Catálogo de publicaciones - libros

Each of the methods described above is explained in detail, with sample evaluation instruments where appropriate, in Chapters 2 through 9. The chapters also include examples of studies that make different theoretical assumptions, address different evaluation questions, and employ different research methodologies. Each study also has important practical policy implications for the organization under study.

Evaluating information systems used in teams requires researchers to understand not only the technical aspects of the system but also the work and interactions of team members who use the system. Researchers using methods such as grounded theory combined with qualitative data collection techniques of observations, interviews, and artifact collection have gained tremendous insight into technology use in teams. Yet, there is still much work to be done. To ensure that information systems effectively support collaboration in teams, we must rigorously evaluate these systems using methods appropriate to studying teams in the healthcare setting.

In summary, the survey methods described in this chapter comprise an essential dimension in a multimethod approach to evaluating the impacts of computers on the functioning of healthcare organizations and the work life of the individuals within them. The chapter and the instruments included in the chapter appendix should provide investigators with standardized instruments, as well as guidance and examples for questionnaire design where no standardized measure exists. Although not intended as a complete review of the literature, this chapter also provides investigators with an overview of topics to consider when planning any investigation of the social impacts of computers in healthcare organizations.

This chapter has briefly reviewed several work-evaluation techniques and attempted to describe in detail the concepts behind work sampling, a technique based on sampling theory. As described throughout this chapter, work sampling studies are not without problems. Even the most thorough study can be severely compromised by the seemingly endless random occurrences that are the rule rather than the exception in health care. One should not be dissuaded for these reasons, because the potential information gained is critical in determining the future role of computers in health care.

Social network analysis can be used to analyze relationships among healthcare providers, departments within healthcare organizations and other organizations. Information obtained using this evaluative methodology can be used to identify influential individuals or opinion leaders who are critical to the successful implementation of medical informatics applications. This methodology can also be used to better understand changes in communication patterns or other interactions over time. Several examples that illustrate this evaluation methodology are presented.

Computer simulation can be used to evaluate complex information systems in situations where traditional methodologies are difficult or too costly to employ. The modeling process is described followed by three examples where computer simulation has been utilized in planning for a wide-area healthcare network and in the use of a hospital information system to reduce costs and errors in order entry.

In summary, study findings indicate that (1) patients have no problem with the use of a computer in the consulting room; and (2) examiners (NPs and PAs) are willing to use a system that they perceive as having benefits for patient care (e.g., enforcing thoroughness in the exam). For systems to be enthusiastically endorsed and used by clinicians, however, they will need to go beyond the mixed benefits of systems such as CompuHx to include features that clearly make practice easier (e.g., easy retrieval of information clinicians need). Physicians in hospitals, for example, are far more likely to use computers to retrieve laboratory results needed to make clinical decisions than to enter their own orders in the computer, which requires additional work on their part. Also essential is a detailed implementation plan that includes adequate time for training and communication with other users, and addresses issues such as the role of the system champion and any performance monitoring concerns. Implementation may also be hindered by the need to identify the monetary value of the cited benefits in light of the additional time initially required to conduct exams using the system, as well as required capital expenditures. Longitudinal research should examine potential changes in clinician tolerance for uncertainty, as well as the impacts of altered communication and consultation patterns, which have been shown to improve productivity in other settings. The project also highlights the importance of research that focuses not only on system outcomes, but also examines the implementation process and includes the necessary information to evaluate factors influencing clinician usage of the system.