Decision Support System Design And Implementation For .
Decision Support System Design andImplementation for Outpatient Prescribing:The Safety in Prescribing StudyAdrianne C. Feldstein, David H. Smith, Nan R. Robertson,Christine A. Kovach, Stephen B. Soumerai, Steven R. Simon,Dean F. Sittig, Daniel S. Laferriere, Mara KalterAbstractBackground: Decision support (i.e., alerts and reminders) at the time ofmedication prescribing has been shown to be an effective method for reducingpotential medication errors for inpatients, but much less is known about theeffects in the outpatient setting. Using qualitative methods to inform our work,medication safety decision support and provider education interventions weredesigned to improve the use of medications in ambulatory care. Methods: Thispaper presents the study rationale, design, development, and implementation ofthe interventions. We include a summary of the qualitative findings, includingusability testing of the decision support, and we describe how the qualitativefindings enhanced the intervention design. We also describe our approach toclinician recruitment. We enumerate limitations of the existing electronic medicalrecord to accommodating recommendations from the qualitative work andusability testing. Results: Our qualitative interviews suggest that clinicians preferdecision support alerts that are clear, concise, and easy to navigate, with minimalinformation in the alert text. In usability testing, we found that decision supportalerts are followed less often when they appear at inappropriate times inworkflow, are difficult to read, add to time pressure, and are canceled beforebeing fully read. Our deliberate approach to clinician recruitment for theeducational sessions achieved an impressive attendance rate of 85 percent.Conclusions: This study shows that careful consideration of alert design andprovider education is critical. Future work will examine the effectiveness of thedecision support and whether our educational intervention improves physicianresponse.IntroductionThe Institute of Medicine (IOM) report on medical errors identifiedcomputerization of medication prescribing as an important patient safetystrategy.1 Decision support through alerts and reminders at the time of prescribinghas been shown to be an effective method for reducing potential medication errorsfor inpatients,2–4 but much less is known about the effects in the outpatient setting.One preliminary study5 examined the effect of basic computerized prescribing onmedication errors in outpatients and concluded that more advanced decisionsupport was necessary. We were unable to identify any studies about outpatient35
Advances in Patient Safety: Vol. 3decision support of significant size and scope, or any assessing whether aneducation program enhances the effectiveness of computerized physician orderentry (CPOE) systems intended to promote medication safety. Whether or notclinician education can improve the effectiveness of alerting systems is animportant question, because educational outreach, often called “academicdetailing,” has been shown to be perhaps the single most consistently successfulintervention for improving prescribing.6Using qualitative methods to inform our work, we designed an interventionstudy of computerized decision support at the time of medication order entry andsupporting clinician education to improve medication safety in ambulatory care.This paper presents the study design and the intervention rationale, development,and implementation for the Safety in Prescribing (SIP) efficacy study and Phase Ieffectiveness trial.7 The objective of the SIP study is to develop patient-specificcomputerized decision support for providers and measure its effectiveness toreduce prescribing errors. The SIP study, begun in 2004, is ongoing; this paperpresents results of the study design process.MethodsSetting and subjectsThe study is being conducted in a large nonprofit group-model healthmaintenance organization (HMO) in the Pacific Northwest (northwest Oregon andsouthwest Washington) that cares for approximately 450,000 members. The grouppractice and decision support intervention includes 669 physicians practicing inall the major specialties of medicine and surgery; they supervise 417 allied healthpractitioners (nurse practitioners and physician assistants) at 20 clinical practicesites. Fifteen primary care practices in adult medicine, employing 281 familypractice and internal medicine practitioners, agreed to participate in the SIP studyrandomized trial. The protocol for the study was approved by the HMO’sinstitutional review board.DatabasesThe HMO has used the EpicCare (Epic Systems, Madison, WI) electronicmedical record (EMR), which includes computerized physician medication andother order entry, for all outpatient contacts since 1996. The EMR provides onlineaccess to medical problem lists, visit diagnoses, procedures, patientdemographics, and visit progress notes. The EpicCare product provides two typesof decision support relevant to this project: (1) drug-specific alerts that appearwhen the target drug is prescribed at the time of CPOE, and (2) patient-specificalerts created using the Best Practice Alerts tool. The latter were tailored withadditional programming, as described below, to generate the alerts for this study.Prescribers had experience with both types of alerts at the time of our qualitativework with them.36
Decision Support for Outpatient PrescribingThe administrative and clinical electronic databases contain information oninpatient admissions, pharmacy dispenses, outpatient visits, laboratory tests, andoutside claims and referrals. All of these databases are linked through a uniquehealth record number that each member receives upon enrollment in the healthplan. Because the study site is a closed, group-model HMO, the databases captureclose to 100 percent of all medical care and pharmacy services members receive.Rates of dispensing errors in the clinical target areas are available through datafrom the outpatient pharmacy dispensing system, and supporting demographic,diagnostic, and procedure information is available through the HMO’s otherdatabases.Intervention rationaleDecision support interventionIn a recent report, Bobb et al.10 found that of 1,111 prescribing errorsidentified in an inpatient setting (62.4 errors per 1,000 medication orders), 64.4percent were rated as likely to be prevented with CPOE (including 43 percent ofthe potentially harmful errors). Bobb et al. also found that 13.2 percent of theerrors were unlikely to have been prevented with CPOE, and 22.4 percent couldpossibly have been prevented with CPOE. Findings were dependent on specificCPOE system characteristics. The authors concluded that incorporating advancedclinical decision support within the order entry routine is vital for achievingmaximum medication safety. In a recent systematic review, Kaushal et al.11 foundfive trials assessing CPOE with clinical decision support that met their criteria. Ofthese studies, two demonstrated a marked decrease in the serious medication errorrate, one showed an improvement in corollary orders (e.g., requesting serumantibiotic levels when antibiotics are ordered), another had an improvement infive prescribing behaviors (i.e., selection of recommended drugs within a class,use of recommended doses and frequencies, reduction of excessive doses, use ofcorollary orders, and compliance with drug use guidelines),12 and one had animprovement in nephrotoxic drug dose and frequency. As noted previously, thereare no comparable data for the outpatient setting.Education interventionWe developed and implemented an educational outreach intervention,popularly called academic detailing, to increase prescriber acceptance ofevidence-based alerts and reminders in a computerized order entry system. Suchprograms draw on adult learning theory and target physicians with clear andprofessionally illustrated educational materials from a credible organization,combined with one-on-one, face-to-face visits.13 Although most early studies ofacademic detailing employed individual meetings between detailers (eitherphysicians or pharmacists) and prescribers, more recently studies havedemonstrated the capacity of group detailing to improve practice behavior. Byusing small groups of four to eight physicians while retaining all other tenets ofacademic detailing, researchers have improved prescribing for various classes of37
Advances in Patient Safety: Vol. 3medications.14–17 In one comparative study, group detailing was as effective asindividual academic detailing in reducing inappropriate prescribing.17Intervention DevelopmentClinical target areas and guidelinesThis project is part of a larger prescribing safety study being conducted by theHMO Research Network, Center for Education and Research on Therapeutics(CERT).18 Therefore, our project investigators participated in and had access toclinical guidelines for prescribing safety that were generated by national projectexpert teams.Locally, we sought to target potential prescribing errors that were ofsignificant frequency and severity and that were preventable. Working withgroups of expert pharmacists and physicians, we reviewed available dataregarding the frequency and severity of potential prescribing errors to help usselect clinical areas to target.During our structured interviews with prescribers, we also elicited theirpreferences. A committee of investigators and expert pharmacists and physicianselected to focus on medications that are generally contraindicated in the elderly,require dose adjustment for renal insufficiency, and are commonly implicated indrug interactions. We assembled pharmacist and physician teams in the threetarget areas to choose specific drugs or drug pairs for intervention and to refinealert text and tools.Organizational commitmentBefore the project could be taken to the physicians for support, four functionalgroups within the HMO—research, applied medical informatics, pharmacy, andinformation technology (IT)—had to agree upon design, accountabilities, resourceallocation, training, implementation, and evaluation. Key committees anddecisionmakers in quality management, patient safety, and IT were enlistedduring the proposal and development phases. Senior management was enlisted toconvene the groups and facilitate discourse to arrive at an integrated approach andresource reassignment as needed.A coordination and implementation oversight committee advised on allaspects of the project and provided IT support. Alert programming and text wentthrough two levels of review and approval. The first review was by a standingcommittee of HMO physicians and IT experts that reviews all alerts andreminders; final approval was through the Regional Formulary and TherapeuticsCommittee. Interestingly, several key managers were hesitant to withhold theeducational intervention from some clinics as part of the study design. We assuredthem that all materials would be available for delivery to the control group afterthe end of the study, should the education be found effective.38
Decision Support for Outpatient PrescribingStructured interviews with prescribersFormative work appears to be an important predictor of intervention successwith practicing clinicians,8 and little literature exists in this area. The methods weused to design the decision support and educational interventions, based on ourformative work, have been presented in detail elsewhere.19 We recruited 20primary care prescribers from family practice and internal medicine, using e-mailand phone followup. Interviews were conducted during paid work time. Wedeveloped a semistructured interview guide with closed and open-ended questionsto elicit prescribers’ prior experience with alerts and education activities related todecision support. We also elicited their opinions about proposed medicationsafety alerts and their preferences about how to learn about new decision supportfunctions.Interviewees provided informed consent. Interviews were transcribed foranalysis. We developed a coding dictionary and used Atlas.ti 4.2 (ScientificSoftware Development, 1997), a qualitative research software package, foranalysis.Alert discount usability testingDiscount usability testing can be used to evaluate computer applicationsaccording to how easy they are to learn and remember, and their efficiency, errorrate, and user satisfaction. This is a useful step before alert implementationbecause although real-time medication prescribing alerts can improve clinicalperformance and patient safety, other investigators have identified flaws in alertlogic. For example, in one study, 37 of 43 (86 percent) sympathomimetic–tricyclic antidepressant interaction alerts were unjustified on the basis of scientificevidence.20 In another study, the data triggering alerts were incomplete. For29 percent (31 of 108) of the critical drug interaction alerts, one of the twointeracting drugs contained “TOP” or “oint” or “shampoo” in the prescription,indicating that the drug was to be given topically rather than taken by mouth orinjected.21There may also be flaws in the alert display. For example, in a study that didnot demonstrate an effect of computerized reminders, the computer systemnotified clinicians only by means of a banner at the bottom of the screen statingthat “there are suggested orders for this patient.” In a subsequent study thatdemonstrated a significant effect, the computer immediately displayed thereminders to the physicians as full, prewritten orders and highlighted thesuggested reminders with a distinctive color scheme, disabled the “escape” key,and set the default to “order,” allowing the physician to accept the item simply bypressing the “enter” key.22For the SIP study, all 11 primary care clinicians from one clinic assigned tothe intervention were asked to participate in a 30-minute usability session in theiroffice. All usability sessions were conducted by trained usability evaluators andaudiotaped for later review and qualitative analysis. Evaluators read a script thatprovided each clinician user with an introduction to the study, obtained their39
Advances in Patient Safety: Vol. 3consent, and described seven hypothetical cases, complete with the patient’s age,gender, and a statement of clinical condition. The clinician was instructed toperform a particular action using the electronic medical record system while“thinking aloud” verbalizing his/her reasoning process while performing theactivity.23 Evaluators recorded the start time, each click made, associatedcomments, and the end time for the activity.Education development processIn developing the educational intervention for this study, we adhered closelyto the key principles of academic detailing.13 The barriers to changing prescribingbehavior identified in our interviews became the principal targets of theeducational program, while the facilitating factors frequently served as usefulcounter-arguments to issues and concerns raised by participants. We establishedcredibility by working through respected organizational sponsors and referencingauthoritative and unbiased sources of information. We collected evidence fromthe peer-reviewed published literature as well as data from the HMO todemonstrate the clinical implications of prescribing alerts. The educationalsessions were intentionally designed to be group discussions, rather than lecturesor presentations.Our qualitative results suggested that physicians preferred to receiveinformation about prescribing safety from internal physician experts. In response,we recruited two physicians, both internists, who were well known and respectedwithin the HMO for their clinical leadership skills and their prior experience aseducators. They were provided with 2 hours of training. The educator trainingfocused on techniques for eliciting barriers to complying with the alerts within theclinical decision support system. We provided the educators with a list of talkingpoints that could be interspersed in the discussion, and we prepared them to“inoculate” the group by proactively addressing issues and concerns that weanticipated would arise. We equipped the educators with a list of arguments to usethrough the discussion. The arguments were derived chiefly from the interviews;counter-arguments were derived from our own research group’s experience andthe literature.An education handout was created to accompany the academic detailingsession. These materials emphasized the graphical and tabular presentation ofinformation and incorporated color and easy-to-read typefaces. This material wasalso posted on a Web site, clearly identified on the organizational homepage,called “Safety in Prescribing.” *The educators presented a practice run of the presentation to the coinvestigator group, and piloted the educational session with the staff at a singleclinic. A focus group followed. The education was generally well received duringthis practice run, and we refined several areas based on the feedback.*The handout is available from the corresponding author.40
Decision Support for Outpatient PrescribingWe recruited primary care clinicians to attend the educational sessions with ane-mail invitation from clinic physician leaders (primary care participants). A localclinician staffing clerk arranged time out of clinic for the meeting and offeredalternate sessions for clinicians not able to attend the main scheduled session.Outcome analysisThe primary outcome measure for SIP is the incidence rate of the targetedprescribing errors. For medications generally contraindicated in the elderly, theoutcome measure is the number of prescriptions of the contraindicatedmedications, divided by the sum of the number of prescriptions of thecontraindicated medication plus the number of prescriptions of the preferredmedication alternatives (specific agents to vary for each medication targeted).For dose adjustment in renal insufficiency, the outcome measure for patientswith study-defined renal insufficiency is the number of prescriptions without doseadjustment divided by the total number of prescriptions of the targetedmedications.For drug interactions, the outcome measure is the number of the targetedcontraindicated coprescribing events, divided by the sum of the number of thecontraindicated coprescribing events plus the desired coprescribing alternatives.ResultsImplementation and outcome measuresFifteen primary care group practices are participating in the randomizedportion of the trial. To ensure baseline comparability, we determined the 15practices’ baseline prescribing rate for several medications generallycontraindicated in the elderly, including tertiary tricyclic antidepressants andlong-acting benzodiazepines. The practice sites were then matched in pairsaccording to baseline prescribing rates, and each pair of clinics was randomized.One clinic of each pair was assigned to receive only the computerized decisionsupport tool, and the other was assigned to receive the decision support tool plusclinician academic detailing.Final target clinical areasFor medications generally contraindicated in the elderly, we chose to targetoral agents that had significant prescribing frequencies in our population betweenJanuary 1, 2000, and June 30, 200l, and that were addressed through the BeersCriteria.24 We selected long-acting benzodiazepines, tertiary tricyclicantidepressants, skeletal muscle relaxants, anti-inflammatory agents, andpropoxyphene. For dose adjustment in renal insufficiency, patients with chronickidney disease25 were those with two measures of glomerular filtration rate (GFR)of 60 ml/min or less,26 separated by at least 90 days. The team then selected drugsto target for intervention that were frequently not adjusted in this population and41
Advances in Patient Safety: Vol. 3for which clear dosing guidelines were available. Allopurinol, ciprofloxacin,colchicine, trimethoprim/Sulfamethoxasole, ga
Adrianne C. Feldstein, David H. Smith, Nan R. Robertson, Christine A. Kovach, Stephen B. Soumerai, Steven R. Simon, Dean F. Sittig, Daniel S. Laferriere, Mara Kalter Abstract Background: Decision support (i.e., alerts and reminders) at the time of medication prescrib
7. Decision Support System (DSS) A decision support system or DSS is a computer based system intended for use by a particular manager or usually a group of managers at any organizational level in making a decision in the process of solving a semi structured decision (Figure 7).
Decision Support System (DSS) Decision Making Every decision and action that humans make is inherently related to decision making. But humans have weaknesses (subjective, bias, forgetful, imprecise, and slow) in many cases of decision making. Therefore humans need to be assisted by computer-based tools called DSS to improve
The Tivoli Decision Support 2.1 User’s Guide describes the features and concepts behind the Tivoli Decision Support 2.1 product. Step-by-step procedures for using the Tivoli Decision Support Discovery Interface are also provided. Hereafter, the Tivoli Decision Support Discover Administrator is called the
1. What is decision theory?.5 1.1 The decision disciplines 5 1.2 Decision processes 7 1.3 Decision matrices 11 1.4 Classification of decision theories 13 1.4.1 Normative and descriptive theories 14 1.4.2 Individual and collective decision-making 15 1.4.3 Degrees of knowledge 16 2.
Oct 18, 2014 · A decision problem is characterized by decision alternatives, states of nature, and resulting payoffs. The decision alternatives are the different possible strategies the decision maker can employ. The states of nature refer to future events, not under the control of the decision maker, which
Decision theory and Decision analysis Decision Analysis De nition (B. Roy):\consists in trying to provide answers to questions raised by actors involved in a decision process using a model" Answers:\Optimal solution" or \Good decision" is absent Models:formalized or not Brice Mayag (LAMSADE) Introduction to Decision Modeling Chapter 0 18 / 36
tables syntax and layout are defined by the DMN standard while Drools native decision tables are defined by the Drools project. Red Hat Decision Manager supports both formats of decision tables, but they are not interchangeable. For more information about Drools decision tables, see Designing a decision service using uploaded decision tables. 1 .
040020205 Decision Support Systems 2014 Ms. Bhoomika A. Patel Page 1 Unit 1 : Decision Making and Computerized Support Short Questions. 1. Define system. List the types of system. 2. Define open system. Give an example of closed system. 3. List out the four steps managers take in making a decision. 4.
