Safe Medication Reconciliation: An Intervention to Improve Residents Medication Reconciliation Skills

Safe Medication Reconciliation: An Intervention to Improve Residents Medication Reconciliation Skills Cherinne Arundel, MD Jessica Logan, MD Ribka Ayana, MD Jacqueline Gannuscio, DNP, ACNP, AACC Jennifer Kerns, MD, MSHS Rebecca Swenson, MD ABSTRACT Background Medication errors during hospitalization are a major patient safety concern. Medication reconciliation is an effective tool to reduce medication errors, yet internal medicine residents rarely receive formal education on the process. Objective We assessed if an educational intervention on quality improvement principles and effective medication reconciliation for internal medicine residents will lead to fewer medication discrepancies and more accurate discharge medication lists. Methods From July 2012 to May 2013, internal medicine residents from 3 academic institutions who were rotating at the Washington DC VA Medical Center received twice-monthly interactive educational sessions on medication reconciliation, using both institutional summary metrics and data from their own discharges. Sessions were led by a faculty member or chief resident. Accuracy of discharge instructions for residents in the intervention group was compared to the accuracy of discharge data from June 2012 for a group of residents who did not receive the intervention. We used v 2 analysis to assess for differences. Results The number of duplicate medications (23% versus 12%, P ¼.01); extraneous medications (14% versus 6%, P ¼.014); medications sorted by disease or indication (25% versus 77%, P,.001); and the number of discrepancies in discharge summaries (34% versus 11%, P,.001) statistically improved. No difference in the number of omissions was found between the 2 groups (17% versus 15%, P ¼.62). Conclusions An educational intervention targeting internal medicine residents can be implemented with reasonable staff and time costs, and is effective in reducing the number of medication discrepancies at discharge. Introduction Medication errors at discharge can lead to adverse outcomes, hospital readmissions, and utilization of additional health care services. 1 8 Medication reconciliation is a valuable tool to reduce these errors. 1,3,6,9 Medical residents are primarily responsible for medication reconciliation, yet rarely receive formal education on the process. To our knowledge, 1 educational intervention and 1 research study were the only previously published studies involving residents, but these studies relied heavily on pharmacy support and focused on high-risk patients. 10,11 We hypothesized that an educational intervention for residents on medication reconciliation and process improvement would reduce the number of medication discrepancies at discharge. Methods Setting and Participants The Washington DC Veterans Affairs Medical Center (DCVAMC) is an academic medical center with 175 DOI: http://dx.doi.org/10.4300/jgme-d-14-00565.1 Editor s Note: The online version of this article contains the pocket card discharge instructions guide and a self-evaluation rubric. inpatient beds and approximately 2400 yearly admissions. Medicine residents affiliated with 3 local universities (Georgetown University, George Washington University, and the Uniformed Services University of the Health Sciences) rotate monthly through our inpatient wards. These residents have no formal baseline instruction on medication reconciliation at their home institutions. Educational Intervention From July 2012 to May 2013, a monthly 2-part intervention focusing on the fundamental principles of medication reconciliation and quality improvement was held during resident teaching conferences. The first session focused on the importance of proper medication reconciliation. A 20-minute didactic session reviewed the literature on medication safety and our own institutional data on medication errors. The residents then divided into their respective teams and practiced the principles of proper medication reconciliation on a simulated patient using a pocket card and a discharge instructions guide (provided as online supplemental material). Residents selected their most recent discharge instructions and assessed their own performance using a self-evaluation rubric (provided as online supplemental material), which Journal of Graduate Medical Education, September 2015 407

FIGURE Medication Reconciliation Rubric 1. Extraneous medications (Y or N) a. Identify extraneous medications, including medications administered in the ER (remain listed as outpatient medications), 1-time medications, completed outpatient antibiotic (which remain active for 1 month despite completion) 2. Duplicate medications (Y or N) a. Any medication listed in duplicate, whether the same dose or different dose (ie, ASA 81 mg and ASA 325 mg), including non-va medications 3. Omitted medications (Y or N) a. Any medications intended to be prescribed at discharge as inferred via final progress notes 4. Discrepancy between medications listed in discharge medication list and discharge summary (Y or N) a. Yes, if lists do not match EXACTLY 5. Absence of pharmacy medication reconciliation (Y or N) a. Yes, if there is no separate note from the pharmacy at the time of discharge 6. Medications sorted by disease/indication (Y or N) a. Yes, if ALL medications are NOT sorted by disease or indication 7. Number of discharge medications a. Numeric count of discharge medications (EXCLUDE insulin supplies and creams/lotions; INCLUDE eye drops/eye medications) 8. Number of admissions the year of review (from the date 1 year prior to the date of review) a. Numeric count of number of admissions during year of review Abbreviations: ER, emergency room; ASA, aspirin; N, no; VA, Veterans Affairs; Y, yes. was developed by the authors based on a modified Medication Discrepancy Tool 12 and prior studies that categorized medication discrepancies. 13 15 Ten days after the first session, authors reviewed the 5 most recent discharge instructions from each team by using a medical reconciliation rubric (FIGURE) and provided a report card to the residents. Session 2, which occurred 3 weeks after the first session, introduced quality improvement principles, such as creating and utilizing aim statements, process maps, outcome measures, and the Plan-Do-Study-Act method of improvement. 16 The session ended with a group discussion on future ideas for continuous process improvement. This resulted in small changes to the intervention, such as scheduling the first session at the beginning of the rotation rather than after the first week. Other suggestions for change, such as hiring more clinical pharmacists and information What was known and gap Medication errors for inpatients pose a significant patient safety threat. What is new An education intervention with feedback to improve medication reconciliation practice, which can be implemented with reasonable staff and time costs. Limitations Single site, single specialty study limits generalizability. Bottom line An intervention improved internal medicine residents discharge instructions and reduced medication discrepancies. technology modifications, were not implemented due to lack of resources. Outcomes The overall impact of the educational intervention was assessed by a preintervention and postintervention chart review. We estimated that 300 total discharge medication instructions were required to attain a power of 80%, assuming an effect size of 20% to 30% (based on medication discrepancies revealed by preliminary chart review at the DCVAMC). A sample of 150 preintervention charts (25 from each of the 6 ward teams, with half of those from experienced interns and half from new interns) was included in June 2012. In January 2013, a total of 150 postintervention charts were selected (25 from each of the 6 teams who had received the intervention). Charts were excluded if they were used as examples in the teaching sessions and in the event of death, observation status, or discharge to another care facility. The specific outcomes assessed included the number of duplicate, extraneous, and omitted medications, as well as grouping medications by indication or disease and whether they matched the discharge summary (FIGURE). Medication omissions were ascertained by review of the assessment and plan portion of the last progress note to determine intended medications at discharge. Outpatient medications at the time of admission and the final progress note served as the gold standard for comparison. Disagreements were reconciled via consensus by 2 investigators using the rubric (FIGURE). The authors collected the mean number of prescribed discharge medications per patient, the mean number of admissions within 1 year per patient, and the percentage of patients who received counseling from a clinical pharmacist. 408 Journal of Graduate Medical Education, September 2015

TABLE Number of Discharge Medication Discrepancies Preintervention and Postintervention a Preintervention, No. (%) Postintervention, No. (%) P Duplicate medications 35 (23) 18 (12).01 Extraneous medications 22 (14) 9 (6).01 Medications by indication or disease 37 (25) 116 (77).001 Medication omissions 26 (17) 23 (15).62 Discrepancy with discharge summary 51 (34) 16 (11).001 a Extraneous medications include medications administered in the emergency room (remain listed as outpatient medications), 1-time medications, and completed outpatient medications (which remain active for 1 month despite completion). The study was approved by the DCVAMC Institutional Review Board. Analysis Chi-square analysis was used to compare differences in outcomes before and after the intervention. The number of prescribed medications per patient, the number of admissions for each patient within 1 year, and the percentage of patients who received bedside counseling about their discharge medications from a clinical pharmacist were compared using a Student t test, with P.05 considered statistically significant. Statistical analysis was performed using SAS software (SAS Institute Inc). Results Eighteen residents per month received the educational intervention. On average, residents rotated at the DCVAMC from 1 to 4 times per year, and some residents received the intervention more than once. The total number of medicine discharges was 326 in June 2012 and 427 in January 2013. The average number of yearly admissions and the number of discharge medications per patient were not statistically significant between the preintervention and postintervention groups (2.37 versus 2.54, P ¼.80; 10.25 versus 10.11, P ¼.58). Similarly, a small percentage of patients received pharmacist counseling in each group (10% versus 10.7%, P ¼.85). In our sample of 300 patients, the mean number of discharge medications was 10.2. Between the pre- and postintervention groups, the number of extraneous medications (14% versus 6%, P ¼.014), duplicate medications (23% versus 12%, P ¼.01), and discrepancies with the discharge summary (34% versus 11%, P.001) were reduced, while medications sorted by indication (25% versus 77%, P.001) were increased. There was no difference in medication omissions (17% versus 15%, P ¼.62; TABLE). Discussion Our study demonstrates that a targeted educational intervention coupled with experiential learning and feedback may reduce medication discrepancies at discharge. Notably, the largest change was in medication sorting and matching the medications in the discharge summary. There was no significant difference in the number of medication omissions. The few published studies on medication reconciliation and residents relied heavily on pharmacy support, focused on high-risk patients, and only involved residents indirectly. 10,11 Our intervention primarily focused on residents, had little pharmacy support, and was not limited to high-risk patients, but similarly resulted in a reduction in medication discrepancies. We speculate that medication sorting allowed for the most convenient way to visualize, organize, and assess the accuracy of the medication list. It is likely that once residents felt confident in the medication list, it was also used in the discharge summary. One possible explanation for the lack of difference in medication omissions is that the reviewers, without full knowledge of the clinical scenario and the possibility of incomplete documentation, were unable to fully discern which medications were indicated on discharge. This measure may also be reflective of residents clinical skills, which was not a target of our intervention. The educational intervention required 3 to 4 hours of faculty time each month in order to take part in teaching sessions and perform a brief assessment of the residents performance using chart review. In regard to the residents, 2 hours monthly of protected academic time was required. Chart review for research purposes was time consuming, with 1 chart taking 5 to 7 minutes to review. However, for practical application of this intervention, a random sample of charts may be adequate and less time consuming to assess efficacy. No additional supplies, pharmacy support, or financial expenditure were required. Overall, the intervention was widely accepted by both faculty and residents Journal of Graduate Medical Education, September 2015 409

and could gain support at other teaching institutions. Sustainability is being addressed by incorporating a version of the intervention into a larger project on transitions of care. Our study has several limitations. Ideally, each intern would have served as his or her own control; however, because the intervention was executed within a few days of arrival, each intern s preintervention number of discharges would have been inadequate to power the study. Also, choosing a mix of experienced and new interns for the June cohort should have theoretically minimized any difference in clinical experience, compared with the January cohort. Medication reconciliation is a new skill not taught at other institutions; therefore, maturation should not have a large impact. Three of 18 residents in the postintervention data analysis group had received the intervention before, which could have had a minimal positive impact. As highlighted in a systematic review by Kwan et al, 17 unintended medication discrepancies are common, yet these discrepancies may not translate to adverse patient outcomes. In our study, medication discrepancies were not classified into discrepancies with a potential for harm or clinically significant discrepancies. Medication reconciliation is complex, and while we had fewer medication discrepancies at discharge, we did not evaluate patients understanding of their medications or adverse clinical outcomes. The next step could be to translate our intervention into clinically relevant outcomes, such as readmission rates, adverse medication events, and patient satisfaction scores. We did not assess the impact of residents participation in the quality improvement aspect of this study, which could be done in the future, including assessing the effect on resident performance on the relevant Accreditation Council for Graduate Medical Education milestones. Longterm retention of this new knowledge and skill set could also be assessed in future studies. Conclusion An educational intervention with performance feedback targeting internal medicine residents on medication reconciliation improves the accuracy of discharge instructions and reduces medication discrepancies. Such an intervention can be implemented with reasonable staff and time costs. References 1. Cornish PL, Knowles SR, Marchesano R, Tam V, Shadowitz S, Juurlink DN, et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005;165(4):424 429. 2. Forster AJ, Clark HD, Menard A, Dupuis N, Chernish R, Chandok N, et al. Adverse events among medical patients after discharge from hospital. CMAJ. 2004;170(3):345 349. 3. Gleason KM, Groszek JM, Sullivan C, Rooney D, Barnard C, Noskin GA. Reconciliation of discrepancies in medication histories and admission orders of newly hospitalized patients. Am J Health Syst Pharm. 2004;61(16):1689 1695. 4. Koehler BE, Richter KM, Youngblood L, Cohen BA, Prengler ID, Cheng D, et al. Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle. J Hosp Med. 2009;4(4):211 218. 5. Jack BW, Chetty VK, Anthony D, Greenwald JL, Sanchez GM, Johnson AE, et al. A reengineered hospital discharge program to decrease rehospitalization: a randomized trial. Ann Intern Med. 2009;150(3):178 187. 6. Pippins JR, Gandhi TK, Hamann C, Ndumele CD, Labonville SA, Diedrichsen EK, et al. Classifying and predicting errors of inpatient medication reconciliation. J Gen Intern Med. 2008;23(9):1414 1422. 7. Tam VC, Knowles SR, Cornish PL, Fine N, Marchesano R, Etchells EE. Frequency, type and clinical importance of medication history errors at admission to hospital: a systematic review. CMAJ. 2005;173(5):510 515. 8. Wong JD, Bajcar JM, Wong GG, Alibhai SM, Huh JH, Cesta A, et al. Medication reconciliation at hospital discharge: evaluating discrepancies. Ann Pharmacother. 2008;42(10):1373 1379. 9. Mueller SK, Sponsler KC, Kripalani S, Schnipper JL. Hospital-based medication reconciliation practices: a systematic review. Arch Intern Med. 2012;172(14):1057 1069. 10. Chan AH, Garratt E, Lawrence B, Turnbull N, Pratapsingh P, Black PN. Effect of education on the recording of medicines on admission to hospital. J Gen Intern Med. 2010;25(6):537 542. 11. Varkey P, Cunningham J, O Meara J, Bonacci R, Desai N, Sheeler R. Multidisciplinary approach to inpatient medication reconciliation in an academic setting. Am J Health Syst Pharm. 2007;64(8):850 854. 12. Coleman E. Medication discrepancy tool. The Care Transitions Program. http://www.caretransitions.org/ mdt_main.asp. Accessed May 18, 2015. 13. Azzi M, Constantino M, Pont L, Mcgill M, Twigg S, Krass I. Medication safety: an audit of medication discrepancies in transferring type 2 diabetes mellitus (T2DM) patients from Australian primary care to tertiary ambulatory care. Int J Qual Health Care. 2014;26(4):397 403. 410 Journal of Graduate Medical Education, September 2015

14. Orrico KB. Sources and types of discrepancies between electronic medical records and actual outpatient medication use. J Manag Care Pharm. 2008;14(7):626 631. 15. Trompeter JM, McMillan AN, Rager ML, Fox JR. Medication discrepancies during transitions of care: a comparison study [published online ahead of print January 13, 2014]. J Healthc Qual. doi:10.1111/jhq. 12061. 16. Langley GL, Moen R, Nolan KM, Nolan TW, Norman CL, Provost LP. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. 2nd ed. San Francisco, CA: Jossey Bass; 2009. 17. Kwan JL, Lo L, Sampson M, Shojania KG. Medication reconciliation during transitions of care as a patient safety strategy: a systematic review. Ann Intern Med. 2013;158(5, pt 2):397 403. Cherinne Arundel, MD, is a Hospitalist, Washington DC Veterans Affairs Medical Center, and Assistant Professor of Medicine, George Washington University, Georgetown University Hospital, and Uniformed Services University of the Health Sciences; Jessica Logan, MD, is a Hospitalist, Washington DC Veterans Affairs Medical Center, and Assistant Professor of Medicine, George Washington University and Uniformed Services University of the Health Sciences; Ribka Ayana, MD, is a Hospitalist, Washington DC Veterans Affairs Medical Center, and Assistant Professor of Medicine, George Washington University and Uniformed Services University of the Health Sciences; Jacqueline Gannuscio, DNP, ACNP, AACC, is Nurse Practitioner, Heart Failure Program, Washington DC Veterans Affairs Medical Center; Jennifer Kerns, MD, MSHS, is a Hospitalist and Codirector of Bariatric Surgery, Washington DC Veterans Affairs Medical Center, and Assistant Professor of Medicine, George Washington University and Uniformed Services University of the Health Sciences; and Rebecca Swenson, MD, is a Hospitalist, Dartmouth Hitchcock Medical Center, and Assistant Professor of Medicine, Geisel School of Medicine at Dartmouth. Funding: The authors report no external funding source for this study. Conflict of interest: The authors declare they have no competing interests. Results from this intervention were presented as a poster at the Society for General Internal Medicine National Meeting in San Diego, California, April 23 26, 2014. The authors would like to thank Samantha McIntosh, MD, and Katherine Chretien, MD, for assisting with study design and editing. Corresponding author: Cherinne Arundel, MD, Veterans Affairs Medical Center, Department of Medicine, Hospitalist Division, 50 Irving Street NW, Washington, DC 20422, 202.745.8000, ext 57067, cherinne.arundel@va.gov Received October 6, 2014; revisions received March 16, 2015, and April 19, 2015; accepted April 28, 2015. Journal of Graduate Medical Education, September 2015 411