into health information systems to provide security? How effective are they? What has been done to educate users about the need for privacy and security and their responsibilities for protecting health information?

· Future mechanisms and best practices:  What other approaches to information privacy and security are worthy of testing in health care organizations? What approaches should be broadly promulgated? How cost-effective are various approaches? What combination of technologies, policies, and standards would help to promote better information security for health-related data? How can highly sensitive aspects of an individual's health care records (e.g., mental health history and HIV status) be better protected?

· Barriers to adoption:  What barriers exist to the adoption of better information security practices and technology (e.g., cost, ease of use)? What incentives are needed to encourage providers to adopt sound information privacy and security practices and to secure health information systems?

Although the focus of the committee's charge was to evaluate practices that individual organizations can use to better protect electronic health information, the committee quickly learned from its research that the primary threats to patient privacy originate from the lack of controls over the legal (and generally legitimate) demands for data made by organizations not directly involved in the provision of care, such as managed care organizations, insurers, public health agencies, and self-insured employers. The committee regarded this larger threat as significant enough to warrant systematic attention. Given the committee's original charge and its composition, however, this report does not make specific recommendations in this area, although it does call for a national debate on these issues.³ Accordingly, this report undertakes the tasks of raising consciousness in the health care industry (and the nation as a whole) regarding privacy and security issues in health care; demonstrating ways in which these issues can be addressed; and providing practical guidance to practitioners in the field of medical informatics and health information management who must continually wrestle with privacy and security concerns.

Recognizing that organizations strive to balance security against other concerns such as cost and access to information, the committee investigated both the efficacy of various privacy and security measures and the implications of such measures for the ability of users to access critical information in a timely manner. In order that its work would have lasting value to the community, the committee attempted, where possible, to project future changes in the uses of health information, the potential threats to such information, and the technologies capable of addressing these threats.

The committee's primary mechanism for gathering information about technical and nontechnical approaches to protecting electronic health information consisted of a series of site visits to six organizations that collect, process, and store electronic health information. Sites were selected on the basis of their reputed leadership in the development of electronic medical records, networked clinical systems, and privacy and security policies. Committee nominations were verified against reports or rankings in several highly regarded health care publications. The selected sites included a large, urban hospital; a tightly integrated health care system; a second tightly integrated health care system affiliated with a community health information network; a more loosely affiliated provider network; a state health care system; and a large insurer. To encourage personnel at the various sites to share their experiences candidly, the committee decided to keep sites' identities confidential.

Because site visits were conducted by different subsets of the committee's members, the committee as a whole developed a standard site visit protocol to ensure some degree of uniformity among the visits (see Appendix A). Prior to each visit, the site visit team gathered information from the site regarding its organizational structure, computer and data security policies, information systems, security mechanisms, confidentiality policies, procedures for releasing medical records, employee training and orientation materials, and disciplinary policies. This information proved valuable not only in orienting committee members to idiosyncrasies of each site, but also in indicating the degree to which the organizations had codified their policies and procedures.

During each one-and-a-half-day visit, the site visit team met with corporate executives; staff from the information systems, health information management (i.e., medical records), human resources, and legal departments; doctors; nurses; and other system users. Where possible, it met with members of health information management committees and of privacy and confidentiality committees. The site visit team discussed a wide range of topics with its hosts on each visit: confidentiality policies, policies regarding data exchanges and uses or releases of aggregated data, means of implementing policies, perceived and experienced threats to patient privacy and system security, training and education programs, information systems, electronic medical records, security mechanisms, users' perceptions of the information systems and security practices, and future needs.

In addition to its site visits, the full committee met five times during the course of the study to plan its work, listen to briefings from relevant stakeholders, and deliberate over its conclusions and recommendations. During these sessions, the committee met with health care providers, insurers, pharmaceutical benefits managers,⁴ vendors of health information systems, experts in computer security (from both the health care and the non-health care communities), privacy advocates and consumer representatives, federal agencies interested in health information systems, insurers, relevant industry associations, and other organizations that maintain health-related databases. The committee also met with groups attempting to develop health care applications of the NII and with researchers who study the uses of medical information, including genetic information. Additional meetings were held with the Massachusetts Health Data Consortium and with representatives of European data commissions to understand the problems they face and the solutions they are implementing (see Appendix B for a complete list of people who briefed the committee).

The site visits and committee meetings provided committee members with numerous opportunities to observe and discuss the confidentiality and security policies, mechanisms, and practices used in a variety of health care organizations and firms in other industries. The visits themselves facilitated extended dialogue with key decision makers within the organizations, allowing the committee to better understand the objectives and motivations of the sites' privacy and security strategies. Many of the practices the committee observed during its site visits were described in its interim report released in September 1996.⁵ This final report provides additional analysis of practices observed during the site visits and describes other practices that have not yet been applied in health care settings; describes the general exchanges of health information throughout the industry and identifies obstacles to and incentives for increased attention to privacy and security concerns; and presents the committee's conclusions and recommendations on the state of practice today, on practices that should be more widely adopted throughout the industry, and on research needs for the future.

The committee recognizes that this report will serve multiple audiences: information systems and operations staffs within medical organizations who are charged with developing and implementing practices to improve privacy and security, government agencies and accrediting bodies with roles to play in overseeing health care organizations and other users of health information, and legislators and other policy makers who are interested in establishing a policy framework for protecing health information while allowing legitimate access. The commiteee hopes that each of these audiences will find useful guidance in this report, both in the detailed practices described in Chapters 4 and 5, and in the findings and recommendations contained in Chapter 6.

ACKNOWLEDGMENTS

The committee members that came together represented a number of different backgrounds and perspectives (Appendix E). The harmony with which people with such diverse points of view worked together is testament to the character of the individual committee members and a reflection upon the importance of the issue this report addresses. Each committee member volunteered a substantial amount of time over the course of the study to meet, conduct site visits, and draft sections of this report. To the extent that this report improves the privacy and security environment by enlightening the public, policy makers who set institutional priorities, daily users of health information, and those who build the systems, the committee members will believe that their effort was worthwhile.

To the CSTB staff, the committee expresses its admiration and gratitude for their faithful capture of the wide-ranging content of its deliberations and for the gentle but effective way they kept things on schedule. It is hard to find people of such talent who are willing to facilitate and support rather than impose their observations and conclusions. Thanks are also extended to those who volunteered to review and critique an early draft of this document, as well as to the numerous briefers who volunteered their time to meet with it and to help the committee better understand their concerns. The committee also owes many thanks to those who hosted the site visits. The committee received honest and open cooperation from a variety of individuals at each site.

Finally, the committee wishes to express appreciation to the sponsors of this report who were willing to invest in developing solutions to societal concerns: Dr. Donald Lindberg and Ms. Betsy Humphreys of the National Library of Medicine; Dr. John Gallin of the Warren Grant Magnuson Clinical Center; and Mr. Elliot Stone of the Massachusetts Health Data Consortium. Such leadership is crucial in motivating more than a fragmented approach to the search for solutions.

 ^{1The terms privacy, confidentiality, and security are used in many different ways to discuss the protection of personal health information. This report uses the term privacy to refer to an individual's desire to limit the disclosure of personal information. It uses the term confidentiality to refer to a condition in which information is shared or released in a controlled manner. Organizations develop confidentiality policies to codify their rules for controlling the release of personal information in an effort to protect patient privacy. Security consists of a number of measures that organizations implement to protect information and systems. It includes efforts not only to maintain the confidentiality of information, but also to ensure the integrity and availability of that information and the information systems used to access it.}

^{2See National Institute of Standards and Technology, 1994, Putting the Information Infrastructure to Work: Report of the Information Infrastructure Task Force Committee on Applications and Technology, NIST Special Publication 857, U.S. Government Printing Office, Washington, D.C., May; Institute of Medicine, 1994, Health Data in the Information Age: Use, Disclosure and Privacy, Molla S. Donaldson and Kathleen N. Lohr (eds.), National Academy Press, Washington, D.C.; Office of Technology Assessment, 1993, Protecting Privacy in Computerized Medical Information, OTA-TCT-576, U.S. Government Printing Office, Washington, D.C., September; National Research Council, 1972, Databanks in a Free Society: Computers, Record Keeping, and Privacy, National Academy of Sciences, Washington, D.C.}

^{3Another study committee convened by the Institute of Medicine was charged to investigate systemic uses of health information and to offer recommendations in this area. See Institute of Medicine. 1994. Health Data in the Information Age: Use, Disclosure, and Privacy, National Academy Press, Washington, D.C.}

^{4 Pharmaceutical benefits managers are organizations such as Merck-Medco Managed Care Inc. and PCS Inc. that offer benefits plans that pay for prescriptions. They typically assist in designing the benefits programs, offer point-of-sale claims processing, and develop formularies of the drugs that participating pharmacies prescribe.}

⁵ Computer Science and Telecommunications Board, National Research Council. 1996. "Observed Practices for Improving the Security and Confidentiality of Electronic Health Information: Interim Report," National Academy Press, Washington, D.C., September.