Florida Family Physician
January 1996 / Volume 46 / Number 1
Extended On-line Article
Conflicting goals are demonstrated by a review of literature citations, and by comparing the outcomes emphasized in Managed Care with those in clinical outcomes research. Superficially, the outcome definitions appear similar. However, they differ markedly in both concept and intentionality. Managed Care providers may be under considerable pressures to define outcomes in ways which bias findings in favor of the provider. Such pragmatism-even though it might be unintentional-is fundamentally at odds with the philosophy underlying the development of the CPR. The primary outcome of interest in Managed Care appears to be cost-effectiveness. While clinical outcome is regularly cited alongside cost measures in articles dealing with Managed Care, there negligible overlap between the literature of Managed Care and that body of literature which has clinical outcomes research as its primary focus.
If CPR developers recognize the conflicting interests between Managed Care and Patient Care, then it should be possible to use this conflict as an opportunity to ensure that the CPR begins to incorporate measures which will allow valid measurement of a variety of outcomes. It is appropriate to involve both professional epidemiologists as well as informed lay men and women so that valid and meaningful measures of outcome can be prospectively defined and incorporated into the data structure of the CPR. Informatiticians must be concerned not only with "Capturing the Data," they must also capture the complex context which allows those data to be meaningfully *interpreted*, *protected* from misuse, and made *available* for proper use.
Introduction
Medicine-in spite of all its recently-acquired technological bells and whistles-remains an essentially premodern artisan's guild, and one which is acutely uncomfortable in an increasingly postmodern age. [56] It is ironic, therefore, that medicine as a field is struggling so vigorously to achieve modernity at this time, stumbling belatedly into the 20th Century just in time to see it close. At the same time, pressures upon health care providers [15] for 'pragmatic' solutions threaten the acceptance of form over substance. Currently, Managed Care is such a driving force; others can be expected to emerge in the future. If those who are shaping the technical tools upon which contemporary medicine will depend recognize these dangers, they will also be in positions to offer alternatives to destructive trends. Such alternatives can reduce the likelihood of these dangers being realized, and provide a much-needed correction to long-standing as well as recent trends within the medical profession.
The introduction of computers into the practice of medicine has had a major impact upon the practice of medicine and the delivery of care. It is ironic, therefore, that a technology known for its capacity to dramatically alter the processes to which it is applied has had-to date-such a limited effect upon the *documentation* of the clinical care which it has affected. The discrepancy between the great increase in technological innovation in medicine and the lag in applying this technology to information management in medicine is conspicuous.
Computers were first introduced into clinical settings not for the purpose of improving patient care, but for the purpose of improving accounting. [9] Accordingly, the impact of such computing on clinical care has been negligible. Nevertheless, expenditures for these mainframe-based accounting and billing systems quickly became routine. Good business management dictated that billing be kept up to date, and the tool for modernity was the Hospital Information System, known-without any documented evidence of affection-as the HIS. The idea that these powerful computers might also serve to directly benefit patient care was novel.
Clinical Computing and the Computer-based Patient Record
Clinical computing, in contrast to administrative computing has had, since its inception, the care of the patient at its center [29]. Ideas which have been advanced for the use of computers in medicine are myriad. They range from the ultra-sophisticated to the absurd [17]. Somewhere in between developed the mundane and imminently practical concept of the Computer-based Patient Record, or CPR. Unlike some applications of computers in medicine (NMR scans, for example), the concept of the CPR was not so much driven by the underlying technology which made it possible to do something new, as by a recognition that computers offered a better way to perform an old and tedious task.
The CPR is not an effort which has sprung from the grassroots of clinical practice [29,52]. There is no evidence that physicians have begun demanding a CPR. Similarly, it has not been a concept or concern of administrators [9]. In spite of this, the CPR is conceptualized as meeting both clinical and administrative needs. It is "designed to be the centerpiece of a comprehensive medical information system [52]." Rather than setting up a dichotomy between administrative and clinical data, the CPR is built upon the notion that truly useful data serve more than one function [18]. Accordingly, "a computer-based medical record must store data according to their *true meaning* (emphasis added), rather than storing them in an interpreted form meeting a preconceived reporting requirement. The data can then be structured and interpreted by one of several output programs to meet the focused needs of the user of the moment. Such an approach increased the variety of users who can be supported by the database and, to a degree, insulates the database from changes in medical thinking about the ways in which patients should be classified [56]."
The scope of the CPR is ambitious, in keeping with its intended role as the Swiss army knife of Medical Informatics. It promises to be all things to all people, which-if other similar claims are predictive-suggests that some of these promises may be difficult to keep. Hammond-one of the acknowledge pioneers and current leaders in CPR development-goes on record to state that the ideal CPR is "a place where we store *everything* (emphasis added) we need to know about a patient and where we keep this information forever so we can do a great job of delivering care [60]." Hammond believes that the CPR is to be a living history of patient care. "For the first time, we will be able to learn everything that is going on with the patient, every medication and test that was ordered by every doctor the patient saw over his lifetime." He continues, "then, we can understand the patient's total health problems [60]."
Clinicians will recognize that the vision described above is compatible with the way practitioners wish that information were organized. Today, in contrast to earlier in this century, medical care is fragmented even for those who are not geographically mobile. It is the exception rather than the rule for a patient to receive medical care from only one provider. Not only was the volume of information lower in pervious generations, continuity of medical care was more often maintained through enduring patient-physician relationships, making the explicit recording of this information less critical than is currently the case.
Generally, there has been little attention paid to the form in which medical data are recorded. Little within the organization of record itself suggests that the information recorded therein is of any genuine or lasting value .
Nevertheless, paper records contain vast amounts of detailed information which is not readily accessible for clinical or research purposes, even when meticulously maintained. The use of even well-organized clinical data, collected with a deliberate view to research, requires sometimes phenomenal effort to abstract and manage the data from the paper record [59]. Major goals of the CPR are to facilitate the collection of data which are 1) highly structured, 2) individualized, and 3) immediately available in an electronic form which makes the data usable. In this way, clinical care and research needs can be met in real time. It is the focus of the CPR on clinical care which promises to give it extended utility in other related areas, such as administration and research.
The CPR--along with all of its genuine promise for improving health care [4],--also holds the possibility of making things worse. While this is true of most human interventions, it is particularly true in the field of medicine. Therefore an emerging field which proposes to change the very way in which clinical information is constructed and handled should be viewed with concern as well as enthusiasm. The effect of the CPR may be "a boon or a hindrance to the practice of medicine [10]." Flaws [32] in some of the implicit and unqualified assumptions (e.g., "structured reporting of medical findings is known to have positive effects on data quality "[33]) underlying the CPR-when recognized-present opportunities for refining the CPR concept, and for improving patient care.
What's Taking So Long?
Until recently, interest in the CPR as a practical and necessary tool was lacking. Several explanations have been proffered as to why widespread implementation of the CPR has been delayed. Rodnick summarized these in a recent editorial [46]. He notes that technological obstacles have not been the limiting step in widespread adoption of the CPR. Rather, he notes diversity among the various health care environments, problems associated with actually entering the data into an electronic form, and the disruptive aspect which a major change in practice pattern would entail dubious cost estimates are noted as contributing to the delay. Now, however, an unprecedented interest in the CPR is evident. Previously, professional presentations on the CPR by international experts in the field were poorly-attended. Now, numerous and well-attended sessions detailing many aspects of development and application of the CPR are common at professional meetings (personal communication, E Hammond). What can explain the sudden interest in an area which, based upon attendance at presentations by leaders in the field, was ignored for so long by most clinicians and administrators?
Managed Care....
While legislative efforts at so-called health care reform have languished in the United States Congress, sweeping trends are evident which are driven by other forces. Chief among these forces is a recent and snowballing phenomenon known as Managed Care[15,16,26,38,51]. The magnitude of this trend is evident in a cursory survey of the medical literature. A Medline search by the medical subheading (MeSH) term "Managed Care" yielded 1197 citations in the 1990-1990 journal database. By contrast, in the 1985-1989 Medline database, only 153 such citations were identified. Currently, a majority of newly insured individuals are entering Managed Care systems [38].
The essence of the trend toward Managed Care is the shifting of responsibility and costs from insurers to providers [3,47]. It is reasoned that, if providers are rewarded for efficient health care delivery, and penalized for inefficiency, that health care costs can be contained [38]. Ironically, the contractual agreements for containment of costs are being implemented in the absence of data about efficiency-or efficacy. Nevertheless, provider systems who agree to provide comprehensive care at a rate lower than the competition are in the best position of securing managed care contracts.
The long-standing inefficiencies of the United States health care system has Chief Financial Officers of major medical centers and clinics deeply concerned. For the first time in history, these executives find themselves in the position of negotiating contractual agreements for health services without any idea of what it will actually cost to provide those services. In spite of this, the rush to secure such managed care contracts has accelerated. But the fiscal uncertainty of these arrangements has been shifted to the provider system, which leaves the providers longing for hard and reliable cost estimates. Few, if any, are available [24 ].
Formerly, insurance carriers sought to protect themselves by insuring large groups in which the financial risks would be well-distributed. The medico-legal risks were assumed by providers. Furthermore, as pragmatic businesses, insurers were often careful if not always transparent in excluding certain groups known to generate high medical costs. Such groups included single adult males, workers in high risk occupations, and persons already known to have had past, serious medical problems. Because the system was financially driven, it tended to be fiscally conservative. Patients were often denied payment, if not actual insurance, because of the associated costs.
The insurer, so long caught in an uncomfortable position, has, under Managed Care, become a mere broker. Decisions about the utilization of health care resources (X-rays, lab tests, surgical procedures, etc.) are in the hands of the provider. However, the *incentive to deny services* now falls to the provider [30], and there are "unavoidable conflicts between medical benefit and cost containment" [21]. Because providers are paid on a fixed-fee per patient basis (capitation), the fewer services rendered to the patient, the better off-financially-the provider will be.
While neatly shifting the financial implications for allocating health care costs to providers, providers remain ethically and legally responsible for the effects of their actions [3,25,47]. Thus-goes the argument of Managed Care-enlightened self-interest on the part of providers will protect the patient from the greed of a system designed to be niggardly with expensive health care services. According to this schema, providers will be pressured to provide the best *necessary* care at the lowest possible cost. Providers who are too expensive will be weeded out, as will those who, for whatever reason, have outcomes which deviate from an aggregate norm. Rather than encouraging progressive innovation in treatment Managed Care may be a regressive force [58].
Managed care requires-at the outset-that providers be able to make accurate cost estimates. Once the process is entered, however, the burden of demonstrating *efficacy of care* is assumed by the provider system. Here, the territory is even more uncertain than the financial. Clinical outcomes research is fraught with methodological and technical problems [2,27,36,40,42,54]. Regarding Managed Care, there is a "paucity of rigorously designed outcome research" [13]. Coupled with the self-interested need to show good outcome *and* cost savings, perhaps it is reasonable to wonder what the outcome of Managed Care outcomes research may be.
Welcome the CPR....
Is it coincidental that recent, wide-spread interest in the CPR--an interest dormant for most of its 25-year history-emerges as Managed Care enters the scene? Managed Care demands what only the CPR can provide: accurate cost estimates of patient care. The CPR--being a clinical instrument--will yield longitudinal data about both individual patients and individual provider practice patterns (in terms of expenditures and clinical treatments), but will provide data which, because their collection will be standardized, can also be analyzed in the aggregate.
Ironically, it is clinical outcomes assessment which is proposed to serve as the basis for evaluating the cost-based implementation of Managed Care. Thus, not only is the language of clinical care co-opted by a system driven by cost controls, but the language of epidemiological research is also employed. It remains to be seen how a system which has a vested interest in demonstrating its own benefit to health care will manage to care for measurement of the outcomes which it will produce. However, the *intent* of Managed Care to utilize any electronically available patient data for
outcomes research' is quite clear [43]. Medical informatiticians expect this, though perhaps the data types being used for such research *would* surprise them [49]. Outcomes will be measured by, among other things, claims data [24,34]. Plans are being laid for the application of epidemiological techniques to measure outcomes based upon analysis of these fiscal data [35].
Although epidemiologists and biostatiticians are quick to point out problems in the use for research of clinical data not designed for such purposes [48], this has not prevented the utilization of such data in the absence of better ones. It has been pointed out that "patient care databases and clinical research databases often share a large portion of data" [18]. It is probably more accurate to state that clinical research databases are often a *subset* of the larger and less structured clinical databases from which they are taken.
The fact that the purposes of the two differ has not prevented such derivations, often through the use of structured data collection instruments applied to medical record data which were collected without any regard to a research question. While acknowledging that "because of their *different purposes* (emphasis added), the record format of the medical and analytical databases are quite different" [18]. Advocates of the CPR clearly intend for it to serve a dual function. This dual functionality is seen as possible via the structuring of data which are necessary for the clinical record. While the CPR elegantly deals with the technical aspect of data abstraction and transfer from a clinical to a research database [18,22,28], it does not address the design issues necessary for rigorous clinical outcomes research. That the "difficulty in classifying managed care plans has implications for health services researchers" is even acknowledged as potentially problematic for outcomes research [38] is, perhaps, reason for some consolation.
Odd bedfellows....
Managed care views the CPR as a means by which accurate patient data-and thus costs-can be estimated. Given the detailed and varied information which the CPR is in fact designed to gather, such estimates can probably be produced from a well-designed CPR. The real and unmeasured cost of Managed Care, however, is the impact which it will have on clinical outcomes. Managed Care clearly intends to measure itself by fiscal standards which it defines, but in the language of clinical outcomes research. Can the data necessary to support valid outcomes research be obtained from the CPR as currently envisioned?
The need for data make the data you have the data you need.
What is the nature of the concerns which Managed Care expresses for Clinical Outcomes Research [27,30]? And, what elements of the CPR are devoted to the measurement of such outcomes? A review of the literature for these three separate areas-Managed Care, the Computer-based Patient Record, and Outcomes Assessment-should reveal significant overlap, if in fact these areas are already linked conceptually and developmentally.
The previously mentioned Medline search for Managed Care references (1990-1994) yielded 1197 citations when the MeSH term "Managed Care" was chosen for the search strategy using OVID at Duke University Medical Library. Similarly, the MeSH term "Medical Records Systems, Computerized," yielded 914 references for the same time frame. 1553 citations for "Outcomes Assessment" were located for the same (1990-1994) time frame. The broader MeSH term, "Treatment Outcome" located 13,335 references.
Limiting the search strategy, sequentially, yielded interesting results. Combing "Managed Care" with "Medical Records, Computerized" via the boolean AND operator gave a total of three (3) citations. Only one appeared in a publication dedicated to Medical Informatics; the other two appeared to be minor contributions to the general clinical literature rather than specific articles about managed care or the CPR. When the limitation was applied to the much larger number of citations (13,335) gained from use of "Treatment Outcomes" as the MeSH term, the overlap tripled to nine (9) citations.
In looking for overlapping citations between literature about "Outcomes Assessment" and "Medical Records, Computerized," four citations were identified. One of these (Shortliffe, 1991) was a relevant contribution by a recognized leader in the field of Medical Informatics. Of note, the combination of the larger category "Treatment Outcomes" and "Medical Records, Computerized" using the boolean AND operator yielded no overlapping citations.
Finally, to demonstrate the area of apparent literature overlap of all three areas, the MeSH search results for "Outcomes Assessment," "Medical Records, Computerized," and "Managed Care" were combined with a boolean AND statement. Not unexpectedly, there was no overlap.
The overall search strategy and results are displayed below:
Unfortunately, the CPR will be also asked to do a task for which it has not been designed, in much the same way that the paper medical record has been asked to provide research quality data from individual records. The CPR can readily be co-opted for such an unintended use, and will be, if past and current uses of clinical data for outcomes research are predictive. In addition, electronic records have been shown to vary widely in the quality of data which they contain, [14,39,53] an issue which seems to get little attention in the Managed Care literature.
Language and Meaning
"When *I* use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean-neither more nor less." "The question is," said Alice, "whether you *can* make words mean so many different things." "The question is," said Humpty Dumpty, "which is to be master-that's all." [12]
The use of outcomes research language by Managed Care does not ensure that outcomes so measured will be valid, much less clinically meaningful. A system with such a vested interest in documenting its own success might be expected to substantially bias to such research [19]. The burden of proof that such bias is absent would certainly fall to the participants of Managed Care: clinicians, administrators, hospitals, and investigators.
As Humpty Dumpty suggests, the use of language may degenerate to the will to power. It is clear that major shifts in power are entailed in the move to Managed Care. From where, then, will come the will and power to resist the potential misuse of the huge amount of highly standardized data which the CPR can provide? For it is very likely that the sheer volume of such data-and the organized manner in which it is collected and presented-will give special credibility to analyses based upon it-no matter how flawed the research design [23].
As Veith has pointed out, postmodern society is conducive to the superficial analysis of the sort which could serve the interests of Managed Care systems [56]. Yet, the philosophy of the CPR, as shown, is fundamentally at odds with such analyses of convenience.
The CPR, as currently conceptualized, is concerned with issues of objectivity and uninterpreted meaning--*true meaning*, according to Stead and Hammond (1988). Unless we are indeed free-as was Humpty Dumpty-to redefine language as suits us, then we are facing real challenges to the integrity of the CPR.
The CPR is based upon the idea that standards are not only essential but central to the electronic collection and exchange of data. Hardware and software standards, as difficult as they are to forge [22,37] are trivial in comparison to the standardization of meaning which the CPR demand [55]. It will serve little purpose to achieve the transparent exchange of data which mean different things to different systems and different people.
Ironically, the notion that clinical data have a "true meaning" as Stead and Hammond suggest (1988) is in danger of being accepted uncritically and of being ignored in practice at the same time. There are reasons for this contradiction. The postmodern world in not uncomfortable with such inconsistency, according to Veith. Some of it is inevitable, in that there is an element of data interpretation which precedes the entry of such data into the CPR. Which data elements, for example, make up the record? Who defines them-Humpty Dumpty or others less likely to rashly place themselves in precarious situations? If it is only-or primarily-a question of power, and not of reason, this would seem to raise real concerns about the ability to advance outcomes research in a manner which would have an enduring meaning and value for patient care [30].
Subjectively Objective and Objectively Subjective
What sorts of outcomes research can truly benefit patient care? Perhaps we should take the unprecedented step of asking some patients. Nothing about the current health care system in the United States suggests that Americans are chiefly interested in reducing costs, particularly when such reductions are made at the expense of reducing choices [1].
The crisis which Managed Care may bring to clinical outcomes research may also bring new opportunity. If clinicians, informaticians, and researchers form alliances with those who have a real interest in advancing legitimate outcomes research-namely patients-then a new direction can be given to such investigative efforts [57]. Too long clinicians have defined satisfactory outcomes without adequate input from their patients. Researchers have operated with relative independence from both patients and clinicians. Medical Informaticians have worked frequently if irregularly with clinicians, less commonly with the patients intended to be served by the CPR, and rarely, it appears, with those who will subsequently be asked to evaluate clinical outcomes on the basis of data to be contained in the CPR.
The design of prospectively defined outcome measures must place a significant emphasis on patient preferences, particularly since it is likely that "laypeople and professionals have conflicting perspectives" regarding this agenda [11]. This does not mean that the patient unilaterally defines satisfactory outcome, but it does mean that clinicians, researchers, and designers of information systems will necessarily work more closely with patients, not only in defining the measures used to determine outcome, but in ensuring that the data necessary for these measures have *shared meaning*. Ironically, work begun in the area of patient-provider attitudes toward the CPR have found more concern about the privacy of data than its accuracy [41,44]. While the making of private data public is clearly unacceptable, the private use of *defective data* for decision making should be equally unacceptable. Legitimate concerns about privacy *and* data quality-in the broadest sense, as well as the appropriate use of such data-must be brought to the attention of the public.
Constructing a truly useful and informed cooperative relationships between designers, users, and beneficiaries of the CPR may be expected to complicate the ongoing design process. It is likely that patient input will not be as easily obtained, categorized and standardized as purely 'clinical' data. Such complications may further delay broad implementation of the CPR, but that would be good if it prevents the imposition of a system which may be technically ripe, but conceptually in need of revision.
A properly implemented CPR-one in keeping with the laudable goals articulated those who have been most active in its development-presents the possibility that the patient and physician can once again be on the same side of the fence, but in a new and healthier relationship. This can serve to mitigate the noxious effect of a system which appears determined to be driven by cost containment strategies. Designers and developers of the CPR can act now to ensure that mutually consensus-based outcomes measures which have both clinical validity and personal value are intrinsic to the CPR. In this way, it can be hoped that the outcomes which will be forthcoming under Managed Care can be measured by standards which are broader and more enduring[11,31,45] than the pragmatic fiscal ones which currently predominate in the Managed Care literature.
Fortunately, the Medical Informatics community is beginning to recognize the need for patient involvement in implementing the CPR, which is, after all, being undertaken to improve clinical care. If the CPR is to be truly patient-centered, then the patient must be meaningfully involved in the construction of the CPR. It is also time to invite those investigators who will without doubt be expected to live with our design decisions: epidemiologists. They might have something to contribute as well. Unless we make the nontechnical design of the CPR a more inclusive team effort, we will find ourselves playing a game for which we are poorly prepared. Otherwise, we risk the triumph of the superficial,56 and not only the further impoverishment of epidemiology [50], but of the clinical care which outcomes research purports to adavance. Informaticians must be concerned not only with "Capturing the Data," they must also capture the complex *context* which allows those data to be meaningfully *interpreted*, *protected* from misuse, and made *available* for proper use.
References
1] Anderson OW. Health Services in the United States: A Growth Enterprise for a Hundred Years. In: Litman TJ, Robins LS (Eds), Health Politics and Policy in Perspective. John Wiley & Sons, NY, 1984, pp.67-80.
2] Anonymous. The issue is not the usability of claims data but the quality of the
indicators [editorial]. Quality Review Bulletin. 16(12):422-3, 1990 Dec.
3] Appelbaum PS. Legal liability and managed care. [Review] American Psychologist. 48(3):251-7, 1993 Mar.
4] Bates DW, O'Neil AC, Boyle D, et al. Potential Identifiability and Preventability of Adverse Events Using Information Systems. J Am Med Informatics Assoc 5(1):404-411.
5] Black DW. Warrack G. Winokur G. Excess mortality among psychiatric patients. The Iowa Record-Linkage Study. JAMA. 253(1):58-61, 1985 Jan 4.
6] Black DW. Warrack G. Winokur G. The Iowa record-linkage study. I. Suicides and accidental deaths among psychiatric patients. Archives of General Psychiatry. 42(1):71-5, 1985 Jan.
7] Black DW. Warrack G. Winokur G. The Iowa record-linkage study. II. Excess mortality among patients with organic mental disorders.Archives of General Psychiatry. 42(1):78-81, 1985 Jan.
8] Black DW. Warrack G. Winokur G. The Iowa record-linkage study. III. Excess mortality among patients with 'functional' disorders. Archives of General Psychiatry. 42(1):82-8, 1985 Jan.
9] Bleich HL, Slack WV. Clinical Computing. MD Computing 6(3):133-135.
10] Bradbury AR. Computerized medical records: defining a standard without the computer. Medical Informatics. 16(3):279-86, 1991 Jul-Sep
11] Brown P. Popular epidemiology and toxic waste contamination: lay and professional ways of knowing. Journal of Health & Social Behavior. 33(3):267-81, 1992
12] Carroll, L. Through the Looking Glass. The Heritage Press, Norwalk, Connecticut, 1969, pp.112-113
13] Chamberlain R. Rapp CA. A decade of case management: a methodological review of outcome research. [Review] Community Mental Health Journal. 27(3):171-88, 1991 Jun.
14] Chan LS. Schonfeld N. How much information is lost during processing? A case study of pediatric emergency department records. Computers & Biomedical Research. 26(6):582-91, 1993 Dec.
15] Coleman RL. Promoting quality through managed care. [Review] American Journal of Medical Quality. 7(4):100-5, 1992 Winter.
16] Corrigan JM. Nielsen DM. Toward the development of uniform reporting standards for managed care organizations: the Health Plan Employer Data and Information Set (Version 2.0). Joint Commission Journal on Quality Improvement. 19(12):566-75, 1993 Dec.
17] Covvey, H. Dominic. A Story of Failure. Chapter 11 in: Concepts and issues in health care computing. Mosby, 1985, pp.118-121.
18] Dozier JA, Hammond WE, Stead WW. Creating a link between medical and analytical databases. Proceedings. Ninth Annual Symposium on Computer Applications in Medical Care. IEEE Computer Society Press, 478-482.
19] Ellenberg JH. Selection bias in observational and experimental studies. [Review] Statistics in Medicine. 13(5-7):557-67, 1994 Mar 15-Apr 15.
20] Faust D. Miner RA. The empiricist and his new clothes: DSM-III in perspective. American Journal of Psychiatry. 143(8):962-7, 1986.
21] Finnerty JJ. Pinkerton JV. Ethical considerations of managed care. [Review] Obstetrical & Gynecological Survey. 48(10):699-706, 1993 Oct.
22] Fitzmaurice JM. Health care data standards are required for medically effective use of workstations. International Journal of Bio-Medical Computing. 34(1-4):331-4, 1994 Jan.
23] Freedland KE. Carney RM. Data management and accountability in behavioral and biomedical research. American Psychologist. 47(5):640-5, 1992 May.
24] Goldfield N. Measurement and management of quality in managed care organizations: alive and improving. Quality Review Bulletin. 17(11):343-8, 1991 Nov.
25] Graddy B Trouble at the gate. Physicians may bear sole liability for poor outcomes when managed care organizations interfere in medical decisions. Texas Medicine 90(7):36-8, 1994.
26] Gray BH. The legislative battle over health services research. Health Affairs. 11(4):38-66, 1992 Winter.
27] Guadagnoli E. McNeil BJ. Outcomes research: hope for the future or the latest rage? Inquiry. 31(1):14-24, 1994 Spring
28] Hammond WE. The role of standards in creating a health information infrastructure. International Journal of Bio-Medical Computing. 34(1-4):29-44, 1994 Jan.
29] Hammond WE. Hospital informations systems: a review in perspective. In: van Bemmel JH, McCray AT (Eds), Yearbook of Medical Informatics, Schattauer Verlagsgesellschft mbH, Stuttgart, Germany, 1994, pp.95-102.
30] Hunter DJ. Doctors as managers: poachers turned gamekeepers?. [Review] Social Science & Medicine. 35(4):557-66, 1992 Aug.
31] Kasper JF. Mulley AG Jr. Wennberg JE. Developing shared decision-making programs to improve the quality of health care. Comment in: QRB Qual Rev Bull 1992 Jun;18(6):182
32] Knight L. Yardley M. Jones A. The dangers resulting from inaccurate computer-based operative records. British Journal of Clinical Practice. 45(1):41-2, 1991 Spring.
33] Kuhn K. Wechsler JG. Zemmler T. Reichert M. Ruetz T. Schwegler V. Ditschuneit H. [Structured documentation of findings and user acceptance--results of a study of abdominal ultrasound]. [German] Original Strukturierte Befunderhebung und Benutzerakzaptanz--Ergebnisse einer Studie fur die abdominale Sonographie. Leber, Magen, Darm. 23(1):25-8, 1993 Jan.
34] Leatherman S. Peterson E. Heinen L. Quam L. Quality screening and management using claims data in a managed care setting. Quality Review Bulletin. 17(11):349-59, 1991 Nov.
35] MacDowell NM. Lutz J. Applications of epidemiology in the operation and marketing of managed care plans. [Review] Journal of Health Administration Education. 11(4):541-50, 1993 Fall.
36] Maklan CW. Greene R. Cummings MA. Methodological challenges and innovations in patient outcomes research. Medical Care. 32(7 Suppl):JS13-21, 1994 Jul.
37] McDonald CJ, Hammond WE. Standard formats for electronic transfer of clinical data. Annals of Internal Medicine 110(5):333-335.
38] Miller RH. Luft HS Managed care plans: characteristics, growth, and premium performance. [Review] Annual Review of Public Health. 15:437-59, 1994.
39] Mullooly JP. The effects of data entry error: an analysis of partial verification. Computers & Biomedical Research. 23(3):259-67, 1990 Jun.
40] O'Connor GT. Plume SK. Wennberg JE. Regional organization for outcomes research. [Review] Annals of the New York Academy of Sciences. 703:44-50; discussion 50-1, 1993 Dec 31.
41] Ornstein S. Bearden A. Patient perspectives on computer-based medical records. Journal of Family Practice.38(6):606-10, 1994 Jun.
42] Paduda J. Outcomes research, analysis, and dissemination: the federal government's role. Medical Interface. 7(2):52-4, 56-7, 1994 Feb.
43] Pollak VE. Pesce A. Kant KS. Continuous quality improvement in chronic disease: a computerized medical record enables description of a severity index to evaluate outcomes in end-stage renal disease. American Journal of Kidney Diseases. 19(6):514-22, 1992 Jun.
44] Rind DM. Safran C. Real and imagined barriers to an electronic medical record. Proceedings - the Annual Symposium on Computer Applications in Medical Care. :74-8, 1993.
45] Reiser SJ The era of the patient. Using the experience of illness in shaping the missions of health care. Comment in: JAMA 1993 Jul 28;270(4):450-1
46] Rodnick JE. Computer-based medical records: time for an upgrade. J Am Board Fam Prac 7(4):359-361.
47] Sederer LI. Judicial and legislative responses to cost containment. [Review] Comment in: Am J Psychiatry 1993 Sep;150(9):1433-4
48] Schlesselman JJ. Case-Control Studies. Design, Conduct, Analysis. Oxford University Press, New York, 1982. pp.90-92.
49] Shortliffe EH. Medical informatics and clinical decision making: the science and the pragmatics. Medical Decision Making. 11(4 Suppl):S2-14, 1991 Oct-Dec.
50] Skrabanek P. The Poverty of Epidemiology. Perspectives in Biology in Medicine 35(2):182-185, 1992.
51] Soffel D. Luft HS. Anatomy of health care reform proposals. Western Journal of Medicine, 159(4):494-500, 1993 Oct.
52] Stead WW, Hammond WE. Computer-based medical records: the centerpiece of TMR. MD Computing 5(5):48-62, 1988.
53] Tate KE. Gardner RM. Computers, quality, and the clinical laboratory: a look at critical value reporting. Proceedings - the Annual Symposium on Computer Applications in Medical Care. :193-7, 1993.
54] Testa MA. Nackley JF. Methods for quality-of-life studies. [Review] Annual Review of Public Health. 15:535-59, 1994.
55] van Ginneken AM. van der Lei J. Moorman PW. Towards unambiguous representation of patient data. Proceedings - the Annual Symposium on Computer Applications in Medical Care. :69-73, 1992.
56] Veith GE. Postmodern Times. Crossway Books, Wheaton, Illinois, 1994.
57] Walker GC. Accountability for quality improvement--patients' right to assess the plan of care. Journal of Professional Nursing. 7(5):269, 1991 Sep-Oct.
58] Westermeyer J. Problems with managed psychiatric care without a psychiatrist-manager. Comment in: Hosp Community Psychiatry 1992 Jun;43(6):648-9
59] Woolson RF. Tsuang MT. Urban LR. Data management in an epidemiological study: experiences from the Iowa 500 field follow-up and family study. Methods of Information in Medicine. 19(1):37-41, 1980 Jan.
60] Sandrick K. A 26-year quest for the ultimate electronic record. Health Data Management July-August 1994:21-23.
61] Blum BI. Clinical Information Systems. Springer-Verlag, New York, 1986, p.376.
Search results from Ovid - Medline (1990 to December 1994)
"Capturing the Data" was the official theme of the 1995 Spring Congress of the American Medical Informatics Association.
This work supported in part by NIH Grant# 5T15LM07071-03