Catálogo de publicaciones - libros

The primary objective of clinical laboratorians is to provide the highest possible quality of service to patients and those who care for patients. High-quality service encompasses accurate and precise analysis, timely, clear, and concise reporting, and delivery of the service to a location in a format most valuable to the user of the service. Quality has also come to include elements of efficiency, effectiveness, analysis of diagnostic utility, and decision support.^1,2 Delivery of the service has come to mean that integration into a comprehensive electronic medical record (EMR) is to be expected.^3,4 Although it is widely accepted that about 70% of the information used in the management of patients comes from the clinical and anatomical pathology laboratories,^5 in one large medical center in which information flow is tracked, about 94% of requests to the EMR are for laboratory results.^4 In addition, a critical quality feature is the transformation of laboratory data into information.^2,5,6–8

It was pointed out in Chapter 1 that the term “laboratory information system” means more than an aggregation of hardware and software. The system is a plan expressed in policies and procedures and implemented in part on a computer. Indeed, computers are not necessary parts of information systems, which have existed under one title or another as long as human activities have been organized. Notebooks, card files, ledgers, and indexing systems all are information systems. The hardware and basic programs and software tools used to support a laboratory information system, although provided by a vendor, must reflect the needs and objectives of the laboratory, as defined by the laboratory and developed as an information plan. The information plan must be coherent, meet the medical service and management objectives of the laboratory and the larger institution, if that is the context in which the laboratory operates, and must make efficient and effective use of the resources of the purchased system. Full, inclusive participation of the laboratory staff in the development of standards and objectives for the system is a sine qua non for success, as is coherent central management of the system. The information system implements a shared goal of improvement in efficiency and effectiveness of services and must be accepted and used by the entire staff of the laboratory. This means that the information system is coherent and uniformly applicable, although not every aspect of laboratory services must be computerized.

The use of computer systems and computerized devices in the laboratory is becoming nearly universal, and increasingly processes once performed by people are being transferred to machines, the inner workings of which may be only dimly understood. The task of the people managing the work of the laboratory is to comprehend the processes by which the computerized laboratory information system (LIS) in use, usually acquired as a package from a vendor, may be shown to be doing the job it is expected to do. The functions of the system must be validated. Useful discussions of regulation^1 and inspection and accreditation^2 of laboratory information systems are available. However, despite its importance there is remarkably little published outside the blood bank literature about the validation of laboratory information systems that is accessible to general users of the systems, that is, addresses specifically laboratory issues in terms and concepts familiar to the laboratorian. Our intention here is to discuss LIS validation as it applies in the laboratory in general, to define the issues related to validation, and to suggest an approach to dealing with them.

The issue of computer system security has changed somewhat over the years, as technology has evolved from reliance on inaccessible, large, centralized mainframes to the use of distributed, linked or networked systems. Modern computer technology based on microcomputers and on-line access and broader use of powerful and user-friendly programs has put computer applications and other data-processing functions previously done only by computer operations experts into the hands of users. This improvement in efficiency and effectiveness presents, however, a serious challenge to achieving adequate data security. Progress in distribution of computer technology has not been accompanied by an increase in the knowledge of users about the vulnerability of data and information to such threats as unauthorized deliberate or accidental modification, disclosure, and destruction. Perhaps the greatest challenge is for all users to be aware of the possible ways that the integrity of a computer system can be compromised and to understand that full and confident use depends on the integrity of the system.

It seems to be well known that a computer-based information system is costly to acquire. Less well understood are the costs associated with maintaining and operating a system over time. Several large laboratory information system vendors had no ready answer, even in general terms, to the question “Do you offer any advice to potential clients about the costs of ownership of your system?” This is not very surprising, as even large, computer-dependent industries often have only the vaguest notion of the true costs of ownership of their systems. According to one estimate, each Microsoft NT workstation typically costs an organization $6,515 per desktop per year,^1 of which capital hardware and software costs account for only 25%. Some of the remaining 75% of costs, which may be overt or hidden, are associated with management and technology support.^1 The Gartner Group, a consulting firm, says that the five-year cost of a personal computer is $44,250.^2 Vendors not are generally expected to provide much information about cost of ownership beyond the services they provide under contract, as costs are mainly related to how owners choose to use and support their system rather than to any particular computer characteristic.

The computer has become the indispensable instrument in the laboratory. Not only are computers necessary for management and the communication of information, but they are usually the core of analytical instruments. No one can claim to be conversant with the production and communication of information from the clinical laboratory without a basic knowledge of the workings of computers. This discussion is an overview of computer systems, architecture, and programs, first focused on the small microcomputer or “personal computer,” then commenting on larger systems and the newer client/server systems. Some of this information will be old to some readers and quite new to others. Bear with us; our objective is to bring all readers to a level of familiarity with computer organization, components, and limitations that will enable them to follow discussions about system selection, integration, and validation, among other topics. Many books have dealt with topics mentioned here in passing, and we do not pretend to be (and feel no obligation to be) comprehensive or exhaustive in this discussion. Topics have been selected for elaboration based on their relevance to computer issues met in the laboratory. The discussion ends with a glossary of terms that must be assimilated if one wishes to follow any discussion of computer systems.

The computer’s position as one of the indispensable instruments in the laboratory logically leads to the necessity of sharing resources and information through the development of computer networks. Computer networking evolved so that people could share devices, resources, files, messages, knowledge, and information. This discussion is intended to provide a basic understanding of computer networks, from a local network to the Internet. Computer networks can bb e relatively simple or very complicated; our objective is to provide a solid foundation of concepts and terminology on which can be built an understanding of how basic computer networks are established and used.

A computer system interface is a means of communication between a user and a computer system, between two or more computer systems, or between computer systems and other analytical systems. Interfaces not only allow a person to use the machine, they also permit one machine to communicate with another, even one running on an entirely different operating system.

A bar code is a machine-readable code in the form of stripes, bars, or squares affixed to an object for the purposes of identification.^1 It is a binary system with several elaborations, including one-dimensional, two-dimensional, and matrix codes. The development of bar coding was a response to the need to rapidly identify objects and incorporate information about them into a format that does not depend on the ability of the human eye to interpret subtle variations in the printing or writing of letters and numbers but rather permits machine reading of symbols and relation of the encoded object to a database. It is generally accepted that manual transcription is the most important source of laboratory errors, especially the transcription of numbers. It is estimated that one in 300 keystrokes is an error. A misspelled word may be obvious, but transposition of digits is not. Also, manual transcription is slow and might have to be repeated several times in the progress of a specimen through the laboratory. Speeding data entry and reducing errors may improve overall laboratory productivity.^2

As the settings in which medical and other health services change, the need for mobility, flexibility, and efficiency in communication increases.^1 Wireless medical technology is in use in diverse settings, including aircraft,^2 ships,^3 and ambulances,^4 as well as the hospital and medical office environments.^5 The replacement of paper records by electronic records changes the way information is collected, stored, and transmitted, requiring changes in procedures that formerly depended on manual data entry and the accumulation and filing of pieces of paper. Wireless communications technology permits an unprecedented degree of mobility of people and transportability of equipment, while retaining functional connection to a central laboratory information system, which may itself be integrated into a larger information structure.^6