1
Introduction

The patient record is the principal repository for information concerning a patient's health care. It affects, in some way, virtually everyone associated with providing, receiving, or reimbursing health care services. Despite the many technological advances in health care over the past few decades, the typical patient record of today is remarkably similar to the patient record of 50 years ago. This failure of patient records to evolve is now creating additional stress within the already burdened U.S. health care system as the information needs of practitioners,1 patients, administrators, third-party payers, researchers, and policymakers often go unmet. As described by Ellwood (1988:1550),

The intricate machinery of our health care system can no longer grasp the threads of experience… Too often, payers, physicians, and health care executives do not share common insights into the life of the patient… The health care system has become an organism guided by misguided choices; it is unstable, confused, and desperately in need of a central nervous system that can help it cope with the complexities of modern medicine.

Patient record improvement could make major contributions to improving the health care system of this nation. A 1991 General Accounting Office (GAO) report on automated medical records identified three major ways in which improved patient records could benefit health care (GAO,

1  

The committee uses the term practitioners to refer to all health care professionals who provide clinical services to patients. These professionals include, but are not limited to, physicians, nurses, dentists, and therapists.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 52
--> 1 Introduction The patient record is the principal repository for information concerning a patient's health care. It affects, in some way, virtually everyone associated with providing, receiving, or reimbursing health care services. Despite the many technological advances in health care over the past few decades, the typical patient record of today is remarkably similar to the patient record of 50 years ago. This failure of patient records to evolve is now creating additional stress within the already burdened U.S. health care system as the information needs of practitioners,1 patients, administrators, third-party payers, researchers, and policymakers often go unmet. As described by Ellwood (1988:1550), The intricate machinery of our health care system can no longer grasp the threads of experience… Too often, payers, physicians, and health care executives do not share common insights into the life of the patient… The health care system has become an organism guided by misguided choices; it is unstable, confused, and desperately in need of a central nervous system that can help it cope with the complexities of modern medicine. Patient record improvement could make major contributions to improving the health care system of this nation. A 1991 General Accounting Office (GAO) report on automated medical records identified three major ways in which improved patient records could benefit health care (GAO, 1   The committee uses the term practitioners to refer to all health care professionals who provide clinical services to patients. These professionals include, but are not limited to, physicians, nurses, dentists, and therapists.

OCR for page 52
--> 1991). First, automated patient records can improve health care delivery by providing medical personnel with better data access, faster data retrieval, higher quality data, and more versatility in data display. Automated patient records can also support decision making and quality assurance activities and provide clinical reminders to assist in patient care. Second, automated patient records can enhance outcomes research programs by electronically capturing clinical information for evaluation. Third, automated patient records can increase hospital efficiency by reducing costs and improving staff productivity. Several sources support these conclusions. The GAO reported that an automated medical record system reduced hospital costs by $600 per patient in a Department of Veterans Affairs hospital because of shorter hospital stays (GAO, 1991). Reductions in the length of inpatient stays were also found in other studies of computerized medical records and medical record summaries (Rogers and Haring, 1979). Other investigators found enhanced care and improved outcome of care for clinic patients (Rogers et al., 1982) and a reduction in medication errors (Garrett et al., 1986). The first step toward patient record improvement is a close examination of the users of the patient record, the technologies available to create and maintain it, and the barriers to enhancing it. To that end, the Institute of Medicine (IOM) of the National Academy of Sciences undertook a study to recommend improvements to patient records in response to expanding functional requirements and technological advances. 2 This report is the product of the multidisciplinary panel's 18-month study of how patient records can be improved to meet the many and varied demands for patient information and to enhance the quality of patient care and the effectiveness and efficiency of health care delivery. The Study The idea for this study originated in discussions between staff at the National Institutes of Health (NIH) and IOM. The NIH staff were involved in patient care, teaching, and research and were motivated by the need to make patient records more useful for all of these purposes. The IOM 2   The IOM committee was originally named the Committee on Improving the Medical Record in Response to Increasing Functional Requirements and Technological Advances. The committee's first action was to change "medical" to "patient" in its name, reflecting its consensus that the medical component of the record does not constitute the total record. Thus, this report generally will refer to what are commonly called medical records as "patient records.'' There are several instances, however, in which the committee continues to refer to medical records rather than patient records. For example, the committee's official charge relates to medical records, and the committee has not undertaken to rename "medical record professionals."

OCR for page 52
--> was considered to be an ideal agent to bring together the diverse perspectives needed to address a broad range of patient record issues. Thus, the participants at a June 1986 IOM program development workshop recommended that the institute conduct a study of the patient record in light of new technologies. Subsequently, staff of the IOM Council on Health Care Technology developed an action plan for the study, which was approved by the National Research Council in July 1987. Efforts to enlist adequate financial support occurred over the ensuing two years. The IOM appointed a study committee in March 1989, and the committee began its deliberations the following September. Among its membership were experts in community and academic medicine, health information services, health services research, hospital services, medical information systems, regulatory functions, and third-party payment. The Committee's Charge In general, the IOM study committee was charged to examine the problems with existing medical record systems and to propose actions and research for their improvement in light of new technologies. Specifically, the committee was asked to: examine the current state of medical record systems, including their availability, use, strengths, and weaknesses; identify impediments to the development and use of improved record systems; identify ways, including developments now in progress, to overcome impediments and improve medical record systems; develop a research agenda to advance medical record systems; develop a plan, design, and/or other provisions for improved medical record systems, including a means for updating these systems and the research agenda as appropriate; and recommend policies and other strategies to achieve these improvements. In addition to addressing the technological issues in its charge, the committee sought to produce a report that would increase the interest of all health care practitioners in improving patient records and health care information management. Involvement of these practitioners in the development of future patient records is required if record improvement efforts are to meet with success. Committee Activities The committee met five times between September 1989 and December 1990. It established three subcommittees—Users and Uses, Technology,

OCR for page 52
--> and Strategy and Implementation (see Appendix A). The subcommittees, each with approximately 15 members, met at least twice and solicited information from more than 70 advisers, including physicians (in both private practice and academic medicine), nurses, dentists, medical record professionals, hospital administrators, researchers, and congressional staff. Also among these advisers were representatives of patient groups, computer software and hardware vendors, third-party payers, government agencies, and professional organizations. Each subcommittee prepared a report that was considered, along with the results of a special workshop and several background papers, by the full committee in its deliberations. Definitions During its work, the committee used the following specific definitions: A patient record is the repository of information about a single patient. This information is generated by health care professionals as a direct result of interaction with a patient or with individuals who have personal knowledge of the patient (or with both). Traditionally, patient records have been paper and have been used to store patient care data. A computer-based patient record (CPR) is an electronic patient record that resides in a system specifically designed to support users by providing accessibility to complete and accurate data, alerts, reminders, clinical decision support systems,3 links to medical knowledge, and other aids. A primary patient record is used by health care professionals while providing patient care services to review patient data or document their own observations, actions, or instructions. A secondary patient record is derived from the primary record and contains selected data elements to aid nonclinical users (i.e., persons not involved in direct patient care) in supporting, evaluating, or advancing patient care.4 Patient care support refers to administration, regulation, and 3   Shortliffe and colleagues define a clinical decision support system as "any computer system designed to help health professionals make clinical decisions" (Shortliffe et al., 1990:469). They identify three types of decision support functions: information management, focusing of attention, and patient-specific consultation. Throughout this report, clinical decision support systems refer to clinical consultation systems that use population statistics or encode expert knowledge to assist practitioners in diagnosis or in formulating treatment plans (Shortliffe et al., 1990). 4   The committee's distinction between primary and secondary patient records parallels, but is not identical to, the approach used by Westin (1976). Westin identified three "zones" through which information flows: (1) direct patient care, (2) supporting activities, and (3) social uses of health data. This report does not address social uses of patient care data that lie outside health care (e.g., law enforcement) and combines supporting activities and health care social uses of patient data into one zone.

OCR for page 52
--> payment functions. Patient care evaluation refers to quality assurance, utilization review, and medical or legal audits. Patient care advancement refers to research. These records are often combined to form what the committee terms a secondary database (e.g., an insurance claims database). A patient record system is the set of components that form the mechanism by which patient records are created, used, stored, and retrieved. A patient record system is usually located within a health care provider setting. It includes people, data, rules and procedures, processing and storage devices (e.g., paper and pen, hardware and software), and communication and support facilities. A patient record system can be part of a hospital information system , which typically handles both administrative and clinical functions, or a medical information system, which has been defined as "the set of formal arrangements by which the facts concerning the health or health care of individual patients are stored and processed in computer" (Lindberg, 1979:9). A patient record system is a type of clinical information system, which is dedicated to collecting, storing, manipulating, and making available clinical information important to the delivery of patient care. The central focus of such systems is clinical data and not financial or billing information. Such systems may be limited in their scope to a single area of clinical information (e.g., dedicated to laboratory data), or they may be comprehensive and cover virtually every facet of clinical information pertinent to patient care (e.g., computer-based patient record systems). Report Organization The remainder of this chapter discusses the current state of patient record systems, including their strengths and weaknesses, and the environment of opportunity that exists for implementing computer-based patient records. Chapter 2 delineates the needs of patient record users and describes how future patient record systems can meet user needs. Chapter 3 identifies technologies essential to future systems and assesses how well existing systems meet future requirements. Chapter 4 describes nontechnological barriers to improving patient records and presents a strategic plan for overcoming them. Finally, Chapter 5 sets forth the committee's recommendations for accelerating the realization of computer-based patient records and suggests an agenda for their implementation and dissemination within a decade. The Patient Record Virtually every person in the United States who has received health care since 1918 has a patient record (MacEachern, 1937). Today, most people

OCR for page 52
--> have multiple patient records—one for each health care provider they have visited. Patient records have proliferated to the extent that some medical centers in large metropolitan areas may now each have more than 4 million paper patient records (Kurland and Molgaard, 1981). Although at any one time these records are not all active, they must be stored for up to 25 years, depending on state laws (Waller, in this volume). Moreover, a given patient may have more than one record even within a particular institution.5 Patient records appear in a variety of forms—for example, as paper; microfilm; a monitor strip; an optical disk; a computer card, tape, or disk; or a combination of these (Amatayakul and Wogan, 1989). They are created and used most frequently in health care provider settings such as physician or dentist offices, hospitals, nursing homes, and public health clinics; but other institutions such as correctional institutions, the armed forces, occupational health programs of employers, and colleges and universities also maintain patient health care records (Westin, 1976).6 For more than a century, the paper patient record has been the primary vehicle for recording patient care information (Huffman, 1981). Yet recent years have seen a trend toward automation of components of patient records (e.g., clinical laboratory test results) or certain patient care functions (e.g., entering physician orders for hospitalized patients; Westin, 1976; Gardner and Perry, 1989; Amatayakul and Sattler, 1990). Some hospitals, health maintenance organizations, and physicians' offices now have prototype elements of computer-based record systems .7 These facilities are still the exception rather than the rule, however, and paper patient records, with their sometimes overlooked strengths and frequently cited weaknesses, are still the norm for most health care providers. Strengths and Weaknesses of Paper Patient Records The committee's literature review did not reveal any substantive documentation of the strengths of paper patient records. This result may be 5   Health care professionals might maintain a separate patient record to protect sensitive data (e.g., psychiatric history) or to support a research interest (i.e., separate records containing detailed data for a research project). 6   Pharmacies capture information pertinent to patient care but do not maintain full patient records. Information on the medications prescribed and the specialties of the physicians writing the prescriptions can provide enough information to determine a patient's medical problems, however, and pharmacy records may thus raise confidentiality issues similar to those associated with patient records. Because the committee focused more closely on traditional patient care records, this report does not address issues related to pharmacy records. 7   Chapter 3 highlights some of the health care provider institutions at the leading edge of patient record technologies.

OCR for page 52
--> explained in part by the facts that the value of maintaining patient records is widely accepted in the health care community and that paper is the most widely used record keeping form. Given the prevalence of paper patient records, the committee noted that support by practitioners for this kind of record keeping should not be underestimated. Time and resource constraints did not permit the committee to survey user attitudes toward paper records; however, committee members identified at least five strengths of such records from the perspective of record users: Paper records are familiar to users who consequently do not need to acquire new skills or behaviors to use them. Paper records are portable and can be carried to the point of care. Once in hand, paper records do not experience downtime as computer systems do. Paper records allow flexibility in recording data and are able to record "soft" (i.e., subjective) data easily.8 Paper records can be browsed through and scanned (if they are not too large). This feature allows users to organize data in various ways and to look for patterns or trends that are not explicitly stated. Criticism of current patient records is sometimes sharp. Burnum (1989:484) states that "medical records, which have long been faulty, contain more distorted, deleted, and misleading information than ever before." Pories (1990:47) relates the story of an engineer who was asked to recommend more efficient use of health care personnel but who instead was "stunned by the disorganization of the medical record and the inefficiencies it imposed on the delivery of care." The engineer concluded that "the redesign of the record offered the most immediate and simple approach for medical cost control and for prevention of malpractice" (p. 47). Pories believes that this situation has not improved and that it is not isolated. "No one has a monopoly on the problem: medical records appear to be equally bad and dangerous throughout the land" (Pories 1990:47). He is not alone in his view that patient records often lack the features needed for their most beneficial use. In a recent survey of internists in academic and private practice, 63 percent of the respondents agreed with the statement that patient records are becoming increasingly burdensome without improving the quality of patient care (Hershey et al., 1989). The weaknesses of patient records, as described in the literature and in the work of the committee, can be subsumed under four main headings: (1) 8   Although flexibility in recording data may be viewed as a strength by the individual recording the information, lack of standard vocabulary and coding can pose problems for subsequent users—including practitioners, administrators, researchers, and third-party payers.

OCR for page 52
--> content; (2) format; (3) access, availability, and retrieval; and (4) linkages and integration. Problems with Patient Record Content Patient record data are often missing, illegible, or inaccurate (Tufo and Speidel, 1971; Zuckerman et al., 1975; Bentsen, 1976; Zimmerman, 1978; Fox et al., 1979; Romm and Putnam, 1981; Gerbert and Hargreaves, 1986; Hsia et al., 1988; Pories, 1990). Data can be missing for at least three reasons: (1) questions were never asked, examinations were never performed, or tests were never ordered; (2) the information was requested and provided, but either it was not recorded by the clinician or delays occurred in placing the information in the record; and (3) the information was requested and delivered but was misplaced or lost. In addition, clinicians, patients, or equipment can all introduce errors into patient records (Burnum, 1989). Many studies have examined the quality of patient record content. Table 1-1 presents the findings of several such investigations. The missing information reported in the various studies often resulted in additional costs of patient care. For example, an estimated 11 percent of laboratory tests in one hospital were ordered to duplicate tests for which findings were unavailable to the physician at the time of the patient visit (Tufo and Speidel, 1971). Although for some records data are missing, in other cases certain data are excessive or redundant (Zimmerman, 1978; Korpman and Lincoln, 1988). The thickness and weight of the records of patients with chronic problems can be imposing, if not daunting, and time constraints may prevent the user from finding and using necessary information. (In one study of paper patient records, the average weight of a clinic record was 1-1/2 pounds [Rogers et al., 1982].) Other issues related to record content include failure to capture the rationale of providers, lack of standardization of definitions of terminology, failure to describe the patient experience, lack of patient-based generic health outcome measures, and incomprehensibility for patients and their families. Problems with Format Several studies have pointed to patient record formats as a problem area that at times impedes record use and effectiveness. The 1980 IOM study cited in Table 1-1 found that the reliability of hospital discharge data depended on the general organization, orderliness, and logic of the medical record. Zimmerman (1978) conducted a survey of physicians who identified poor organization of medical records as a deficiency that contributed

OCR for page 52
--> TABLE 1-1 Selected Parameters and Findings of Studies of Patient Record Content Author(s)/Parameters Study Data Tufo and Speidel (1971) Purpose Evaluate record availability, missing data, recording of laboratory results, incomplete physician narrative, and data collected for general health evaluations Sample 1,149 patient visits in five outpatient U.S. Army facilities Findings 11% of patients had no past medical data available   5–20% of charts had information missing:   75% of missing data were laboratory test results or reports of radiologic examinations   25% of missing data were lost, incomplete, or illegible data from previous visits   13–79% of laboratory results were not placed in the record   10–49% of visits did not have a well-defined problem in the record   6–49% of visits did not have a well-defined treatment in the record   40–73% of records did not have evidence of general medical information useful for preventive medicine Dawes (1972) Purpose Determine presence of 18 data elements Sample 1,628 medical records (the last episode of disease) in general practices Findings 10% of patient ages were not recorded   30% of episodes had no therapeutic agent recorded; of those recorded, 75% were missing the amount prescribed, and 80% were missing dosages   40% of episodes had no diagnosis recorded   60% of males and 77% of females had no occupation recorded   99% of males and 21% of females had no marital status recorded Zuckerman et al. (1975) Purpose Appraise the extent to which records document adequately the content of verbal communication between physicians and patients Sample 51 tape-recorded physician-patient encounters in pediatric clinic Findings Percent present on tape and absent on record:   0% of chief complaints   6% of reason for visit   10% of degree of disability   12% of allergies   22% of compliance data   31% of indications for follow-up   51% of cause of illness Bentsen (1976) Purpose Assess the validity of data in the information system of the department Sample 59 patient encounters in family medicine clinics Findings 41% of problems identified by observers were not recorded

OCR for page 52
--> Author(s)/Parameters Study Data IOM (1980) Purpose Assess the reliability of data collected as part of the National Hospital Discharge Survey Sample 3,313 medical records from 66 hospitals that participated in the National Hospital Discharge Survey Findings 75% of face sheets had no discharge disposition   48% of face sheets were inadequate for determining principal diagnosis   15% of face sheets and discharge summaries were inadequate for determining principal diagnosis Romm and Putnam (1981) Purpose Document extent of agreement between the record and the verbal content of the physician-patient encounter Sample 55 patient encounters in general medicine clinics Findings Percent agreement between record and observation of encounter:   29% for other medical history   66% for therapy   71% for information related to current illness   72% for tests   73% for impression/diagnosis   92% for chief complaint Hsia et al. (1988) Purpose Verify coding of diagnosis-related groups (DRGs) Sample 7,050 medical records of Medicare patients from 239 hospitals Findings 20.8% of the discharges were coded incorrectly (in the direction of higher weighted DRGs) to ineffective use. The format of the usual paper record does not lend itself easily to dealing with multiple problems over a long period of time. For example, Pories (1990) noted that the physical form of paper records (i.e., binders or charts) is often unmanageable; data are fragmented within the record and not sorted for relevance. Fries (1974) showed that information could be found four times faster in a structured flow sheet than in a traditional paper medical record and that 10 percent of items in the traditional record could not be found. Traditionally, patient records are organized according to the sources and chronology of data (Feinstein, 1970), although several alternative record formats have been developed. For example, the problem-oriented medical record (POMR) is assembled according to the patient's problems to support a more organized approach to clinical problem solving and management

OCR for page 52
--> (Weed, 1968). The summary time-oriented record (STOR) is an outpatient medical record system that consists of a concise summary of a patient's clinical data that can be used either in conjunction with the traditional medical record or by itself (Whiting-O'Keefe et al., 1985). Problems with Access, Availability, and Retrieval Record unavailability and difficulties in accessing records when they are available are frequent problems for patient record users (Pories, 1990). Tufo and Speidel (1971) documented in their study that medical records were unavailable in up to 30 percent of patient visits. They attributed this rate of unavailability to several possible causes: patients being seen in two or more clinics on the same day, charts not being forwarded, physicians keeping records in their offices or removing them from their offices, and records being misfiled in the file room. One hospital in the GAO study on automated medical records reported that it could not locate medical records 30 percent of the time (GAO, 1991). Even when records are readily available, the amount of time required to retrieve necessary information from a record can frustrate users (Fries, 1974; Zimmerman, 1978; Pories, 1990). For researchers, access to paper records can be problematic and is generally resource intensive (Davies, 1990). Identifying records that contain needed data, retrieving needed records, reviewing records, collecting data, and entering data into data sets for analysis are time-consuming, expensive tasks. Yet access to existing computer-based records can also prove difficult for researchers because documentation on how to use systems may be lacking. Further, data aggregation can be hampered by lack of compatibility among systems. Problems with Linkages and Integration One of the major criticisms of the U.S. health care system is the discontinuity of care among providers (Rulin et al., 1988; Case and Jones, 1989). This discontinuity extends to patient records, whose lack of integration of inpatient and outpatient information is a significant deficiency. Paper patient records offer little hope of improving the coordination of health care services within or among provider institutions. Moreover, the inadequacy of patient record interfaces with other clinical data, administrative information, or medical knowledge impedes optimal use of record information in providing patient care. Several health care systems and institutions have developed records that overcome many of the problems associated with traditional paper records, but even these improved records suffer from their lack of easy transferability to other health care provider systems or institutions.

OCR for page 52
--> Outpatient Records Attention is frequently focused on patient records in hospitals rather than in outpatient settings. (An inpatient record is used by many different individuals during an episode of illness, so its weaknesses can appear quite pronounced.) Yet outpatient records are greater in number, are scattered among individual physician offices, and may exhibit even greater variance in quality than inpatient records. There are no established standards or review organizations for outpatient records as there are for inpatient records. As a result, outpatient records often serve as files for ''correspondence and reports rather than as a well organized chronology of health care" (Pories, 1990:49). Ambulatory care records frequently contain poorly organized data, lack documentation of key aspects of care, and exhibit inaccurate diagnostic coding (IOM, 1990c). Health care researchers and clinicians who conduct retrospective studies using such records are likely to identify at least four weaknesses: lack of standardization in content and format, inaccessibility (except in some hospitals or large health plans), incompleteness, and inaccuracies (Davies, 1990). Information Management Challenges An Information-Intensive Industry Providing high-quality health care services is an information-dependent process. Indeed, the practice of medicine has been described as being "dominated" by how well information is processed or reprocessed, retrieved, and communicated (Barnett, 1990). An estimated 35 to 39 percent of total hospital operating costs have been associated with patient and professional communication activities (Richart, 1970). Physicians spend an estimated 38 percent (Mamlin and Baker, 1973) and nurses an estimated 50 percent (Korpman and Lincoln, 1988) of their time writing up patient charts. Information-processing activities associated with providing health services to patients are extremely varied. Clinicians obtain and record information about patients, consult with colleagues, read scientific literature, select diagnostic procedures, interpret results of laboratory studies, devise strategies for patient care, instruct allied health professionals, discuss care plans with patients and families, and document the progress of patients. In addition, health care practitioners must distill knowledge, interpret data, apply knowledge, and manage the complexities of medical decision making (Haynes et al., 1989; Greenes and Shortliffe, 1990). Thus, health care professionals routinely need access to appropriate compilations of thorough, up-to-date knowledge and advice to make prompt, informed decisions regarding

OCR for page 52
--> patient care (Haynes et al., 1989; Saba et al., 1989; Greenes and Shortliffe, 1990). Furthermore, a wide range of information-processing tasks supports patient care, including performing laboratory tests, processing medical imaging data, capturing patient demographic information, filling prescription orders, monitoring quality and appropriateness of services provided, and billing (Martin, 1990). An Information Explosion For the past several years, health care practitioners have faced an explosion in medical knowledge. For instance, MEDLINE9 now indexes approximately 360,000 new articles each year from those published in the biomedical literature (National Institutes of Health, 1989). Scientific and technological advances that have contributed to improving the health of the public have also resulted in more complexity in medical practice; clinicians must track ever more numerous diagnoses, procedures, diagnostic tests, clinical processes, devices, and drugs. The increase in available technologies places an additional burden on physicians: they must read and synthesize the literature and try to decide whether and how new technologies should be applied (Brook, 1989). Often, a gap occurs between the information physicians need and the information available to them at the time of providing patient care. According to Covell and colleagues (1985), an estimated 70 percent of physician information needs are unmet during the patient visit. In addition to more knowledge, there are more data for health care professionals to manage. The volume and complexity of information per patient have increased owing to a greater number of patient encounters (as patients live longer and experience chronic disease), higher patient acuity of illness (both in and out of hospitals), more kinds of clinical data elements arising from new diagnostic technologies, and developments in the delivery system that result in many patients receiving care at multiple sites. Concomitant with increases in information, however, have been efforts to reduce unit costs of health care provider institutions, which create pressures on health care professionals to be more productive and to see more patients. This paradoxically reduces the time and energy available for the functions associated with management and communication of information. Health care professionals are thus placed in a frustrating, difficult, and perhaps untenable position. 9   MEDLINE is an on-line bibliographic database of medical information. It covers 25 years and includes citations to more than 6 million articles from approximately 3,500 journals. MEDLINE is part of the National Library of Medicine Medical Literature Analysis and Retrieval System (MEDLARS), which includes specialized databases in health administration, toxicology, cancer, medical ethics, and population studies (National Institutes of Health, 1989).

OCR for page 52
--> Increasing Demand for Data Even as patient care data become more voluminous and complex, the demand by multiple users for access to patient care data is increasing (Barnett, 1990). Information must be shared among the multiplicity of health care professionals who constitute the "health care team." These professionals represent the physician specialties, as well as nurses, dentists, therapists, pharmacists, technicians, social workers, and other health care providers. Patients may also require access to records; some providers advocate greater patient input into the process of care through patient identification of preferences among treatments, patient contributions to the record (particularly history and functional status; Davies and Ware, 1988; Donnelly, 1988; Tarlov et al., 1989), patient reading and validation of record data, and patient control and transport of pertinent parts of the record (Tufo et al., 1977; Bronson et al., 1978). Administrators and managers of health care institutions require information to manage the quality of care provided and to allocate resources (e.g., labor, supplies, equipment, and facilities) according to the institution's patient case mix. Managers of provider institutions seek to link financial and patient care information to develop meaningful budgets, measure productivity and costs, and evaluate market position. Long-term institutional planning for personnel recruitment, equipment acquisition, and facilities development depends on anticipated trends in patient population needs. Quality assurance activities constitute another information need. Such activities are a requirement for accreditation of hospitals by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO); in addition, third-party payers carry out various quality monitoring and evaluation efforts. The best known is probably the Medicare peer review organization program administered by the Health Care Financing Administration (IOM, 1990c). Public and private third-party payers, medical professional societies, and researchers have been exploring practice guidelines and outcomes management as tools for improving care (IOM, 1990a,b). The risk management programs established by many health care institutions in response to the recent history of medical malpractice litigation add another level of information use (IOM, 1990c). Patient care information to adjudicate claims for reimbursement made to third-party payers is an additional area of data needs. As expenditures related to health care have risen10 and as third-party payers have sought to 10   Health care expenditures accounted for 7.3 percent of the nation's gross national product (GNP) in 1970 (Levit et al., 1990). In 1989, health care spending amounted to 11.6 percent of GNP (U.S. Department of Health and Human Services, 1990).

OCR for page 52
--> contain costs, payers have also increased their demands for data. Patient data now are used for coverage decisions (e.g., preadmission review) as well as for payment. Cost-containment approaches such as utilization management rely on individual patient data for making short-term decisions; they also rely on aggregated data to make judgments about the effectiveness of medical services in the long-term (IOM, 1989). A recently developed hybrid of quality assurance and cost containment—"value purchasing"—refers to the effort by purchasers, providers, and consumers to promote "the idea of value or quality within the context of cost" (IOM, 1990c:36). Value purchasing requires access to information on both costs and outcomes of care. The public health arena is also seeking more and better data. The 1988 IOM report on the future of public health recommended that a uniform national data set be established to permit valid comparison of local, state, and national health data to facilitate progress toward achieving national health objectives (IOM, 1988). More recently, the U.S. Public Health Service identified public health surveillance as a primary means of supporting the national disease prevention objectives for the year 2000 (U.S. Public Health Service, 1990). Data are needed to understand the health status of the U.S. population and to develop, administer, and evaluate public health programs aimed at controlling and preventing adverse health events. The Public Health Service report specifically mentions data on (1) mortality, morbidity, and disability from acute and chronic conditions; (2) injuries; (3) personal, environmental, and occupational risk factors associated with illness and premature death; (4) preventive and treatment services; and (5) costs. The issues of quality, cost, effectiveness, and appropriateness largely frame the questions that today's clinical and health services researchers pursue. Efficacy and safety are no longer sufficient criteria for assessing a technology whose purchasers also want to know (1) whether it is effective and safe outside the controlled environment of clinical trials, (2) whether it is cost-effective, and (3) whether it produces desired outcomes. "[L]ongitudinal observations of natural variations in use and outcomes of economically and clinically important medical technologies in different practice situations" are sought to support utilization management (IOM, 1989:158).11 Late in 1989, the Omnibus Budget Reconciliation Act (P.L. 101–239) established the Agency for Health Care Policy and Research (AHCPR) to enhance "the quality, appropriateness, and effectiveness of health care services, and access to such services, through the establishment of a broad 11   One expert has suggested that "[w]hat is needed is a new kind of trial, one that combines randomized prescription of approved drugs and hands-off follow-up with recording of medical outcomes and determination of costs from routinely generated computerized patient records" (Paterson, 1988:112).

OCR for page 52
--> base of scientific research and through the promotion of improvements in clinical practice and in the organization, financing and delivery of health care services" (U.S. Congress, 1989). In addition to its mandate to "conduct and support research, demonstration projects, evaluations, training, and the dissemination of information, on health care services and on systems for the delivery of such systems," AHCPR is specifically directed to support the improvement and supplementation of existing databases and the design and development of new databases for use in outcomes and effectiveness research. More recently, several IOM studies have endorsed efforts to support, expand, and improve research and the knowledge base on efficacy, effectiveness, and outcomes of care (IOM, 1990b,c). Such efforts are part of a systematic endeavor to develop clinical practice guidelines and standards of care. All of these activities underscore the vastly increased demand for patient data that has emerged during the 1980s. Maintaining Confidentiality In contrast to these trends—an increased supply of and demand for patient data—is the absolute necessity to protect patient privacy. 12 The ancient principle of confidentiality—the obligation of health care professionals to avoid violating a patient's right to privacy—is affirmed by the American Medical Association (AMA) Council on Ethical and Judicial Affairs (1989).13 Thus, a significant challenge in creating future patient record systems is to achieve an appropriate balance between confidentiality and access by users with a need to know.14 12   Privacy is the right of individuals to be left alone and to be protected against physical or psychological invasion or the misuse of their property. It includes "freedom from intrusion or observation into one's private affairs; the right to maintain control over certain personal information; and the freedom to act without outside interference" (Peck, 1984). 13   "And whatsoever I shall see or hear in the course of my profession, as well as outside of my profession in my intercourse with men, if it be what should not be publicized abroad, I will never divulge, holding such things to be holy secrets" (Hippocratic Oath, as quoted in Small, 1989). According to the AMA, "[t]he information disclosed to a physician during the course of the relationship between physician and patient is confidential to the greatest extent possible. The patient should feel free to make a full disclosure of information to the physician in order that the physician may most effectively provide needed services. The patient should be able to make this disclosure with the knowledge that the physician will respect the confidential nature of the communication" (AMA Council on Ethical and Judicial Affairs, 1989:21). 14   Concerns about the privacy of information are not limited to health care data. The 1977 report, Personal Privacy in an Information Society , cited consumer credit, depository, insurance, employment, and education data as all needing protection (Privacy Protection Study Commission, 1977).

OCR for page 52
--> Information Management Opportunities Some health care provider institutions are well on their way toward developing new electronic approaches to patient records. This progress has been possible because of advances in computer technology and successful research on the application of such technology to medicine. These advances include, but are not limited to, hand-held computers; picture archiving and communications systems that permit electronic storage, transmission, and display of medical images; disk drives that offer mean time between failures of 60,000 or more hours; and high-speed telecommunications networks that can carry 1.7 billion characters per second over a 100-kilometer link. New technologies and applications such as these can potentially improve the quality of patient care, advance the scientific basis of medicine, moderate the costs of health care services, and enhance the education of health care professionals. Computer-based patient records, as defined by the committee, could positively affect the quality of patient care in at least four ways. First, they offer a means of improving both the quality of and access to patient care data. Second, they allow providers to integrate information about patients over time and between settings of care. Third, they make medical knowledge more accessible for use by practitioners when needed. Fourth, they provide decision support to practitioners. Research efforts could also benefit from computer-based patient record keeping in two key ways. First, improved data and access to those data would be available to researchers. Second, research findings could be communicated to practitioners through computer-based patient record systems Computer-based patient records could assist efforts to moderate the costs of health care in three ways. First, improved information could reduce redundant tests and services that are performed in response to the unavailability of test results. Second, administrative costs could be reduced by electronic submission of claims and the ability to generate routine reports automatically. Third, the productivity of practitioners could be enhanced by (1) reducing the time needed to find missing records or to wait for records already in use; (2) reducing, if not eliminating, the need for redundant data entry; and (3) reducing the time needed to enter or review data in records that have been streamlined to eliminate unnecessary information. Health care professionals not only encourage self-directed learning among their students (Gastel and Rogers, 1989) but are viewed as having a responsibility for continuing lifelong education (AAMC, 1984). Students thus require skills in organizing information and solving problems. Computer-based patient records can support information management and independent learning by health care students and professionals in both patient care and clinical research settings. Tools for such learning include clinical decision

OCR for page 52
--> support systems, bibliographic and knowledge links (including links to clinical practice guidelines), and statistical software. Beyond Technology Meeting the challenge of managing health care information depends on more than technological advances. The usefulness of any technology depends on how well it and its progeny are applied. In addition to technology, a study of the improvement of patient records must address how the use of those records might be improved, a question that raises potentially sensitive issues.15 "Improving records" and "improving clinical reasoning" are topics inevitably connected to one another because ideally the record reflects the clinical reasoning process. If better record systems are to be created in the future, the user must be recognized as part of the system, and the problem solving activities of practitioners must be examined. In addition to technological and behavioral opportunities for improving patient records, certain strategic issues must be addressed. Other information-intensive industries (e.g., banking) have successfully implemented widespread computer-based information management technologies. Understanding the factors that have slowed the development and diffusion of such technologies in health care is a first step toward achieving more rapid advances in the future. Why Now? Many attempts have been made over the years to advance clinical computing, to reform the patient record, and to encourage health care professionals to maintain the record more conscientiously (e.g., by entering necessary clinical data). Why should or how could renewed efforts to establish the routine use of new computer-based record systems succeed now when previous attempts have failed? Why might this report have a significant impact? The committee believes that five conditions of the environment in which its strategic plan might be implemented increase the likelihood of achieving widespread use of computer-based patient records. First, current demands for patient information throughout the health care sector will not diminish; indeed, they will probably increase. Second, technologies essential to computer-based patient records are becoming more powerful and less expensive 15   Examination of the role of patient records in the clinical process, as manifested in the debate surrounding the problem-oriented record, has been under way for more than 20 years (Weed, 1968; Goldfinger, 1972; Margolis, 1979).

OCR for page 52
--> Third, patients and practitioners gradually are becoming accustomed to the use of computers in virtually all facets of everyday life. Fourth, an aging and mobile population results in more information to be managed and demands for improved transferability or portability of that information. Finally, the committee believes that those components of needed reform in health care that require evaluation, consolidation of data, and improved communication will not easily be achieved without reforms in the scope, use, and automation of the patient record. References AAMC (Association of American Medical Colleges). 1984. Physicians for the Twenty-first Century. Report of the Panel on the General Professional Education of the Physician. Washington, D.C.: AAMC. Amatayakul, M., and A. R. Sattler. 1990. Computerization of the medical record: How far are we? Proceedings of the Fourteenth Annual Symposium on Computer Applications in Medical Care. Los Alamitos, Calif.: IEEE Computer Society Press. Amatayakul, M., and M. J. Wogan. 1989. Fundamental Considerations Related to the Institute of Medicine Patient Record Project. Paper prepared for the Institute of Medicine Committee on Improving Patient Records in Response to Increasing Functional Requirements and Technological Advances. American Medical Association Council on Ethical and Judicial Affairs. 1989. Current Opinions. Chicago, Ill.: American Medical Association. Barnett, O. 1990. Computers in medicine. Journal of the American Medical Association 263:2631–2633. Bentsen, B. G. 1976. The accuracy of recording patient problems in family practice. Journal of Medical Education 51:311. Bronson, D. L., A. S. Rubin, and H. M. Tufo. 1978. Patient education through record sharing. Quality Review Bulletin December:2–4. Brook, R. H. 1989. Practice guidelines and practicing medicine: Are they compatible? Journal of the American Medical Association 262:3027–3030. Burnum, J. F. 1989. The misinformation era: The fall of the medical record. Annals of Internal Medicine 110:482–484. Case, C. L., and L. H. Jones. 1989. Continuity of care: Development and implementation of a shared patient data base. Cancer Nursing 12:332–338. Covell, D. G., G. C. Uman, and P. R. Manning. 1985. Information needs in office practice: Are they being met? Annals of Internal Medicine 103:596–599. Davies, A. R. 1990. Health Care Researchers' Uses and Requirements of the Patient Record. Unpublished draft memorandum to the Users and Uses Subcommittee of the Institute of Medicine Committee on Improving Patient Records in Response to Increasing Functional Requirements and Technological Advances. March. Davies, A. R., and J. E. Ware. 1988. Involving consumers in quality assessment. Health Affairs 7:33–48. Dawes, K. S. 1972. Survey of general practice records. British Medical Journal 3:219–223.

OCR for page 52
--> Donnelly, W. J. 1988. Righting the medical record: Transforming chronicle into story. Journal of the American Medical Association 260:823–825. Ellwood, P. M. 1988. Shattuck Lecture—Outcomes management: A technology of patient experience. New England Journal of Medicine 23:1549–1556. Feinstein, A. R. 1970. Quality of data in the medical record. Computers and Biomedical Research 3:426–435. Fox, L. A., G. Stearns, and W. Imbiorski. 1979. The Administrator's Guide to Evaluating Records and the Medical Record Department, ed. The Beacon Group. Chicago, Ill.: Care Communications. Fries, J. F. 1974. Alternatives in medical record formats. Medical Care 12:871–881. GAO (General Accounting Office). 1991. Medical ADP Systems: Automated Medical Records Hold Promise to Improve Patient Care. Washington, D.C.: GAO. Gardner, E., and L. Perry. 1989. Assessing computer links' risks and rewards. Modern Healthcare November 24:20–22, 26, 28. Garrett, L. E., Jr., W. E. Hammond, and W. W. Stead. 1986. The effects of computerized medical records on provider efficiency and quality of care. Methods of Information in Medicine 25:151–157. Gastel, B., and D. E. Rogers. 1989. Clinical Education and the Doctor of Tomorrow. Proceedings of the Josia Macy, Jr., Foundation National Seminar on Medical Education: Adapting Clinical Medical Education to the Needs of Today and Tomorrow. New York: New York Academy of Medicine. Gerbert, B., and W. A. Hargreaves. 1986. Measuring physician behavior. Medical Care 24:838–847. Goldfinger, S. E. 1972. The problem-oriented record: A critique from a believer. New England Journal of Medicine 288:606–608. Greenes, R. A., and E. H. Shortliffe. 1990. Medical informatics: An emerging academic discipline and institutional priority. Journal of the American Medical Association 263:1114–1120. Haynes, R. B., M. Ramsden, K. A. McKinnon, C. J. Walker, and N. C. Ryan. 1989. A review of medical education and medical informatics. Academic Medicine 64:207–212. Hershey, C. O., M. H. McAloon, and D. A. Bertram. 1989. The new medical practice environment: Internists' view of the future. Archives of Internal Medicine 149:1745–1749. Hsia, D. C., W. M. Krushat, A. B. Fagan, J. A. Tebbutt, and R. P. Kusserow. 1988. Accuracy of diagnostic coding for Medicare patients under the prospective-payment system. New England Journal of Medicine 318:352–355. Huffman, E. K. 1981. Medical Record Management. Berwyn, Ill.: Physicians' Record Company. IOM (Institute of Medicine). 1980. Reliability of National Hospital Discharge Survey Data. Report of a study. Washington, D.C.: National Academy of Sciences. IOM. 1988. The Future of Public Health. Washington, D.C.: National Academy Press. IOM. 1989. Controlling Costs and Changing Patient Care? The Role of Utilization Management, ed. B. H. Gray and M. J. Field. Washington, D.C.: National Academy Press.

OCR for page 52
--> IOM. 1990a. Clinical Practice Guidelines: Directions for a New Program , ed. M. J. Field and K. N. Lohr. Washington, D.C.: National Academy Press. IOM. 1990b. Effectiveness and Outcomes in Health Care, ed. K. A. Heithoff and K. N. Lohr. Washington, D.C.: National Academy Press. IOM. 1990c. Medicare: A Strategy for Quality Assurance, vols. 1 and 2, ed. K. N. Lohr. Washington, D.C.: National Academy Press. Korpman, R. A., and T. L. Lincoln. 1988. The computer-stored medical record: For whom? Journal of the American Medical Association 259:3454–3456. Kurland, L. T., and C. A. Molgaard. 1981. The patient record in epidemiology. Scientific American 245:54–63. Levit, K. R., M. S. Freeland, and D. R. Waldo. 1990. National health care spending trends: 1988. Health Affairs 9:171–184. Lindberg, D. A. B. 1979. The Growth of Medical Information Systems in the United States. Lexington, Mass.: Lexington Books. MacEachern, M. T. 1937. Medical Records in the Hospital. Chicago, Ill.: Physicians' Record Company. Mamlin, J. J., and D. H. Baker. 1973. Combined time-motion and work sampling study in a general medicine clinic. Medical Care 11:449–456. Margolis, C. Z. 1979. Problem-oriented record: A critical review. Paediatrician 8:152–162. Martin, J. B. 1990. The environment and future of health information systems. Journal of Health Administration Education 8:11–24. National Institutes of Health. 1989. National Library of Medicine MEDLARS: The World of Medicine at Your Fingertips. NIH Pub. No. 89–1286. Bethesda, Md.: U.S. Department of Health and Human Services. Paterson, M. L. 1988. The challenge to technology assessment: An industry viewpoint. Pp. 106–125 in Quality of Care and Technology Assessment, ed. K. N. Lohr and R. A. Rettig. Washington, D.C.: National Academy Press. Peck, R. S. 1984. Extending the constitutional right to privacy in the new technological age. Hofstra Law Review 12:893–912. Pories, W. J. 1990. Is the medical record dangerous to our health? North Carolina Medical Journal 51:47–55. Privacy Protection Study Commission. 1977. Personal Privacy in an Information Society. Washington, D.C.: U.S. Government Printing Office. Richart, R. H. 1970. Evaluation of a medical data system. Computers and Biomedical Research 3:415–425. Rogers, J. L., and O. M. Haring. 1979. The impact of a computerized medical record summary system on incidence and length of hospitalization. Medical Care 6:618–630. Rogers, J. L., O. M. Haring, P. M. Wortman, R. A. Watson, and J. P. Goetz. 1982. Medical information systems: Assessing impact in the areas of hypertension, obesity and renal disease. Medical Care 20:63–74. Romm, F. J., and S. M. Putnam. 1981. The validity of the medical record. Medical Care 19:310–315. Rulin, M. C., T. T. Hayashi, and D. M. Badway. 1988. Continuity of ambulatory care in an obstetrics and gynecology residency program. Obstetrics and Gynecology 71:787–790.

OCR for page 52
--> Saba, V. K., D. M. Oatway, and K. A. Rieder. 1989. How to use nursing information sources. Nursing Outlook 37:189–195. Small, S. A. 1989. Confidentiality issues for health administrators. Hospital and Health Services Administration 34:591–598. Shortliffe, E. H., L. E. Perreault, L. M. Fagan, and G. Wiederhold, eds. 1990. Medical Informatics Computer Applications in Health Care . Reading, Mass.: Addison-Wesley Publishing Company. Tarlov, A. R., J. E. Ware, A. Greenfield, E. C. Nelson, E. Perrin, and M. Zubkoff. 1989. The Medical Outcomes Study: An application of methods for monitoring the results of medical care. Journal of the American Medical Association 262:925–930. Tufo, H. M., and J. J. Speidel. 1971. Problems with medical records. Medical Care 9:509–517. Tufo, H. M., R. E. Bouchard, A. S. Rubin, J. C. Twitchell, H. C. VanBuren, and L. Bedard. 1977. Problem-oriented approach to practice. 2. Development of the system through audit and implication. Journal of the American Medical Association 237:502–505. U.S. Congress. 1989. Congressional Record, vol. 135, no. 165 (pt. 3). November 21. U.S. Department of Health and Human Services. 1990. HHS News. December 20 U.S. Public Health Service. 1990. Healthy People 2000: National Health Promotion and Disease Prevention Objectives. DHHS Pub. No. (PHS) 90–50212. Washington, D.C.: U.S. Department of Health and Human Services. Weed, L. L. 1968. Medical records that guide and teach. New England Journal of Medicine 12:593–600, 652–657. Westin, A. F. 1976. Computers, Health Records and Citizen Rights. National Bureau of Standards Monograph No. 157. Washington, D.C.: U.S. Government Printing Office. Whiting-O'Keefe, Q. E., D. W. Simborg, W. V. Epstein, and A. Warger. 1985. A computerized summary medical record system can provide more information than the standard medical record. Journal of the American Medical Association 254:1185–1192. Zimmerman, J. 1978. Physician Utilization of Medical Records: Preliminary Determinations. St. Louis, Mo.: Washington University, Biomedical Computer Lab. Zuckerman, A. E., B. Starfield, C. Hochreiter, and B. Kovasznay. 1975. Validating the content of pediatric outpatient medical records by means of tape-recording doctor-patient encounters. Pediatrics 56:407–411.