Introduction

Information technology (IT) plays a crucial role in the practice of evidence-based medicine (EBM) by allowing health care practitioners to access and evaluate clinical evidence as they formulate their patient care strategies. This oftentimes involves an analysis of a large amount of complex information. Hence, health care administrators responsible for implementing information systems and EBM policies and guidelines will need to be familiar with EBM concepts, as well as the capabilities and limitations of available IT solutions employed to perform EBM.

Purpose

Problem

Background

Objectives

Literature Review

Practicing EBM involves using the best clinical evidence, obtained using a rigorous methodology, to make decisions for an individual patient or population (Bartkowiak, 2004, p. 254; Eddy, 2005, p. 16; Hurwitz, Tornetta, & Wright, 2006, p, 1873; Mendelson & Carino, 2005, p. 133; Straus et al., 2005, p. 1). The rigorous EBM methodology entails the use of a “problem solving algorithm [involving] (1) formulating answerable questions, (2) gathering evidence, (3) evaluating the evidence, (4) putting evidence into practice, and (5) evaluating the results of putting evidence into practice” (Hurwitz, et al., 2006, p. 1873). Evaluating evidence and putting evidence to practice requires clinicians to understand and apply “statistics, probability, clinical research, guided inquiry, systematic reviews, and levels of evidence” (Hurwitz, et al., 2006, p. 1873). Bartkowiak (2004, 2005a, & 2005b), Hurwitz, et al. (2006), and Straus, et al. (2005) provide instruction on how to use these tools to apply the EBM methodology. They each describe how health information technology can facilitate this process, particularly with regard to formulating the question, gathering evidence, and evaluating evidence for validity and relevance.

Web sites (Table 1) are also available, such as those from the Health Services Library of the University of Carolina (“The Well-Built Clinical Question”) and the University of Manchester (“Evidence 4U”), to assist practitioners with formulating well-constructed questions (Bartkowiak, 2004). Other web sites contain specific databases and search engines, some free of charge (PubMed/MEDLINE with the Clinical Queries feature) and others requiring a fee (Cochrane Library, Bandolier, Ovid's Evidence Based Medicine Reviews, Clinical Evidence and TRIP Database Plus), that can be accessed by practitioners to review clinical study outcomes and systematic reviews (Bartkowiak, 2005a; Straus, et. al, 2005).

Figure 1

Table 1: Useful Evidence-Based Medicine URLs

Bartkowiak (2005b) cogently distills the critical evaluation of evidence into three questions: “1) is it valid?; 2) what are the results?; and 3) is it applicable to the patient?” (p. 113). Several web sites are available to assist with this evaluation (Bartkowiak, 2005b; Hurwitz, et al., 2006; and Straus, et al., 2005). For example, the Critical Appraisal Skills Programme (CASP) section of the United Kingdom's National Health Service web site provides several resources, including forms that can be used to critically appraise evidence (Bartkowiak, 2005b, p. 113). Worksheets can also be accessed at The Centre for Evidence-Based Medicine web site, as well as a step-by-step approach to evidence appraisal (Bartkowiak, 2005b, p. 114). Evidence-based practice guidelines have also been developed by several organizations and working groups. These guidelines provide a review of the literature and rate the overall quality of evidence based on a grading scale (Hurwitz, et al., 2006, p. 1878). Hurwitz, et al. (2006) and Straus, et al. (2005) provide several web sites, such as those for the Cochrane Collaboration, Ovid's Evidence Based Medicine Reviews and the British Medical Journal's “Clinical Evidence,” where various guidelines can be accessed.

Several groups are evaluating computer-based clinical decision-support systems for practicing EBM. It is envisioned that these tools will generate patient-specific recommendations based on a comparison of the individual patient characteristics to a computerized knowledge base (Garg, et al., 2005, p. 1223). Apkon, et al. (2005) performed a patient-level randomized trial of the use of Problem-Knowledge Couplers, a decision support tool proposed for the Department of Defense health information network, on quality of care. According to Apkon, et al. (2006), the trial did not provide strong evidence to support the utility of the tool, but it did “demonstrate the value of rigorous evaluation of decision support information technology” (p. 2388). Additionally, Garg, et al. (2005) performed a systematic review of 100 controlled trials of the use of clinical decision support systems, published from 1998 through September 2004, and concluded that “the majority of available systems are not yet ready for mainstream use” (p. 1236).

Health care practitioners require access to the Internet to efficiently retrieve evidence-based information. Many health care practitioners are looking to mobile handheld devices to provide more real-time access at the patient's bedside (Cocosila & Archer, 2006; Krauskipt, 2006; Shannon, Feied, Smith, Handler & Gillam, 2005; Lu, Xiao, Sears, & Jacko, 2005). These include wireless devices and devices that are equipped to be synchronized through wired connectivity in a stationary environment. According to Lu, Xiao, Sears, and Jacko (2005), 40 percent of physicians in the United States owned a personal digital assistant (PDA) in 2003. Cocosila and Archer (2006) have identified many of the important issues associated with mobile device implementation in the health care environment, including interference with telemetry equipment, security and privacy, compatibility and integration, acceptance, and ease of use.

Analysis

Physicians and other health care practitioners have historically relied on their extensive medical training, individual experience and expert opinion to make decisions concerning the medical management of their patients. Very little evidence-based clinical information was available, and much of the evidence that was available was based on limited or poorly conducted clinical investigations (Eddy, 2005, p. 10; Mendelson & Carino, 2005, p. 134). However, this began to change in the 1980s and 1990s as the variability in medical practice and significant occurrence of preventable medical errors became public knowledge. Incorporating information based on valid and relevant clinical evidence does appear to be a valid approach to improving the medical decision-making process and ultimately, clinical outcomes.

Generating robust clinical evidence and incorporating the rigor of EBM into mainstream medical practice will require significant cost and a shift in the way medical decisions are made. Health care practitioners performing EBM must be skilled at formulating proper questions and critically evaluating the available evidence based, for many, on unfamiliar statistical concepts. Recognizing this, EBM methods are being incorporated into medical education programs (Straus, 2005). Regulatory agencies are also requiring medical product companies to produce more evidence based on properly conducted randomized controlled trials (Mendelson & Carino, 2005, p. 134). Even so, as aptly stated by Hurwitz, et al. (2006), practicing evidence-based medicine requires health care practitioners to “adopt better evidence by changing their opinions and more importantly, their practice, when confronted with good evidence. This step may be the most difficult... ” (p. 1878). Furthermore, it is also notable that there is little evidence to support the benefit of EBM (Mendelson & Carino, 2005, p. 133). Nevertheless, the significant cost of medical errors and growing public demand for quality health care will likely increase the requirement for and acceptance of EBM.

Based on the literature, current computer-based clinical decision-support systems are limited in application. According to Short, Frischer, and Bashford (2004), the adoption of current systems has been hindered due to several factors, including physicians' limited skills and confidence in the system recommendations; limited understanding of the statistical concepts, such as risk reduction; and time pressures. Still, many physicians have a positive outlook on the potential for these systems, particularly relating to practitioner performance (Apkon, et al., 2005, p. 2393; Garg, et al., 2005, p. 1229; Short, et al., 2004, p. 359). The randomized trial conducted by Apkon, et al. (2005) demonstrates the value of rigorously evaluating these tools. Hence, future improved decision-support systems should be similarly scrutinized prior to broad adoption.

Internet access is required to incorporate EBM into the medical environment. Bringing EBM to the bedside using wireless devices, however, is more complex. Although it seems reasonable that practicing EBM at the bedside using mobile devices would increase efficiency, several barriers to bedside implementation have been identified. These include privacy and security concerns, time constraints, negative patient perceptions, and ease of use challenges.

Serious concerns exist surrounding the security and privacy of wireless communications in health care organizations. Cocosila and Archer (2006) quote Albright when stating, “healthcare executives point out that their privacy and security concerns are comparable to those in the financial services industry. Even so, the heathcare industry lags behind in dealing with sensitive data” (p. 241). Cocosila and Archer (2006) explain that the goal for wireless communications is to provide “wired equivalent security” (p. 241). This goal has not yet been realized, as even those institutions at the forefront of wireless adoption do not allow wireless access to patient data (Cocosila & Archer, 2006, p. 241; Krauskopf, 2006, p. 49).

Lack of time to reference clinical evidence in the busy practice setting in addition to concerns regarding negative patient perceptions are also notable (Cocosila & Archer, 2006; Short, Frischer & Bashford, 2004). However, Lu, et al. (2005) suggest, based on their review of the literature, that using PDAs actually “saves clinicians time in regard to accessing, retrieving, and recording data... allowing them to do more direct patient care” (p. 414). Lu, et al.'s (2006) research also suggests that the physicians are more concerned than patients about negative patient perceptions of PDA use (23% of physicians were concerned versus 10% of patients).

Radio frequency interference of wireless devices with telemetry monitoring equipment is another important concern (Cocosila & Archer, 2006, p. 240). This concern can be alleviated by restricting wireless technology to Wireless Local Area Networks (WLAN). Additionally, most medical telemetry monitoring devices are migrating to a reserved communications band, the Wireless Monitoring Telemetry Service (WMTS) band at 608-614 MHz (Cocosila & Archer, 2006, p. 240).

The screen size, keypad size and screen resolution of many mobile devices cause these devices to be more difficult to use. Screen size and resolution, in particular, limit the ability of health care practitioners to read full evidence-based medicine articles and clearly see graphics and pictures (Cocosila & Archer, 2006, p. 243). Hence, practitioners using smaller mobile devices will have to depend on summaries developed specifically for these devices.

Recommendations for Health Care Administrators

Given the current health care climate with increasing demands for health care organizations to manage patients according to established clinical evidence and practice guidelines, health care administrators should incorporate at least some aspects of EBM, including the associated expenses, into their IT strategic plans. If possible, administrators should provide Internet access to health care providers, although this will not always be possible in some rural locations. As with any IT system, the requirements, comfort level and experience level of potential IT users should be assessed prior to designing or implementing an EBM system. Available EBM systems include free-access to electronic database and search engines to subscriptions to specialized EBM services or clinical decision-support systems. In general, the literature does not support the broad implementation of currently-available clinical decision-support systems. However, if the health care organization prides itself in being at the forefront of technology adoption, health care administrators should consider planning to evaluate clinical decision-support systems for future implementation. Similarly, it does not appear that wireless and other mobile devices should be implemented for bedside EBM practice alone. However, EBM capabilities should be considered as a part of any broader mobile device implementation initiative. Currently, privacy and security issues are the most significant limitations to broad wireless mobile device implementation.

Conclusion

Incorporating valid and relevant clinical evidence into medical management decisions will continue to be emphasized as efforts are made to reduce medical care errors and variability. Since IT plays a crucial role in EBM implementation, health care administrators will need to be familiar with EBM concepts and incorporate these into their strategic IT plans. Several web sites and recent literature are available to describe and assist with performing EBM. Those practicing EBM will need to be able to assess the relevance and validity of available evidence, which may require new skills. Health care IT will continue to be an essential component of EBM. Future IT will likely include mobile devices and clinical decision-support tools.

Acknowledgement

Glossary

Clinical Decision Support System (CDSS): Any system designed to improve clinical decision making related to diagnostic or therapeutic processes of care. (Agency for Healthcare Research and Quality Patient Safety Network, http://psnet.ahrq.gov/glossary.aspx)

Evidence-based medicine (EBM): the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients. The practice of evidence-based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research. (Centre for Evidence-Based Medicine, http://www.cebm.net/glossary.asp)

Information Technology (IT): Processing information by computer. IT is an umbrella term for the entire computer industry and its latest moniker, which took hold in the 1990s. It actually took 40 years before the industry settled on what to call itself. First it was “electronic data processing” (EDP), which was followed by “management information systems” (MIS) and then “information systems” (IS). (TechWeb Technoencyclopedia, http://www.techweb.com/encyclopedia/)

Personal Digital Assistant (PDA): A handheld computer for managing contacts, appointments and tasks. It typically includes a name and address database, calendar, to-do list and note taker, which are the functions in a personal information manager. Wireless PDAs may also offer e-mail, Web browsing and cellular phone service. Data are synchronized between the PDA and desktop computer via a cabled connection or wireless. (TechWeb Technoencyclopedia, http://www.techweb.com/encyclopedia/)