Taming the EHR (Electronic Health Record) - There is Hope.

With increasing diffusion of EHR technology over the last half decade, clinician burnout is rising. As healthcare is a complex and highly regulated field, the rapid and mass adoption of EHR technology has created disruption for highly skilled workers such as clinicians. Although, much has been written about dissatisfaction with the EHR (electronic health record), a paucity of immediate solutions exists in the literature. This article suggests three actionable steps health systems and clinicians can make to expedite gains from and mitigate the effect of the EHR on clinical practice.

Discussion of "Combining Health Data Uses to Ignite Health System Learning".

This article is part of a For-Discussion-Section of Methods of Information in Medicine about the paper "Combining Health Data Uses to Ignite Health System Learning" written by John D. Ainsworth and Iain E. Buchan [1]. It is introduced by an editorial. This article contains the combined commentaries invited to independently comment on the paper of Ainsworth and Buchan. In subsequent issues the discussion can continue through letters to the editor. With these comments on the paper "Combining Health Data Uses to Ignite Health System Learning", written by John D. Ainsworth and Iain E. Buchan [1], the journal seeks to stimulate a broad discussion on new ways for combining data sources for the reuse of health data in order to identify new opportunities for health system learning. An international group of experts has been invited by the editor of Methods to comment on this paper. Each of the invited commentaries forms one section of this paper.

Safe Implementation of Computerized Provider Order Entry for Adult Oncology.

BACKGROUND: Oncology has lagged in CPOE adoption due to the narrow therapeutic index of chemotherapy drugs, individualized dosing based on weight and height, regimen complexity, and workflows that include hard stops where safety checks are performed and documented. OBJECTIVES: We sought to establish CPOE for chemotherapy ordering and administration in an academic teaching institution using a commercially available CPOE system. METHODS: A commercially available CPOE system was implemented throughout the hospital. A multidisciplinary team identified key safety gaps that required the development of a customized complex order display and a verification documentation workflow. Staff reported safety events were monitored for two years and compared to the year prior to go live. RESULTS: A workflow was enabled to capture real-time provider verification status during the time from ordering to the administration of chemotherapy. A customized display system was embedded in the EMR to provide a single screen view of the relevant parameters of chemotherapy doses including current and previous patient measurements of height and weight, dose adjustments, provider verifications, prior chemotherapy regimens, and a synopsis of the standard regimen for reference. Our system went live with 127 chemotherapy plans and has been expanded to 189. Staff reported safety events decreased following implementation, particularly in the area of prescribing and transcribing by the second year of use. CONCLUSIONS: We observed reduced staff reported safety events following implementation of CPOE for inpatient chemotherapy using an electronic verification workflow and an embedded custom clinical decision support page. This implementation demonstrates that CPOE can be safely used for inpatient chemotherapy, even in an extremely complex environment.

Drug interaction alert override rates in the Meaningful Use era: no evidence of progress.

BACKGROUND: Interruptive drug interaction alerts may reduce adverse drug events and are required for Stage I Meaningful Use attestation. For the last decade override rates have been very high. Despite their widespread use in commercial EHR systems, previously described interventions to improve alert frequency and acceptance have not been well studied. OBJECTIVES: (1) To measure override rates of inpatient medication alerts within a commercial clinical decision support system, and assess the impact of local customization efforts. (2) To compare override rates between drug-drug interaction and drug-allergy interaction alerts, between attending and resident physicians, and between public and academic hospitals. (3) To measure the correlation between physicians' individual alert quantities and override rates as an indicator of potential alert fatigue. METHODS: We retrospectively analyzed physician responses to drug-drug and drug-allergy interaction alerts, as generated by a common decision support product in a large teaching hospital system. RESULTS: (1) Over four days, 461 different physicians entered 18,354 medication orders, resulting in 2,455 visible alerts; 2,280 alerts (93%) were overridden. (2) The drug-drug alert override rate was 95.1%, statistically higher than the rate for drug-allergy alerts (90.9%) (p < 0.001). There was no significant difference in override rates between attendings and residents, or between hospitals. (3) Physicians saw a mean of 1.3 alerts per day, and the number of alerts per physician was not significantly correlated with override rate (R2 = 0.03, p = 0.41). CONCLUSIONS: Despite intensive efforts to improve a commercial drug interaction alert system and to reduce alerting, override rates remain as high as reported over a decade ago. Alert fatigue does not seem to contribute. The results suggest the need to fundamentally question the premises of drug interaction alert systems.

Towards meaningful medication-related clinical decision support: recommendations for an initial implementation.

SUMMARY: Clinical decision support (CDS) can improve safety, quality, and cost-effectiveness of patient care, especially when implemented in computerized provider order entry (CPOE) applications. Medication-related decision support logic forms a large component of the CDS logic in any CPOE system. However, organizations wishing to implement CDS must either purchase the computable clinical content or develop it themselves. Content provided by vendors does not always meet local expectations. Most organizations lack the resources to customize the clinical content and the expertise to implement it effectively. In this paper, we describe the recommendations of a national expert panel on two basic medication-related CDS areas, specifically, drug-drug interaction (DDI) checking and duplicate therapy checking. The goals of this study were to define a starter set of medication-related alerts that healthcare organizations can implement in their clinical information systems. We also draw on the experiences of diverse institutions to highlight the realities of implementing medication decision support. These findings represent the experiences of institutions with a long history in the domain of medication decision support, and the hope is that this guidance may improve the feasibility and efficiency CDS adoption across healthcare settings.

Computer order entry system decreased use of sliding scale insulin regimens.

OBJECTIVES: Despite evidence documenting their ineffectiveness, sliding scale insulin is a commonly used regimen for glucose management for hospitalized patients with diabetes mellitus. At the Veterans Affairs Puget Sound Medical Center, where computer order entry has been mandated, we tested the hypothesis that an evidence-based minimal intervention order (supplemental insulin only when fasting serum glucoses exceeded 400 mg/dl) would decrease the use of sliding scale insulin orders. METHODS: Using a computerized order entry system, providers were initially offered a traditional sliding scale order or their own ad hoc orders for glycemic control of inpatients. After 34 weeks providers were offered a third option; a "minimal intervention order" with supplemental insulin only for glucose > 400 mg/dl. We extracted all regular insulin orders and performed a retrospective review of insulin sliding scale orders written between December 1, 1998 and November 16, 1999. We compared the frequency of traditional insulin sliding scale orders before and after the introduction of the minimal intervention order. RESULTS: Nearly all orders in the first 34 weeks were traditional insulin sliding scales. We found a significant decrease in the number of traditional insulin sliding scale orders in the 16 weeks after the introduction of a computerized quick-order for minimal intervention, from 978/1007 (97.1%) to 254/398 (63.8%) (P < 0.001). CONCLUSIONS: A simple, evidenced-based quick-order in a computer order entry system rapidly and significantly reduced use of sliding scale insulin regimens for glycemic control of inpatients.

Evaluation of a command-line parser-based order entry pathway for the Department of Veterans Affairs electronic patient record.

OBJECTIVE: To improve and simplify electronic order entry in an existing electronic patient record, the authors developed an alternative system for entering orders, which is based on a command- interface using robust and simple natural-language techniques. DESIGN: The authors conducted a randomized evaluation of the new entry pathway, measuring time to complete a standard set of orders, and users' satisfaction measured by questionnaire. A group of 16 physician volunteers from the staff of the Department of Veterans Affairs Puget Sound Health Care System-Seattle Division participated in the evaluation. RESULTS: Thirteen of the 16 physicians (81%) were able to enter medical orders more quickly using the natural-language-based entry system than the standard graphical user interface that uses menus and dialogs (mean time spared, 16.06 +/- 4.52 minutes; P=0.029). Compared with the graphical user interface, the command--based pathway was perceived as easier to learn (P<0.01), was considered easier to use and faster (P<0.01), and was rated better overall (P<0.05). CONCLUSION: Physicians found the command- interface easier to learn and faster to use than the usual menu-driven system. The major advantage of the system is that it combines an intuitive graphical user interface with the power and speed of a natural-language analyzer.

Extending the VA CPRS electronic patient record order entry system using natural language processing techniques.

An automated practitioner order entry system was recently implemented at the VA Puget Sound Health Care System. Since the introduction of this system, we have experienced various problems, among them an increase in time required for practitioners to enter orders. In order to improve usability and acceptance of the order entry, an alternate pathway was built within CPRS that allows direct natural language based order entry. Implementation of the extension in CPRS has been made possible because of the three layers CPRS architecture and its strong object oriented models. This paper discusses the advantages and needs for a natural language based order entry system and its implementation within an existing order entry system.

Use of a clinical event monitor to prevent and detect medication errors.

Errors in health care facilities are common and often unrecognized. We have used our clinical event monitor to prevent and detect medication errors by scrutinizing electronic messages sent to it when any medication order is written in our facility. A growing collection of medication safety rules covering dose limit errors, laboratory monitoring, and other topics may be applied to each medication order message to provide an additional layer of protection beyond existing order checks, reminders, and alerts available within our computer-based record system. During a typical day the event monitor receives 4802 messages, of which 4719 pertain to medication orders. We have found the clinical event monitor to be a valuable tool for clinicians and quality management groups charged with improving medication safety.

Computer decision support systems.

Computer decision support systems are computer applications designed to aid clinicians in making diagnostic and therapeutic decisions in patient care. They can simplify access to data needed to make decisions, provide reminders and prompts at the time of a patient encounter, assist in establishing a diagnosis and in entering appropriate orders, and alert clinicians when new patterns in patient data are recognized. Decision support systems that present patient-specific recommendations in a form that can save clinicians time have been shown to be highly effective, sustainable tools for changing clinician behavior. Designing and implementing such systems is challenging because of the computing infrastructure required, the need for patient data in a machine-processible form, and the changes to existing workflow that may result. Despite these difficulties, there is substantial evidence from trials in a wide range of clinical settings that computer decision support systems help clinicians do a better job caring for patients. As computer-based records and order-entry systems become more common, automated decision support systems will be used more broadly.

The transition to automated practitioner order entry in a teaching hospital: the VA Puget Sound experience.

We recently installed an automated practitioner order entry system on our busiest inpatient wards and critical care units. The installation followed 20 months preparation in which we created the workstation, network, and host infrastructure, developed requisite policies, recruited personnel to support the system, and installed the software in areas where the pace of order entry was less intense. Since implementing automated order entry, we have experienced problems such as an increase in time required for practitioners to enter orders, workflow changes on inpatient units, difficulties with patient transfers, and others. Our user support system has been heavily used during the transition period. Software tailoring and enhancements designed to address these problems are planned, as is installation of the order entry system in remaining clinical units in our medical centers.

Development of a clinical event monitor for use with the Veterans Affairs Computerized Patient Record System and other data sources.

We are developing an event monitor to operate with the Veterans Affairs Computerized Patient Record System (CPRS). The event monitor is designed to receive messages when important patient events such as posting of new results, patient movement, and orders occur. Our design separates the event monitor from CPRS itself, using communication via a network connection to receive HL7 messages, to access other data needed to run rules, and to communicate with providers by message display, electronic mail and other mechanisms. Results from operation of the event monitor using patient data in our test account show that a wide variety of data can be accessed by the event monitor with acceptable response times.

Practicing population-based care in an HMO: evaluation after 18 months.

OBJECTIVE: To determine whether population-based care in a primary care practice results in improvement in compliance with patient care guidelines. DESIGN: Time series analysis. SETTING: One primary care practice in Group Health Cooperative of Puget Sound (GHC). PARTICIPANTS: Approximately 1500 enrollees cared for by the practice. INTERVENTIONS: An ongoing approach to aid clinical planning at the level of the primary care team--population-based care--that depends on clinical guidelines, a computing system to provide epidemiologic data on guideline performance in the practice and reminders, and a process whereby the practice team analyzed and designs interventions for specific clinical problems. MAIN OUTCOME MEASURES: We compared compliance with practice guidelines for preventive care and chronic illness management at baseline and after 18 months in the intervention population with other patients in the same clinic and with patients in GHC as a whole. RESULTS: Compliance with breast cancer screening and colorectal cancer screening guidelines in the intervention population increased from baseline 32% and 18% respectively. These increases were significantly greater than in the remainder of the clinic or in GHC as a whole. CONCLUSIONS: The availability of practice-based data, clinical guidelines and a local intervention design process resulted in significant improvements in compliance with patient care guidelines.

The role of the information architect at King Faisal Specialist Hospital and Research Centre.

Successful implementation of integrated clinical information system requires modification of the institution's long range strategic plans and its personnel's behavior. The changes warrant a concerted effort on the part of many different individuals; this paper describes the role of the Information Architect whose primary functions are to steer the process to fulfill stated objectives and build consensus where divergent forces are at work. The workings of the Architect is presented in context of a unique Middle-Eastern institution currently undergoing automation of clinical information.

A comparison of four schemes for codification of problem lists.

We set out to evaluate the completeness of four major coding schemes in representation of the patient problem list: the Unified Medical Language System (UMLS, 4th edition), the Systematized Nomenclature of Medicine (SNOMED International), the Read coding system (version 2), and the International Classification of Diseases (9th Clinical Modification)(ICD-9-CM). We gathered 400 problems from patient records at primary care sites in Omaha and Seattle. Matching these against the best description found in each of the coding schemes, we asked five medical faculty reviewers to rate the matches on a five-point Likert scale assessing their satisfaction with the results. For the four schemes, we computed the following rates of dissatisfaction, satisfaction, and average scores: [table: see text] From this analysis, we conclude that UMLS and SNOMED performed substantially better in capturing the clinical content of the problem lists than READ or ICD-9-CM. No scheme could be considered comprehensive. Depending on the goal of systems developers, UMLS and SNOMED may offer different, and complementary, advantages.

How useful is the UMLS metathesaurus in developing a controlled vocabulary for an automated problem list?

We are developing a set of problem list phrases to be used in the automated problem list of a prototype clinical computing system. Because of the large number of terms in the Unified Medical Language System (UMLS) and the links between them, we are experimenting with the use of the UMLS as the foundation for our problem list phrase set. We have found the UMLS to be very useful for this project, but that it lacks many phases clinicians wish to include in the problem list. Internal linkages between phrases provided in the UMLS are not well suited to our needs. We plan to continue our use of the UMLS but to add problem list phrases and linkages between phrases to support browsing and decision support applications.

The ambulatory medical record project at Group Health Cooperative: what did a decade of experience teach us?

Computer-based record systems are documented to improve patient care (Barnett, 1984; McDonald & Tierney, 1988) and their importance in the future is widely accepted. The report of the Institute of Medicine on patient computer-based record systems (Detmer, 1991) will help guide the development of future computer-based record systems and will likely stimulate renewed interest in them. We believe computing systems have great value to an HMO but understand that the benefits do not come without the risk of setbacks. We plan to build on what we have learned from our decade of experience.

How well does ICD9 represent phrases used in the medical record problem list?

We are considering using the International Classification of Diseases with Clinical Modifications, Ninth Revision (ICD9) as the basis for an automated problem list for a clinical information system. To determine physician satisfaction with an ICD9 representation of phrases used in the medical record problem list, we asked 6 physicians to evaluate ICD9 representations of 332 phrases taken from medical record problem lists, using a scale of 1(extremely dissatisfied) to 5 (extremely satisfied). The mean score was 3.0; intraclass correlation for 25 phrases given to all 6 evaluators was 0.47. In 45% of the phrases the physicians were dissatisfied with the ICD9 representation. In developing an automated problem list it is desirable to improve the level of satisfaction of clinician users above this level. This could be done by modifying ICD9, using a different vocabulary to represent the problem list, or improving the method of assigning ICD9 codes.