Chapter 1 - Approaching Clinical Decision Support in Medication Management
CHAPTER 1 SECTIONS
- Overview of CDS Five Rights
- Applying CDS to Medication Management
- Types of CDS Interventions
- How Key Terms Are Used in this Book
- Typical State of Medication Management Today
- A Vision for Optimal, CDS-Enabled Medication Management
- A Peek at the Literature on Medication Use and CDS
- Concluding Comments
- Begin by developing a clear picture of central processes and concepts, such as the medication management cycle, CDS intervention types, and challenges and opportunities related to medication management.
- Consider how the “CDS Five Rights” apply to improving processes related to medication use and outcomes throughout the medication management cycle.
- Take a broad perspective on opportunities to support end users with CDS; the toolkit contains many important CDS tools in addition to interruptive alerts and reminders.
- Go into your efforts bearing in mind what optimal, CDS-facilitated medication management might look like.
How well are tasks related to the medication use cycle typically accomplished in healthcare delivery today? Because you are reading this book, you probably sense that there is vast opportunity for improvement. Each step in the medication management process is fraught with opportunities for suboptimal outcomes, even with the best trained/educated/intended participants. CDS interventions offer promise for shoring up many of these weak spots. In typical practice today, however, CDS is used in relatively few components of the medication management process and is generally not optimally executed. A recent report from The Institute of Medicine (IOM) (see Summary of the 2006 IOM Report on Medication Errors) hints at the extensive opportunity for improvement.
To set the stage for the CDS implementation guidance offered in the remainder of this book, this chapter will walk you through the following related elements:
- The Five Rights of CDS – a framework for supporting clinical decisions to improve outcomes
- How CDS can be applied within the medication management cycle
- The types of CDS interventions that can be used to support medication management
- How key terms are used in this book
This chapter also briefly summarizes the current state of medication management to help you place your environment and challenges within a broader context. We provide a glimpse of a future characterized by more optimal CDS-enhanced medication management to inspire and guide your improvement efforts. We conclude this chapter with a literature sampling on medication errors and CDS in medication management as a springboard for those interested in deeper research on these topics.
CDS interventions can be applied throughout the medication management cycle to optimize medication safety and other pertinent outcomes. A useful framework for achieving success in this effort is the “CDS Five Rights” approach. This should not be confused with the Five Rights of medication use, which speak to ensuring that the right patient gets the right drug at the right dose, via the right route, and the right time. 
The CDS Five Rights model states that we can achieve CDS-supported improvements in desired healthcare outcomes if we communicate:
- The right information: evidence-based, suitable to guide action, pertinent to the circumstance
- To the right person: considering all members of the care team, including clinicians, patients, and their caretakers
- In the right CDS intervention format: such as an alert, order set, or reference information to answer a clinical question
- Through the right channel: for example, a clinical information system (CIS) such as an electronic medical record (EMR), personal health record (PHR), or a more general channel such as the Internet or a mobile device
- At the right time in workflow: for example, at time of decision/action/need
For each step in the medication management process, one can consider how to apply these CDS Five Rights to ensure that the step is negotiated with optimal effectiveness, safety, and resource use. Chapters 3 through 5 will explore in detail how to address these parameters to achieve specific objectives associated with medication management, such as decreasing drug-drug interactions (DDIs), and supporting proper drug selection and dosing.
The following outlines tasks in the medication management cycle and related CDS opportunities; typical parties responsible for each step in the cycle are listed in parentheses. This outline underpins further discussion on applying CDS to improving medication use and outcomes in this and subsequent chapters.
Medication Selection/Reconciliation (prescriber, nurse, pharmacist)
- Select, for a clinical condition, a medication that is safe and effective, appropriate for a specific patient and circumstance, and available within an organizational formulary. The selection should be based on clinical evidence, best practice, patient characteristics, and cost-effectiveness.  The selection process requires access to pertinent patient information (such as clinical history, weight, height and age), as well as pertinent disease management knowledge (for example, evidence-based best practice treatment approaches) and drug information (addressing dosing, side effects, costs, interactions, contraindications and the like).
- Support medication reconciliation with an accurate list of the patient’s medications, along with medication identification and therapeutic use information.
- Create a medication order/prescription (ideally linked to the indication) for the patient to take a drug or for the drug to be administered.
- Provide dosing recommendations, ideally specific to patient and clinical condition.
- Conduct automatic checks (or at least communicate reference information when appropriate) for contraindications, duplications, DDIs, drug-lab interactions, clinically significant allergies, right dose/route/frequency.
Verification/Dispensing (pharmacist, pharmacy staff)
- Double check for interactions, appropriateness/contraindications, right dose/route/frequency/timing.
- Match prescription/order to correct dose and dose form.
- Check for proper concentration and volume to minimize pump programming errors, incompatibilities, and dispensing waste — important especially in pediatrics (for example, using 500-mL instead of 1000-mL bags when appropriate, and the like).
Administration (patient or caretaker, nurse, and/or other clinician)
- Make positive medication and patient identification.
- Assess patient and document pertinent parameters (such as blood pressure, heart rate, blood glucose, pain level) prior to administration.
- Check for incompatibilities/interactions, such as between parenteral medications, between medications and foods, etc.
- Recheck right dose/route/frequency, administration technique and timing, monitoring guidelines.
- Provide reminder/guidance when medications are not administered at the appropriate time or are delayed or missed.
Education (patient or caretaker, pharmacist, nurse, prescriber, other clinicians)
- Engage patient in effective medication use; help patient understand how and why to properly take medications (including indication, administration, and desired effects), how to appropriately store and handle medications, and potential adverse effects to be vigilant for and how to address them; ensure patient understanding of information (whether communicated via discussion, handouts, patient portals/kiosks, PHRs, and/or audio/video material and other media). Engage caregiver/parent when needed to support patient.
Monitoring (patient or caregiver, nurse, other clinicians, pharmacist, prescriber, health system)
- Verify proper patient adherence to the medication regimen.
- Anticipate and monitor individual desired and adverse effects, for example, through history/symptoms, examination, and check of appropriate labs with notification of critical labs/adverse effects.
- Track adverse events across populations, for example, (ideally) via robust structured data-reporting system that incorporates medication error taxonomy standards, and updating the patient record with any new allergy/side effect/interaction.
- Provide feedback and input about patient medication use — across care settings — into the medication reconciliation/selection step, and thus help close the medication management loop.
Figure 1-1 illustrates how all the individuals and components of the medication management cycle are connected. It is important to remember these connections going forward to avoid the tendency to isolate key tasks in silos; this approach plagues the traditional medication management process and supports errors and inefficiency. You do not want to create or continue an environment in which people take a narrow view of this interdependent process and respond with, "That’s not my job", or, "The pharmacist or nurse will take care of that", thereby creating a breeding ground for poor outcomes. As we will show in this chapter and in Chapters 3 through 5, well-executed CDS can support knowledge and data flows that help optimize care quality and efficiency throughout this cycle.
Figure 1-1: Medication Management Cycle
The medication management cycle just described defines the "when," or workflow step, and the "who" for applying CDS; the palate of available CDS intervention types addresses the "how," or format for delivering information to support decisions.
Table 1-1 outlines several major CDS intervention types and subtypes along with their benefits. These categories are not meant to be rigid or entirely mutually exclusive. Rather, the intent is to paint a rich picture of opportunities for guiding decisions through targeted information delivery. In practice, CDS interventions often combine several elements from these basic types. For example, an order set might highlight—through a non-interruptive alert—an essential intervention that should routinely be ordered and provide an infobutton link to more detailed reference information that supports the clinical recommendation.
Table 1-1: Clinical Decision Support Intervention Types
||Patient self-assessment forms||
|Clinician patient assessment forms||
|Clinician encounter documentation forms||
|Departmental/multidisciplinary clinical documentation forms||
|Data flowsheets (usually a mixture of data entry form and relevant data presentation, see next entry)||
||Relevant data for ordering, administration, or documentation||
|Retrospective/aggregate reporting or filtering||
|Environmental parameter reporting||
|Practice status display||
||Single-order completers including consequent orders||
|Tools for complex ordering||
||Stepwise processing of multi-step protocol or guideline||
|Support for managing clinical problems over long periods and many encounters ||
||Alerts to prevent potential omission/commission errors or hazards||
|Alerts to foster best care||
Table 1-2 illustrates how the material discussed in the previous sections on the CDS Five Rights, the medication management loop, and CDS intervention types can provide a framework for developing a CDS intervention strategy to improve medication outcomes. Information in the table is not intended to be comprehensive and include every CDS option and objective but rather serves as a strawman to consider as you develop your approach to aligning CDS interventions with targeted goals (as discussed in Chapters 2 through 5).
Table 1-2: Using CDS to Improve Medication Use and Outcomes
|MEDICATION MANAGEMENT CYCLE STEPS|
|WHY (GOAL)||Optimize: EBM/Quality/Regulatory, Cost, Safe Transition||Safer Use: DDI, dosing, allergies, etc.||Safety/Appropriatenes Check||Safe Administration||Optimize Patient Self-Care||Track Intention/Unintentional Effects|
|WHO (PERSON)||Prescriber, Nurse, Pharmacist, (Patient)||Prescriber||Pharmacist||Nurse, Other Clinician, (Patient)||Clinicians, Patient||Clinicians, Patient, Health System|
|WHAT (INFORMATION)||Reference on drugs (selection, dosing, ID, pricing, etc.), diseases (treatment), condition-specific recommendations||Condition-specific Order Sets and Order Sentences; Order Checks and References||Reference/alerts on dosing/interactions||Reference information (e.g., administration, IV compatibility)||Patient-oriented reference (drug, disease, lab)||Reference drugs (effects monitoring), Diseases (course), Labs (interpretation); Effect monitoring|
|HOW (FORMAT)||Order Sets, Reference (lookup/InfoButton)||Reference (lookup/InfoButton), Order Sets/Sentences||Reference (lookup/InfoButton), Unsolicited alerting||Reference (lookup/InfoButton)||Reference (lookup/InfoButton)||Reference (lookup/InfoButton); Rule checking/unsolicited alerting/relevant data|
|WHERE (CHANNEL)||Internet, EMR/CPOE, Mobile, Med Rec Applications, Formulary Tools||CPOE, EMR, Internet, Mobile, Paper/electronic Order Forms||Pharmacy system, Internet, EMR||eMAR, EMR, Bar coding, Dispensing cabinets, IV pumps, Internet, Mobile, PHR||Internet, EMR, PHR||EMR/Surveillance systems, PHR, Internet, Mobile|
Many specialized clinical and information technology (IT) terms come into play in addressing the subject matter of this book, some of which may be jargon to those unfamiliar with a particular topic being discussed. Readers can consult online or print-based dictionaries  for clarification of such terms. There has been a national effort to standardize definitions of several key health IT terms . We have generally followed these definitions, though they have not yet been widely adopted.
In CDS and medication management, words shape thinking, which translates to action that changes the environment in which we operate. We therefore define several key terms that are fundamental to the subject at hand. Although these terms can be used differently in other contexts, the definitions presented here serve as the foundation for the guidance provided in this book.
Clinical decision support. We use this term, broadly defined, to encompass a wide variety of approaches—increasingly but not exclusively computer-based—for delivering clinical knowledge and intelligently filtered patient information to clinicians and/or patients for the purpose of improving healthcare processes and outcomes. CDS includes knowledge-delivery interventions, such as targeted documentation forms and templates, relevant data presentation, order and prescription creation facilitators, protocol and pathway support, reference information and guidance, and alerts and reminders (see Table 1-1).
Clinical information systems. This term broadly refers to computer-based systems that manage patient-related data, for example, EMR, PHR, CPOE (computerized practitioner order entry), and the like. As previously noted, there are early attempts to standardize the definitions of selected key systems, although widely accepted definitions do not generally exist.
Medication errors. The National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP)  has defined a medication error as "...any preventable event that may cause or lead to inappropriate medication use or patient harm, while the medication is in the control of the healthcare professional, patient, or consumer. Such events may be related to professional practice, healthcare products, procedures, and systems, including prescribing; order communication; product labeling, packaging and nomenclature; compounding; dispensing; distribution; administration; education; monitoring; and use."   NCC MERP has developed a taxonomy for categorizing medication errors and the associated harm to patients, summarized next.  This taxonomy informs medication error reporting programs, such as that provided by U.S. Pharmacopeia (USP) and the Institute for Safe Medication Practices (ISMP): 
- No Error
- Category A - Circumstances or events that have the capacity to cause error.
- Error, no harm
- Category B - Error did not reach patient (an “error of omission” does reach the patient).
- Category C - Error reached patient but did not cause harm.
- Medication reaches patient and is administered.
- Medication reaches patient and is not administered.
- Category D - Error reached patient and requires intervention to preclude harm (i.e., monitoring, lab tests).
- Error, harm
- Category E - Error may have contributed or resulted in harm that required intervention.
- Category F - Error may have contributed or resulted in harm that required initial or prolonged hospitalization.
- Category G - Error contributed or resulted in permanent patient harm.
- Category H - Error contributed or resulted in intervention necessary to sustain life.
- Category I - Error contributed or resulted in death.
Medication management. This term includes the full medication-use cycle in inpatient, ambulatory care and/or home settings, with emphasis on opportunities for CDS (see discussion and Figure 1-1 on page xx). Note that, in addition to helping optimize these process steps individually, a medication-management CDS program should continually measure and improve how CDS-mediated information delivery optimizes overall medication-related outcomes, such as pertinent morbidity, mortality, cost-effectiveness, and efficiency (see Chapter 7).
Medication use (appropriate). Ideally medications are used in a manner that minimizes, to the greatest extent possible, preventable harm caused by both errors of omission (such as failing to obtain appropriate follow-up data about a patient’s response to a medication) and commission (such as providing a medication to a patient for which they have a known life-threatening allergy). It likewise includes using medications effectively and efficiently to optimize health (such as ensuring that patients receive medications that are strongly indicated to minimize disease complications). This appropriate use is accomplished, in part, by applying the best evidence, clinical information, and care practices to decisions and actions throughout the medication use cycle — all of which can be supported through well-executed CDS.
Outcomes related to medication use. For the purposes of this book, medication use outcomes include the extent to which the therapeutic (or diagnostic) intent for the medication is achieved; the safety with which medications are used — for example, avoiding preventable adverse events - and the financial implications of medication use — for example, cost-effectiveness compared with alternatives.
As noted previously in this chapter, each step in the medication management process is fraught with opportunities for suboptimal outcomes, even with the best trained/educated/intended participants. The current state of affairs is roughly described as follows:
- The most common form of CDS applied to medication management is typically DDI and drug-allergy checking modules. Alert fatigue and general dissatisfaction with these interventions is common.
- Relatively limited diffusion of advanced CISs, such as CPOEs and EMRs, further limits the opportunities for fully optimizing the CDS Five Rights. Even when organizations have these systems in place, interoperability barriers across systems—for both patient data and CDS interventions/knowledge—frustrates successful implementation.
- Standard terminologies for clinical information, such as LOINC®,  SNOMED-CT®,  and RxNorm  are still underutilized across CIS, and further limit data and knowledge interoperability across pertinent systems.
- Medication-error reporting channels exist but capture only a small fraction of errors.
- USP-ISMP  , MEDMARX®, and Federal (MedWatch) are prominent examples among others. There has been progress in standardization and terminology (defining contributing factors, nodes, NCC MERP error categories) and electronic submission into large databases at national, regional, and hospital levels. However, federal mandatory error reporting only contributes a relatively few reportable sentinel events. Hence, there is still dependency on voluntary reporting mechanisms that are highly variable, with reporting volume depending on the care quality and culture across different delivery organizations.
- Sharing information about successful strategies for applying CDS to improving medication-related outcomes is relatively limited.
- Many institutions customize their CDS assets, such as DDI databases, to address current problems with excessive alerting; however these customized—and potentially more practical—solutions are not readily available to other institutions due to lack of interoperability standards and other barriers.
- The lack of widespread, plug and play interoperability of CDS interventions with CIS typically makes CDS deployment more difficult than it should be and slows the pace of innovation. These factors exacerbate the potential workflow difficulties that arise when CIS and CDS and systems are not developed and deployed with end-user needs and constraints in mind.
- Historically, misaligned financial incentives, an unclear business case, low capital availability, and/or fragmentation between quality/medication efforts and CDS activities have prevented a coordinated and systematic approach to CDS and medication management.
- In some cases, clinicians resist use of medication decision support tools for fear they will reduce autonomy or increase liability, or because of the perception (often justified) that the interventions (typically unsolicited alerts) are more annoying than useful.
- Conflicts between guidelines from multiple sources regarding appropriate medication use are not infrequent, which creates challenges for implementers attempting to facilitate guideline-based care with CDS.
The future state outlined next is adapted from the Roadmap for National Action on CDS. 
- The most current and applicable evidence and best practices for proper medication use underpin guidance throughout the medication management cycle. This guidance is easily accessed and accurate and is used appropriately as needed.
- Clear, evidence-based information about treatments that work best for specific patients and circumstances is readily available. 
- Widely used, standardized, and practical formats are available for expressing specific health-related knowledge and medication-specific interventions in both human and machine-readable form. These formats are used to create a plug-and-play environment wherein CDS guidance is readily deployed in CISs and sharable across systems and organizations. 
- Knowledge is readily customizable for location and facility-specific needs.
- Data needed for medication-related CDS (for example, drug terminologies, and patient drug sensitivities, allergies, clinical problems, and the like) are managed using richly expressive and widely adopted standards. 
- Repositories make CDS knowledge more readily accessible for incorporation into CDS interventions and information systems (IS).
- Certified CISs and related tools that support medication management are widely implemented throughout the medication management cycle with high adoption rates by clinicians and patients.  This use provides a rich substrate for deploying CDS interventions that drive desired outcomes, such as eliminating preventable medication-related harm.
- Principles of user-centered design are fully applied by developers in producing CIS and CDS systems, fostering more rapid and successful workflow integration and high end-user acceptance.
- Electronic prescribing functionality is widely adopted. A 2005 white paper  has outlined a current state and future state related to electronic prescribing and environmental changes — standards, certification requirements, etc. — that need to be implemented to realize the future state for e-prescribing systems. These include:
- Accurate, complete, patient medication list available at all times in all settings.
- Accurate, complete, allergy list with associated reactions readily available.
- Implementation of standard medical vocabularies across all systems and data assets.
- Drug of choice suggestion or lookup function tied to diagnostic impression Benefits and formulary confirmation.
- Use of TALLman lettering and predefined order strings to prevent transcription and interpretation errors.
- Drug of choice refinement based on other diagnoses, problems, age, weight, physiologic status (renal, cardiac, pulmonary, cognitive function), and medications.
- Drug-interaction checking (for example, drug-drug, drug-lab, drug-food, etc.).
- Monitoring activities (such as recommended surveillance of laboratory parameters) suggested by system and initiated at time of prescription.
- Electronic prescription transmission to pharmacy.
- Patient education communicated at encounter with clinician review, at prescription dispensing with pharmacist review, and via online link to further information and educational resources.
- CDS functionality — especially within ISs that underpin workflow — provides a favorable return on investment (ROI) for system purchasers and is welcomed and widely used by all pertinent recipients.
- The Five CDS Rights are optimized to ensure this information delivery is workflow/user-friendly and supports improved outcomes.
- Such effective deployments produce data that demonstrate the ability of CDS to reduce medication errors, increase user efficiency and satisfaction, and improve other key outcomes of great interest to stakeholders. This further drives the business case for, as well as uptake and adoption of, effective approaches.
- Local implementations are characterized by a variety of key elements for success, including:
- Organizational structures to provide governance and gain clinician buy-in.
- Robust knowledge assets, such as evidence-based content, rules, and order sets, etc. that are explicitly focused on addressing priority targets.
- Rich knowledge management (KM) processes and tools to manage these assets and ensure their appropriateness (currency, evidence base, consistency, etc.) and value.
- Substantial attention to end-user needs and constraints to optimize acceptance and usability.
- Rigorous attention to measuring deployed CDS performance, including rich feedback loops with affected stakeholders, and to seeking opportunities for program enhancement.
- A continuous-improvement cycle of knowledge related to medication management and corresponding CDS deployment — within and across many institutions nationally/globally — drives rapid and widespread CDS-supported advances in desired outcomes.
- Anonymous reports of medical errors and near-misses are used to identify points at which medication-focused interventions should be applied.
- Medication errors are reported and stored in databanks to monitor for error spikes that can trigger appropriate responses, including new or modified CDS interventions that could be applied locally where spikes occur, or more widely as indicated.
- Experiences from deploying CDS (including incorporating these interventions into CISs) to improve medication-related outcomes are systematically tracked and synthesized, further adding to the dynamic body of best implementation practices. These best practices are widely disseminated and continually enhanced, for example through mechanisms such as this book and extensions of this work as outlined in the preface.
There is fairly rich evidence that details challenges associated with medication use and application of CDS in specific improvement opportunities. Journals such as those in general internal medicine , medical informatics , and other disciplines are increasingly devoting theme issues and other coverage to such topics. A comprehensive literature review is beyond this guide’s scope, but a sampling helps set the stage for the guidance that follows.
Medication errors, which indicate breakdown in the medication use cycle and may cause bad outcomes, are increasingly the focus of public attention and local and national improvement efforts. A recent report from the IOM highlights the challenges and opportunities.
In 2006, IOM released a landmark report on medication safety entitled Preventing Medication Errors, as part of their Quality Chasm series.  The report looked at both prescribing and administration errors. Prescribing error rates ranged from 12.3 to 1,400 errors per 1,000 admissions. The wide distribution reflects varying methods for finding errors—estimates at the lower end were generally based on error reports filed by clinicians. The study that estimated the error rate at 1,400 errors per 1,000 admissions was based on chart review and estimated approximately 0.3 errors per patient per day.  Although most of these errors were unlikely to result in patient harm, 7.5% of them either resulted in a preventable ADE or could have resulted in one.
Administration errors also occur with significant frequency. The IOM report reviewed five studies that estimated administration error rates as between 2.4 and 11.1 errors per 100 medication administrations. The IOM committee estimated that between all sources of errors, a hospital patient is subject, on average, to one medication error per day.
The IOM report also synthesized research on medication errors in the outpatient setting, citing studies which found that 21% of all prescriptions contain at least one error , 3% of doses in an outpatient chemotherapy unit contained an error , and between 1.7% and 24% of community pharmacy prescriptions were dispensed incorrectly. 
Beyond error incidence, the IOM report also reviewed research on the incidence of ADEs. They found three studies that met their inclusion criteria. These studies estimated the preventable ADE rate at 1.2 preventable ADEs per 100 admissions , 1.8 preventable ADEs per 100 non-obstetric admissions , and 5.57 ADEs per 1,000 patient days.  Preventable ADEs in hospitalized patients increased length of stay (LOS) by 4.6 days and total costs by $5857 (1993 cost data).  In ambulatory Medicare patients at least 65 years old, the cost (in 2000 dollars) per preventable ADE is estimated at $1983. 
Based on this literature synthesis, the IOM report concludes that, in the United States, 1.5 million people suffer preventable injury every year as a result of medication errors. Roughly 530,000 preventable drug-related injuries occur each year among Medicare recipients in outpatient clinics alone. The extra annual medical costs for treating patient injuries that occur in hospitals alone is $3.5 billion.
"The Future Is Already Here, It’s Just Unevenly Distributed" 
As outlined previously in this chapter, CDS holds great promise for addressing the pressing challenges related to medication use that have been identified by IOM and others. A growing collection of studies demonstrates that this promise is currently being realized in leading organizations to varying degrees. In developing this guidebook, an effort has been made to base implementation recommendations on evidence from such literature on successful practices; therefore references are included throughout. Given the practical focus and largely volunteer nature of this effort, the literature analysis has been more opportunistic than exhaustive.
Perhaps a more important barrier to providing strongly evidence-based implementation guidance than extensive literature review is that, although there have been a number of studies on medication-related CDS, this literature only covers, and in a somewhat limited manner, the full medication management cycle. Even in relatively well-studied areas, such as CPOE to reduce ADEs, major gaps still exist.  Further, much of the published literature is derived from leading academic organizations with locally developed systems; 70% of the studies in this review examining ADE rates used homegrown CPOE with CDS. 
There is much to be learned from these pioneers, but the CDS functionality and results from custom-crafted systems in academic settings may not be directly transferred elsewhere. In this guide, we have tried to extrapolate lessons from groundbreaking efforts to healthcare delivery settings in which vendor supplied systems are the norm and organizational dynamics may be different. Where evidence is not specifically cited, the guidance provided is drawn from the experience of the editors, contributors, and reviewers, based on success strategies (or at least thoughtful approaches) gleaned in the course of their efforts.
In subsequent chapters, we occasionally reference CDS studies to reinforce specific recommendations. For example, Chapter 5 contains references to studies exploring the issues in applying CDS to various specific targets, such as reducing drug-allergy, DDI, and drug dosing problems. In addition, the following is a sampling from the literature on applying CDS to medication management. Studies of this sort have been loosely summarized in various sources such as the "Roadmap for National Action on CDS,"  the IOM report, "Preventing Medication Errors,"  and systematic reviews of CPOE with CDS as previously mentioned. 
Sampling of Literature on CDS in Medication Management
- Utilizing CDS has been shown to improve adherence to guidelines. Traditionally, clinical guidelines are left "on the shelf" and, as a result, errors and suboptimal care persist. Using CDS to communicate the guidance within clinician workflow, and hence increase adherence, has resulted in an absolute increase in influenza and pneumococcal vaccination of 12% and 20%, respectively.  Another study demonstrated a 3.3% absolute decrease in the primary endpoint of DVT or PE within 90 days after hospitalization. 
- CPOE with CDS has improved physician prescribing practices, formulary adherence, and cost savings , and produced an 86% absolute reduction in non-intercepted serious medication errors. 
- CDS for empiric preoperative antibiotic selection decreased postoperative wound infections. 
- CDS-supported medication dosing produced a 21% increase in appropriate medication prescribing in renal insufficiency and a 4.5% reduction in length of stay (LOS). 
- EMR use does not appear to be associated with improved performance on ambulatory care quality indicators without focused CDS features  and basic computer prescribing (which improves legibility and completeness) was not associated with a reduced rate of errors in an outpatient setting unless combined with advanced systems with CDS (drug-drug, drug allergy, and drug-interaction checking).  Similarly, even hospitals with highly computerized medication management processes experience high rates of ADEs without CDS focused specifically on drug selection, dosing and monitoring.  To realize benefits of CDS within CPOE, authors have suggested a two-stage approach: (1) basic CDS (drug-allergy testing, basic dosing guidance, duplication checking, and DDI checking); followed by (2) advanced CDS (dose-lab, renal dosing, drug-disease checking). 
- A systematic review found that CPOE reduces medication errors and ADEs.  Extensive tables featured in this review outline the CDS functionality examined in various studies, whether the systems were homegrown or commercial, and the risk ratio for ADEs with the system.
We will take a closer look at literature-suggested CDS targets in Chapter 2.
Deploying CDS in a manner that effectively addresses priority objectives doesn’t generally occur in a piecemeal manner (for example, through entirely separate initiatives for each target, such as medication use, regulatory compliance, and the like). Typically, a variety of targets are addressed in a somewhat integrated fashion within a broader organizational CDS program. Nonetheless, since improving medication management with CDS is such a widespread (perhaps universal) and important healthcare goal, we are focusing this book on this specific topic. Keep in mind, however, that insights conveyed here may be applicable to your broader CDS efforts, and likewise, pearls you may have gleaned from CDS initiatives focused on other targets may well help accelerate your medication management CDS efforts.
In any case, a systematic performance improvement (PI) approach based on careful attention to the CDS Five Rights, the full spectrum of CDS intervention types, and the interdependent nature of the medication management cycle, should help provide solid footing for your efforts. Inspiration from a vision of what optimal CDS-enabled medication management could be should further help you enhance your program so that it delivers optimal benefit today and in the future.
We will take a closer look at literature-suggested CDS targets in Chapter 2.
- Institute for Healthcare Improvement. The Five Rights of Medication Administration; accessed 9/30/2008.
- Institute for Safe Medical Practices (ISMP) stresses that if all of the other aspects mentioned are achieved, then the drug will be cost effective.
- Tang PC, Young CY. ActiveGuidelines: Integrating Web-based guidelines with computer-based patient records. AMIA Proceedings. 2000; 843-847.
- Maviglia SM, Zielstorff RD, Paterno M, et al. Automating complex guidelines for chronic disease: lessons learned. J Am Med Inform Assoc. 2003; 10(2):154-165.
- Background on infobuttons and the version of the HL7 standard approved in May 2008 -- Health Level 7. HL7 V3 Infobutton, R1; accessed 9/30/2008.
- For example, Wikipedia, accessed 10/1/08, and the Medline Plus Medical Dictionary.
- NAHIT releases HIT Definitions; accessed 12/12/2008.
- National Coordinating Council for Medication Error Reporting and Prevention; accessed 9/30/2008.
- The Institute for Safe Medication Practices. Frequently Asked Questions (FAQ); accessed 9/30/2008.
- National Coordinating Council for Medication Error Reporting and Prevention. About Medication Errors; accessed 9/30/2008.
- National Coordinating Council for Medication Error Reporting and Prevention. About Medication Errors. Taxonomy of Medication Errors Now Available (see items 30-34); accessed 9/30/2008. See also links to diagrams under NCC MERP Index for Categorizing Medication Errors; accessed 12/12/2008.
- USP-ISMP Medication Errors Reporting Program; accessed 12/12/2008.
- Standard terminology for laboratory information. See Logical Observation Identifiers Names and Codes (LOINC); accessed 9/30/2008.
- A very rich, standardized, multi-lingual clinical healthcare terminology. See Systematized Nomenclature of Medicine-Clinical Terms (SNOMED CT); accessed 9/30/2008.
- A standard terminology for clinical drugs. See U.S. National Library of Medicine. Unified Medical Language System: RxNorm; accessed 9/30/2008.
- Osheroff JA, Teich JM, Middleton B, et al. A roadmap for national action on clinical decision support. J Am Med Inform Assoc. 2007; 14(2):141-145. See also American Medical Informatics Association. CDS Roadmap; accessed 9/30/2008.
- The Institute of Medicine has recently developed a detailed roadmap for this. See Institute of Medicine. Knowing What Works in Health Care: A Roadmap for the Nation; accessed 9/30/2008.
- The standards development organization HL7 - accessed 10/2/2008 - has a Clinical Decision Support Technical Committee that produces standards for CDS capabilities such as rules and alerts, infobuttons, and order sets; further work is needed to produce a robust suite of widely used standards for the full suite of CDS interventions.
- The Health Information Technology Standards Panel (HITSP) - accessed 10/1/2008 - has begun work on this. See, for example, The Medication Management Interoperability Specification; accessed 10/1/2008. This work is supporting the medication management use case. See Health Information Technology. Medication Management Use Case; accessed 10/1/2008. Developed by Office of the National Coordinator for Health IT.
- The Certification Commission for Health Information Technology (CCHIT) has begun this work. See, for example, the clinical decision support requirements related to medication management in the latest inpatient and ambulatory certification requirements and test scripts, available at CCHIT; accessed 10/1/2008.
- Teich JM, Osheroff JA, Pifer EA, et al. Clinical decision support in electronic prescribing: recommendations and an action plan: report of the joint clinical decision support workgroup. J Am Med Inform Assoc. 2005; 12(4):365-376.
- See, for example, Clancy CM, White PJ. Introduction to the JGIM special issue on health information technology. J Gen Intern Med. 2008; 23(4):353-4; accessed 10/1/2008.
- See, for example, Focus on Safe e-Prescribing. Journal of the American Medical Informatics Association. 2008; 15(4); accessed 10/1/2008.
- Institute of Medicine. Preventing Medication Errors: Quality Chasm Series. Washington, DC: National Academies Press; 2006; accessed 10/1/2008.
- Bates DW, Boyle DL, Vander Vliet MB, et al. Relationship between medication errors and adverse drug events. J Gen Intern Med. 1995; (10)4:199-205.
- Shaughnessy AF, Nickel RO. Prescription-writing patterns and errors in a family medicine residency program. J Fam Pract. 1989; 29(3):290-295.
- Gandhi TK, Bartel SB, Shulman LN, et al. Medication safety in the ambulatory chemotherapy setting. Cancer. 2005; 104(11):2477-2483.
- Allan EL, Barker KN, Malloy MJ, et al. Dispensing errors and counseling in community practice. Am Pharm. 1995; NS35(12):25-33.
- Flynn EA, Barker KN, Carnahan BJ. National observational study of prescription dispensing accuracy and safety in 50 pharmacies. J Am Pharm Assoc (Wash). 2003; 43(2):191-200.
- Classen DC, Pestotnik SL, Evans RS, et al. Adverse drug events in hospitalized patients. Excess length of stay, extra costs, and attributable mortality. JAMA. 1997; 277(4):301-306.
- Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA. 1995; 274(1):29-34.
- Jha AK, Kuperman GJ, Rittenberg E, et al. Identifying hospital admissions due to adverse drug events using a computer-based monitor. Pharmacoepidemiol Drug Saf. 2001; 10(2):113-119.
- Bates DW, Spell N, Cullen DJ, et al. The costs of adverse drug events in hospitalized patients. Adverse Drug Events Prevention Study Group. JAMA. 1997; 277(4):307-311; accessed 10/2/2008.
- Field TS, Gilman BH, Subramanian S, et al. The costs associated with adverse drug events among older adults in the ambulatory setting. Med Care. 2005; (12):1171-1176; accessed 10/2/2008.
- Quotation attributed to William Gibson, Nov 30, 1999; accessed 10/1/2008.
- Eslami S, Abu-Hanna A, de Keizer NF. Evaluation of outpatient computerized physician medication order entry systems: a systematic review. J Am Med Inform Assoc. 2007; 14(4):400-6; accessed 10/2/2008.
- Wolfstadt JI, Gurwitz JH, Field TS, et al. The effect of computerized physician order entry with clinical decision support on the rates of adverse drug events: a systematic review. J Gen Intern Med. 2008; 23(4).
- Dexter PR, Perkins SM, Maharry KS, et al. Inpatient computer-based standing orders vs. physician reminders to increase influenza and pneumococcal vaccination rates: a randomized trial. JAMA. 2004; 292(19):2366-2371.
- Kucher N, Koo S, Quiroz R, et al. Electronic alerts to prevent venous thromboembolism among hospitalized patients. N Engl J Med. 2005; 352(10):969-977.
- Teich JM, Merchia PR, Schmiz JL, et al. Effects of computerized physician order entry on prescribing practices. Arch Intern Med. 2000; 160(18):2741-2747.
- Bates DW, Teich JM, Lee J, et al. The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999; 6(4):313-321.
- Evans RS, Classen DC, Pestotnik SL, et al. Improving empiric antibiotic selection using computer decision support. Arch Intern Med. 1994; 154(8):878-884.
- Chertow GM, Lee J, Kuperman GJ, et al. Guided medication dosing for inpatients with renal insufficiency. JAMA. 2001; 286(22):2839-2844.
- Linder JA, Ma J, Bates DW, et al. Electronic health record use and the quality of ambulatory care in the United States. Arch Intern Med. 2007; 167(13):1400-1405.
- Gandhi TK, Weingart SN, Seger AC, et al. Outpatient prescribing errors and the impact of computerized prescribing. J Gen Intern Med. 2005; 20(9):837-841.
- Nebeker JR, Hoffman JM, Weir CR, et al. High rates of adverse drug events in a highly computerized hospital. Arch Intern Med. 2005; 165(10):1111-1116; accessed 10/2/2008.
- Kuperman GJ, Bobb A, Payne TH, et al. Medication-related clinical decision support in computerized provider order entry systems: a review. J Am Med Inform Assoc. 2007; 14(1):29-40.
- Ammenwerth E, Schnell-Inderst P, Machan C, et al. The effect of electronic prescribing on medication errors and adverse drug events: a systematic review. J Am Med Inform Assoc. 2008; 15(5):585-600.