Research in Gerontological Nursing

Empirical Research 

Medication Problems Occurring at Hospital Discharge Among Older Adults with Heart Failure

Janice B. Foust, PhD, RN; Mary D. Naylor, PhD, RN; M. Brian Bixby, MSN, CRNP; Sarah J. Ratcliffe, PhD

Abstract

Medication reconciliation problems are common among older adults at hospital discharge and lead to adverse events. The purpose of this study was to examine the rates and types of medication reconciliation problems among older adults hospitalized for acute episodes of heart failure who were discharged home. This secondary analysis of data generated from a transitional care intervention included 198 hospital discharge medical records, representing 162 patients. A retrospective chart review comparing medication lists between hospital discharge summaries and patient discharge instructions was completed to identify medication reconciliation problems. Most hospital discharges (71.2%) had at least one type of reconciliation problem and frequently involved a high-risk medication (76.6%). Discrepancies were the most common problem (58.9%), followed by incomplete discharge summaries (52.5%) and partial patient discharge instructions (48.9%). More attention needs to be given to the quality of discharge instructions, and the problem of vague phrases (e.g., “take as directed”) can be addressed by adding it to “do not use” lists to promote safer transitions in care.

Abstract

Medication reconciliation problems are common among older adults at hospital discharge and lead to adverse events. The purpose of this study was to examine the rates and types of medication reconciliation problems among older adults hospitalized for acute episodes of heart failure who were discharged home. This secondary analysis of data generated from a transitional care intervention included 198 hospital discharge medical records, representing 162 patients. A retrospective chart review comparing medication lists between hospital discharge summaries and patient discharge instructions was completed to identify medication reconciliation problems. Most hospital discharges (71.2%) had at least one type of reconciliation problem and frequently involved a high-risk medication (76.6%). Discrepancies were the most common problem (58.9%), followed by incomplete discharge summaries (52.5%) and partial patient discharge instructions (48.9%). More attention needs to be given to the quality of discharge instructions, and the problem of vague phrases (e.g., “take as directed”) can be addressed by adding it to “do not use” lists to promote safer transitions in care.

Hospital discharge is a high-risk transition, and reconciliation of medication changes is a daunting and complex task (Greenwald, Denham, & Jack, 2007). Medication discrepancies are common at hospital discharge (39.6% to 70.7%) (Nickerson, MacKinnon, Roberts, & Saulnier, 2005; Schnipper et al., 2006; Vira, Colquhoun, & Etchells, 2006; Wong et al., 2008) and during the post-hospital transition (14.1% to 94%) (Coleman, Smith, Raha, & Min, 2005; Corbett, Setter, Daratha, Neumiller, & Wood, 2010). Among post-hospital adverse events, medications were the most common problem (66% to 72%), and nearly all post-hospital adverse drug events (ADEs) involved a new medication or dosage change (Forster et al., 2004; Forster, Murff, Peterson, Gandhi, & Bates, 2003, 2005). Additionally, inadequate time for proper discharge teaching and an abrupt shift in responsibility from professionals to patients are other potential factors contributing to medication errors at the time of discharge (Foust, Naylor, Boling, & Cappuzzo, 2005).

Background of the Problem

Medication changes during the transition from hospital to home are the norm. Unroe et al. (2010) reported that nearly all hospitalized patients (96%) experienced at least one medication change when comparing their discharge and prior home regimens. Pippins et al. (2008) reported that more than six medication changes from hospital admission to discharge was a risk factor for potential ADEs. Others have found medication changes were common (64%) among frail older adults transferred between hospitals and nursing homes (Boockvar et al., 2004). More specifically, Boockvar et al. (2004) reported an average 14-day span between medication changes made in the hospital to the ADE occurrence in the nursing home that highlights the risk of three coinciding events (i.e., medication change, delayed effect, and move to another setting). The Joint Commission (TJC) introduced medication reconciliation as a National Patient Safety Goal (NPSG) to address discrepancies occurring when medication regimens change as patients move through the health care system. In 2009, this NPSG was temporarily suspended as a scored item during accreditation surveys, and recently, TJC released a revised process for reconciling medication information with an effective date of July 2011 (TJC, 2010).

Incomplete medication information and inadequate patient education were identified as two critical barriers to providing continuity of medication information at hospital discharge (Bayley, Savitz, Rodriguez, Gillanders, & Stoner, 2005). These authors described inadequate patient education as a “high-probability, high-severity failure” and identified causes such as limited time to prepare patients or incomplete orders (e.g., “resume home medications”) (p. 93). Health literacy, a factor of high relevance to patient education, was found to be the strongest predictor of outpatients’ knowledge of medication (Marks, Schectman, Groninger, & Plews-Ogan, 2010). In addition, people were less knowledgeable about their medications if they were men, older than 70, had 8 years of education or less, or had someone assisting them with medication administration (Marks et al., 2010). These studies highlight the importance of comprehensive discharge instructions and the need to address individuals’ learning needs to improve their understanding of medication regimens.

Unfortunately, medication changes are often not clearly communicated to patients and their family caregivers, who must manage their medications once patients return to their homes. Only 44% of patients who were discharged home on new or altered medications received written alerts of these changes, and even fewer (12%) were given written information indicating they should stop taking a discontinued medication they took prior to being hospitalized (Unroe et al., 2010). Although the reasons are not stated, a majority of patients (71%) interviewed within a week of hospital discharge reported taking their medications differently than recorded at discharge (Schnipper et al., 2006). Providing some context for such differences, home health care nurses who reconciled medication lists identified intentional and non-intentional nonadherence as two of the most common patient-level discrepancies after patients were discharged from the hospital (Corbett et al., 2010). Non-intentional nonadherence was identified for 30% of the patients because they did not understand how to take their medications. These studies underscore the importance of patient medication education throughout hospitalization, at discharge, and during the post-hospital period to promote better medication management.

Greater attention to discharge medications and better communication with patients were two strategies identified in a report of the Medicare Payment Advisory Commission (MedPAC, 2007) to potentially reduce hospital readmissions. The diagnosis of heart failure was used in this report as an example of a patient population that would likely benefit from such efforts. More recently, researchers reported patients with heart failure were the most likely to be readmitted within 30 days (26.6%); this patient group represented 7.6% of all rehospitalizations during a 15-month period between 2003–2004 (Jencks, Williams, & Coleman, 2009).

Caring for patients with heart failure has become more complex over the past decades. Hector et al. (2010) reported that stays for patients hospitalized with heart failure decreased from 8.8 days in 1993 to 6.3 days in 2006. During this same period, the authors reported a 4.2% decrease in hospital mortality and a nearly 3% increase in 30-day readmission rate. In addition, coexisting conditions such as hypertension (52.2%) or diabetes (38.4%) were relatively common, posing additional clinical challenges (Hector et al., 2010). Other researchers described a similar trend of increasing complexity specific to heart failure medication regimens (Masoudi et al., 2005). In particular, patients with a primary discharge diagnosis of heart failure took an average of 6.8 medications at a cost of approximately $3,100 annually in 1998–1999. Twenty-seven months later (2000–2001), a second cohort of patients took an average of 7.5 medications costing approximately $3,800 a year. These studies exemplify three serious challenges to effectively preparing patients hospitalized with heart failure to return home to safely manage their medications: (a) shorter hospital stays, (b) increasingly complex and costly medication regimens, and (c) higher rates of hospital readmissions.

Study Purpose

To the best of our knowledge, this is the first reported study that examined the medication reconciliation problems and the adequacy of written discharge instructions among heart failure patients who are at the greatest risk for rehospitalization. The purpose of this study was to describe discharge medication reconciliation problems during critical transitions in health, with the specific aim to describe discrepancies found between hospital discharge summaries and patient discharge instructions among older adults with heart failure who were discharged home. Specific medications linked to medication reconciliation problems were noted to further demonstrate the potential impact on patient outcomes.

Method

Study Sites and Population

This study was a secondary analysis of medical records collected during a randomized controlled trial (RCT) testing a comprehensive transitional care intervention for older adults with heart failure (Naylor et al., 2004). Patients were eligible for study participation if they were alert and oriented, spoke English, could be reached by telephone, and resided within 30 miles of study sites. Patients with end-stage renal disease were excluded. The original study took place between February 1997 and January 2001. Patients were enrolled from one of six academic or community hospitals in the Philadelphia area. An advanced practice nurse (APN) led the transitional care intervention, which involved collaboration with health care professionals across care sites. It is important to note that although the APN protocol emphasized continuity of medication information, the APNs were not directly responsible for preparing the final discharge instructions or dictating the discharge summaries that were the basis of this study. The hospital staff completed these documents as expected within their organization. Patient consent was obtained during the original RCT and included collection of medical records from their index hospitalization and subsequent rehospitalizations as part of their consent to participate. No additional data were collected for this study. Patient medical records for the first 3 months of follow up (i.e., index and rehospitalization) were blinded with removal of all personal health information—except for age—before they were copied for use in this secondary analysis. The University of Pennsylvania’s Institutional Review Board approved the study.

Data Collection

Each medical record (N = 324) was reviewed to verify that the patient was discharged home and that legible discharge documents were available. Discharge records were excluded if the patient was transferred to another facility (n = 23), died during hospitalization (n = 6), if records were not available (n = 2), or if either the hospital discharge summary or patient discharge instructions were not available (n = 74). Two investigators (J.B.F., M.B.B.) verified illegibility of medical records (n = 21) before exclusion. A total of 126 discharge records were excluded (38.9%), representing 62 unique patients.

Each of the six hospitals had site-specific patient discharge instructions that were sent home with the patient. These forms varied by the number of lines allocated for medications (0 to 11) and whether or not sections were labeled (e.g., dosage, frequency). At the time of the study, all discharge instructions were completed by hospital professionals via handwriting. One patient instruction form had checkboxes with general statements such as “resume medications taken before admission” or “no medication necessary.” All the patient instruction forms remained the same throughout the study period. All hospital discharge summaries were typewritten.

To address interrater agreement, more than 10% of hospital discharge records (n = 28) were randomly selected and coded by an abstractor who was not otherwise involved in the study. The abstractor was trained by the principal investigator (J.B.F.) in the chart abstraction procedure, sources of data, operational definitions, and use of standardized forms (Gilbert, Lowenstein, Koziol-McLain, Barta, & Steiner, 1996). To determine level of agreement, two investigators (J.B.F., M.B.B.) met with the abstractor and compared codes in each of the selected medical records. Initial complete agreement was achieved for 62% of the records. The 11 remaining records had at least one disagreement requiring review of the medical record and discussion. Consensus was easily reached on all disagreements by reviewing the medical record together. Most of the disagreements involved clarifying codes used with vague terms. The principal investigator subsequently reviewed the database for consistency of codes.

Definitions of Study Variables

The hospital discharge summary was compared with the patient discharge instructions for consistency of medication information (i.e., medication name, dosage, and frequency) in order to identify the types of medication reconciliation problems. Data were collected using a standardized chart abstraction tool. A codebook of definitions (Table 1) was developed to account for different types of conflicting information, missing information, or the use of vague terms (e.g., “resume prior medications”) and on which of the two documents the problem occurred. Specifically, medication discrepancies were identified when both the discharge summary and patient instructions listed medications, but at least one medication was omitted and/or a dosage or frequency was inconsistent or dosage units were not comparable. The remaining two categories—partial patient discharge instructions or incomplete hospital discharge summary—captured missing information (i.e., dosage or frequency) or the use of vague terms. For example, the discharge instructions may have indicated to “resume prior medications” or in other instances, a specific medication (e.g., warfarin [Coumadin®]) was ordered to be taken “as directed.” The medications involved with all reconciliation problems were documented. Medications were identified as “high risk” if they fell within one of the six medication classifications associated with a majority (87%) of post-hospital ADEs (Forster et al., 2005). The six drug classifications were: (a) cardiovascular medications; (b) anticoagulant agents; (c) anti-infective agents; (d) analgesic agents, including narcotic agents; (e) anti-epileptic medications, and (f) corticosteroid drugs. Dr. Forster reviewed and verified the consistency of medication classification used in this study with his work (A.J. Forster, personal communication, September 19, 2006).

Definition of Medication Reconciliation Problems

Table 1: Definition of Medication Reconciliation Problems

Statistical Analyses

Analyses were conducted using SPSS version 15.0. Patient demographic characteristics were summarized; comparisons between those included and excluded in the study were made using chi-square or Mann-Whitney U-tests, as appropriate. The primary analyses were conducted on hospital discharges (N = 198), not patients (n = 162). The lack of standardized systems and the frequency of medication changes require that health care professionals complete medication reconciliation during all transitions. Further, different professionals could have completed the forms from the same patient at different discharges; thus, any residual association between one visit and the next for the same patient would be small. Therefore, discharges were deemed to be independent units for analysis, as discharge reconciliation problems from one discharge should be independent from problems on a subsequent discharge. Types of problems were summarized via frequencies and percentages. Chi-square, Fisher’s exact, or Mann-Whitney U-tests were used to compare discharges with and without problems, and between those involving high-risk medications (yes or no).

Results

After reviewing the charts for eligibility, 162 patients accounted for the hospital discharges included in this study (N = 198). As noted earlier, some patients were hospitalized more than once during the 3-month post-discharge follow-up period. Of the discharges included, 8 people had three hospital discharges, 20 were discharged home twice, and the remainder (n = 134) were discharged once. The patient characteristics are described in Table 2. Patients excluded from the study (n = 62) did not significantly differ from those included.

Baseline Sociodemographic and Health Characteristics of the Sample (N = 162)

Table 2: Baseline Sociodemographic and Health Characteristics of the Sample (N = 162)

Primary Findings

A majority of hospital discharges (71.2%) had at least one type of medication reconciliation problem, with an average of 1.3 problems per discharge (Table 3). The most common reconciliation problem was medication discrepancies (58.9%), followed by incomplete hospital discharge summary (52.5%) and partial patient discharge instructions (48.9%). Specifically, inconsistent dosages and/or frequencies were the most common type of medication discrepancy (62.7%), including five instances when dosage units were not consistent (e.g., tablets versus milligrams). Other subcategories are described in Table 3. For those patients who received partial discharge instructions, three types of problems existed: a missing dosage or frequency (49.3%), use of a vague phrase to refer to the dosage or frequency (40.6%), or a general statement about continuing medications taken prior to hospitalization or “as directed” (17.4%). In two cases, the discharge instructions did not list any medications (2.9%). The number of medications on hospital admission did not differ between those with discharge reconciliation problems and those without a problem (mean = 7 versus 6.6 medications, p = 0.694). Some discharge records had more than one reconciliation problem (Table 4).

Medication Reconciliation Problems

Table 3: Medication Reconciliation Problems

Number of Reconciliation Problems Per Discharge

Table 4: Number of Reconciliation Problems Per Discharge

High-Risk Drugs

Of the 141 discharges with a problem, 76.6% (n = 108) involved at least one high-risk medication. This is likely an underestimation, since information about high-risk drugs was not complete due to 26 of the discharges having a discharge document that did not list any medications, used vague directions about medications (e.g., “resume prior medications”), or a page was missing from the discharge summary. Of those with known issues related to high-risk medications, the average number involved was 1.6 (range = 1 to 7).

Of the 69 discharges with partial instructions, 75.3% (n = 53) involved at least one high-risk medication. High-risk medications were involved in 76.5% (n = 26) of instances when instructions were missing a dosage or frequency. Nitroglycerin was the cardiovascular drug most frequently listed without a dosage or frequency (n = 7). Of note, other discharge instructions were missing a dosage or frequency for respiratory inhalers (n = 5), insulin (n = 3), or oral potassium supplements (n = 1). A majority (71.4%, n = 20) of the discharge instructions using a vague phrase to refer to dosage or frequency (e.g., “take as directed”) involved high-risk medications. In these instances, warfarin was the most common medication (n = 10) followed by furosemide (Lasix®, n = 4). Other similar, vague phrases were used with the instructions to take hypoglycemic agents (i.e., insulin or oral medications, n = 2) or oral potassium chloride supplements (n = 2). High-risk medications were involved in all or nearly all instances when patient instructions included a general statement to resume prior medication (91.7%) or did not list any medications at all when compared with the hospital discharge summary (100%).

Discussion

Our study of hospital discharge medication reconciliation found that a majority of older adults with heart failure were discharged home with inconsistent or incomplete medication information. A majority of these problems involved high-risk medications previously associated with post-hospital ADEs (Forster et al., 2005). Medication discrepancies between hospital discharge summaries and patient discharge instructions were the most common problem, followed by problems of incomplete hospital discharge summaries and partial patient discharge instructions. On average, each hospital discharge had 1.3 reconciliation problems, which is within the range of other studies reporting mean discrepancies of 1.2 to 3.26 per patient at discharge or during the post-hospital period (Coleman et al., 2005; Corbett et al., 2010; Vira et al., 2006).

The overall rates of medication reconciliation problems reported here are comparable to other hospital discharge reconciliation studies, although other studies used different labels to describe inconsistencies between medication lists. We found more medication discrepancies (58.9%) than other research teams who reported 41.3% of the patients had an actual unintentional discrepancy at discharge (Wong et al., 2008). Although worded differently, Wong et al. (2008) reported similar rates of “unintentional discrepancies” (55.3%), which are comparable to our findings of incomplete discharge summaries (52.5%) or partial discharge instructions (48.9%). Other investigators reported fewer medication discrepancies (14.1% to 49%) among patients discharged from general medical or family practice units (Coleman et al., 2005; Nickerson et al., 2005; Schnipper et al., 2006; Vira et al., 2006).

The differences among reported rates may be attributed to study designs and populations. Specifically, our retrospective study may have found more reconciliation problems than prospective or interventional studies that used clinical pharmacists to clarify problems when reconciling discharge medications and/or were designed to provide medication counseling among more heterogeneous populations (Nickerson et al., 2005; Schnipper et al., 2006; Vira et al., 2006; Wong et al., 2008). Our findings may be an underestimate of the prevalence of high-risk medications among this population because insulin or oral hypoglycemic agents were not among the classes of medications associated with post-hospital ADEs (Forster et al., 2005) used in this study. Yet, these medication classes are listed as specific high-alert medications by the Institute for Safe Medication Practices (ISMP, 2011) and have been associated with preventable ADEs in the community (Thomsen, Winterstein, Søndergaard, Haugbølle, & Melander, 2007).

A unique contribution of our study is the description of the rates and types of medication reconciliation problems found on specific discharge documents, each of which has different implications. Patient discharge instructions that are incorrect or missing information most directly affect the patient and/or family caregivers who presumably use them to guide how they take their medications correctly, whereas inaccurate hospital discharge summaries are more problematic to post-hospital clinicians who receive them. Our study found that more than half (52.5%) of the hospital discharge summaries were missing medication information. A systematic review identified communication deficits between hospital and primary physicians, such as discharge summaries that lacked critical information or were not available for the first post-hospital visit (Kripalani et al., 2007). The revised TJC (2010) standard highlights reconciling medication information between current hospital regimens and the list of medications the patient was taking before admission to the hospital. However, we found inconsistencies between two discharge documents that should have been the same because they represent the same moment in time, which underscores the complexity of the discharge medication reconciliation process. Our study findings suggest that hospital medication reconciliation processes should be instituted to ensure both patient instructions and discharge summaries are consistent as a critical step to promote continuity of medication information.

Incomplete discharge instructions were a common problem across studies. We found nearly half (48.9%) of the patient discharge instructions were missing at least one component of the medication order. In two instances, the patient instructions did not list any medications at all. Other studies reported similar problems of discharge prescriptions requiring clarification (49.5%); incomplete, inaccurate, or illegible discharge instructions (16.1% to 46.5%); inadequate instructions about medication changes (51%); or conflicting information between sources (35.6%) (Coleman et al., 2005; Corbett et al., 2010; Vira et al., 2006; Wong et al., 2008). Bayley et al. (2005) highlighted specific system failures of medication handoffs at hospital discharge that included the use of vague phrases (e.g., “resume medications”), which occurred in 19% of charts they reviewed. Our findings support a similar rate of using vague phrases; however, we present more detail on its frequency on either patient discharge instructions or hospital discharge summaries and which high-risk medications were involved (e.g., warfarin, nitroglycerin).

Discharge medication teaching was described as a common, high-risk activity (Bayley et al., 2005), and it is a common part of nursing practice. The MedPAC (2007) report identified the need for greater attention to discharge medications and patient and family communication as two strategies that may prevent rehospitalizations. Both strategies need to be incorporated into the interdisciplinary design and implementation of revised medication reconciliation processes. The use of vague phrases in discharge documents can cause misunderstandings for post-hospital clinicians and patients alike. Current “do not use” lists (ISMP, 2011; TJC, 2009) focus on abbreviations or dosage symbols that should be avoided to prevent confusion with prescriptions. Our study findings suggest that vague phrases should be added to these “do not use” lists to improve medication communication and patient understanding, which are essential elements to safe, high-quality transitional care.

A limitation of this study was use of medical records that predated the TJC medication reconciliation initiative. Despite this limitation, this secondary analysis provides valuable and detailed descriptions of missing information (e.g., medication, dosage, frequency), use of vague terms, and which types of medications and on which document the problem occurred. These findings are potentially useful to the current dialogue and efforts to implement the newly revised NPSG of medication reconciliation (TJC, 2010), patient safety initiatives, and specific transitional care strategies to reduce hospital readmissions. Medication reconciliation has been a challenging patient safety issue for several years now. Given our findings and the subsequent national effort to address medication reconciliation (e.g., electronic discharge instructions, collaborative team efforts), the authors recommend a replication of this study.

A second limitation was the use of six hospital sites that had site-specific discharge documents and practices. For example, we do not know by whom and when the discharge instructions were completed or how consistent the process was across individuals within sites. However, the prevalence of similar problems across sites underscores quality discharge instructions as a common issue. A third limitation was the homogeneous sample of older adults discharged home with heart failure. These patients represent a chronically ill population at risk for hospitalization (Jencks et al., 2009; Naylor et al., 1999) and a group with increasingly complex medication regimens (Masoudi et al. 2005). As a result, our findings may overestimate the medication reconciliation problems found in a younger or less chronically ill population. However, they emphasize the substantial challenges and risks embedded in preparing a vulnerable population to safely manage their medication regimen at home.

Implications and Conclusion

Our study highlights the need to consider hospital patient instructions and discharge summaries as two strategic documents that must be reconciled to provide continuity of medication information to two distinct groups (i.e., patients and their family caregivers and post-hospital clinicians). In addition, we are recommending that vague phrases (e.g., “take as directed”) be added to “do not use” lists to promote patient safety and facilitate better written support to patients and their family caregivers once they are at home, as well as improve communication to post-hospital clinicians. Chronically ill older adults, such as those living with heart failure, are more likely to have complicated medication regimens that change during hospitalization. With frequent medication changes occurring during hospitalization, it takes a team of patients, family caregivers, and clinicians to provide continuity of medication information during transitions in care. Clear, consistent hospital discharge medication lists are needed to highlight intentional changes from prior regimens.

Nurses practicing in hospitals are typically involved in discharge teaching of patients and their family caregivers. As such, they can make a tremendous difference by devoting time to ensure clear medication instructions that include the time a medication was last given, alerting patient and family caregivers to changes in prior medication regimens, and clarifying vague terms. For patients with complex medication regimens and clinical needs, home health nurses can be partners who build on such discharge instructions to reinforce and tailor patient teaching in ways that promote safer medication management in the home. In addition, home health nurses can play a vital and collaborative role with primary care providers because they can see which medications are actually in the home and assess patients’ clinical response that leads to more effective medication management, and ideally, improves patient outcomes.

References

  • Bayley, K.B., Savitz, L.A., Rodriguez, G., Gillanders, W. & Stoner, S. (2005). Barriers associated with medication information handoffs. In Agency for Healthcare Research and Quality, Advances in patient safety: From research to implementation (Vol. 3). Retrieved from http://www.ahrq.gov/downloads/pub/advances/vol3/Bayley.pdf
  • Boockvar, K., Fishman, E., Kyriacou, C.K., Monias, A., Gavi, S. & Cortes, T. (2004). Adverse events due to discontinuations in drug use and dose changes in patients transferred between acute and long-term care facilities. Archives of Internal Medicine, 164, 545–550. doi:10.1001/archinte.164.5.545 [CrossRef]
  • Coleman, E.A., Smith, J.D., Raha, D. & Min, S.-J. (2005). Posthospital medication discrepancies: Prevalence and contributing factors. Archives of Internal Medicine, 165, 1842–1847. doi:10.1001/archinte.165.16.1842 [CrossRef]
  • Corbett, C.F., Setter, S.M., Daratha, K.B., Neumiller, J.J. & Wood, L.D. (2010). Nurse identified hospital to home medication discrepancies: Implications for improving transitional care. Geriatric Nursing, 31, 188–196. doi:10.1016/j.gerinurse.2010.03.006 [CrossRef]
  • Forster, A.J., Clark, H.D., Menard, A., Dupuis, N., Chernish, R., Chandok, N. & van Walraven, C.,… (2004). Adverse events among medical patients after discharge from the hospital. Canadian Medical Association Journal, 170, 345–349.
  • Forster, A.J., Murff, H.J., Peterson, J.F., Gandhi, T.K. & Bates, D.W. (2003). The incidence and severity of adverse events affecting patients after discharge from the hospital. Annals of Internal Medicine, 138, 161–167.
  • Forster, A.J., Murff, H.J., Peterson, J.F., Gandhi, T.K. & Bates, D.W. (2005). Adverse drug events occurring following hospital discharge. Journal of General Internal Medicine, 20, 317–323. doi:10.1111/j.1525-1497.2005.30390.x [CrossRef]
  • Foust, J.B., Naylor, M.D., Boling, P.A. & Cappuzzo, K.A. (2005). Opportunities for improving post-hospital home medication management among older adults. Home Health Care Services Quarterly, 24(1–2), 101–122. doi:10.1300/J027v24n01_08 [CrossRef]
  • Gilbert, E.H., Lowenstein, S.R., Koziol-McLain, J., Barta, D.C. & Steiner, J. (1996). Chart reviews in emergency medicine research: Where are the methods?Annals of Emergency Medicine, 27, 305–308. doi:10.1016/S0196-0644(96)70264-0 [CrossRef]
  • Greenwald, J.L., Denham, C.R. & Jack, B.W. (2007). The hospital discharge: A review of a high risk care transition with highlights of a reengineered discharge process. Journal of Patient Safety, 3, 97–106. doi:10.1097/01.jps.0000236916.94696.12 [CrossRef]
  • Hector, B., Ross, J.S., Wang, Y., Chen, J., Vidán, M.T., Normand, S.-L.T. & Krumholz, H.M.,… (2010). Trends in length of stay and short-term outcomes among Medicare patients hospitalized for heart failure, 1993–2006. Journal of the American Medical Association, 303, 2141–2147. doi:10.1001/jama.2010.748 [CrossRef]
  • Institute for Safe Medication Practices. (2011). ISMP’s list of error-prone abbreviations, symbols, and dose designations. Retrieved from http://www.ismp.org/tools/errorproneabbreviations.pdf
  • Jencks, S.F., Williams, M.V. & Coleman, E.A. (2009). Rehospitalizations among patients in the Medicare fee-for-service program. New England Journal of Medicine, 360, 1418–1428. doi:10.1056/NEJMsa0803563 [CrossRef]
  • The Joint Commission. (2009). Official “do not use” list. Retrieved from http://www.jointcommission.org/assets/1/18/dnu_list.pdf
  • The Joint Commission. (2010, December8). Revised NPSG on medication reconciliation is approved. Joint Commission Online. Retrieved from http://www.jointcommission.org/assets/1/18/jconline_Dec_8_10.pdf
  • Kripalani, S., LeFevre, F., Phillips, C.O., Williams, M.V., Basaviah, P. & Baker, D.W. (2007). Deficits in communication and information transfer between hospital-based and primary care physicians: Implications for patient safety and continuity of care. Journal of the American Medical Association, 297, 831–841. doi:10.1001/jama.297.8.831 [CrossRef]
  • Marks, J.R., Schectman, J.M., Groninger, H. & Plews-Ogan, M.L. (2010). The association of health literacy and socio-demographic factors with medication knowledge. Patient Education and Counseling, 78, 372–376. doi:10.1016/j.pec.2009.06.017 [CrossRef]
  • Masoudi, F.A., Baillie, C.A., Wang, Y., Bradford, W.D., Steiner, J.F., Havranek, E.P. & Krumholz, H.M.,… (2005). The complexity and cost of drug regimens of older patients hospitalized with heart failure in the United States, 1998–2001. Archives of Internal Medicine, 165, 2069–2076. doi:10.1001/archinte.165.18.2069 [CrossRef]
  • Medicare Payment Advisory Commission. (2007). Payment policy for inpatient readmissions. In Report to the Congress: Promoting greater efficiency in Medicare (pp. 103–120). Retrieved from http://www.medpac.gov/documents/Jun07_EntireReport.pdf
  • Naylor, M.D., Brooten, D.A., Campbell, R.L., Jacobsen, B.S, Mezey, M.D., Pauly, M.V. & Schwartz, J.S. (1999). Comprehensive discharge planning and home follow-up of hospitalized elders: A randomized clinical trial. Journal of the American Medical Association, 281, 613–620. doi:10.1001/jama.281.7.613 [CrossRef]
  • Naylor, M.D., Brooten, D.A, Campbell, R.L., Maislin, G., McCauley, K.M. & Schwartz, J.S. (2004). Transitional care of older adults hospitalized with heart failure: A randomized controlled trial. Journal of the American Geriatrics Society, 52, 675–684. doi:10.1111/j.1532-5415.2004.52202.x [CrossRef]
  • Nickerson, A., MacKinnon, N.J., Roberts, N. & Saulnier, L. (2005). Drug-therapy problems, inconsistencies and omissions identified during a medication reconciliation and seamless care services. Healthcare Quarterly, 8(Special Issue), 65–72.
  • Pippins, J.R., Gandhi, T.K., Hamann, C., Ndumele, C.D., Labonville, S.A., Diedrichsen, E.K. & Schnipper, J.L.,… (2008). Classifying and predicting errors of inpatient medication reconciliation. Journal of General Internal Medicine, 23, 1414–1422. doi:10.1007/s11606-008-0687-9 [CrossRef]
  • Schnipper, J.L., Kirwin, J.L., Cotugno, M.C., Wahlstrom, S.A., Brown, B.A., Tarvin, E. & Bates, D.W.,… (2006). Role of pharmacist counseling in preventing adverse drug events after hospitalization. Archives of Internal Medicine, 166, 565–571. doi:10.1001/archinte.166.5.565 [CrossRef]
  • Thomsen, L.A., Winterstein, A.G., Søndergaard, B., Haugbølle, L.S. & Melander, A. (2007). Systematic review of the incidence and characteristics of preventable adverse drug events in ambulatory care. Annals of Pharmacotherapy, 41, 1411–1426. doi:10.1345/aph.1H658 [CrossRef]
  • Unroe, K.T., Pfeiffenberger, T., Riegelhaupt, S., Jastrezembski, J., Loknygina, Y. & Colón-Emeric, C. (2010). Inpatient medication reconciliation at admission and discharge: A retrospective cohort study of age and other risk factors for medication discrepancies. American Journal of Geriatric Pharmacotherapy, 8, 115–126. doi:10.1016/j.amjopharm.2010.04.002 [CrossRef]
  • Vira, T., Colquhoun, M. & Etchells, E. (2006). Reconcilable differences: Correcting medication errors at hospital admission and discharge. Quality and Safety in Health Care, 15, 122–126. doi:10.1136/qshc.2005.015347 [CrossRef]
  • Wong, J.D., Bajcar, J.M., Wong, G.G., Alibihai, S.M.H., Huh, J.-H., Cesta, A. & Fernandes, O.A.,… (2008). Medication reconciliation at hospital discharge: Evaluating discrepancies. Annals of Pharmacotherapy, 42, 1373–1379.

Definition of Medication Reconciliation Problems

Type of Problem Subtypes
Medication discrepancies (at least one or more medication order varied between the two documents)

Medication(s) omitted from the hospital discharge summary

Medication(s) omitted from patient discharge instructions

Dosage and/or frequency varied

Incomparable units used (e.g., tablets versus milligrams)

Partial patient discharge instructions (at least one or more medications were missing dosage or frequency, used vague terms, or no medications were listed)

Missing dosage and/or frequency

Use of a vague phrase to refer to taking dosage and/or frequency as before

General statement about taking medications as before

No medications listed

Incomplete hospital discharge summary (at least one or more medications were missing dosage or frequency, used vague terms, no medications listed, or page was missing)

Dosage and/or frequency missing

Use of a vague phrase to refer to taking dosage and/or frequency as before

General statement about taking medications as before

No medications listed

Page 2 missing

Baseline Sociodemographic and Health Characteristics of the Sample (N = 162)

Characteristic Mean (SD)
Age (years) 76.1 (6.8)
Index length of stay (days) 4.8 (2.3)
Number of medications on admission 6.7 (2.9)
Number of comorbid conditions 6.4 (2.2)
n(%)
Women 98 (60.5)
Race
  Caucasian 104 (64.2)
  African American 58 (35.8)
Educational level
  High school or higher 95 (58.6)
Marital status
  Married 70 (43.5)
  Widowed 68 (42.2)
  Other 24 (14.8)
Annual income
  <$10,000 56 (34.6)
  $10,000 to $19,999 38 (23.5)
  ⩾$20,000 28 (17.3)
  Missing data 40 (24.7)

Medication Reconciliation Problems

At Least One Problem of Any Type (n = 141, 71.2%) Problem Involving High-Risk Drugs (n = 108, 76.6%)b
Reconciliation Problema n(%) n(%)b
Medication discrepancy 83 (58.9) 69 (83.1)
  Dosage and/or frequency varied 52 (62.7) 45 (86.5)
  Omitted from hospital discharge summary 36 (43.4) 29 (80.6)
  Omitted from patient discharge instructions 21 (25.3) 19 (90.5)
Incomplete hospital discharge summary 74 (52.5) 58 (78.4)
  Missing dosage and/or frequency 47 (63.5) 33 (70.2)
  Use of a vague phrase to refer to taking dosage and/or frequency as before 15 (20.3) 10 (66.7)
  No medications listed 10 (13.5) 10 (100)
  General statement about taking medications as before 4 (5.4) 4 (100)
  Page 2 missing 3 (4.1) 3 (100)
Partial patient discharge instructions 69 (48.9) 53 (76.8)
  Missing dosage and/or frequency 34 (49.3) 26 (76.5)
  Use of a vague phrase to refer to taking dosage and/or frequency as before 28 (40.6) 20 (71.4)
  General statement about taking medications as before 12 (17.4) 11 (91.7)
  No medications listed 2 (2.9) 2 (100)

Number of Reconciliation Problems Per Discharge

No. of Reconciliation Problems No. of Discharges (%)
0 57 (28.8)
1 62 (31.3)
2 45 (22.7)
3 25 (12.6)
4 8 (4)
5 1 (0.5)
Authors

Dr. Foust is Assistant Professor, University of Massachusetts Boston, Department of Nursing, Boston, Massachusetts; Dr. Naylor is Marian S. Ware Professor in Gerontology and Director, NewCourtland Center for Transitions and Health, Mr. Bixby is Advanced Practice Nurse, University of Pennsylvania, School of Nursing, and Dr. Ratcliffe is Associate Professor of Biostatistics, Department of Biostatistics and Epidemiology, University of Pennsylvania, Perelman School of Medicine, Philadelphia, Pennsylvania. Dr. Foust completed portions of this work while at the Center for Home Care Policy and Research, Visiting Nurse Service of New York, and the University of New Hampshire, Department of Nursing.

The authors disclose that they have no significant financial interests in any product or class of products discussed directly or indirectly in this activity. The authors gratefully acknowledge funding from the John A. Hartford Foundation (JAHF) Building Academic Geriatric Nursing Capacity Program. Dr. Foust is a 2003–2005 JAHF Claire M. Fagin Postdoctoral Fellow, Hartford Center of Geriatric Nursing Excellence, University of Pennsylvania, School of Nursing. The authors thank J. Sanford Schwartz, MD, Principal Investigator, National Institutes of Health grant RO1-NR07616 for his support. Portions of this work have been presented at annual meetings of the Gerontological Society of America in November 2003, 2005, and 2008.

Address correspondence to Janice B. Foust, PhD, RN, Assistant Professor, University of Massachusetts Boston, 100 Morrissey Boulevard, Boston, MA 02125; e-mail: janice.foust@umb.edu.

Received: May 18, 2011
Accepted: October 11, 2011
Posted Online: December 29, 2011

10.3928/19404921-20111206-04

Sign up to receive

Journal E-contents