R E V I E W S & A N A LY S E S Medication Errors Involving Overrides of Healthcare Technology Matthew Grissinger, RPh, FISMP, FASCP INTRODUCTION Manager, Medication Safety Analysis The use of technologies to prevent and detect medication errors has been increasing Pennsylvania Patient Safety Authority over the past decade. A stratified random sample survey of pharmacy directors at ABSTRACT 1,435 general and children’s hospitals in the United States found that the large major- Users can bypass many of the safety ity (97.1%) of hospitals use automated dispensing cabinets (ADCs) in their medication features incorporated in medication-use distribution systems, 88.4% use bar-coded medication administration (BCMA) systems technologies that provide warnings about to verify patient identity and electronically check doses administered by nurses, 80.9% possible unsafe conditions or errors. use computerized prescriber order entry (CPOE) systems, and 80.5% use smart infusion Analysts reviewed medication error event pumps (infusion pumps that incorporate medication safety software and can contain a reports that indicated the use of overrides comprehensive library of drugs, usual concentrations, dosing units [e.g., mcg/kg/min, submitted from January 2013 through units/hr], and dose limits [minimum/maximum] that can be set according to institu- December 2014 to the Pennsylvania tion-established parameters1).2 Patient Safety Reporting System. Of the Each of these healthcare technologies provide the capability to alert users to possible 583 event reports related to the use of unsafe conditions or errors with the use of a medication. Many technologies can record overrides, the most commonly mentioned the number and types of alerts presented to users, the alert overrides, and the user’s technology was automated dispensing stated explanation for overriding the alert (from a standard list or a free-text explana- cabinets (77.0%, n = 449), followed tion). While technologies employed in the medication-use process can generate reports by computerized prescriber order entry delineating overrides, these reports do not always capture all of the factors that led to (8.2%, n = 48) and bar-code medication the decision to override, or provide the result of an error in which an override played a administration devices (7.5%, n = 44). role. Other systems, such as an organization’s internal event reporting system, also can The most common classes of medica- be used to help capture the factors contributing to an event. tions cited were antibiotics (12.0%, Analysis of events reported through the Pennsylvania Patient Safety Reporting System n = 70), opioids, (12.0%, n = 70), and (PA-PSRS) has identified medication errors which involve an override of healthcare anticoagulants (7.4%, n = 43); and technology. This analysis reviews reports submitted to the Pennsylvania Patient Safety 26.4% (n = 154) of the reports involved Authority mentioning the use of overrides to delineate factors that led to overriding an at least one high-alert medication. alert and the results related to the use of an override. Organizations may consider develop- ing criteria for alerts that focus on real METHODS chances of patient harm while preventing alert fatigue, minimizing the need for or When reviewing events reported through PA-PSRS, Authority analysts can further clas- use of overrides. (Pa Patient Saf Advis sify reports using a tag for future query opportunities. Analysts queried the PA-PSRS 2015 Dec;12[4]:141-8.) database for events reported as medication errors to the Authority from January 2005 through December 2014 that had been tagged as events involving overrides. Analysts also queried the database using the keyword search terms “overri*,” “overro*,” “overid*,” and “overo*,” where the asterisk represents a wild-card to include multiple endings to each search term. The query yielded 5,399 medication error reports. The medication name, route, patient care area, event description, and harm score, adapted from the National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) harm index,3 were provided by the reporting facility. When a medication name data field was left blank but the name was provided in the event description, an analyst adjusted the medication name field. The reports were evaluated to determine what fac- tors were associated with medication errors involving the use of healthcare technology and an override. Authority analysts focused on the last two calendar years, which yielded 790 reports. Two hundred seven (26.2%) of these reports were excluded because the error did not result from the use of an override, because “override” was used in the event description of an event that clinically warranted the use of an override (e.g., to obtain medication during an emergent situation), or because an override could not be performed. Five Vol. 12, No. 4—December 2015 Pennsylvania Patient Safety Advisory Page 141 ©2015 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S hundred eighty-three reports remained Figure 1. Harm Scores for Events Involving Overrides, as Reported to the Pennsylva- for qualitative analysis. Reports were nia Patient Safety Authority, 2013 through 2014 (N = 583) analyzed and assigned a category or type of error (e.g., type of technology involved, NO. OF REPORTS 350 345 potential cause of error) based on the analyst’s interpretation of the event description. Analysts made note of events 300 involving a high-alert medication, based on the ISMP List of High-Alert Medications 250 in Acute Care Settings.4 RESULTS 200 Categorizing the reports by harm score 150 shows that more than 75% of the events reached the patient (harm score = C 100 90 through I) and only 0.3% (n = 2) resulted 68 in patient harm (harm score = E through I) 62 (Figure 1). 50 16 Overall, 57 unique types of care areas 2 0 0 0 0 were associated with events involving 0 A B1 B2 C D E F G H I an override; the most common were MS15629 medical-surgical units. Intensive care Incident (99.7%, n = 581) Serious Event (0.3%, n = 2) units (ICUs) and emergency departments (EDs)—care areas that have patients with HARM SCORE more acute conditions for which there may Reports of events that did not reach the patient (25.0%, n = 146) be a greater need to override an alert in Reports of events that reached the patient (75.0%, n = 437) order to obtain medications emergently— accounted for less than a quarter of the care areas cited in reports (Figure 2). The most common type of event involv- situations in which medications were with- ing overrides of ADCs were unauthorized drawn “on override” before a pharmacist More than half of all reports involved medications (e.g., obtaining a medication reviewed the order or when the pharmacy elderly patients (65 years of age or older), for a patient with no prescribed order was closed. and only 5.5% (n = 32) of the events for the patient), followed by wrong-drug involved a pediatric patient. The following are examples of events in events and wrong dosage form events which a healthcare practitioner obtained The most common classes of medica- (e.g., selecting a sustained-release product high-alert medications from an ADC tions cited were antibiotics and opioids, instead of the immediate-release form, using an override.* with slightly more than a quarter of the selecting an oral formulation instead events involving at least one high-alert of the injection) (Figure 4). A majority Nursing thought that the warfarin medication.4 Among events involving of the unauthorized medication events was ordered per the warfarin com- high-alert medications, the three classes specifically stated there were no orders ment on the MAR [medication most commonly cited were opioids (e.g., for the medication, and over 30% of the administration record]. There was morphine), anticoagulants (e.g., warfarin, unauthorized medication events involved no order written for the warfarin on heparin), and insulin; combined, these a high-alert medication. the MAR, and the nurse mistook medication classes represented 78.6% the 2 in the warfarin comment as a While most of the wrong-drug and wrong 2 mg dose. The error was discovered (n = 121 of 154) of the events involving dosage form events did not include a high-alert medication. by the pharmacy when reviewing the enough detail to determine additional When looking at the types of technology causative factors for the event, 16.4% that were overridden by users, over 75% of (n = 26 of 159 wrong-drug and wrong * The details of the PA-PSRS event narratives override events involved ADCs (Figure 3). dosage form reports) mentioned in this article have been modified to preserve confidentiality. Page 142 Pennsylvania Patient Safety Advisory Vol. 12, No. 4—December 2015 ©2015 Pennsylvania Patient Safety Authority Figure 2. Care Areas for Events Involving Overrides, as Reported to the Pennsylvania Patient Safety Authority, 2013 through 2014 (N = 583) CARE AREA Medical/surgical unit 111 (19.0%) Intensive care unit 80 (13.7%) Medical unit 63 (10.8%) Telemetry 46 (7.9%) Emergency department 45 (7.7%) Acute specialty rehab unit 34 (5.8%) Rehabilitation unit 29 (5.0%) Psychiatric unit—adult 23 (3.9%) Medical/surgical/cardiology 15 (2.6%) intermediate unit Pharmacy 15 (2.6%) Labor and delivery 10 (1.7%) Other care areas 112 (19.2%) 0 20 40 60 80 100 120 MS15630 NO. OF REPORTS profile override report the next day. The pharmacist noted that this was and wrong dose/overdosage (Figure 5). The pharmacy called the unit to ask unusual [concurrent prescriptions for High-alert medications were reported if there was an order written for this two extended-release opioids] and put in 31.4% (n = 22 of 70) of these events; dose, and there was not. Upon review the order on pending status until clari- anticoagulants (50.0%, n = 11 of 22) was of the MAR and speaking to the fied. The nurse told the pharmacist the class of high-alert medications most nurse involved, it was discovered that that since the patient was having often involved. the warfarin comment was mistaken pain, she had overridden and adminis- Almost 20% (n = 13 of 70) of the high- for an actual dose. tered the oxyCODONE ER without alert medication reports mentioned Patient was ordered sliding scale insu- pharmacy verification at a time when overrides of both CPOE and pharmacy lin using Humulin® R [insulin human the pharmacy was open. Upon clari- order entry system alerts for a given order, injection, USP (rDNA origin)] insulin. fying with the patient’s pharmacy, with a prescriber overriding an alert and The nurse removed Humalog® [insulin the pharmacists determined that the pharmacist also overriding the same lispro injection, USP (rDNA origin)] the patient was actually on oxyCO- type of alert. Of the reports that cited insulin from the ADC machine on DONE immediate release 40 mg po only CPOE systems (n = 48), 12.5% (n = override and administered this instead TID prn [as needed for] pain. 6 of 48) mention practitioners other than of Humulin R. In 12.0% (n = 70 of 583) of the events, prescribers (e.g., nurses, unit secretaries) Order entered for oxyCODONE ER overrides occurred during the use of entering the orders into the system. [extended release] 40 mg po TID CPOE and/or pharmacist order entry The following are examples of reports of [by mouth three times a day] as well systems. The most common types of errors associated with overrides involving as morphine ER 100 mg po TID. alerts that were overridden were those electronic order entry systems. for drug allergies, duplicate drug therapy, Vol. 12, No. 4—December 2015 Pennsylvania Patient Safety Advisory Page 143 ©2015 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S Patient had a standing order for Figure 3. Technologies Overridden in Events Involving Overrides as Reported to the Coumadin® [warfarin sodium] 4 mg. Pennsylvania Patient Safety Authority, 2013 through 2014 (N = 583) The doctor ordered a 7.5 mg tablet Smart pump but left the 4 mg order active. A n = 12, 2.1% Electronic medication duplicate therapy alert was generated administration record and was overridden by the physician Pharmacy order n = 10, 1.7% as “Not clinically significant.” Phar- entry systems macist discontinued the existing 4 mg n = 35, 6.0% Unknown dose to avoid duplication of therapy, n = 2, 0.3% which could have resulted in patient Bar-code medication getting 11.5 mg of Coumadin. administration n = 44, 7.5% Lovenox® [enoxaparin sodium] treat- Automated ment dose x1 ordered for patient in Computerized dispensing cabinet the ED. The physician received an prescriber order n = 449, 77.0% alert since allergy field had heparin entry systems and related preparations. Physician n = 48, 8.2% entered override reason = “Give not a true allergy.” Order verified and drug MS15631 sent [by pharmacy] and was charted as given. Rapid response called the fol- lowing morning. Platelets dropped to Note: Numbers add up to more than 583 because in 17 reports, multiple forms of 30,000 platelets per microliter (mcL) technology were mentioned as being overridden within the same report. [which is below the normal range]. However, after platelet value returned events reported through PA-PSRS. The ing that this was not for the right (and patient transferred off unit), most common types of these events were patient, but she continued on and pharmacist noticed the allergy field. wrong drug, wrong dosage form, and overrode the warning. The patient Patient’s weight was accidentally wrong dose/underdose (Figure 6). One did get the correct drug and dose, but entered into dose field by pharma- out of four reports involving BCMA only because the two patients were cist. Patient received 74 units/kg/ involved a high-alert medication. ordered the same thing. The other hr of heparin instead of 13 units/ nurse ended up having to call us for Following are examples of reports of kg/hr, which exceeded maximum a missing dose. errors associated with overrides involving rate. Patient received 5,476 units BCMA. OxyCODONE [immediate release] of heparin over an hour instead of Patient received twice daily morning 15 mg was dispensed from pharmacy the ordered 962 units. Unclear why medications early prior to dialysis, instead of MS Contin 15 mg to pharmacist overrode dose alert warn- including MS Contin® [morphine [ADC]. Nurse removed [wrong] medi- ing that fired. Nurse attempted to sulfate extended release] and Cell- cation from Accudose and overrode autoprogram pump but received alert cept® [mycophenolate mofetil]. When the bar-code scan that indicated it and programmed pump manually. patient returned from dialysis, the was the wrong product. A nurse took a telephone order from nurse gave the morning medications, the doctor for potassium chloride DISCUSSION including those that had already been 20 mEq po bid [by mouth twice a administered. Early dose warnings ADCs can be linked to pharmacy com- day], with the first dose given now. had fired, but nurse overrode warning. puter systems (or “profiled”) so that a There was an override comment of The nurse gave the patient the pharmacist must review the appropri- “provider approved” entered, and the 5 mg dose of Coumadin that was ateness of a medication order prior to pharmacist verified the order without for another patient. She did scan the administration—most notably identify- questioning if the patient should be patient, but she scanned the label ing drugs to which patients are allergic, on 2 separate potassium orders. on the bag instead of scanning the unsafe doses, or unrecognized food or Overrides with the use of BCMA were drug interactions—and approve that order drug. She did receive a warning stat- cited in 7.5% (n = 44 of 583) of the Page 144 Pennsylvania Patient Safety Advisory Vol. 12, No. 4—December 2015 ©2015 Pennsylvania Patient Safety Authority Figure 4. Common Event Types of Reports Involving Overrides of Automated CPOE and pharmacy order entry systems Dispensing Cabinets, as Reported to the Pennsylvania Patient Safety Authority, have clinical decision support (CDS) 2013 through 2014 (N = 449) systems, which can provide warnings about wrong dosages or other related % OF TOTAL OVERRIDE REPORTS prescribing conflicts, interactive computer 100 programs, or other tools that are designed to assist physicians and other healthcare professionals with decision making.7 80 CDS systems provide various forms and levels of alerts to indicate possible issues with medication orders, such as allergies 60 to the prescribed medications, excessive doses, and therapeutic duplications. Unfortunately, little attention may be 40 given to how the accuracy of these alerts should be best aligned with their appear- 25.2 (n = 113) 23.6 (n = 106) ance and degree of interruption. Many of these alerts are “soft stop” alerts, which 20 11.8 (n = 53) can be interruptive and the user can dis- 4.5 (n = 20) miss by providing a simple override of the 2.2 (n = 10) warning. The display of excessive numbers 0 of alerts can lead to the phenomenon No order Other/unknown Order discontinued often referred to as “alert fatigue.” The Unauthorized medication* Wrong drug† Wrong way alerts are prioritized and presented (31.8%, n = 143 of 449) dosage form† to the user may be as important as which alerts are presented. Alerts for very serious MS15632 EVENT TYPE clinical situations (i.e., true positive alerts) may be ignored when lost in a sea of less * 30.8% (n = 44 of 143) of the unauthorized medication events involved a high-alert medication. clinically important or irrelevant ones † 16.4% (n = 26 of 159) of the reports mentioned situations in which medications were (i.e., false-positives). withdrawn “on override” before a pharmacist reviewed the order or when the phar- macy was closed. In a study involving adult primary care practices affiliated with a teaching hospi- before a nurse is able to remove the medica- accessible medications available on over- tal, Weingart et al. showed that physicians tion from the ADC.5 These devices also ride in locations such as the ED and overrode 91.2% of drug allergy alerts and allow for the use of overrides to bypass the ICU. Therefore, the term “override” takes 89.4% of high-severity drug interaction pharmacist’s review of a medication order on a different meaning with this technol- alerts.8 The physician reviewers in this when assessment of the patient indicates ogy, as practitioners are not overriding study determined that 36.5% of the alerts that a delay in obtaining a medication or bypassing a clinical alert presented were inappropriate. Slight et al. conducted from the ADC (e.g., to wait for a pharma- to them but are removing a medication a study (which included primary care prac- cist’s review of the order) would harm the before the pharmacist’s review of an tices affiliated with two Harvard teaching patient. order. This practice can be unsafe when hospitals with over 1,700 prescribers) that this crucial clinical review is routinely evaluated the appropriateness of provid- Organizations have developed lists, ers’ drug-drug interaction alert overrides, commonly called an “override list,” of bypassed for convenience (“normalized deviance”) or to remedy process problems the reasons why they chose to override medications that can be removed from these alerts, and what actions they took as an ADC without a pharmacy review of such as excessive order turnaround time. After a review of 470 medication over- a consequence of the alert.9 The authors the order. If there is an urgent clinical found that 68.2% of the drug-drug inter- need for administering a drug before a rides, Kester et al. noted that 11.7% of overrides involved variances with written action alert overrides were considered pharmacist can reasonably be expected appropriate. In addition, a detailed chart to review the order and/or dispense orders, and 85.5% of those variances were not appropriately documented.6 review revealed that of the appropriate the drug, it is important to have readily alert overrides for which the provider Vol. 12, No. 4—December 2015 Pennsylvania Patient Safety Advisory Page 145 ©2015 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S Figure 5. Types of Alerts Overridden in Computerized Prescriber Order Entry and LIMITATIONS Pharmacy Computer Systems, as Reported to the Pennsylvania Patient Safety In-depth analysis by the Authority of Authority, 2013 through 2014 (N = 70) overrides associated with the use of % OF REPORTS technology occurring in Pennsylvania hospitals is limited by the information 100 reported through PA-PSRS, including the event descriptions and reasons why the event occurred. As a result, additional 80 override events and associated causes may have been reported but were not identi- fied by the query and analysis. 60 RISK REDUCTION STRATEGIES 34.3 The medication error events submitted to 40 (n = 24) 31.4 (n = 22) the Authority involving the use of over- rides when using technology reveal the 15.7 14.3 complex nature and variety of factors that 20 (n = 11) (n = 10) contribute to errors. Some of those factors 1.4 1.4 1.4 were an extension of the unique challenges (n = 1) (n = 1) (n = 1) associated with the use of each type of tech- 0 nology; however, many of the factors were ra te e n se er d do e/ dr zed tim tio the plica similar across all forms of technology. It is all ente do py er os gy se ug dic ri ng ov g d tho tra also important to understand that the use m Du ain ro Ex cu n au W ro ntr of overrides is not a primary problem with Do Un W MS15633 Co the use of healthcare technology but rather ALERT TYPE a symptom of a larger problem of poor decision support design. Unfortunately, indicated they would “monitor as recom- BCMA systems allow overrides if emer- most of the reports did not provide much mended,” only 35.5% actually did. gency administration of a medication is explanatory information about the errors, necessary, if the bar code on the medica- causes, and contributing factors. Even BCMA technology can improve medi- tion’s package is not recognized by the so, these reports, observations from the cation safety through several levels of bar-code scanner, if the bar code is miss- Institute for Safe Medication Practices, functionality.10 At the most basic level, ing or unreadable, or if the patient’s and recommendations in the literature do the system helps to verify that the right corresponding identification band cannot offer strategies that healthcare facilities may drug is being administered to the right be scanned. In response to alerts, users consider to decrease risk in the medication- patient in the right dose and at the right either change their actions (e.g., find cor- use process. time. When one of these items does rect patient or medications) or override not match the patient record and order, alerts and document their reasons for General Strategies most systems alert the practitioner prior overriding the alerts. However, problems to administration. For example, an alert Strategies that can be applied to all of may occur when an alert is overridden. may be presented when the patient does these technologies include the following: In a review of BCMA use at five hospitals not have an active order or is allergic to that included analyzing BCMA over- —— Improve the positive predictive the scanned medication or if the dosage ride log data, Koppel et al. found that value (e.g., the number of true strength scanned is higher than what nurses overrode BCMA alerts for 4.2% of positive alerts compared with all was ordered. If BCMA systems detect patients charted and for 10.3% of medica- positive alerts) of alerts, and adjust mismatches between patient and medi- tions charted.3 Possible consequences of the presentation of the alerts (e.g., cation or medication and medication those workarounds included administra- interruptive versus noninterruptive) order, audible and/or visual alerts tion of wrong medications, wrong doses, according to how accurate they are. are triggered.11 wrong times, and wrong formulations. Page 146 Pennsylvania Patient Safety Advisory Vol. 12, No. 4—December 2015 ©2015 Pennsylvania Patient Safety Authority —— Develop a mechanism to identify Figure 6. Types of Alerts Overridden in Bar-Coded Medication Administration and remove alerts that provide little Systems, as Reported to the Pennsylvania Patient Safety Authority, 2013 through or no clinical value, which may con- 2014 (N = 44) tribute to alert fatigue. % OF REPORTS —— Solicit an explanation of the reasons or rationale for an override of alerts 100 that are of high severity. Limit this strategy, as requiring an explanation for all alerts could further contribute 80 to alert fatigue. —— Assess staff competency related to the safe use of technology and over- 60 rides, and provide education when indicated. 36.4 —— Review and approve all override 40 (n = 16) policies through the pharmacy and therapeutics (P&T) committee, 18.2 (n = 8) 13.6 11.4 medication safety committee, or an 20 (n = 6) (n = 5) 9.1 6.8 equivalent group. (n = 4) (n = 3) 2.3 2.3 —— Review override reports to identify (n = 1) (n = 1) and address barriers to the safe use 0 of healthcare technologies. Incor- rm ug se e nt se n e tim os tio tie do do dr fo rd tra porate additional means to identify pa er ng ge ng tra ve en nd ng ro /o sa ro Ex override hazards by reviewing the nc /u W W ro do se co se W do organization’s medication error do ng ng ng MS15634 report data and external sources ro ro ng ro W W ro ALERT TYPE W of information; conducting direct W observation of the use of technology; and implementing conversations circumstances.12 Additional strategies * Use a process whereby the drug with end users to determine when include the following: and dose are checked against the and why staff use overrides. —— Developing clearly stated organi- patient’s allergies, and weight as appropriate, to determine if the Automated Dispensing Cabinets zational policies and criteria for system overrides that limit access drug and dose are appropriate. The use of ADC overrides should be to medications before orders have * Provide preparation instructions situationally dependent and should not been reviewed and approved by a if the nurse is required to recon- occur merely because the desired medica- pharmacist13 stitute or dilute a medication. tion is on a list of medications for which overrides are sometimes indicated.12 —— Implementing strategies to reduce * Require an independent double While there may be a list of drugs with the risk of an error when an override check with another licensed the potential to be obtained emergently, is used, such as the following:11,14 healthcare provider when using there may be many other situations when * Limit the quantity and number the override function to remove there is sufficient time for the pharmacist of drug concentrations available. an organization-identified high- to review the medication prior to a nurse alert medication. * Minimize the use of multidose retrieving the dose. Establish criteria for containers. system overrides that allow emergency Computerized Provider Order * Ensure medications available Entry Systems access in circumstances in which wait- for override are unit specific and ing for a pharmacist to review the order To realize the benefits of CDS, CPOE removed only when there is an before accessing the medication could and pharmacy order entry systems need to emergent need. adversely impact the patient’s condition, be implemented correctly and used effec- but limit access before review in other tively. Too many alerts could lead to the Vol. 12, No. 4—December 2015 Pennsylvania Patient Safety Advisory Page 147 ©2015 Pennsylvania Patient Safety Authority R E V I E W S & A N A LY S E S use of overrides, system rejection, or unan- implications of implementing medication- medication ordering and administration ticipated outcomes such as an increased related CDS.2 If organizations are in the for a variety of reasons. Analysts identified number of errors or adverse events.15-17 process of implementing a CPOE system, 583 medication error events submitted Decreases in the volume of nuisance alerts consider and evaluate CDS components to the Authority from 2013 through 2014 have been shown to yield greater attention before CPOE implementation, keeping in involving an override of technology that paid to potentially more valuable alerts.18 mind that a high number of interruptive resulted in an error. A majority of event Consider examining the alerts currently alerts may even threaten clinician accep- reports mentioned that these errors took active in CPOE and pharmacy order entry tance of CPOE. Prescribers, pharmacists, place when healthcare practitioners were systems, and evaluate if any may be turned and other practitioners, as appropriate, allowed to simply bypass a warning, with off or relegated to a lower severity tier. should participate in the development of no other strategies in place to catch a Although vendor systems may allow alerts medication-related CDS and should work resulting error. to be tiered, there is typically a significant with medical leadership—either through Risk reduction strategies provided in this the P&T committee, an informatics amount of work necessary to vet any analysis may help organizations minimize committee, or another interdisciplinary changes and carry out the technical work the occurrence of override-related adverse committee—to decide how and when involved in the customization. events. Organizations may also consider medication-related CDS will be custom- The combination of pharmacists’ clinical providing criteria for the development ized and implemented.2 knowledge of drugs and their experience of alerts (in any form of technology) that with the interruptive alerts that have focus on real chances of patient harm CONCLUSION been present in pharmacy information while preventing alert fatigue. Healthcare practitioners use overrides systems for years provide pharmacists when using various technologies related to with a unique understanding of the many NOTES 1. Cohen MR, ed. Medication errors. 2nd ed. hospitals and health systems. Am J Health of automated dispensing devices. Am J Washington (DC): American Pharmaco- Syst Pharm 2011 Mar 15;68(6):e9-31. Also Health Syst Pharm 2010 Mar 15;67(6):483- logical Association; 2007:437. available at http://www.ashp.org/DocLi- 90. Also available at https://www.ashp. 2. Pederson CA, Schneider PJ, Scheckelhoff brary/BestPractices/AutoITGdlCPOE.aspx org/DocLibrary/BestPractices/AutoITG DJ. ASHP national survey of pharmacy 8. Weingart SN, Toth M, Sands DZ, et al. dlADDs.aspx practice in hospital settings: dispensing Physicians’ decisions to override comput- 14. Kowiatek JG, Weber RJ, Skledar SJ, et and administration—2014. Am J Health erized drug alerts in primary care. Arch al. Assessing and monitoring override Syst Pharm 2015 Jul 1;72(13):1119-37. Intern Med 2003 Nov 24;163(21):2625-31. medications in automated dispensing 3. National Coordinating Council for Medi- 9. Slight SP, Seger DL, Nanji KC, et al. Are devices. Jt Comm J Qual Patient Saf 2006 cation Error Reporting and Prevention. we heeding the warning signs? Examining Jun;32(6):309-17. NCC MERP index for categorizing medi- providers’ overrides of computerized drug- 15. Koppel R, Metlay JP, Cohen A, et al. Role cation errors [online]. 2001 Feb [cited drug interaction alerts in primary care. of computerized physician order entry 2015 Aug 6]. http://www.nccmerp.org/ PLoS One 2013 Dec 26;8(12):e85071. systems in facilitating medication errors. medErrorCatIndex.html 10. Pennsylvania Patient Safety Authority. JAMA 2005 Mar 9;293(10):1197-203. Also 4. Institute for Safe Medication Practices. Medication errors occurring with the use of available at http://jama.jamanetwork. ISMP list of high-alert medications in bar-code administration technology. Pa com/article.aspx?articleid=200498 acute care settings [online]. 2014 [cited Patient Saf Advis [online] 2008 Dec [cited 16. Han YY, Carcillo JA, Venkataraman ST, 2015 Aug 6]. http://www.ismp.org/ 2015 Aug 6]. http://www.patientsafety et al. Unexpected increased mortality Tools/institutionalhighAlert.asp authority.org/ADVISORIES/Advisory after implementation of a commercially 5. Institute for Safe Medication Practices. Library/2008/Dec5%284%29/Pages/ sold computerized physician order entry JCAHO has it right – pharmacists should 122.aspx system. Pediatrics 2005 Dec;116(6):1506-12. review all non-urgent drug orders prior to 11. Koppel R, Wetterneck T, Telles JL, et al. 17. McCoy AB, Waitman LR, Lewis JG, et al. administration. ISMP Med Saf Alert Acute Workarounds to barcode medication admin- A framework for evaluating the appropri- Care 2000 Jul 26;5(15):1. istration systems: their occurrences, causes, ateness of clinical decision support alerts 6. Kester K, Baxter J, Freudenthal K. Errors and threats to patient safety. J Am Med Inform and responses. J Am Med Inform Assoc associated with medications removed Assoc 2008 Jul-Aug;15(4):408-23. 2012 May-Jun;19(3):346-52. Also available from automated dispensing machines 12. Institute for Safe Medication Practices. at http://jamia.oxfordjournals.org/ using override function. Hosp Pharm 2006 Guidance on the interdisciplinary safe use content/19/3/346 Jun;41(6):535-7. Also available at http:// of automated dispensing cabinets [online]. 18. Simpao AF, Ahumada LM, Desai BR et al. archive.hospital-pharmacy.com/doi/ 2008 [cited 2015 Aug 6]. http://www. Optimization of drug-drug interaction pdf/10.1310/hpj4106-535 ismp.org/Tools/guidelines/ADC_Guide- alert rules in a pediatric hospital’s elec- 7. American Society of Health-System Phar- lines_Final.pdf tronic health record system using a visual macists. ASHP guidelines on pharmacy 13. American Society of Health-System Phar- analytics dashboard. J Am Med Inform planning for implementation of com- macists. ASHP guidelines on the safe use Assoc 2015 Mar;22(2):361-9. puterized provider-order-entry systems in Page 148 Pennsylvania Patient Safety Advisory Vol. 12, No. 4—December 2015 ©2015 Pennsylvania Patient Safety Authority PENNSYLVANIA PATIENT SAFETY ADVISORY This article is reprinted from the Pennsylvania Patient Safety Advisory, Vol. 12, No. 4—December 2015. The Advisory is a publication of the Pennsylvania Patient Safety Authority, produced by ECRI Institute and ISMP under contract to the Authority. Copyright 2015 by the Pennsylvania Patient Safety Authority. This publication may be reprinted and distributed without restriction, provided it is printed or distributed in its entirety and without alteration. Individual articles may be reprinted in their entirety and without alteration provided the source is clearly attributed. This publication is disseminated via e-mail. To subscribe, go to http://visitor.constantcontact.com/ d.jsp?m=1103390819542&p=oi. To see other articles or issues of the Advisory, visit our website at http://www.patientsafetyauthority.org. Click on “Patient Safety Advisories” in the left-hand menu bar. THE PENNSYLVANIA PATIENT SAFETY AUTHORITY AND ITS CONTRACTORS The Pennsylvania Patient Safety Authority is an independent state agency created by Act 13 of 2002, the Medical Care Availability and Reduction of Error (Mcare) Act. Consistent with Act 13, ECRI Institute, as contractor for the Authority, is issuing this publication to advise medical facilities of immediate changes that can be instituted to reduce Serious Events and Incidents. For more information about the Pennsylvania Patient Safety Authority, see the Authority’s website at http://www.patientsafetyauthority.org. An Independent Agency of the Commonwealth of Pennsylvania ECRI Institute, a nonprofit organization, dedicates itself to bringing the discipline of applied scientific research in healthcare to uncover the best approaches to improving patient care. As pioneers in this science for nearly 50 years, ECRI Institute marries experience and indepen- dence with the objectivity of evidence-based research. More than 5,000 healthcare organizations worldwide rely on ECRI Institute’s expertise in patient safety improvement, risk and quality management, and healthcare processes, devices, procedures and drug technology. The Institute for Safe Medication Practices (ISMP) is an independent, nonprofit organization dedicated solely to medication error prevention and safe medication use. ISMP provides recommendations for the safe use of medications to the healthcare community including healthcare professionals, government agencies, accrediting organizations, and consumers. ISMP’s efforts are built on a nonpunitive approach and systems-based solutions. Scan this code with your mobile device’s QR reader to subscribe to receive the Advisory for free.