Medication Reconciliation Challenges at Discharge from Hospital using an Electronic Medication Management System and Electronic Discharge Summaries

RESEARCH Medication Reconciliation Challenges at Discharge from Hospital using an Electronic Medication Management System and Electronic Discharge Summaries Carmen Ng, Susan A Welch, Jane Luddington, Diana Bui, Elizabeth Glasson, Katrina L Richardson ABSTRACT Background: Continuity of care can be facilitated by reconciling changes made to electronic discharge medicines profiles (edps) in the electronic medication management system (emms) and electronic discharge summaries (edss). Aim: To review edps containing discrepancies and to discern the proportion of discrepancies reconciled in the emms and edss. Method: All edps presented to the pharmacy during 4 one-week periods: September/October 2009, March 2011, January, February/March were reviewed retrospectively. edps with changes were identified and the number, type of changes and potential severity assessment were recorded. An eds was introduced in January, with discharge medicines listed in the emms populating the edss. edps identified by the pharmacist as requiring edits by the prescriber were cross checked with the final version in the emms and edss. Results: Number of edps reviewed were: 67 (2009), 150 (2011) and 382 (). Percentage of edps with changes made were: 40% (2009), 33% (2011) and 20% (). Percentage of items on edps with changes not reconciled in the emms were: 100% (2009), 81% (2011) and 31% () (p < 0.0001). Of the edps that had changes made, 59% (34) had accurate edss in compared with 11% of handwritten discharge summaries in 2009. Conclusion: While there has been a substantial improvement in the reconciliation of discharge medicines in the emms, further work is required to ensure these are imported into the edss so that an accurate list is maintained in both systems. J Pharm Pract Res 2013; 43: 25-8. INTRODUCTION Medication reconciliation is the process of obtaining, verifying and documenting an accurate list of a patient s current medicines on admission and comparing it to the admission, transfer and discharge orders to identify and resolve discrepancies. 1,2 A recent editorial highlighted some startling statistics about the challenges surrounding a patient s discharge from hospital and the importance of medication reconciliation at the point of discharge. 3 The chance of a patient continuing the same medications Carmen Ng, BPharm, GradDipClinPharm, Pharmacist, Susan A Welch, BPharm, CertClinPharm, Senior Pharmacist (Research), St Vincent s Hospital; Honorary Associate Lecturer, Faculty of Pharmacy, The University of Sydney, Jane Ludington, BPharm, DipHospPharm, MHPEd, GradCertDiabEd, CDE, Pharmacist, Diana Bui, BPharm, Pharmacist, Elizabeth Glasson, BPharm, Inpatient Dispensary Manager, St Vincent s Hospital, Katrina L Richardson, BPharm, DipHospPharm, Pharmacist emedicines Management, Information Technology Services Centre, St Vincent s and Mater Health, Sydney, New South Wales Corresponding author: Susan Welch, Pharmacy Department, St Vincent s Hospital, Darlinghurst NSW 2010, Australia. E-mail: swelch@stvincents.com.au at discharge as those on admission is less than 10%. 4 On average, 28% to 40% of a patient s medications are discontinued during hospitalisation, and 45% of medications prescribed at discharge are initiated during the hospital stay. 5,6 In another study, more than 60% of patients had three or more changes to their medications during their hospital stay. 7 This, together with the fact that errors occur at discharge and that the number of errors increases with the number of medications at discharge, makes medication reconciliation at discharge an important focal area for improving continuity of care. 8 Previous studies have shown that 48% of electronic discharge forms faxed to general practitioners (GPs) contained medication errors when compared to the pharmacy copy of the electronic discharge forms. 9 While a subsequent study reported 17% of items on electronic discharge forms had a manual change made to them by a pharmacist and/or a prescriber, 77% were not reconciled in the electronic medication management system (emms). 10 Medication reconciliation at discharge from hospital, involves preparing a complete and accurate list of medicines the patient is to continue taking and details of any changes made during the episode of care. 1,11-14 A coordinated and shared responsibility by doctors and pharmacists for medication reconciliation at discharge ensures independent validation and checks. 2,15 St Vincent s Hospital, Sydney, a tertiary referral hospital, implemented the emms, MedChart, in 2005. MedChart is used across all inpatient wards, theatres and intensive care units with the exception of the emergency department. On discharge, an electronic discharge medicines profile (edp) is generated within the emms, which serves as a take-home list and as a prescription for supply through the hospital pharmacy to ensure continuity of treatment until the patient is able to obtain further supplies outside the hospital. 16 An electronic discharge summary (eds) was introduced in January. Since the discharge medicines section of the eds is populated by the edp, a complete and accurate edp is paramount to ensure medications are accurate on discharge from hospital. This prompted a review of the medical and pharmacy business processes required to prepare an accurate edp which could be imported into the discharge medicines section of the eds. Using an edp to populate the discharge medicines section in the eds should reduce the errors associated with handwritten and typed discharge summaries, e.g. illegibility, transcription errors, unintentional omissions. 17,18 Failure to correct discrepancies in the edp has the potential to cause harm and negatively impact on continuity of care if these are imported into the eds or remain unreconciled in the eds and are referred to if the patient is readmitted to hospital. 15 Journal of Pharmacy Practice and Research Volume 43, No. 1, 2013. 25

ELECTRONIC DISCHARGE SUMMARIES The following key aspects of the medication reconciliation process were highlighted and addressed during the 7-month lead-up period of introducing the hospital s Medical (Electronic) Discharge Summary and Discharge Medications Procedure: 1. The Medication on Admission (emoa) module of the emms was adopted as the source-of-truth for documenting admission medication histories. Pharmacists were the sole health professionals trained to use the emoa module for recording medication histories as the emoa had restricted functionality and usability for doctors. The emoa provides a list available to any clinician during the patient s episode of care and at discharge. Other clinicians continue to use medical notes to record medication histories. 2. Identification of emoa showstoppers that restricted doctor buy-in, such as the lack of a comprehensive prescribing function on the emoa. Functionality enhancements were requested, such as easy one-click prescribing on the inpatient chart and prescribing plan options, to enable automated medication reconciliation across the episode of care. 3. Increased management of discharge prescriptions in the emms and reduced paper-based prescriptions. A bulletin was issued by the Director of Clinical Governance advising doctors that every patient must have an edp completed in the emms regardless of whether these medications were being supplied by the hospital pharmacy. 4. Creation of an online tool by the Information Technology Services Centre to facilitate reconciliation of the admission with the discharge medicines. This tool displays the admission and discharge medicine lists side by side, enabling visual medication reconciliation and identification of medications on admission that were withheld and needed recommencing at discharge, and to communicate any changes. 5. Medication reconciliation by pharmacists of edp revealed discrepancies that required amendments by doctors. If the pharmacist notes a discrepancy on the edp, this is discussed with the prescriber who must edit and finalise the edp in the emms. The Director of Clinical Governance supported pharmacy business requirements that discharge medicines would not be dispensed from the edp until the final reconciliation had taken place in the emms. 6. Outlining the process of manually importing the edp into the eds by doctors. For edps which required corrections in the emms by the prescriber, the process of reimporting the corrected edp into the eds was also stipulated in the Medical (Electronic) Discharge Summary and Discharge Medications Procedure. 7. A free text Details of Medication Change in the eds was created for doctors to communicate reasons for ceased or changed medicines on discharge. A limited roll out of the eds and the Medical (Electronic) Discharge Summary and Discharge Medications Procedure began in late December 2011 to a few prescribers. New junior medical officers were trained to use the eds in the third week of January and the Medical (Electronic) Discharge Summary and Discharge Medications Procedure was formally implemented in February. This study aimed to review edps containing discrepancies and to discern the proportion of discrepancies reconciled in the emms and edss. METHOD Ethics approval for this study was obtained from St Vincent s Hospital Human Research Ethics Committee. All edps presented to the hospital pharmacy during four 1-week periods: September/October 2009, March 2011, January (the limited roll out period) and February/ March (1 month after the formal launch of the eds) were reviewed retrospectively. edps with changes were identified and the number and type of changes were recorded. Changes recorded included: frequency, drug, dose, form, duration and any additional or deleted orders made to the prescription. A cross check of the original edp with the final version in the emms and the corresponding eds (in ) was performed to determine whether these changes were reconciled in the emms. Chi-square tests were conducted to assess whether the proportion of edps with changes not reconciled in the emms differed across time periods as the data were categorical. The NSW Health Severity Assessment Code (SAC) Matrix 2005 was also applied to the errors identified in to determine the potential consequences of these medication errors had they not been identified. 19 A SAC score of 1 is classified as a high-risk event requiring immediate action, whereas a SAC score of 4 is a low-risk event which can be managed with routine procedures. Each item was reviewed independently by two pharmacists, the resulting scores were averaged and rounded to the nearest whole number. RESULTS The edps reviewed over the four 1-week periods (2009 to ) that had changes made are presented in Figure 1. The complete data set is presented in Table 1. The most common changes to medications are outlined in Table 2. The percentage of items on edps with changes that were not reconciled in the emms over the four 1-week periods (2009 to ) is presented in Figure 2. In January, 68% (27/40) of edps had an eds available for review, with 67% (18/27) containing the reconciled edp. In February/March, 86% (31/36) of edps had an eds available for review, with 52% (16/31) containing the reconciled edp. Of the edps that had 27 (40%) Oct-09 Number of edps with changes made 67 49 (33%) Mar-11 150 Jan-12 Number of edps reviewed 40 (21%) 36 (19%) Feb/Mar- Figure 1. edps reviewed from 2009 to (edp = electronic discharge medicines profile). 187 195 26 Journal of Pharmacy Practice and Research Volume 43, No. 1, 2013.

Table 1. Changes made to the edp in the emms (Jan) 2011 2009 No. of edp 195 187 150 67 edp with changes Items for review Items with changes Items with changes not reconciled in emms 36 40 49 27 1584 1471 1279 510 58 (4%) 71 (5%) 73 (6%) 54 (11%) 20 (35%) 20 (28%) 59 (81%) 54 (100%) edp = electronic discharge medicines profile. emms = electronic medication management system. 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Table 2. Types of changes made to medications not reconciled on the edp in the emms Types of change (Jan) 2009 Drug changed 1 (2%) 4 (5%) 7 (7%) Omission 13 (22%) 25 (33%) 17 (17%) Needed to be deleted 11 (18%) 8 (10%) 21 (21%) Duplication 0 2 (3%) 0 Frequency total 11 (18%) 6 (8%) 26 (26%) Dose total 5 (8%) 13 (17%) 12 (12%) Duration total 7 (12%) 9 (12%) 13 (13%) Presentation/formulation 12 (20%) 10 (13%) 4 (4%) edp = electronic discharge medicines profile. emms = electronic medication management system. 100% 81% 28% 35% Oct-09 Mar-11 Jan-12 Feb/Mar- Figure 2. Percentage of items on the electronic discharge medicines profiles with changes that were not reconciled in the electronic Medication Management System (p < 0.0001). changes made, 59% (34/58) in had accurate edss compared with 11% of handwritten discharge summaries in 2009 (Table 3). The SAC scores for errors made in January and February/March were primarily 3 or 4, with an average score of 3 for each period. However, 9.5% (14/146) of errors were rated 2, indicating that potentially high-risk consequences could have occurred from these errors. Medications involved were: allopurinol, enoxaparin, hydrocortisone, prednisolone, risperidone, tacrolimus, topiramate and warfarin. Table 3. Changes made to the edps in the eds (Jan) 2011 2009 No. of eds 31 27 - - Items with changes in eds 53 46 - - Items not reconciled in eds 20 13 - - Items not reconciled in emms Changes not reconciled in eds Changes not reconciled in emms eds with all items reconciled eds with reconciled medication list Handwritten discharge summaries with reconciled medication list 12 10 - - 38% 28% - - 23% 22% - - 16 18 - - 52% 67% - - - - 19% 11% edp = electronic discharge medicines profile. eds = electronic discharge summary. emms = electronic medication management system. DISCUSSION The edps reviewed in 2009 identified that 100% of changes made to discharge medicines were not reconciled in the emms. Given the potential adverse consequences from these discrepancies, the results highlighted the need to provide education and training and to address any barriers contributing to this poor level of reconciliation in the emms. These measures were then reassessed with audits in 2011 and, which showed a substantial improvement in the percentage of edps with changes reconciled. The major findings of this study compared to 2009 were a 70-fold increase in the reconciliation of discharge medicines in the emms and a 4-fold increase in accurate medicines listed within the discharge summary. Although there was an improvement in accuracy of discharge medicines in the emms, not all of the reconciled edps were imported into the eds. This may be due to the influx of new junior medical officers learning an unfamiliar procedure. Other possible reasons may be that the eds requires the doctor to manually import the edited and reconciled edp. This may be carried out incorrectly or incompletely due to technical issues with access to the emms or when late discharge planning occurs. Despite support and mandate from the Director of Clinical Governance that pharmacy would not dispense discharge medicines from the edp until the reconciliation was complete, in practice pharmacy encountered barriers which effected the dispensary work flow and patient flow. When discrepancies were noted by the pharmacist, the medical team was contacted via telephone to determine the corrections needed on the edp, i.e. the pharmacist obtained a telephone order from the medical team. Since pharmacists are unable to make changes to the edp, dispensing discharge medicines from the hand-annotated edp commenced in the dispensary but was not released until the amended edp was provided. In some instances, when the reconciled version did not eventuate in time, the pharmacy dispensed from the phone order approved edp due to pressure from patients wanting to be discharged and the wards needing the beds for new patients. Journal of Pharmacy Practice and Research Volume 43, No. 1, 2013. 27

Our review provides a snapshot of the number of edps which are reviewed by a pharmacist over four 1-week periods. One limitation of using a snapshot view is that the sample size is dependent on the number of patients being discharged within pharmacy operating hours. In practice, not all edps are reviewed by a pharmacist, i.e. when no supply is required and after hours. Therefore, our numbers do not account for all edps available within our institution. Our results highlighted that there has been a positive improvement since 2009 in the level of reconciliation made to edps. Our results were limited by the number of eds available for review given that the eds was introduced in. Our results provide a baseline for future audits to assess our progress in improving the level of reconciliation in the emms. There were some difficulties in applying the SAC matrix to these errors. Primarily, because the SAC score is reliant on the consequences which may arise from the error as well as the frequency that the error is likely to reoccur. While there may have been an extreme potential outcome for the discrepancy identified in our study, the pharmacists took a pragmatic approach in determining the SAC score based on the available information and their clinical experience. To improve medication reconciliation processes at discharge we propose the following recommendations: 1. Maintain independent clinical pharmacy review of edps and liaison with doctors to correct discrepancies in a timely manner. This shared responsibility enhances patient care and safety by ensuring edps are complete and accurate to minimise medication-related incidents and errors. Future automation possibilities using web services which may improve communication between doctors and pharmacy should be explored, such as notifications when a discharge prescription has been finalised in the emms and is ready for clinical pharmacy review. 2. Provide ongoing education and training to doctors to improve the medication reconciliation process so that all changes made to edps are re-imported into the eds to provide accurate information to support continuity of care. 3. Develop fully automated medication reconciliation functionality in the emms. Future areas of development include enhancement of the emoa so that doctors may select a prescribing plan to continue, change, cease, withhold or suspend medications. The use of notification flags via web services and appropriate timed import of the finalised reconciled edp into the eds should be developed. This would reduce discrepancies and remove the time impost and reliance on doctors to manually import the final reconciled edp into the eds. 4. Perform regular audits/compliance reviews to monitor and report on accurate medication reconciliation for discharge medicines and discharge summaries and to identify opportunities for improvement. In conclusion, while there has been a substantial improvement in the reconciliation of discharge medicines in the emms, further work is required to ensure these are imported into the eds so that an accurate list is maintained in both systems. References 1. Duguid M. The importance of medication reconciliation for patients and practitioners. Aust Prescr ; 35: 15-19. 2. St Vincent s Hospital Management Committee. St Vincent s Hospital operational procedure: medical (electronic) discharge summary and discharge medications procedure. Darlinghurst: St Vincent s Hospital Management Committee;. 3. Thompson-Moore N, Liebl MG. Health care system vulnerabilities: understanding the root causes of patient harm. Am J Health Syst Pharm ; 69: 431-5. 4. Mansur N, Weiss A, Beloosesky Y. Relationship of in-hospital medication modifications of elderly patients to postdischarge medications, adherence, and mortality. Ann Pharmacother 2008; 42: 783-9. 5. Himmel W, Tabache M, Kochen MM. What happens to long-term medication when general practice patients are referred to hospital? Eur J Clin Pharmacol 1996; 50: 253-7. 6. Beers MH, Dang J, Hasegawa J, Tamai IY. Influence of hospitalization on drug therapy in the elderly. J Am Geriatr Soc 1989; 37: 679-83. 7. Himmel W, Kochen MM, Sorns U, Hummers-Pradier E. Drug changes at the interface between primary and secondary care. Int J Clin Pharmacol Ther 2004; 42: 103-9. 8. Cornu P, Steurbaut S, Leysen T, De Baere E, Ligneel C, Mets T, et al. Discrepancies in medication information for the primary care physician and the geriatric patient at discharge. Ann Pharmacother ; 46: 983-90. 9. Ward-Panckhurst L, Bula N. Assessing the accuracy of electronic discharge summaries [abstract]. Proceedings of the 35th National Conference of The Society of Hospital Pharmacists of Australia; 2009 Nov 5-8; Perth, Western Australia. Perth: The Society; 2009. p.152. 10. Patel B, Liddy S. An audit of the accuracy of discharge medicines information using an electronic medication management system [abstract]. Proceedings of the 35th National Conference of The Society of Hospital Pharmacists of Australia; 2009 Nov 5-8; Perth, Western Australia. Perth: The Society; 2009. p.214. 11. Herrero-Herrero JI, Garcia-Aparicio J. Medication discrepancies at discharge from an internal medicine service. Eur J Intern Med 2011; 22: 43-8. 12. Agrawal A, Wu W, Khachewatsky I. Evaluation of an electronic medication reconciliation system in inpatient setting in an acute care hospital. Stud Health Technol Inform 2007; 129: 1027-31. 13. Australian Pharmaceutical Advisory Council. Guiding principles to achieve continuity in medication management. Canberra: Australian Pharmaceutical Advisory Council; 2005. 14. Australian Commission on Safety and Quality in Health Care. Medication reconciliation. Canberra: Australian Commission on Safety and Quality in Health Care; 2011. 15. Callen J, McIntosh J, Li J. Accuracy of medication documentation in hospital discharge summaries: a retrospective analysis of medication transcription errors in manual and electronic discharge summaries. Int J Med Inform 2010; 79: 58-64. 16. NSW Ministry of Health. Medication handling in NSW public hospitals. Sydney: NSW Ministry of Health; 2007. 17. Motamedi SM, Posadas-Calleja J, Straus S, Bates DW, Lorenzetti DL, Baylis B, et al. The efficacy of computer-enabled discharge communication interventions: a systematic review. BMJ Qual Saf 2011; 20: 403-15. 18. Murphy EM, Oxencis CJ, Klauck JA, Meyer DA, Zimmerman JM. Medication reconciliation at an academic medical centre: implementation of a comprehensive program from admission to discharge. Am J Health Syst Pharm 2009; 66: 2126-31. 19. NSW Health. Severity assessment code (SAC) matrix 2005. Sydney: NSW Health; 2005. Available from <www0.health.nsw.gov.au/pubs/2005/pdf/sac_ matrix.pdf> Received: 11 September Revisions requested after external review: 29 November Revised version received: 4 March 2013 Accepted: 13 March 2013 This activity has been accredited for 1 hour of Group 2 CPD activity (or 2 CPD credits) suitable for inclusion in an individual pharmacist s CPD plan, if online questions are completed and submitted. No: S2013/20 Competing interests: None declared. 28 Journal of Pharmacy Practice and Research Volume 43, No. 1, 2013.

Copyright of Journal of Pharmacy Practice & Research is the property of Society of Hospital Pharmacists of Australia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.