Consult a clinical decision support system workbook for optimal implementation.

How can you arm clinicians with convenient checks and balances for enhanced care delivery? Consider a clinical decision support system (CDSS)-computerized applications that offer care providers clinical knowledge and information to enhance patient care. CDSSs help detect dangerous or potentially dangerous laboratory values; drug-to-drug interactions; patient allergies, risk profiles, or treatment contraindications; and errors of omission. Nurses use them to gather important information to support their decisions and to provide reminders and recommendations at the right place at the right time.

CDSS development and evaluation for the nursing profession is in its infancy. Nursing literature indicates the development of systems to support patient preference-based care planning, critical care pharmacology applications, cancer pain reporting and management, and cancer pain management.1-4 Early CDSSs failed because they weren't integrated into the hospital information system or clinical workflow.5 Improvements in CDSSs have garnered better results, enabling facilities to substantially decrease adverse patient outcomes, such as falls and pressure ulcers.6

STEP-BY-STEP GUIDE

Although you no doubt have great ideas for improving care processes and patient safety, the task of actually designing and implementing a CDSS is daunting. The Decision Support Workgroup, comprising a nationally representative group of clinicians and informaticists from the Healthcare Information and Management Systems Society Patient Safety Task Force, recognized this challenge and developed a new Web-based workbook to help others create and implement clinical decision support.7 Before publication, a multidisciplinary panel of experts in information and library science, medicine, and nursing reviewed the workbook. Nurses can view the Clinical Decision Support Implementer's Workbook (CDSIW), free of charge, at http://www.himss.org/asp/cdsworkbook.asp.

The workbook contains six major steps to implementation. Consider these key elements of those major steps:

1. Identify stakeholders and goals. CDSS developers must identify the key persons and committees with a stake in proposing, validating, supporting, communicating, or using the CDSS. Together, developers and stakeholders will determine the organizational goals, objectives, clinical goals, and clinical actions for the program. Three worksheets are available to guide this activity.

2. Catalog available information systems. The CDSS developer must review all existing information systems to determine which tools already exist that could help in implementation and which tools must be added. Note the functionality and content of each system along with the type of data each system contains and how they're coded, communicated, and aggregated. This step helps the developer identify existing systems that may propel or hinder the CDSS implementation process. Potential facilitators include information systems that share common data elements or have standardized, coded data aggregated in a clinical data repository or data warehouse. Barriers include data collected on paper in various, nonstandard formats or electronic data housed in disparate systems.8

3. Select CDSS interventions. This important, complex step links your organization's goals and objectives with its available information infrastructure. Decision support developers must map their objective-to-objective classes and intervention types, and determine their place in the workflow. Next, the implementer determines where the system fits in the workflow, considering the setting, time frame, and available information infrastructure. This step could include setting in-patient, ambulatory, home care-and time frame-previsit, patient intake, or ordering. The implementer must address who'll use the system and when, how, where, and what they'll see.

4. Validate and finalize the program. With the CDSS plan in hand, the implementer must seek validation with appropriate stakeholders. The stakeholders will review the plan for acceptability, workflow fit, likelihood of achieving desired outcomes, and adverse impact on other healthcare operations. Provide a mechanism for receiving feedback about the proposed system's impact and use the feedback for system improvement. Consider the impact on current operations inside and outside the organization. For example, our hospital discharge referral system may greatly impact the number of referrals sent to home care.

5. Put interventions into action. Developers must undertake careful planning, testing, and implementation to ensure users accept and operate the CDSS appropriately. Before launching the program, pilot test it with appropriate stakeholders to ensure it performs as expected. Use a wide variety of test cases to evaluate the system's impact on the clinical operations and end-user workflow, modifying as needed. This section of the workbook contains four worksheets to guide implementers as they put the CDSS into action.

6. Monitor results and refine the program. Implementers must continuously monitor the CDSS to ensure it's meeting its intended goals and objectives. Continuously gather and integrate user feedback, evaluate clinical outcomes related to the system, and maintain current clinical content. This iterative process requires revisiting many of the implementation steps previously completed. This section of the workbook contains four worksheets to assist with system monitoring and refinement.

7. Make future plans and enhancements. The workbook authors invite users to share insights and suggestions for workbook additions and enhancements. Further, they plan to hold a forum for implementer/vendor information exchange.

BEYOND THE WORKBOOK

The Clinical Decision Support Implementer's Workbook is the first published comprehensive guide to implementing decision support. As clinicians and informaticists work together to improve patient care, they're strongly urged to take advantage of this comprehensive guide. The science of decision support is in its early stages, and with the help of valuable tools such as this workbook, more rapid and accurate development, deployment, and testing of CDSS may occur.

REFERENCES