Nurses and other health

Infusing Clinical Decision Support I n t e rventions into Electronic Health R e c o rds Jane M. Bro k e l Nurses and other health c a re providers use the p a t i e n t s documented h i s t o ry, p re f e re n c e s, a s s e s sment, and diagnostic re s u l t s to make decisions while pro v i d- ing care within a clinic, while hospitalized, or within the home. C u rrent clinical information systems capture a large volume of patient data and are capable of s u p p o rting clinical decisions or suggesting interventions. This a rticle provides examples of diff e rent clinical decision support interventions nurses may encounter within new health care i n f o rmation technologies. Clinical decision support has recently been defined as pro v i d- ing clinicians or patients with clinical knowledge and patientrelated information, intelligently f i l t e red or presented at appro p r i- ate times, to enhance patient care Jane M. B ro k e l, P h D, R N, is an Assistant P r o fe s s o r, the University of Iowa, College of Nursing, Iowa City, IA; a member of the State of Iowa Electronic Health Info rm a t i o n E xe c u t i ve Committee and Advisory Council; an Info rmatics Researcher, Trinity Health, N ovi, MI; and is the Guest Editor of this issue of Urologic Nursing. Statement of Discl o s u r e : The author r e p o rted no actual or potential conflict of interest in relation to this continuing nu r s i n g education activity. N o t e : O b j e c t i ves and CNE Evaluation Fo rm appear on page 353. E l e c t ronic medical records and health records provide a variety of clinical decision support interventions to guide or support the cl i n i - cal user. These interventions are design features to guide users in n ext steps, o f fer useful evidence-based know l e d ge, or provide patient i n formation relevant to a decision. O b j e c t i v e s 2009 Society of Urologic Nurses and Associates U rologic Nurs i n g, p p. 3 4 5-3 5 3. Key Wo rd s : Clinical decision support (CDS), clinical decision support interventions, clinical decision support applications, electronic health record (EHR). 1. Define clinical decision support. 2. List several types of clinical decision support interventions. 3. Discuss applications for clinical decision support interventions in nursing. ( O s h e ro ff, Pifer, Teich, Sittig, & Jenders, 2004, p. 3). This support p rovides the nurse with the appropriate clinical evidence re s o u rc e s associated with accurate patient assessments and test results. These data are quickly available, p roviding the ability to make the right decision at the right time within the right place and to the right person. If information and re s o u rces are not available, the nurse may be unable to make a decision and implement care i n t e rventions in a timely manner. All clinical disciplines that have access to the electronic health re c o rd will enter and/or use information. These data are integrated within an electronic health re c o rd. Thus, health care decisions are m u l t i d i s c i p l i n a ry and include the p a t i e n t s perspective. Types of Clinical Decision Support Several types of clinical decision support (CDS) interv e n t i o n s a re available within electro n i c health re c o rds (EHRs) and elect ronic medical re c o rds (EMRs) that provide support for those delivering care (Kawamoto, Houlihan, Balas, & Lobach, 2005; O s h e ro ff et al., 2007). These CDS i n t e rventions may or may not be used with the EMR of a health c a re facility (such as a hospital or clinic), but are often available a c ross health care delivery are a s UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5 345

within a larger health system s EHR. When the EMR or EHR is p u rchased, CDS interv e n t i o n s a re not routinely provided unless requested and subsequently designed with some level of customized specifications for the health system or local facility. The EHR and EMR re p o s i t o ry of data is used within many CDS i n t e rventions. These data are s e c u red by passwords, and access is traceable, thus pro v i d- ing a level of security and confidentiality to the health care information with any clinical decision support interv e n t i o n. E l e c t ronic Form The first type of CDS intervention is an electronic form and is common in many health care settings. The electronic form can be designed to guide patient data e n t ry (see Figure 1). The instru c t- ing guidance provides support when completing interviews and o b s e rvations at admission, during daily care, when performing focused assessments, and implementing interventions. The user is able to access the full scope of assessments and patient care activities available within the EMR/EHR electronic form. The user is given tips for when a section of the form should be completed and how to complete it. Typically, the EMR/EHR uses point and click responses, and thus, requires minimal typing skills. Instructions and tips within e l e c t ronic forms are most helpful for the novice, new orienting staff, and nursing students. Physicians and other disciplines may use the same sections of electronic form s, and thus, data entry is share d. This CDS intervention support s documentation of care by using reusable templates that are available to all appropriate health care p roviders wherever they work. The reusable templates are sections within electronic forms that allow data to be integrated fro m all settings. The section on uro l o g- ic assessments would be an example of a reusable template. Figure 1. Example of Electronic Forms with Instructional Statements Figure 2. Example of Order Set with Order Details for Clinical Condition P rovider Order Sets Another type of CDS intervention is the provider order sets (see Figure 2). The order sets contain order sentences to first guide the provider selecting the ord e r details during computerized p rovider order entry, and next guide the nurses and therapists who need to interpret and perf o rm the intervention. An example is the set of standardized care i n t e rventions and patient care o rders for a urologic pro c e d u re designed for the physician or nurse practitioner to eff i c i e n t l y place orders. The provider selects the pre-designed urologic ord e r set that has the common ord e r sentences pre-selected and optional selections for additional orders. All pre-selected and optional orders are organized under common categories, such as medications orders, continuous intravenous orders, and therapy o rd e r s. 346 UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5

This CDS intervention facilitates standard specifications for order sentences, including required and optional order details for individual orders. R e q u i red details ensure a more complete order that meets the needs of evidence-based practices. A standardized order set p rovides complete order details for medication orders, diagnostic tests, diet, activity ord e r, and therapies following evidence. After the provider enters orders with an o rder set, nurses, pharm a c i s t s, therapists, or dietitians immediately see the complete order details as part of the patient s plan of care. These can be used to ensure that care is uniformly d e l i v e red and eliminates phone calls to clarify or discuss what was omitted. For example, the o rder sets specific to urologic proc e d u res would provide details on s t a rt time, end time, dose, re a s o n or condition for perf o rming the procedures as intermittent catheterization, or continuous catheterization, which re m o v e s the ambiguity often associated with written and incomplete details with an ord e r. When the order sets are developed for physicians, nurses and all clinical depart m e n t s responsible for executing that o rder correctly need to be involved within the design and development of order sets. T h e re f o re, nurses within the urologic specialty will need to review all developed physician u rologic order sets for the EMR/EHR before approved for use. Nurses should ensure that each order is clear, understandable, and complete, with all o rder details present in the ord e r sentences to guide those in training and new staff. P rescriptive Plan of Care A third type of CDS interv e n- tion is a prescriptive plan of care to identify priority problems and patient pre f e rences. The plan of c a re or care set is pre-designed so the nurse can quickly select and Figure 3. Example of a Plan of Care Set to Support Options for Care modify as necessary (see Figure 3). The pre-designed plan of care is based on best evidence and s t a n d a rds. The nurse identifies the plan appropriate for the patient respective of his or her d e s i red pre f e rences and outcomes. This pre-packaged plan p rovides the ability to quickly identify pre-selected care ord e r s, add optional care orders, or unselect interventions. As a re s u l t, this CDS intervention pro v i d e s guidance by linking nursing diagnoses, patient outcomes, and nursing interventions and assessments to evaluate the expected outcomes. This CDS interv e n t i o n functions similar to the physic i a n s order set and aligns the care within the patient s EMR/EHR re c o rd. This clinical decision supp o rt reflects the established linkage between nursing diagnosis, outcome expected, and nursing i n t e rv e ntion (American Nurses Association [ANA], 2005). Branching pathways are designed with evidence-based linkages to guide choices for nursing interventions that have been documented to have an effect to change the condition/nursing diagnosis. D rug Databases The fourth type of CDS intervention provides all clinical disciplines with drug databases that intersect with the EHR. It monitors patient medication(s) for d ru g - a l l e rgy interactions, dru g - d rug interactions, drug-food interactions, and therapeutic duplications (see Figure 4). This CDS tool p rovides instant access to dru g i n f o rmation for all clinicians and for patient education. Most EMRs and EHRs will use a database to generate alerts. The accuracy of the nurse s early admission assessments and documentation of allergies, home medications, and height and weight are helpful data to support the interaction checking and dose range checking for all disciplines. In many situations, the nurse s documentation is needed to allow the CDS intervention to function for all health c a re providers. Bar Coding The fifth type of CDS inter- UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5 347

Figure 4. Example of Drug Database Alert for Drug-Allergy Interactions Figure 5. I n formation Button to Provide Definition or Reference Tex t vention is primarily used by nurses to administer medications. This involves a bar code medication scanning device that ensure s the right patient, drug, dose, ro u t e and time occur. Medications are often delivered once a day to the nursing unit in unit dose, bulk dose, or unit stock. Using the device re q u i res that the nurse check the bar code on the medication to the patient s arm band. The device then determines if this is the right medication and dose for the patient s profile. If data do not match, a short message is displayed, allowing the nurse to recheck the information and avoid an error in administration. The device often links with the drug database, giving instant access to drug inform a t i o n. Evidence Button The sixth type of CDS intervention provides information, re f- e rence text or a definition for the clinical condition, normal range, i n t e rvention, or pro c e d u re (see F i g u re 5). The information buttons are designed to link nursing or medical definitions, descriptions, or related details to pro v i d e clinicians with re f e rences at the bedside (Bates et al., 2003; O s h e ro ff, 2009). This CDS provides information from nursing re s e a rch manuscripts and pro f e s- sional nursing organizations about anticipated drug therapeutic effects and side effects. E l e c t ronic forms, orders, and alert messages may have an associated i n f o rmation button to click for evidence and articles. If Then Logic Module The last type of CDS interv e n- tion is the most complex in design and re q u i res broad re p re s e n t a t i o n of experts to develop. CDS is provided through an if then logic module. This re q u i res the use of critically appraised evidence and e x p e rts to design the use of data within the work steps of the user. W h e n and if the conditions exist, a synchronous or asynchro n o u s step occurs within the EHR or 348 UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5

Figure 6. Clinical Decision Support Rule with If-Then Logic to S u p p o rt a Decision EMR. When a synchronous action occurs, the user will receive a message detailing what actions a re possibilities (see Figure 6). When an asynchronous action occurs, the user will not immediately receive a message or notice, but will receive orders, actions, communications, or a potential p roblem added to the list. This logic module, when triggere d, assesses conditions before pro v i d- ing a message that details options and actions or recommends an assessment or intervention. Advanced nurse practitioners often help by identifying the evidence to guide the development of the logic for the clinical decision supp o rt ru l e. Clinical pro g r a m m e r s will insert data re q u i rements for the CDS rules per the oversight of e x p e rt clinicians. Nurse expert s will often work with clinical programmers to ensure that re s p o n s- es from the CDS are appro p r i a t e and reflect the particular needs of the institution, and are evident for this patient population. The specifications needed for the CDS if then logic model are achieved through the collaborative eff o rt of expert clinicians and the EMR/EHR staff. Other CDS Interventions Other types of CDS assimilate patient information using a userfriendly display screen, pro v i d e t rend views with charts, and display coloration of data outside the n o rmal parameters, critical high and low laboratory values, or noticeable deviation from set parameters. The assimilated patient information appears as a patient summary view or dashb o a rd providing a historical summ a ry necessary to make decisions. The summary is compiled by running re p o rts on the found data and presenting a dashboard of the summary findings, such as intake and output, mean and range for glucose readings, heart rate, and blood pre s s u re, as well as the 24-hour dose levels for medications, current medications, and interventions. Applications for Use in Nursing CDS interventions may or may not be included in EHR/ EMR systems. These tools can s u p p o rt a nurse throughout the nursing process. Several examples have been employed within EMRs and EHRs to support the practice of nursing and could be c o n s i d e red. Assessments The documentation or data collection of patient assessments a re supported by the development of the electronic flow sheet and/or electronic forms. For example, urinary functional assessments (Fung, 2006) have been designed into an electro n i c f o rm to guide collection of concepts for elimination (such as freq u e n c y, urg e n c y, and sensation) and characteristics (such as odors, colors, and volume). The input of e x p e rt nurses is necessary to test the EHR/EMR electronic form or flow sheet designs. This ensure s nurses will be able to use assessments using a standard set of data elements and observable indicators, thus allowing appro p r i a t e and accurate judgment of a UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5 349

Direct links to measuring a patient s progress and decline over time will require further development and the input of nurses to improve the design of existing EMRs and EHRs. p a t i e n t s pro g ression or decline with a longitudinal re c o rd. Diagnosing Problems and Identifying Pre f e rences After these data are collected, CDS interventions are used to o rganize assessments and other clinical findings to display the t rends and summarize the p a t i e n t s history. A CDS interv e n- tion organizes the patient s information and provides instru c t i o n s that can be used to guide the next steps, which may include identifying priority problems, selecting nursing diagnoses, or developing a problem list. This not only will e n s u re problems are identified but will provide patient-centere d c a re. Several types of CDS interventions are available to help nurses diagnose problems and identify patient-desired outcomes. EMR/EHR applications include CDS logical rules, summ a ry displays, and usability guides with evidence-based re f- e rences, coloration, or tre n d c h a rts. Health care institutions include the documentation of the p a t i e n t s expectations, values, and beliefs, as well as the famil y s assets (Agency for Healthcare R e s e a rch and Quality [AHRQ], 2006). Patient-centered inform a- tion is often part of the discharg e planning process and instru c- tions in preparation for discharg e f rom a hospital, homecare serv i c- es, or ambulatory visit. Planning Care Once priority problems have been identified, two types of CDS i n t e rventions are available: 1) o rder sets and 2) plan of care sets. Either intervention provides pre - selected and optional nursing i n t e rventions, as well as specific details related to care for patient p roblems and urologic diagnoses. Each order set and care plan is pre - c o o rdinated (re a d y for use), with evidence-based re f- e rence knowledge to support the i n t e rventions (Bates et al., 2003; Kaushal et al., 2006). Uro l o g i c physicians may also use pre - c o o rdinated order sets to quickly select orders and edit details for a p a t i e n t s specific conditions and c o - m o r b i d i t y. Nurses will use p re - c o o rdinated care sets (care plans) with pre-selected interventions supported by evidencebased guidelines and optional i n t e rventions to individualize for patients with co-morbid conditions. Pre-selected orders, when used, should be supported by a higher appraised level of evidence, such as meta-analysis and p rospective clinical trials, or systematic reviews from re s e a rc h studies (Bobb, Payne, & Gro s s, 2007). When evidence support s the use of care interventions for specific problems, care interv e n- tions are designed to be re q u i re d or pre-selected, while other interventions may be optional choices for the nurse to individualize the p a t i e n t s care. P e rf o rming Interventions In using the EMR/EHR, each health care provider can quickly review and manage the integrated plan of care with other disciplines. Multidisciplinary patientc e n t e red care is possible with s h a red electronic plans of care and electronic forms. Some EMR/EHRs link documentation to orders to facilitate followt h rough with care. Several CDS i n t e rventions include the use of info buttons and re f e re n c e texts, which link to definitions, descriptions, and evidence-based p ro c e d u re details. Info buttons a re a re s o u rce for the new staff member and others perf o rm i n g i n f requent interventions. Guides on steps for the pro c e d u re or actions to take are re m i n d e r s. Students often benefit from having the re f e rence guides. Evaluating Outcomes F o remost, the nursing process needs to be supported by CDS interventions that re q u i re a c l i n i c i a n s response to the patient status on a regular interval. CDS i n t e rventions that include branching pathways linking the indicated outcomes to measure as re l e- vant to the pro b l e m / d i a g n o s i s and the interventions are not as available in current EMRs/EHRs. These outcome assessments provide indicators for measuring the status for the patient s desire d outcome. Direct links to measuring a patient s pro g ress and decline over time will re q u i re f u rther development and the input of nurses to improve the design of existing EMRs and EHRs. I m p l i c a t i o n s CDS interventions are being developed and put into operation within hospitals, clinics, and community settings, and are not yet perfect systems to support continuous nursing care pro c e s s- es across shifts and settings. Nurses will need to provide feedback when defects, issues, or delays hindering care are experienced. These challenges may include the inability to document patient assessments or i n t e rventions, the inability to compile the necessary inform a- 350 UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5

Ta ble 1. Types of Requests for CDS Interv e n t i o n s P r event Erro rs Bar code medication administration applications with alert messages about wrong dru g, r o u t e, and dose when giving medication to patients A l e rt messages for duplicate thera py and interactions (dru g - a l l e r g y, dru g - d rug, dru g - fo o d i n t e ra c t i o n s ) P romote Safe t y D i s p l aying documented precautions and risks (fa l l s, pressure ulcers, deep vein thrombosis, u ri n a ry tract infe c t i o n s, bleeding) A l e rts for labora t o ry results (creatinine levels) before contrast medium A d verse drug event alerts for labora t o ry values or vital sign measures C o m munication and E x ch a n ge of Data Embed Evidence-Based P r a c t i c e s Electronic plan of care with organized standardized prescribed orders Controlled list of safe abb r eviations and standardized language for medical, nursing, t h e ra py, health and drug concepts Automatic notifications of clinical disciplines based on the patient s clinical findings Documentation templates for urology assessment, intake and output self-calculating d i s p l ay, discharge template R e ference guidance through URL links, info rmation buttons with the intervention activities Renal dosing guidance Order sets and care sets based on best research findings tion in one view for decision making, the inability to find i n f o rmation or evidence if located in an obscure location, and the occurrence of too many alert s that interrupt care. When the EHR or EMR results in the inability to make patient care decisions, nurses should not accept the current workflow, but rather, re p o rt the issue (or gap) thro u g h the institution s tracking pro c e s s to re p o rt problems in using the t e c h n o l o g y. Nurses can impro v e the EMR and EHR systems by re p o rting defects, errors, and issues to the informatics nurses or health IT analysts who supp o rt the systems 24/7. These logs of defects and issues lead to i m p rovement of technical and i n f o rmation gaps in curre n t EMRs and EHRs. CDS interventions will provide the right information (evidence-based interventions and valid assessment tools) to make health care decisions if nursing specialists assist in the CDS development. The CDS interv e n- tion may include a wide variety of solutions to support the nurse with the use of messages, ord e r s, w a rnings, reminders, alert s, s c reen display instructions, and real-time re p o rts that compile findings into a dashboard. The nurse or other health care p roviders will have more access to knowledge-based re s o u rc e s t h rough personal digital assistant devices or through secure I n t e rnet or intranet-accessible guidelines that connect with the EMR/EHR. Foremost, the CDS i n t e rvention must be observ e d (not ignored) and used within a routine daily workflow of care activities on the unit or clinic ( G a rg et al., 2005). The CDS intervention will need to be pre s e n t e d in the right location and to the right person when a decision or action is re q u i red. CDS interv e n- tions are in their infancy, and t h e re f o re, will need continuous evaluation and improvements as nurses use data in EMRs/EHRs. Overall, CDS interventions are c o n s i d e red a benefit if they supp o rt the efficiency and quality of the decision with more valued i n f o rmation (Kaushal et al., 2006). The ultimate goal would be for complication pre v e n t i o n or improved patient outcomes within a reasonable period of time. It is important to know that CDS interventions are available to assist the nurse in making health care decisions when using volumes of data. The purpose for a CDS intervention request may be to prevent errors, pro m o t e s a f e t y, communicate inform a- tion, or provide evidence-based practices. Examples of interv e n- tions used to support a purpose a re presented in Table 1. Health c a re providers specializing in u rology will need to request the clinical decision support intervention and provide the evidence to support the re q u e s t. Requests should state the rationale, such as to support dosing of medications when there is re n a l decline, to remove urinary catheters when no longer indicated to reduce risk of infections, or to communicate to the consulting physician, enterostomal therapist, or social worker when findings re q u i re their services. Conclusion In conclusion, nurses have been known to adapt rapidly to changes with the EHR. A variety of types of interventions may be used to support clinical decisions. Nurses should consider tracking displays, electronic documentation templates and/or flow sheets, care sets, pro t o c o l branching pathways, alert mes- UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5 351

sages, and information links using buttons on the screen as tools to support decisions. U n f o rt u n a t e l y, not all tools are fully applied within current systems due to the re s o u rc e s re q u i red to design, maintain, and monitor their effectiveness. Further re s e a rch is necessary to p rove the effectiveness of tools b e f o re re-application of CDS i n t e rventions is realized on a b road scale in the health care i n d u s t ry. New tools for the health care d e l i v e rysystem exist but are not fully realized. Many org a n i z a- tions do not understand the benefits of the various CDS applications needed to be a part of EHR implementation and long-term development. R e f e re n c e s Agency for Healthcare Research and Q u a l i t y. (AHRQ). (2006). Evolution of state health information exchange: A study of vision, strategy, and pro g re s s. Rockville, MD: The Agency for H e a l t h c a re Research and Quality, U.S. D e p a rtment of Health and Human S e rvices, AHRQ Pub. 06-0057. American Nurses Association (ANA). (2005). Principles for documenta - tion. PDI-05. Silver Springs, MD: ANA Publishing. Bates, D.W., Kuperman, G.J., Wang, S., Gandhi, T., Kittler, A., Volk, L., et al. (2003). Ten commandments for e ffective clinical decision support : Making the practice of evidencebased medicine a re a l i t y. J o u rnal of American Medical Inform a t i c s Association, 10(6), 523-530. Bobb, A.M., Payne, T.H., & Gross, P. A., (2007). Viewpoint: Contro v e r s i e s s u rrounding use of order sets for clinical decision support in computerized provider order entry. J o u rn a l of the American Medical Inform a t i c s Association 14(1), 41-47. Fung, C.H. (2006). Computerized condition-specific templates for impro v- ing care of geriatric syndromes in a p r i m a ry care setting. J o u rnal of General Internal Medicine, 21( 9 ). 9 8 9-9 9 4. G a rg, A.X., Adhikari, N.K.J., McDonald, H., Rosas-Arellano, M.P., Devere a u x, P.J., Beyene, J., et al. (2005). Eff e c t s of computerized clinical decision s u p p o rt systems on practitioner perf o rmance and patient outcomes: A systematic re v i e w. J o u rnal of the American Medical Association, 2 9 3(10), 1223-1238. Kaushal, R., Jha, A.K., Franz, C., Glaser, J., S h e t t y, K.D., Jaggi, T., et al. (2006). R e t u rn on investment for a computerized physician order entry system. J o u rnal of the American Medical I n f o rmatics Association, 13(3), 261-2 6 6. Kawamoto, K., Houlihan, C.A., Balas, E.A., & Lobach, D.F. (2005). I m p roving clinical practice using clinical decision support systems: A systematic review of trials to identify features critical to success. B r i t i s h Medical Journal, 330, 765-773. O s h e ro ff, J.A., Pifer, E.A., Teich, J.M. Sittig, D. F., & Jenders, R.A. (2004). C l i n i c a l decision support implementers workbook. Chicago: HIMSS. O s h e ro ff, J.A., Teich, J.M., Middleton, B., Steen, E.B., Wright, A., & Detmer, D.E. (2007). A roadmap for national action on clinical decision support. J o u rnal of the American Medical I n f o rmatics Association, 14(2), 141-145. O s h e ro ff, J.A. (2009). I m p roving medica - tion use and outcomes with clinical decision support: A step-by-step g u i d e. Chicago: HIMSS. Additional Reading O s h e ro ff, J.A., Teich, J.M., Middleton, B.F., Steen, E.B., Wright, A., & Detmer, D.E. (2006). A roadmap for national action on clinical decision support. For AMIA as a work product for ONC Contract HHSP233200500877P. June 13, 2006. U rologic Nursing Editorial Board Statements of Discl o s u r e In accordance with ANCC-COA gove rning rules Urologic Nursing E d i t o rial Board statements of disclosure are published with each CNE offe ri n g. The statements of disclosure fo r this offe ring are published below. K aye K. G a i n e s, M S, A R N P, C U N P, disclosed that she is on the Speake r s Bureau fo r P f i ze r, Inc., and Nova rtis Oncology. Susanne A. Q u a l l i ch, A N P - B C, N P - C, C U N P, disclosed that she is on the Consultants Bureau for Coloplast. All other U rologic Nurs i n g E d i t o rial Board members reported no actual or potential conflict of interest in relation to this continuing nursing education art i c l e. 352 UROLOGIC NURSING / September-October 2009 / Volume 29 Number 5