Clinical Decision Support: a 25 Year Retrospective and a 25 Year Vision.

OBJECTIVE: The objective of this review is to summarize the state of the art of clinical decision support (CDS) circa 1990, review progress in the 25 year interval from that time, and provide a vision of what CDS might look like 25 years hence, or circa 2040. METHOD: Informal review of the medical literature with iterative review and discussion among the authors to arrive at six axes (data, knowledge, inference, architecture and technology, implementation and integration, and users) to frame the review and discussion of selected barriers and facilitators to the effective use of CDS. RESULT: In each of the six axes, significant progress has been made. Key advances in structuring and encoding standardized data with an increased availability of data, development of knowledge bases for CDS, and improvement of capabilities to share knowledge artifacts, explosion of methods analyzing and inferring from clinical data, evolution of information technologies and architectures to facilitate the broad application of CDS, improvement of methods to implement CDS and integrate CDS into the clinical workflow, and increasing sophistication of the end-user, all have played a role in improving the effective use of CDS in healthcare delivery. CONCLUSION: CDS has evolved dramatically over the past 25 years and will likely evolve just as dramatically or more so over the next 25 years. Increasingly, the clinical encounter between a clinician and a patient will be supported by a wide variety of cognitive aides to support diagnosis, treatment, care-coordination, surveillance and prevention, and health maintenance or wellness.

New Unintended Adverse Consequences of Electronic Health Records.

Although the health information technology industry has made considerable progress in the design, development, implementation, and use of electronic health records (EHRs), the lofty expectations of the early pioneers have not been met. In 2006, the Provider Order Entry Team at Oregon Health & Science University described a set of unintended adverse consequences (UACs), or unpredictable, emergent problems associated with computer-based provider order entry implementation, use, and maintenance. Many of these originally identified UACs have not been completely addressed or alleviated, some have evolved over time, and some new ones have emerged as EHRs became more widely available. The rapid increase in the adoption of EHRs, coupled with the changes in the types and attitudes of clinical users, has led to several new UACs, specifically: complete clinical information unavailable at the point of care; lack of innovations to improve system usability leading to frustrating user experiences; inadvertent disclosure of large amounts of patient-specific information; increased focus on computer-based quality measurement negatively affecting clinical workflows and patient-provider interactions; information overload from marginally useful computer-generated data; and a decline in the development and use of internally-developed EHRs. While each of these new UACs poses significant challenges to EHR developers and users alike, they also offer many opportunities. The challenge for clinical informatics researchers is to continue to refine our current systems while exploring new methods of overcoming these challenges and developing innovations to improve EHR interoperability, usability, security, functionality, clinical quality measurement, and information summarization and display.

Validation of a Crowdsourcing Methodology for Developing a Knowledge Base of Related Problem-Medication Pairs.

BACKGROUND: Clinical knowledge bases of problem-medication pairs are necessary for many informatics solutions that improve patient safety, such as clinical summarization. However, developing these knowledge bases can be challenging. OBJECTIVE: We sought to validate a previously developed crowdsourcing approach for generating a knowledge base of problem-medication pairs in a large, non-university health care system with a widely used, commercially available electronic health record. METHODS: We first retrieved medications and problems entered in the electronic health record by clinicians during routine care during a six month study period. Following the previously published approach, we calculated the link frequency and link ratio for each pair then identified a threshold cutoff for estimated problem-medication pair appropriateness through clinician review; problem-medication pairs meeting the threshold were included in the resulting knowledge base. We selected 50 medications and their gold standard indications to compare the resulting knowledge base to the pilot knowledge base developed previously and determine its recall and precision. RESULTS: The resulting knowledge base contained 26,912 pairs, had a recall of 62.3% and a precision of 87.5%, and outperformed the pilot knowledge base containing 11,167 pairs from the previous study, which had a recall of 46.9% and a precision of 83.3%. CONCLUSIONS: We validated the crowdsourcing approach for generating a knowledge base of problem-medication pairs in a large non-university health care system with a widely used, commercially available electronic health record, indicating that the approach may be generalizable across healthcare settings and clinical systems. Further research is necessary to better evaluate the knowledge, to compare crowdsourcing with other approaches, and to evaluate if incorporating the knowledge into electronic health records improves patient outcomes.

Medical informatics on the Internet: creating the sci.med. informatics newsgroup.

A Usenet newsgroup, sci.med.informatics, has been created to serve as an international electronic forum for discussion of issues related to medical informatics. The creation process follows a set of administrative rules set out by the Usenet administration on the Internet and consists of five steps: 1) informal discussion, 2) request for formal discussion, 3) formal discussion, 4) voting, and 5) posting of results. The newsgroup can be accessed using any news reader via the Internet.

Computer-based physician order entry: the state of the art.

Direct computer-based physician order entry has been the subject of debate for over 20 years. Many sites have implemented systems successfully. Others have failed outright or flirted with disaster, incurring substantial delays, cost overruns, and threatened work actions. The rationale for physician order entry includes process improvement, support of cost-conscious decision making, clinical decision support, and optimization of physicians' time. Barriers to physician order entry result from the changes required in practice patterns, roles within the care team, teaching patterns, and institutional policies. Key ingredients for successful implementation include: the system must be fast and easy to use, the user interface must behave consistently in all situations, the institution must have broad and committed involvement and direction by clinicians prior to implementation, the top leadership of the organization must be committed to the project, and a group of problem solvers and users must meet regularly to work out procedural issues. This article reviews the peer-reviewed scientific literature to present the current state of the art of computer-based physician order entry.

Work-sampling: a statistical approach to evaluation of the effect of computers on work patterns in healthcare.

An increasing number of health-care institutions are in the process of implementing clinical computing systems. The need for an accurate assessment of the clinical, administrative, social, and financial effects of such systems has been recognized. Techniques have been developed to evaluate these effects on the work patterns of health-care workers including: time-motion analysis, subjective evaluations, review of departmental statistics, personal activity records, and work-sampling. This study reviews these techniques, discusses both positive and negative aspects, and presents a step-by-step description of work-sampling.

A quantitative ranking of the Biomedical Informatics serials.

We have developed a quantitative serial ranking system based on multiple citation analysis techniques, library use statistics, expert opinion, and selected distinguishing publication characteristics. Evaluation criteria categories include: average Science Citation Index (Impact Factor, Immediacy Index, Total citations) rankings from 1987 to 1992; citation source counts of multiple "core" biomedical informatics publications; a questionnaire sent to American College of Medical Informatics Fellows; publication delay; distinguishing characteristics (e.g., subscription cost, total circulation, year established, places indexed, affiliation with a professional society, major biomedical resource library holdings); and the total number of interlibrary loan requests to the U. S. National Library of Medicine. The top serials were Computers and Biomedical Research, MD Computing, Methods of Information in Medicine, Medical Decision Making and Computers in Biology and Medicine.

The multi-trellis software architecture and the intelligent cardiovascular monitor.

A real-time, intelligent cardiovascular monitor is complex. It must process multiple waveforms, recognize artifacts, extract pertinent parameters, recognize a patient's clinical state, analyze the problem and formulate a response. This paper presents the multi-trellis (a collection of process trellises), a software architecture for building such a monitor. A process trellis is a uniform hierarchical framework for heterogeneous program modules. The multi-trellis extension allows one to compile several process trellis programs with widely varying run-time requirements into a single executable program that it is efficient, predictable and usable. Our prototype consists of two process trellises. The lower trellis contains processes to analyze three different analog signals: the blood pressure from a non-invasive monitor and an arterial catheter, and the ECG. The upper trellis contains processes to help detect evolving hemodynamic trends, identify abnormalities, and present a succinct summary to the clinician. Our prototype shows that the multi-trellis is a demonstrably useful software architecture for building these real-time, intelligent monitors.

A survey of patient-provider e-mail communication: what do patients think?

Communication between patients and providers forms the backbone of the patient-provider relationship. Often such communication is strained due to time and space limitations on the part of both patients and providers. Many healthcare organizations are developing secure e-mail communication facilities to allow patients to exchange e-mail messages with their providers. Providers are worried that opening such lines of communication will inundate them with vast quantities of e-mail from their patients. Patients are worried that their messages will be intercepted and read by unauthorized people. In an attempt to determine how a group of internet-active, e-mail-ready patients currently use, or potentially view, the ability to exchange e-mail messages with their health care providers, we distributed a survey via e-mail to over 9500 patients. After determining each patient's e-mail activity level (based on the number of messages sent each day), we asked questions such as: "Have you ever sent e-mail to your provider?" "What issues or concerns have prevented you from sending e-mail messages to your provider?" "If your provider were to tell you that someone in his/her office may screen, read or perhaps reply to your message before he/she sees it, to what extent would you be concerned about this?" and "How would you rate your overall satisfaction with the use of e-mail to communicate with your provider?" Results from the survey indicate that nearly 85% of the patients surveyed send at least one e-mail message per day, but that very few (i.e. 6%) of the patients have actually sent an e-mail message to their provider. Interestingly, over half of the patients indicated that they would like to send their providers e-mail, but that they do not know their provider's e-mail address.

A computer-based outpatient clinical referral system.

The process of generating a clinical referral for a patient, and the resulting transfer of information from the primary care physician to the specialist and back again, are key components in the struggle to deliver less costly and more effective clinical care. We have created a computer-based, outpatient clinical referral application that facilitates: (1) identifying an appropriate specialist; (2) collecting the clinical, demographic, and financial data required to generate a referral; and (3) transferring the information between the specialist and the primary care physician (PCP). This article describes the development of the application itself and several of the knowledge bases that were created to facilitate this process. Preliminary results indicate that the new computer-based referral process is faster to use than conventional methods.

Physiologic trend detection and artifact rejection: a parallel implementation of a multi-state Kalman filtering algorithm.

Using a parallel implementation of the multi-state Kalman filtering algorithm, we have developed an accurate method of reliably detecting and identifying trends, abrupt changes, and artifacts from multiple physiologic data streams in real-time. The Kalman filter algorithm was implemented within an innovative software architecture for parallel computation: a parallel process trellis. Examples, processed in real-time, of both simulated and actual data serve to illustrate the potential value of the Kalman filter as a tool in physiologic monitoring.

Developing interactive computer-based simulations: an object-oriented development methodology enhances computer-assisted instruction.

The purpose of this research was to investigate the application of object-oriented technology and AI techniques to enhance development of computer-based training simulations. Towards that end, a comprehensive computer-assisted instructional unit was developed to teach the skills and concepts of window-based applications, the OS/2 desktop, and the use of a patient care information system. By taking advantage of sophisticated computer graphics for the visual representation of objects and the behavioral modeling capabilities of the object-oriented language, domain knowledge modeling and human-computer interactions were implemented without complex natural language processing techniques. The results of this research indicate that nurses and physicians are able to learn the basic skills and concepts of computer systems and how to query for patient information. The new methodology described for building these computer-assisted instructional simulations significantly eased the training and teaching of large numbers of nurses and physicians and simplified their transition to a complex, computer-based hospital information system environment.

BIO-SPEAD: a parallel computing environment to accelerate development of biologic signal processing algorithms.

We have created BIO-SPEAD (pronounced speed), a BIOlogical Signal Processing Environment for Algorithm Development. BIO-SPEAD is designed to accelerate development of complex algorithms which integrate information derived from single or multiple physiologic waveforms. BIO-SPEAD currently performs all of the basic analyses of several arterial blood pressure waveforms, and allows the user to utilize the results of those low-level analyses for development of more complex algorithms. We utilized a parallel programming architecture called the Process Trellis which keeps the different tasks, or processes, within BIO-SPEAD independent of each other. Additionally, we have developed a graphics interface to enable the user to visualize the waveform under analysis, the low-level system analysis, and the internal workings of the algorithm under development. The system has been used for several algorithm development projects and has demonstrated its utility.

Computerized management of patient care in a complex, controlled clinical trial in the intensive care unit.

Acute respiratory distress syndrome (ARDS) is often not responsive to conventional supportive therapy and the mortality rate may exceed 90%. A new form of supportive care, extracorporeal carbon dioxide removal (ECCO2R), has shown a dramatic increase in survival (48%). A controlled clinical trial of the new ECCO2R therapy versus conventional continuous positive pressure ventilation (CPPV) is being initiated. Detailed care protocols have been developed by 'expert' critical care physicians for the management of patients. Using a blackboard control architecture, the protocols have been implemented on an existing hospital information system and will direct patient care and help manage the controlled clinical trial. Therapeutic instructions are automatically generated by the computer from data input by physicians, nurses, respiratory therapists, and the laboratory. Preliminary results show that the computerized protocol system can direct therapy for acutely ill patients.

Grand challenges in dental informatics.

When truly significant scientific challenges are overcome, it profoundly changes the daily activities, as well as the future research activities, of everyone involved in the related field. By identifying and describing the grand challenges facing a scientific field, we can help funding agencies identify and prioritize projects for support, stimulate and encourage new investigators to work on these intellectual and technological challenges, and help define the field itself. In this article, we present an informatics-oriented, future-patient-care scenario, then describe a series of applications and the related informatics grand challenges facing the dental field today. New techniques and technologies to help us overcome these challenges would facilitate the development of truly monumental applications, such as a comprehensive electronic oral health record, an automated dental treatment planning system for all diagnoses, or a system to profile patient risk for chronic oral diseases.

Design and development of a computer-based clinical referral system for use within a physician hospital organization.

The process of creating a clinical referral for a patient and the transfer of information from the primary care physician to the specialist and back again is a key component in the struggle to deliver less costly and more effective clinical care. We have created a computer-based clinical referral application which facilitates 1) identifying an appropriate specialist; 2) collecting the clinical, demographic, and financial data required to generate a referral; and 3) transferring the information between the specialist and the primary care physician. Preliminary results indicate that the new computer-based process is faster.

Components of the optimal ambulatory care computing environment.

We present here a framework of core components of an ambulatory care computing environment, based on clinical and functional needs and workflow scenarios. We have established this framework through the use of two study devices: a vision of the clinical office of the future, and a survey of possible computer applications, both designed to help clinicians and practice directors communicate their information needs to systems designers. Clinicians prioritize applications based on strategic and practice goals: support for clinical users' workflow, improved quality of care, reduced cost of care, and the ability to measure performance and status. By reorganizing the needed functionality from a clinical viewpoint into a technical viewpoint, we are able to identify core information components for systems design. Based on this analysis, information needs in the ambulatory environment can be divided into five primary functions: patient data retrieval, documentation, communication, knowledge resources, and aggregate reporting. Three other fundamental processes--knowledge-based interventions, information integration, and confidentiality--run through all of these front-line functions. Component applications and data structures built with this framework in mind will afford a maximum combination of functionality and flexibility to handle future changes in the clinical environment.

Death, taxes and advance directives.

Suboptimal care at the end-of-life can be due to lack of access or knowledge of patient wishes. Ambiguity is often the result of non-standardized formats. Borrowing digital technology from other industries and using existing health information infrastructure can greatly improve the completion, storage, and distribution of advance directives. We believe several simple, low-cost adaptations to regional and federal programs can raise the standard of end-of-life care.

State of the art in clinical informatics: evidence and examples.

OBJECTIVE: The field of clinical informatics has expanded substantially in the six decades since its inception. Early research focused on simple demonstrations that health information technology (HIT) such as electronic health records (EHRs), computerized provider order entry (CPOE), and clinical decision support (CDS) systems were feasible and potentially beneficial in clinical practice. METHODS: In this review, we present recent evidence on clinical informatics in the United States covering three themes: 1) clinical informatics systems and interventions for providers, including EHRs, CPOE, CDS, and health information exchange; 2) consumer health informatics systems, including personal health records and web-based and mobile HIT; and 3) methods and governance for clinical informatics, including EHR usability; data mining, text mining, natural language processing, privacy, and security. RESULTS: Substantial progress has been made in demonstrating that various clinical informatics methodologies and applications improve the structure, process, and outcomes of various facets of the healthcare system. CONCLUSION: Over the coming years, much more will be expected from the field. As we move past the "early adopters" in Rogers' diffusion of innovations' curve through the "early majority" and into the "late majority," there will be a crucial need for new research methodologies and clinical applications that have been rigorously demonstrated to work (i.e., to improve health outcomes) in multiple settings with different types of patients and clinicians.

Rapid assessment of clinical information systems in the healthcare setting: an efficient method for time-pressed evaluation.

OBJECTIVE: Recent legislation in the United States provides strong incentives for implementation of electronic health records (EHRs). The ensuing transformation in U.S. health care will increase demand for new methods to evaluate clinical informatics interventions. Timeline constraints and a rapidly changing environment will make traditional evaluation techniques burdensome. This paper describes an anthropological approach that provides a fast and flexible way to evaluate clinical information systems. METHODS: Adapting mixed-method evaluation approaches from anthropology, we describe a rapid assessment process (RAP) for assessing clinical informatics interventions in health care that we developed and used during seven site visits to diverse community hospitals and primary care settings in the U.S. SETTING: Our multidisciplinary team used RAP to evaluate factors that either encouraged people to use clinical decision support (CDS) systems or interfered with use of these systems in settings ranging from large urban hospitals to single-practitioner, private family practices in small towns. RESULTS: Critical elements of the method include: 1) developing a fieldwork guide; 2) carefully selecting observation sites and participants; 3) thoroughly preparing for site visits; 4) partnering with local collaborators; 5) collecting robust data by using multiple researchers and methods; and 6) analyzing and reporting data in a structured manner helpful to the organizations being evaluated. CONCLUSIONS: RAP, iteratively developed over the course of visits to seven clinical sites across the U.S., has succeeded in allowing a multidisciplinary team of informatics researchers to plan, gather and analyze data, and report results in a maximally efficient manner.