Problems and Solutions in the Process of Informating Patient Wards in Third World Hospitals Villela Jr, G. F. Center for Biomedical Informatics, State University of Campinas , Caixa Postal 6005, 13081 Campinas SP, Brazil. ABSTRACT In recent years there has been a visible trend among hospitals worldwide towards an increasing use of computers in their information systems (HIS's). This trend occurs also in the hospitals of Third World countries, though with some delay and to a lesser extent than in their First World counterparts. In developing countries, however, the process of hospital informating stops in most cases at the administrative and financial routines, therefore not reaching the patient ward, from where virtually all the information in the hospital environment ultimately originates. Many factors explain this apparent paradox, mostly related to the ward's essential features, which make its informating difficult and costly. However, HIS's are continually evolving, and may ripen to a point where it will be necessary to informate the patient ward. In Third World hospitals there may be some additional problems to cope with. These problems and the strategies for their management are the subject of this paper. 1. Introduction Though somewhat late when compared to the process of informating hospitals in developed countries, Third World hospitals are becoming increasingly informated. There are, of course, many different approaches which may be followed, varying from case to case; however, there are some steps which are usually followed. The automation of a Third World hospital's information system (HIS) usually starts with administrative and financial routines, and more commonly with those routines which can be found in any institution, be it a hospital or otherwise: for example, payroll, accounting, inventory, etc. In a later stage are developed or implemented applications which are specific to the hospital environment, but still within the domain of administrative and/or financial applications, most typically in connection with the billing of patients. In a third stage, there appear operational applications, such as the control of internment and release of patients, hospital census, ambulatory agenda, etc. There may be seldom a later stage in which medical records are informated; more rarely still, there may be applications oriented towards the medical and nursing practice per se [1,2]. Consequently, the informating of hospitals in the Third World very rarely reaches the infirmary, which contains physically the patient. This represents a paradox, since the ultimate purpose of a hospital is the care of patients, but the automated part of the information flow in the hospital is kept physically distant from what ultimately generates virtually all the information that exists in the HIS. However, the dissemination of information processing technologies is a worldwide phenomenon, which excludes neither Third World countries nor their hospitals. The institutional culture of the latter is becoming increasingly influenced by automation, and there will almost inevitably be a point when patient wards will have to be informated, in order to sustain the growth and the stability of the larger HIS. This paper tries to discuss some problems to be expected, and to propose some general principles to guide the development and implementation of automated information systems in the infirmary environment. 2. The Infirmary within the HIS Environment 2.1. General Considerations For the purposes of this paper, the terms "infirmary" and "(patient) ward" will be interchangeably used, and defined as any places within a hospital where there are supposed to be patients receiving medical and/or nursing care. This includes places such as operating rooms, Intensive Care Units, examination rooms and other facilities which are not commonly called "infirmaries" or "wards" in hospital terminology. This definition may also eventually conflict with other established definitions for the same terms, but will be adopted nonetheless, for expressing perhaps more adequately than any other words the object of this work. Next, let us state once more that the infirmary, since it contains physically the patient, is the place of origin, destination and/or transit of all the information generated during and by the stay of the patient in the hospital, however indirectly. This basic assumption is the main reason why, in spite of the harsh difficulties posed, the informating of the infirmary is inevitable in the long- term. The infirmary represents a node of convergence which simply cannot be ignored when analyzing a hospital's information flow. As for the patient, although he is the main fact from where the information derives, he does not directly feed the HIS, automated or not. It is the hospital personnel who actually does this, and more specifically, due to the infirmary's nature, the medical, paramedical and nursing personnel, who spend a substantial share of their time dealing with information in various forms. The time of these professionals is obviously very precious, but in developing countries they are very rarely aided by information processing technologies to concentrate their time in the care of the patient. Their place of work, i.e., the infirmary, generally has a low priority in the hospital's long-term informating policy. This is actually not a mistake; on the contrary, there are very sound reasons why patient wards should not, in most cases, be informated in an early stage, particularly in developing countries. This will be discussed in the next two topics. 2.2. The Infirmary and the Evolution of the HIS An infirmary does not work as an independent entity; it is part of a larger system, which is the hospital itself. For this reason, there is a huge amount of information flowing from the infirmary to the other parts of the hospital, and vice versa. Just to give a few examples: admission (occupancy control), supplies and pharmacy (requests for drugs and miscellaneous material), accounting (billing), Nutrition and Dietetics (diet requirements for individual patients), radiology and clinical laboratory (various clinical tests) and, given the comprehensive definition of "infirmary" adopted here, other infirmaries (patient transfers, the performing of various diagnostic and therapeutic procedures, expert evaluation, occupancy of operating rooms, etc.). The Medical Records Division is the ultimate destination of most, if not all of this information, and therefore also has a strong information link with the infirmary. Last, but not least, there is a considerable extent of internal communication among the members of the infirmary staff. The infirmary has its own internal information system, in which physicians must communicate with each other and with the nurses, the latter must do conversely the same, and both categories have to process an enormous amount of clinical, operational and administrative information. So, if the infirmary has such a complex information system, and with equally complex interactions with other parts of the hospital's organization and information system, why not informate it in the earliest stages? There are two main reasons why this should not be. Both can be said to apply to any hospital, but pose additional problems in Third World hospitals. The first reason is the low priority given to patient wards in the hospital's long-term informating plan. This is actually a correct approach. Upon informating any institution, a hospital or otherwise, it is easier and gives a better cost/benefit ratio to start with those areas which must process large amounts of information, but in a simple, known, repetitive way. It is also convenient that these areas be terminal points in the organization's information flow, so as not to disturb the work of other areas, especially during the difficult initial implementation phase, when it is not unusual for one to see information systems nearly collapsing, in the turmoil of the transition from a manual processing-oriented culture to an automated one [3]. Ideally, the activities of the areas to be informated at first must not be critical to the activities of the institution as a whole, and preferably unrelated to them. This minimizes initial risks while giving a broader choice of software packages. The areas which come closer to this ideal situation are the administrative and financial areas, and this is why the automation of the HIS usually starts with them. The infirmary, in contrast, is almost the antithesis of this ideal situation. Although the amount of information it has to process is often too large for the number of people who must do that, it is small in absolute terms, when compared to applications such as the hospital's payroll or accounting, in a ratio that becomes more distant the larger the hospital. Moreover, the infirmary environment's information handling procedures are numerous, distinct, complex, unstandardized and highly variable in time and space. The infirmary is also an originating, not a terminal point in the hospital's information flow, and its functions are highly specialized. These features make its premature informating, prior to the consolidation of an automatic information processing culture in the hospital, potentially disastrous. The considerations above are practically universal, though enhanced in the conditions of developing countries, where the process can be expected to be very slow, due chiefly to the low availability of financial resources to be applied to information processing technologies, and to the relatively low dissemination of these technologies in society as a whole. Nevertheless, those hospitals which choose the path of informating will have their information systems irreversibly changed and in fast evolution, up to a point where they may be ready to top their HIS's with informated wards. It is then that the second major reason not to informate patient wards prematurely, and a major difficulty to do so in later stages, will appear: the human resource problem, discussed next. 2.3. Human Resources and the Informating of Infirmaries There are two kinds of problems related to human resources in the conditions of developing countries, concerning the implementation of automated HIS's in hospitals in general, and in patient wards in particular. The first is the hospital personnel's low degree of familiarity with information processing technologies. The dissemination of computers and allied technologies is a recent phenomenon, which started later in the Third World than in the First, and is being incorporated by the former's society at a much slower pace, which may also vary according to local peculiarities. This has consequences in the hospital environment and its people. Established professionals must learn new concepts and new methods of work. This can be a formidable feat for some of them, complicated on their part by a considerable amount of prejudice, rejection and irrational fears [4-7]. As for future professionals, that is, those who are currently attending graduation courses in the health sciences at universities of the developing world, they will suffer from the present lack of concern of most such institutions, which, except in a few selected cases, have hardly realized the importance of training future health professionals to work in an increasingly informated environment. There is no shortterm solution for this problem. In the long term, medical societies, especially those devoted to Medical Informatics, will have to do their best to foster the teaching of Informatics in the universities of the Third World, if the gap is to be bridged. Those universities which are already concerned with the problem, having established Medical Informatics curricula, should also make an effort to disseminate their experiences. It should be stressed, however, that solutions must at any rate be local, suitable to the conditions of each country and region [8]. Meanwhile, hospitals that wish to automate their HIS's should invest massively in the training of their health professionals. This training should not wait until patient wards are about to be informated. It should begin as soon as the first operational routines of the hospital are automated. The already implemented systems will benefit from a better understanding of their nature and requirements by the health professionals, who directly or indirectly are their main feeders. In this way, when time comes for the informating of these professionals' own working environment, i.e., the infirmary, the process will be conducted more smoothly, because health professionals will already be committed to the betterment of the HIS [9,10]. Another kind of problem is the scarcity of qualified Informatics professionals in developing countries. The availability of ready software packages for hospitals in these countries has been growing fast, but options are still comparatively few, so that it may be better for many hospitals to develop their own applications, lengthy and costly as it may be. Anyway, when one talks about general hospital applications, ordinary systems analysts and programmers will usually suffice, though they may be quite scarce in some countries. When it comes to the infirmary, however, Informatics professionals are required to have a high degree of qualification, in order to be able to manage the environment's peculiarities and idiosyncracies, which include highly specialized functions. They will have to interact effectively with physicians, nurses and allied professionals, whose participation is mandatory. In this way, Medical Informatics specialists will be an essential requirement, which few institutions in even fewer developing countries will be able to fulfill. Once more, the solution will arrive only in the long term, as such specialists become more common. And once more the solution depends on the long-term investment in the formation and training of these professionals in developing countries. In many cases, international aid will be necessary and welcome, though it still remains a basic premise that each country and region must have its own resources and solutions. 3. Final Recommendations and Conclusions Any hospital that wants to informate its wards will have to follow some basic steps in order to succeed. First of all, the expected result will have to be precisely outlined, in a realistic basis which should take into account both the hospital and the infirmary's needs, as well as the actual feasibility of reaching the specified goal. The infirmary may be initially informated just as an entry point for data bound for the larger HIS. These data, being closer to their source, would thus be more accurate and enter the system more quickly. Or, the infirmary may receive informated support for its own clerical and administrative functions, such as drug prescription, queries for the patients' clinical data, generation of statistical and productivity reports, etc. The infirmary's system could even provide clinical diagnostic and therapeutic decision support, which in extreme situations could go as far as medical workstations with hypermedia, intelligent image and signal processing and AI features, even though this scenario would be extremely unlikely and hardly feasible at all in the typical conditions of developing countries. But, whatever the reasons for informating an infirmary, and the correspondent approach chosen, the hospital must be conscious of the expected final results, and be realistic about the cost and difficulties of getting there [2,8]. Once the expected goal is outlined, the hospital must arrange to finance the whole process, including hardware, software and, last but not least, the training of its staff, as emphasized in topic 2.3, above. Finally, the hospital must plan the whole process, trying to minimize its impact on the institution's activities and to foresee the outcome of the future evolution of the system. An Organization and Methods (O&M) study can be especially helpful in paving the way for informating the infirmary; though usually expensive, O&M studies may save many times their direct cost by increasing the efficiency of work, saving many indirect, often "invisible" costs. It is extremely important that health professionals, who will be the ultimate users of the infirmary's information system, not only be heard but also participate actively in every step of the system's development and/or implementation. This is a mandatory requisite for success. In hospitals of developing countries, there is expected to be little money available for spending in Informatics, and limited options to spend it. One must priorize its needs and make the best out of the limited resources available. In this way, one can expect that informated wards will not be commonplace in Third World hospitals for some time. However, as stressed above, the dissemination of information processing technologies is not restricted to industrialized or postindustrialized countries. These technologies are becoming increasingly disseminated in developing countries as well, especially in the so-called "newly industrialized countries" (such as Brazil, Mexico, Singapore, South Korea, etc.). Their health institutions are also part of this trend; HIS's are becoming more and more informated, in an irreversible way. Their future evolution will almost inevitably reach the patient wards, where virtually all the information in the hospital environment ultimately comes from. This paper was intended to point out some issues which will inevitably have to be addressed by all those involved in this process - hospital administrators and planners, health and Informatics professionals - to whom it is dedicated. 4. References [1] Arthur D. Little International, Inc. Computer System Development Plan for Hospital de Cl¡nicas da Faculdade de Ciˆncias M‚dicas da Universidade Estadual de Campinas. New York: Arthur D. Little International, Inc., 1977. [2] Barry CP and Gibbons LK. Information Systems Technology: Barriers and Challenges to Implementation. Nurse Adm 1990, 20:40-42. [3] Willems JL. Relation Between Organization and Hospital Information System. In: Towards New Hospital Information Systems, Nijmegen, IMIA, Working Group 10, 1988: VII.1-6. [4] M”hr JR. Training and Education in Medical Informatics. In: MEDINFO 83. Van Bemmel J H, Ball M J, Wigertz O (eds). Amsterdam: North-Holland, 1983:1030-1035. [5] Reichertz PL. Preparing for Change: Concepts and Education in Medical Informatics. Comput Meth Progr Biomed 1987, 25:89-102. [6] Gr‚my F. Persons and Computers in Medicine and Health. Meth Inform Med 1988, 27:3-9. [7] Friedman BA. The Potential Role of Physicians in the Management of Hospital Information Systems. Clin Lab Med 1990, 10:239-250. [8] Rienhoff OA. A Curriculum for Short-term Training im Medical Informatics for Developing Countries, In: Health Informatics in Developing Countries: Experiences and Viewpoints. De Talens AFP, Ravetto EM, Shires DB (Ed.). Amsterdam: North Holland, 1982: 295-304. [9] Rodrigues RJ. Experience with the design and implementation of a large medical information system in a developing country. In: MEDINFO 80. Lindberg D A B, Kaihara S (eds). Amsterdam: North-Holland, 1980. [10] Lun K C, Watanabe R and Kaihara S. Hospital Information Systems in Japan. Meth Inform Med 1986, 25:4-14. Published in: Proceed. 11th International Congress of the European Federation of Medical Informatics. London, Freund Publ., p. 249-253, 1993. (c) 1993 by Freund Publish. Co. Ltd., London