PRACTICE REPORT Care transitions service
PRACTICE REPORT
Care Transitions Service: A pharmacy-driven program for medication reconciliation through the continuum of care Jessica R. Conklin, John C. Togami, Allison Burnett, Melanie A. Dodd, and Gretchen M. Ray
T
ransitions from one care setting to another are known to be critical junctures that put patients at high risk for experiencing medication errors. Approximately 22% of potentially preventable medication errors occur during admissions, 66% during a transition to or from the intensive care unit, and 12% during discharge.1,2 A reported 38.6% of medication discrepancies that occur in the inpatient setting have the potential to cause moderate-to-severe discomfort or clinical deterioration.3 Medication errors are not isolated to the inpatient setting. A recent study by Haynes et al.4 found that 50.8% of discharged patients experienced one or more clinically important medication errors during the 30 days after hospital discharge. It is suspected that many factors contribute to postdischarge medication errors, including but not limited to nonadherence, potential adverse drug events, physician and system errors, outpatient pharmacy discrepancies, patient safety issues, and discharge
Purpose. A quality-improvement program at University of New Mexico Hospital (UNMH) encompassing admission, discharge, and postdischarge medication reconciliation activities is described, with a report on initial assessments of the program’s impact on rates of medicationrelated problems (MRPs). Methods. Pharmacists conducted a five-month evaluation of the UNMH Care Transitions Service (CTS), which serves inpatients admitted to the hospital’s family medicine service, providing medication reconciliation and targeted MRP interventions. Selected patients who received CTS services from November 2012 through March 2013 (n = 191) were included in the analysis. The study endpoints were the rates and types of MRPs identified, the most commonly implicated medication classes, and predictors of MRPs. Postdischarge MRP rates during a two-month
medication confusion, including patient health literacy concerns.1-5 In 2004, the Institute of Medicine estimated that 90 million adults in the United States may have trouble
Jessica R. Conklin, Pharm.D., PhC, is Visiting Assistant Professor, Department of Pharmacy Practice & Administrative Sciences, University of New Mexico (UNM) College of Pharmacy, Albuquerque; at the time of the study, she was Postgraduate Year 2 (PGY2) Ambulatory Care Pharmacy Resident, UNM College of Pharmacy. John C. Togami, Pharm.D., is PGY2 Ambulatory Care Pharmacy Resident, UNM College of Pharmacy; at the time of the study, he was Postgraduate Year 1 Pharmacy Resident, UNM Hospital, Albuquerque. Allison Burnett, Pharm.D., PhC, is Clinical Pharmacist, UNM Hospital. Melanie A. Dodd, Pharm.D., PhC, BCPS, is Associate Professor; and Gretchen M. Ray, Pharm.D., PhC, BCPS, is Assistant
802
Am J Health-Syst Pharm—Vol 71 May 15, 2014
trial of CTS services at a UNMH outpatient clinic were also evaluated. Results. During the five-month evaluation of inpatient CTS services, a total of 1140 MRPs were identified (an average of 6 per patient), about 70% of which were resolved independently of provider review using pharmacy-driven protocols. During the twomonth pilot test of CTS outpatient services (n = 16), a total of 28 MRPs were identified; in over 80% of cases, there was a decline in the number of MRPs from the admission to the postdischarge medication reconciliation. Conclusion. MRPs were identified through the continuum of care. The majority of MRPs identified in both the inpatient and outpatient settings involved patient variables and patient nonadherence. Seventy percent of inpatient MRPs were resolved independently by the CTS team under pharmacy-driven protocols. Am J Health-Syst Pharm. 2014; 71:802-10
understanding and acting on health information.6 The Newest Vital Sign (NVS), available in both Englishand Spanish-language versions, is a clinical tool for measuring health
Professor, Department of Pharmacy Practice & Administrative Sciences, UNM College of Pharmacy. Address correspondence to Dr. Ray ([email protected]). Dr. Dodd was the chair of the American Society of Health-System Pharmacists Council on Public Policy and has received travel funding to attend council meetings in Bethesda, MD; she is also a director and secretary of the New Mexico Pharmaceutical Care Foundation. The other authors have declared no potential conflicts of interest. Copyright © 2014, American Society of Health-System Pharmacists, Inc. All rights reserved. 1079-2082/14/0502-0802$06.00. DOI 10.2146/ajhp130589
PRACTICE REPORT Care transitions service
literacy with demonstrated reliability (Cronbach’s coefficient alpha, > 0.76). 7 The NVS assesses numeracy and literacy skills by asking a patient to answer six questions about an ice cream nutrition label. Pharmacy-driven models of medication reconciliation during care transitions have been studied.8,9 Pharmacists are uniquely qualified to perform medication reconciliation because of their formal training in obtaining medication histories and extensive medication expertise. Pharmacists have been shown to obtain medication histories that are more accurate and comprehensive than those obtained by other health care professionals.10,11 Also, pharmacists often have training in both hospital and retail pharmacy settings, which can give them an advantage over other health care providers in navigating various medication distribution systems. Medication errors are often evaluated from two distinct perspectives: that of the inpatient/hospital setting and that of the postdischarge/ ambulatory care setting. According to a consensus statement published by Greenwald et al.,12 a “comprehensive reconciliation system is needed across the continuum of care.” To the best of our knowledge, few studies have analyzed medication discrepancies across the continuum of care (defined as the movement of a patient from hospital admission through the postdischarge period). The purpose of the study described in this article was to analyze an ongoing inpatient-focused, pharmacy-driven program—the Care Transitions Service (CTS)—and its impact on identifying and resolving medication-related problems (MRPs). We also evaluated the implementation of a pilot extension of this service into the outpatient setting and its impact on MRP identification and resolution through the continuum of care. Methods Study design. A single-center descriptive study was conducted within
University of New Mexico Hospital (UNMH) and a predetermined outpatient UNMH family medicine clinic. The study was approved by the university’s Human Research Review Committee. Analyses of the inpatient service were conducted over a five-month period (November 2012– March 2013), and analyses of the outpatient service were conducted over a two-month period (January–February 2013). Patients were included in the study if they received medication reconciliation services from CTS during the predefined study periods. CTS description. CTS is an ongoing pharmacy-driven program that provides three phases of inpatient services, with a fourth, pilot phase of services provided in the outpatient clinic setting (Figure 1). CTS services are offered to UNMH family medicine inpatients who are at least 18 years of age and speak English or Spanish. Patients are ineligible for CTS if they do not have an extensive
medication history (defined as at least three long-term medications) or if there are plans for a readmission. Patients who have a previous admission within the last 14 days and already received CTS services, are unwilling to undergo a medication interview, or have been discharged to a skilled nursing facility, long-term care facility, or prison are also ineligible. The outpatient extension of the service was offered to all patients who received inpatient CTS services, were scheduled for a postdischarge outpatient family medicine clinic appointment, and were independently able to answer questions about their medications. Phase 1 of the program provides admission medication reconciliation services in addition to those performed by a nonpharmacist provider. Phase 2 provides a discharge medication review and a hospitalto-community pharmacist handoff, which involves obtaining consent
Figure 1. Flow chart of University of New Mexico Hospital’s Care Transitions Service program.
Phase 1 Admission medication reconciliation (inpatient)
Phase 4 Follow-up appointment medication reconciliation (outpatient)
Phase 2 Discharge medication review and hospitalto-community pharmacist handoff (inpatient)
Phase 3 Follow-up phone call reminder (inpatient)
Am J Health-Syst Pharm—Vol 71 May 15, 2014
803
PRACTICE REPORT Care transitions service
from a family medicine provider for CTS staff to call the patient’s community pharmacy and authorize discontinuation of any prescriptions no longer needed after hospital discharge. Phase 3 provides a followup phone call within 72 hours of discharge to assess adherence to and tolerance of discharge medications and to remind patients to take all medication bottles to their follow-up appointment. Phase 4 (the pilot extension of the service) consisted of an outpatient medication reconciliation conducted by a pharmacist at the first postdischarge appointment. Program phases 1–3 were conducted by a CTS team comprising a lead inpatient pharmacist, four postgraduate year 1 (PGY1) pharmacy residents, and six advanced pharmacy practice experience (APPE) clerkship pharmacy students. Phase 4 was conducted by a lead pharmacist and a postgraduate year 2 ambulatory care pharmacy resident. The APPE students and PGY1 residents were provided extensive training to ensure consistency of CTS processes. All members of the CTS team were responsible for identification and documentation of MRPs. All documentation was summarized in a written clinical note and posted in the electronic medical record (EMR). All APPE students were supervised by the PGY1 residents as MRPs were identified and documented. Weekly quality checks were completed by the lead pharmacist. Data collection. MRPs were identified and documented by a CTS member during the admission reconciliation process. Each MRP was classified only once. Interventions were performed independently of provider review according to protocols agreed on by the pharmacy department and inpatient family medicine service. Recommendations requiring additional provider action or review were communicated to the responsible family medicine provider. Multiple methods were used to 804
communicate with providers, including phone calls, pages, and alerts in the EMR. Additional MRPs identified in the outpatient setting were documented, including medication continuance despite discontinuation orders at discharge and patient selfdiscontinuation of medications despite continuation orders at discharge. During the outpatient visit, each patient’s NVS score was assessed and recorded. Definitions. High-alert medications were identified based on the currently available list maintained by the Institute for Safe Medication Practices.13 Medications requiring monitoring were defined as any medication with an associated laboratory value that is used for therapeutic monitoring. The MRP classifications were based on the Health Resources and Services Administration classification scheme (Appendix A).14 Pharmacy interventions and recommendations in response to MRPs are listed in Appendix B. Study outcomes. The primary endpoint was to evaluate the numbers and types of MRPs identified by pharmacists at hospital admission (CTS phase 1). Secondary endpoints for the inpatient service evaluation included the medication classes most commonly associated with MRPs and patient-specific predictors of MRPs. For the evaluation of outpatient services, secondary endpoints included the types and frequencies of new or persistent MRPs occurring from hospital discharge to the first follow-up appointment and the association of health literacy and MRPs among CTS patients. Statistical analysis. Data analyses were conducted using SPSS, version 19.0 (IBM Corporation, Armonk, NY) and Microsoft Access (Microsoft Corporation, Redmond, WA). Descriptive analyses were conducted to characterize the types and rates of medication reconciliation discrepancies between the inpatient and
Am J Health-Syst Pharm—Vol 71 May 15, 2014
outpatient settings and the health literacy of CTS patients. Univariate Poisson regression models were used to determine variables that would be retained for testing in the final model, and an a priori level of significance of 0.2 was chosen in order to prevent the elimination of variables too soon. A multivariate Poisson regression model with an a priori level of significance of 0.05 was used as the final model to determine patient-specific predictors of MRPs. Results Baseline characteristics. Baseline characteristics of patients included in the evaluated inpatient and outpatient CTS samples are listed in Table 1. Inpatient CTS sample. A total of 191 patients received admission medication reconciliation (CTS phase 1) over the five-month inpatient evaluation period; their mean age was 61 years, and 48% were male. Approximately 50% of patients were of Hispanic/Latino ethnicity, which is similar to the ethnic mix of the general population in New Mexico. Ninety-two patients (48.2%) received discharge medication review and hospital-to-community pharmacist handoff (CTS phase 2), and 46 patients (24.1%) received a follow-up phone call (CTS phase 3). Outpatient CTS sample. Thirty patients were eligible for phase 4 of the CTS during the two-month pilot testing period. Seven of the 30 patients failed to attend their scheduled postdischarge appointment. Three patients were ineligible for the service because they were unable to independently answer questions. One patient deferred the service. Missed opportunity, due to external factors (e.g., unavailability of the pharmacist to perform a medication reconciliation), occurred in 3 patients. Therefore, a total of 16 patients received outpatient CTS services and were included in the analysis. Those patients had a mean age of 57.8 years, 56%
PRACTICE REPORT Care transitions service
were male, and 50% were Hispanic/ Latino. Medication reconciliation sources. Inpatient CTS sample. One hundred forty-one patients (73%) had their medication list reconciled via three sources (EMR, patient interview, and outpatient pharmacy data). Thirtynine patients (20%) had their medication list reconciled via two sources (EMR and outpatient pharmacy data). The remainder of the patients had only one source available to use for reconciliation. The mean time to complete the phase 1 admission medication reconciliation was 45 minutes. Outpatient CTS sample. Adherence barriers to therapy (cognitive, physical, financial, or transportation factors or patient beliefs) were identified in 62.5% of patients. Approximately 32% of patients in the CTS phase 4 evaluation were receiving a high-alert medication; no patients were taking medications that required monitoring. Summary data on the results of inpatient and outpatient medication reconciliation by the CTS team are shown in Table 2. Identified MRPs. Inpatient CTS sample. A total of 1140 MRPs were identified during admission medication reconciliation (CTS phase 1). The mean number of MRPs identified per patient was 6. Forty-nine percent of MRPs were attributable to patient variables and patient nonadherence, with 25%, 21%, and 5% attributable to miscellaneous, safety, and provider variables (issues of appropriateness or effectiveness), respectively. The medication class most commonly associated with MRPs was cardiovascular agents (22%), followed by gastrointestinal–genitourinary, central nervous system (CNS), analgesic, and endocrine–metabolic agents. These classes comprised nearly 70% of all medications associated with MRPs. Outpatient CTS sample. Overall, the number of MRPs identified, per
Table 1.
Baseline Patient Characteristicsa Characteristic
No. (%)b,c
Inpatients (n = 191) Mean age, yr Male Languaged English Spanish Ethnicity White, non-Hispanic Hispanic/Latino African American Native American Other Health insurance Private Public County indigent None Marital status Single Married Divorced Widowed Living with partner Source of admission Home Caregiver’s home LTCF/SNF Other Median ± S.D. Charlson Comorbidity Index score Outpatients (n = 16) Mean age, yr Male Language English Spanish Ethnicity White, non-Hispanic Hispanic/Latino African American Native American Health insurance Private Public County indigent None
61.0 92 (48) 168 (88) 21 (11) 61 (32) 95 (50) 10 (5) 15 (8) 10 (5) 39 (20) 71 (37) 64 (33) 17 (9) 79 (41) 54 (28) 28 (15) 28 (15) 4 (2) 175 (92) 4 (2) 7 (4) 5 (3) 2.0 ± 1.8 57.8 9 (56) 14 (87) 2 (12) 6 (37) 8 (50) 1 (6) 1 (6) 5 (31) 4 (25) 6 (37) 1 (6)
LTCF = long-term care facility, SNF = skilled nursing facility. All data are no. (%) unless specified otherwise. c For some characteristics, percentages do not sum to 100 due to rounding. d Not reported by two patients. a
b
Am J Health-Syst Pharm—Vol 71 May 15, 2014
805
PRACTICE REPORT Care transitions service
patient, from admission medication reconciliation (CTS phase 1) to postdischarge medication reconciliation (phase 4) decreased. A total of 28 MRPs were identified in the phase 4 pilot (range, 0–3 per patient), with an average of 1.56 per patient. The maximum number of MRPs for a single patient was 3 (this occurred in 37.5% of patients), whereas 18.8% of patients had no identified MRP. Approximately 75% of the MRPs identified were related to patient variables and nonadherence. In over 80% of patients, fewer MRPs were identified at the follow-up appointment than had been identified during their admission to the hospital (Figure 2). The specific MRPs were tracked through the continuum of care. Eighty-six percent of all postdischarge MRPs were newly identified MRPs, while 14% had persisted through the continuum of care. Pharmacy interventions and recommendations. A pharmacy intervention or recommendation was documented in response to each MRP. Eight hundred seven
MRPs (70%) were resolved through a pharmacy intervention performed independently of the provider under pharmacy-driven protocols; 43% of pharmacy interventions involved the removal of drugs from the medication list, and 29% involved the addition of drugs to the medication list. The remaining roughly 30% of MRPs required provider follow-up; the majority of these recommendations included recommendations to clarify the patient’s medication regimen (25%), consider removing a drug from the medication list (22%), and add an indication for a prescribed medication (18%). A breakdown of pharmacy interventions per protocol and pharmacy recommendations can be found in Figures 3 and 4, respectively. Predictors of inpatient MRPs. Certain marital status, insurance coverage, and age variables, as well as the presence of medications requiring monitoring and the use of a higher number of medications before CTS phase 1 medication reconciliation, were patient-specific predictors of
Table 2.
Medication Reconciliation Findings Finding Inpatients (n = 191) Adherence barriersa ≥1 None Presence of medication(s) requiring monitoring Yes No Presence of high-alert medication(s) Yes No Mean no. medications prior to medication reconciliation Mean no. medications after medication reconciliation Outpatients (n = 16) Adherence barriersa ≥1 None Presence of high-alert medication(s) Yes No Includes cognitive, physical, transportation, financial, and patient-belief barriers.
a
806
Am J Health-Syst Pharm—Vol 71 May 15, 2014
No. (%)
111 (58.1) 80 (41.9) 28 (14.7) 163 (85.3) 69 (36.1) 122 (63.9) 12 11
10 (62.5) 6 (37.5) 5 (31.2) 11 (68.8)
MRPs (Table 3). Patients with medications requiring drug monitoring had a 35% higher incidence rate of MRPs than patients not prescribed such medications (incidence rate ratio [IRR], 1.35; 95% confidence interval [CI], 1.15–1.159). For each additional medication above three on a patient’s medication list, there was a 6% increase in the incidence rate of MRPs (IRR, 1.06; 95% CI, 1.05–1.07). Patients having no insurance (IRR, 1.48; 95% CI, 1.18–1.85) or private insurance (IRR, 1.20; 95% CI, 1.03–1.41) had 48% and 20% higher MRP incidence rates, respectively, than patients enrolled in the UNM Care indigent-care program. There was a 1% decrease in the MRP incidence rate for every one-year increase in age (IRR, 0.99; 95% CI, 0.98–0.99). Health literacy. The majority of patients (69%) who completed the NVS health literacy assessment during the phase 4 outpatient service evaluation were found to have a high likelihood of limited health literacy (defined as an NVS score of 0 or 1). Thirty-one percent of patients were found to have adequate health literacy (an NVS score or 4 or 5). MRPs were identified in patients with limited health literacy as well as those with adequate health literacy. Discussion This single-center descriptive study of an ongoing pharmacydriven inpatient service and pilot outpatient service demonstrated the need for medication reconciliation at each transition of care. The 1140 MRPs identified in CTS phase 1 were found after the admitting team had already conducted medication reconciliation as part of routine care, indicating the need for a pharmacist at the initial admission medication reconciliation. Several factors contributing to MRPs identified during CTS phase 1 were identified, which in the future may help to identify the patients most in need of a care tran-
PRACTICE REPORT Care transitions service
sitions service. Our study found that for each additional medication above three on a patient’s medication list, there was a 6% increase in the MRP rate, suggesting the total number of medications can be used as a means to identify patients at highest risk.15,16 We found a 35% higher rate of MRPs in patients receiving medications that required monitoring, indicating another group with a greater need for CTS-type services. In addition, patients with either no insurance or private insurance were found to have 48% and 20% higher rates of MRPs, respectively, relative to patients in the UNM Care program, which provides discounted medications to New Mexico indigent patients. Because the UNM Care program is specific to UNMH, these findings could be explained by
prescriber awareness of and familiarity with the UNM Care formulary as opposed to differing formularies of private insurance carriers. We also found that age appeared to have a protective effect against MRPs, which was not expected and differed from the findings in some published studies. 16-18 Pippens et al. 19 and Kripalani et al.,18 however, found that elderly patients were at lower risk for potential adverse drug events identified during both inpatient and postdischarge medication reconciliation; this effect persisted when adjusted for multiple patient characteristics and could be explained by the wide age range (21–99 years) of patients included in the studies or, perhaps, the greater diligence of outpatient providers in maintaining accurate medication lists in older patients. In
our study, medication classes such as cardiovascular, gastrointestinal– genitourinary, and CNS agents were frequently involved in MRPs, which was consistent with the findings in other studies.20-22 There were a number of important limitations to our study. First, patients readmitted to the hospital within 14 days did not receive inpatient CTS services because they received those services during the prior hospitalization; thus, we were unable to discern if any of these readmissions were due to an MRP. Second, CTS services are offered only to patients on the UNMH inpatient family medicine service, so these results only represent a small subgroup of the patient population within a large academic teaching hospital. Third, the CTS is an ongoing service staffed
Figure 2. Number of medication-related problems (MRPs) per Care Transitions Service (CTS) patient (n = 16) at admission medication reconciliation (green bars) and at the first postdischarge follow-up visit (black bars) during a two-month evaluation of CTS outpatient services.
Patient No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 9 No. 10 No. 11 No. 12 No. 13 No. 14 No. 15 No. 16 0
2
4
6
8
10
12
14
16
18
20
No. MRPs
Am J Health-Syst Pharm—Vol 71 May 15, 2014
807
PRACTICE REPORT Care transitions service
Figure 3. Pharmacy interventions per protocol conducted by the Care Transitions Service (CTS) team, by type. A total of 1140 medicationrelated problems among 191 patients were identified during the five-month evaluation of CTS inpatient services. DME = durable medical equipment. 400
350
No. Interventions
350 300 231
250 200 150 100
62
50
39
50
28
22
13
12
n re
r/t ea
ct
,a
lle
pr
rg
ov i
y/
de
in ue nt co
ad
d
C
dr
on
ug
ta
D
y/
D
la
rif
ov e
C
R
R
ac tio
m
ug dr
lis t n is
ed m m
M
M
E
od
fro
io
ify
na
do
li
se
nt
/fr
er
eq
ic at io
ue
th e
ve n
O
tio
n
lis t n at io at Ed
em
em
ov e
Ad
d
dr
ug
dr
uc
ug
fro
to
m
m
m
ed
ed
ic
ic
at io
n
lis t
r
nc y
0
Intervention Type
Figure 4. Pharmacy recommendations, by type, made by the Care Transitions Service (CTS) team during the five-month evaluation of CTS inpatient services. The team made a total of 333 recommendations targeting 1140 identified medication-related problems. 90
82 73
No. Recommendations
80 70
60
60
47
50 40 30
20
20
19 11
10
8
6
4
3
Re
m
ov e
dr u
Cl
g
ar
ify
do s
e do /fre q s ag ue fro e nc m fo y/ m rm ed ica Ad tio n d lis in di t ca Ad tio n d fo dr rd ug ru to g m ed ica tio n lis Di M t od sc on ify tin re ue gi m re en ta il p re sc rip tio Re n vie w in te ra So ct io cia n lw or k O re rd fe er rra la l bs /le M od ve ls/ ify te do st sa s ge fo rm /ro Do ut sa e ge ad ju st m en t
0
808
Recommendation Type
Am J Health-Syst Pharm—Vol 71 May 15, 2014
PRACTICE REPORT Care transitions service
by multiple pharmacists, which can lead to inconsistencies in data collection. However, a standardized database was used and all CTS pharmacists received training on MRP classification in the database. Among other study limitations, patient and caregiver recall and reporting of medications during the interview could have introduced bias into our results; however, the majority of medication lists (73%) were reconciled using three sources, minimizing the impact of this potential bias. Due to the short time frame of this analysis, we were unable to classify MRPs based on their potential to cause harm (i.e., clinically significant discrepancies) or to assess the clinical impact of the identified MRPs on endpoints such as readmission rates, patient outcomes, or health care utilization. This is an important area of future research through the continuum of care. It must also be emphasized that the outpatient service (CTS phase 4) was a small pilot extension of the ongoing inpatient service. Because only patients discharged to a predetermined family medicine clinic during the two-month pilot testing period were eligible for the outpatient service, the sample size was very small, which limited the statistical analyses; however, we still found a clinically significant number of MRPs in the outpatient setting at the first postdischarge appointment. Moreover, it is important to note that the interventions made by the CTS team among inpatients were associated with an overall reduction in the number of MRPs identified in the outpatient setting (Figure 2). Despite this, new MRPs that occurred during the time between discharge and the postdischarge appointment were identified. The presence of these newly identified MRPs revealed the need for a pharmacist to complete the final medication reconciliation in the outpatient clinic setting and thus enable improved MRP identification
and resolution across the continuum of care. Additionally, only patients who had received inpatient CTS services were eligible to receive outpatient CTS services at the clinic. A health literacy assessment was conducted for all patients targeted for CTS phase 4 services; however, due to the small sample size, we were unable to identify an association between health literacy scores and the presence of MRPs. We identified MRPs in patients with limited health literacy as well as those with adequate health literacy. Future studies are needed to evaluate associations of health literacy and MRPs. Lastly, it is important to note that some patients also received CTS inpatient phase 2 services (discharge
medication review and hospital-tocommunity pharmacist handoff ) and phase 3 services (postdischarge follow-up phone contact), which may have also reduced the number of MRPs identified, further supporting the need for medication reconciliation at several points throughout the continuum of care. The majority of MRPs identified in both the inpatient and outpatient settings involved patient variables and patient nonadherence. Seventy percent of inpatient MRPs were resolved independently by the CTS team as a result of pharmacy-driven protocols. Despite identification and resolution of MRPs during hospital admission, new MRPs developed between the time of discharge and the first follow-up appointment. How-
Table 3.
Predictors of Inpatient Medication-Related Problemsa Characteristic Marital status Widowed Living with partner Divorced Married Singleb Health insurance None Private Public (Medicare/Medicaid) Indigent (UNM Care)b Adherence barriers ≥1 Noneb Presence of high-alert medication(s) on admission reconciliation list Yes Nob Age Presence of medication(s) requiring monitoring on admission reconciliation list Yes Nob No. medications prior to medication reconciliation
Incident Rate Ratio (95% CI) 1.17 (0.95–1.43) 0.94 (0.63–1.40) 0.74 (0.60–0.92)c 1.28 (1.11–1.48)c . . .d 1.48 (1.18–1.85)c 1.20 (1.03–1.41)c 0.98 (0.84–1.14) . . . 1.09 (0.96–1.24) . . .
0.90 (0.79–1.03) . . . 0.99 (0.98–0.99)c
1.35 (1.15–1.159)c . . . 1.06 (1.05–1.07)c
UNM = University of New Mexico. Reference condition. c Statistically significant (p ≤ 0.05). d Not applicable. a
b
Am J Health-Syst Pharm—Vol 71 May 15, 2014
809
PRACTICE REPORT Care transitions service
ever, in the majority of patients, the admission medication reconciliation (phase 1) was associated with a lower number of MRPs after transition to the outpatient setting. Conclusion MRPs were identified through the continuum of care. The majority of MRPs identified in both the inpatient and outpatient settings involved patient variables and patient nonadherence. Seventy percent of inpatient MRPs were resolved independently by the CTS team under pharmacydriven protocols.
iation patient centered, clinically relevant and implementable: a consensus statement on key principles and necessary first steps. J Hosp Med. 2010; 5:477-85. Institute for Safe Medication Practices. ISMP’s list of high-alert medications. www.ismp.org (accessed 2012 Jun 15). Chen S, for Healthcare Communities. Medication therapy intervention and safety documentation form (version 8, 9/12/11). Available with registration at http://healthcarecommunities.org/ Comm_DocsDetails.aspx?groupid= 2&id=12345&tid=81321 (accessed 2011 Oct 12). Matsen Picone D, Titler MG, Dochterman J et al. Predictors of medication errors among elderly hospitalized patients. Am J Med Qual. 2008; 23:115-27. Tschantz Unroe K, Pfeiffenberger T, Riegelhaupt S et al. Inpatient medication reconciliation at admission and discharge: a retrospective cohort study of age and other risk factors for medication discrepancies. Am J Geriatr Pharmacother. 2010; 8:115-26. Bedell SE, Jabbour S, Goldberg R et al. Discrepancies in the use of medications, their extent and predictors in an outpatient practice. Arch Intern Med. 2000; 160:2129-34. Kripalani S, Price M, Vigil V et al. Frequency and predictors of prescriptionrelated issues after hospital discharge. J Hosp Med. 2008; 3:12-9. Pippens JR, Gandhi TK, Hamann C et al. Classifying and predicting errors of inpatient medication reconciliation. J Gen Intern Med. 2008; 23:1414-22. Steurbaut S, Leemans L, Leysen T et al. Medication history reconciliation by clinical pharmacists in elderly inpatients admitted from home or a nursing home. Ann Pharmacother. 2010; 44:1596-603. Gleason KM, McDaniel MR, Feinglass J et al. Results of the Medication At Transitions and Clinical Handoffs (MATCH) study: an analysis of medication reconciliation errors and risk factors at hospital admission. J Gen Intern Med. 2010; 25:441-7.
13. 14.
15.
16.
References 1. National Quality Forum. Preventing medication errors. www.qualityforum.org/ N P P / . . . / P r e v e n t i n g _ Me d i c a t i o n _ Errors (accessed 2012 Jun 15). 2. Santell JP. Reconciliation failures lead to medication errors. Jt Comm J Qual Patient Saf. 2006; 32:225-9. 3. Cornish PL, Knowles SR, Marchesano R et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005; 165:424-9. 4. Haynes KT, Oberne A, Cawthon C et al. Pharmacists’ recommendations to improve care transitions. Ann Pharmacother. 2012; 46:1152-9. 5. Davis TC, Wolf MS, Bass PF III et al. Literacy and misunderstanding prescription drug labels. Ann Intern Med. 2006; 145:887-94. 6. Institute of Medicine. Health literacy: a prescription to end confusion. In Nielsen-Bohlman L, Panzer A, Kindig DA, eds. Washington, DC: National Academies Press; 2004:1-17. 7. Weiss BD, Maysm MZ, Martz W et al. Quick assessment of literacy in primary care: the Newest Vital Sign. Ann Fam Med. 2005; 3:514-22. 8. Kwan JL, Lo L, Sampson M et al. Medication reconciliation during transitions of care as a patient safety strategy: a systematic review. Ann Intern Med. 2013; 158:397-403. 9. Mueller SK, Cunningham Sponsler K, Kripalani S et al. Hospital-based medication reconciliation practices: a systematic review. Arch Intern Med. 2012; 172:105769. 10. Reeder TA, Mutnick A. Pharmacistversus physician-obtained medication histories. Am J Health-Syst Pharm. 2008; 65:857-60. 11. Nester TM, Hale LS. Effectiveness of a pharmacist-acquired medication history in promoting patient safety. Am J HealthSyst Pharm. 2002; 59:402-5. 12. Greenwald JL, Halasyamani L, Greene J et al. Making inpatient medication reconcil-
810
17.
18.
19.
20.
21.
22. Quélennec B, Beretz L, Paya D et al. Potential clinical impact of medication discrepancies at hospital admission. Eur J Intern Med. 2013; 24:530-5.
Appendix A—Medication-related problem classifications14 Appropriateness/effectiveness Inadequate dosing regimen Monitoring standards not followed Untreated medical problem Home medication not resumed Safety Allergy/food/drug interaction Contraindication Dosage adjustment needed (renal or hepatic) Drug dosing excessive for treatment Incorrect dose/frequency Intended use differs from actual use Laboratory test indicated, not ordered Missing dose/frequency No indication for drug prescribed Therapy duplication Patient variables/nonadherence Adherence divergence Dosage form/route inappropriate Medication continued despite discontinuation orders Patient self-discontinuation despite continuation orders Patient not taking medication due to financial reasons Medication on medication list, patient no longer taking Medication not on medication list, patient taking Miscellaneous Durable medical equipment more than two years old and patient no longer using Medication more than one year old and patient no longer taking Medication allergy/reaction unclear Hospital/retail medication dose/frequency mismatch Over-the-counter/herbal medication missing from medication list Other
Appendix B—Pharmacy interventions and recommendationsa,b Intervention per Protocol Add drug to medication list Remove drug from medication list Contact provider/team Modify dose/frequency Discontinue drug Clarify/add drug allergy/reaction Educational intervention Remove DME from medication list Other
Recommendation Add drug to medication list Remove drug from medication list Add indication for drug Clarify dose/frequency/dosage form Discontinue retail prescription Dosage adjustment (renal or hepatic) Modify dose/frequency/duration Modify dosage form/route of administration Order laboratory tests/drug levels/diagnostic tests Review drug/allergy/food interactions Social work referral
DME = durable medical equipment. Interventions per protocol were performed independently by the pharmacist. Recommendations were made to the admitting team. a
b
Am J Health-Syst Pharm—Vol 71 May 15, 2014
Copyright of American Journal of Health-System Pharmacy is the property of American Society of Health System Pharmacists 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.
PRACTICE REPORT
Care Transitions Service: A pharmacy-driven program for medication reconciliation through the continuum of care Jessica R. Conklin, John C. Togami, Allison Burnett, Melanie A. Dodd, and Gretchen M. Ray
T
ransitions from one care setting to another are known to be critical junctures that put patients at high risk for experiencing medication errors. Approximately 22% of potentially preventable medication errors occur during admissions, 66% during a transition to or from the intensive care unit, and 12% during discharge.1,2 A reported 38.6% of medication discrepancies that occur in the inpatient setting have the potential to cause moderate-to-severe discomfort or clinical deterioration.3 Medication errors are not isolated to the inpatient setting. A recent study by Haynes et al.4 found that 50.8% of discharged patients experienced one or more clinically important medication errors during the 30 days after hospital discharge. It is suspected that many factors contribute to postdischarge medication errors, including but not limited to nonadherence, potential adverse drug events, physician and system errors, outpatient pharmacy discrepancies, patient safety issues, and discharge
Purpose. A quality-improvement program at University of New Mexico Hospital (UNMH) encompassing admission, discharge, and postdischarge medication reconciliation activities is described, with a report on initial assessments of the program’s impact on rates of medicationrelated problems (MRPs). Methods. Pharmacists conducted a five-month evaluation of the UNMH Care Transitions Service (CTS), which serves inpatients admitted to the hospital’s family medicine service, providing medication reconciliation and targeted MRP interventions. Selected patients who received CTS services from November 2012 through March 2013 (n = 191) were included in the analysis. The study endpoints were the rates and types of MRPs identified, the most commonly implicated medication classes, and predictors of MRPs. Postdischarge MRP rates during a two-month
medication confusion, including patient health literacy concerns.1-5 In 2004, the Institute of Medicine estimated that 90 million adults in the United States may have trouble
Jessica R. Conklin, Pharm.D., PhC, is Visiting Assistant Professor, Department of Pharmacy Practice & Administrative Sciences, University of New Mexico (UNM) College of Pharmacy, Albuquerque; at the time of the study, she was Postgraduate Year 2 (PGY2) Ambulatory Care Pharmacy Resident, UNM College of Pharmacy. John C. Togami, Pharm.D., is PGY2 Ambulatory Care Pharmacy Resident, UNM College of Pharmacy; at the time of the study, he was Postgraduate Year 1 Pharmacy Resident, UNM Hospital, Albuquerque. Allison Burnett, Pharm.D., PhC, is Clinical Pharmacist, UNM Hospital. Melanie A. Dodd, Pharm.D., PhC, BCPS, is Associate Professor; and Gretchen M. Ray, Pharm.D., PhC, BCPS, is Assistant
802
Am J Health-Syst Pharm—Vol 71 May 15, 2014
trial of CTS services at a UNMH outpatient clinic were also evaluated. Results. During the five-month evaluation of inpatient CTS services, a total of 1140 MRPs were identified (an average of 6 per patient), about 70% of which were resolved independently of provider review using pharmacy-driven protocols. During the twomonth pilot test of CTS outpatient services (n = 16), a total of 28 MRPs were identified; in over 80% of cases, there was a decline in the number of MRPs from the admission to the postdischarge medication reconciliation. Conclusion. MRPs were identified through the continuum of care. The majority of MRPs identified in both the inpatient and outpatient settings involved patient variables and patient nonadherence. Seventy percent of inpatient MRPs were resolved independently by the CTS team under pharmacy-driven protocols. Am J Health-Syst Pharm. 2014; 71:802-10
understanding and acting on health information.6 The Newest Vital Sign (NVS), available in both Englishand Spanish-language versions, is a clinical tool for measuring health
Professor, Department of Pharmacy Practice & Administrative Sciences, UNM College of Pharmacy. Address correspondence to Dr. Ray ([email protected]). Dr. Dodd was the chair of the American Society of Health-System Pharmacists Council on Public Policy and has received travel funding to attend council meetings in Bethesda, MD; she is also a director and secretary of the New Mexico Pharmaceutical Care Foundation. The other authors have declared no potential conflicts of interest. Copyright © 2014, American Society of Health-System Pharmacists, Inc. All rights reserved. 1079-2082/14/0502-0802$06.00. DOI 10.2146/ajhp130589
PRACTICE REPORT Care transitions service
literacy with demonstrated reliability (Cronbach’s coefficient alpha, > 0.76). 7 The NVS assesses numeracy and literacy skills by asking a patient to answer six questions about an ice cream nutrition label. Pharmacy-driven models of medication reconciliation during care transitions have been studied.8,9 Pharmacists are uniquely qualified to perform medication reconciliation because of their formal training in obtaining medication histories and extensive medication expertise. Pharmacists have been shown to obtain medication histories that are more accurate and comprehensive than those obtained by other health care professionals.10,11 Also, pharmacists often have training in both hospital and retail pharmacy settings, which can give them an advantage over other health care providers in navigating various medication distribution systems. Medication errors are often evaluated from two distinct perspectives: that of the inpatient/hospital setting and that of the postdischarge/ ambulatory care setting. According to a consensus statement published by Greenwald et al.,12 a “comprehensive reconciliation system is needed across the continuum of care.” To the best of our knowledge, few studies have analyzed medication discrepancies across the continuum of care (defined as the movement of a patient from hospital admission through the postdischarge period). The purpose of the study described in this article was to analyze an ongoing inpatient-focused, pharmacy-driven program—the Care Transitions Service (CTS)—and its impact on identifying and resolving medication-related problems (MRPs). We also evaluated the implementation of a pilot extension of this service into the outpatient setting and its impact on MRP identification and resolution through the continuum of care. Methods Study design. A single-center descriptive study was conducted within
University of New Mexico Hospital (UNMH) and a predetermined outpatient UNMH family medicine clinic. The study was approved by the university’s Human Research Review Committee. Analyses of the inpatient service were conducted over a five-month period (November 2012– March 2013), and analyses of the outpatient service were conducted over a two-month period (January–February 2013). Patients were included in the study if they received medication reconciliation services from CTS during the predefined study periods. CTS description. CTS is an ongoing pharmacy-driven program that provides three phases of inpatient services, with a fourth, pilot phase of services provided in the outpatient clinic setting (Figure 1). CTS services are offered to UNMH family medicine inpatients who are at least 18 years of age and speak English or Spanish. Patients are ineligible for CTS if they do not have an extensive
medication history (defined as at least three long-term medications) or if there are plans for a readmission. Patients who have a previous admission within the last 14 days and already received CTS services, are unwilling to undergo a medication interview, or have been discharged to a skilled nursing facility, long-term care facility, or prison are also ineligible. The outpatient extension of the service was offered to all patients who received inpatient CTS services, were scheduled for a postdischarge outpatient family medicine clinic appointment, and were independently able to answer questions about their medications. Phase 1 of the program provides admission medication reconciliation services in addition to those performed by a nonpharmacist provider. Phase 2 provides a discharge medication review and a hospitalto-community pharmacist handoff, which involves obtaining consent
Figure 1. Flow chart of University of New Mexico Hospital’s Care Transitions Service program.
Phase 1 Admission medication reconciliation (inpatient)
Phase 4 Follow-up appointment medication reconciliation (outpatient)
Phase 2 Discharge medication review and hospitalto-community pharmacist handoff (inpatient)
Phase 3 Follow-up phone call reminder (inpatient)
Am J Health-Syst Pharm—Vol 71 May 15, 2014
803
PRACTICE REPORT Care transitions service
from a family medicine provider for CTS staff to call the patient’s community pharmacy and authorize discontinuation of any prescriptions no longer needed after hospital discharge. Phase 3 provides a followup phone call within 72 hours of discharge to assess adherence to and tolerance of discharge medications and to remind patients to take all medication bottles to their follow-up appointment. Phase 4 (the pilot extension of the service) consisted of an outpatient medication reconciliation conducted by a pharmacist at the first postdischarge appointment. Program phases 1–3 were conducted by a CTS team comprising a lead inpatient pharmacist, four postgraduate year 1 (PGY1) pharmacy residents, and six advanced pharmacy practice experience (APPE) clerkship pharmacy students. Phase 4 was conducted by a lead pharmacist and a postgraduate year 2 ambulatory care pharmacy resident. The APPE students and PGY1 residents were provided extensive training to ensure consistency of CTS processes. All members of the CTS team were responsible for identification and documentation of MRPs. All documentation was summarized in a written clinical note and posted in the electronic medical record (EMR). All APPE students were supervised by the PGY1 residents as MRPs were identified and documented. Weekly quality checks were completed by the lead pharmacist. Data collection. MRPs were identified and documented by a CTS member during the admission reconciliation process. Each MRP was classified only once. Interventions were performed independently of provider review according to protocols agreed on by the pharmacy department and inpatient family medicine service. Recommendations requiring additional provider action or review were communicated to the responsible family medicine provider. Multiple methods were used to 804
communicate with providers, including phone calls, pages, and alerts in the EMR. Additional MRPs identified in the outpatient setting were documented, including medication continuance despite discontinuation orders at discharge and patient selfdiscontinuation of medications despite continuation orders at discharge. During the outpatient visit, each patient’s NVS score was assessed and recorded. Definitions. High-alert medications were identified based on the currently available list maintained by the Institute for Safe Medication Practices.13 Medications requiring monitoring were defined as any medication with an associated laboratory value that is used for therapeutic monitoring. The MRP classifications were based on the Health Resources and Services Administration classification scheme (Appendix A).14 Pharmacy interventions and recommendations in response to MRPs are listed in Appendix B. Study outcomes. The primary endpoint was to evaluate the numbers and types of MRPs identified by pharmacists at hospital admission (CTS phase 1). Secondary endpoints for the inpatient service evaluation included the medication classes most commonly associated with MRPs and patient-specific predictors of MRPs. For the evaluation of outpatient services, secondary endpoints included the types and frequencies of new or persistent MRPs occurring from hospital discharge to the first follow-up appointment and the association of health literacy and MRPs among CTS patients. Statistical analysis. Data analyses were conducted using SPSS, version 19.0 (IBM Corporation, Armonk, NY) and Microsoft Access (Microsoft Corporation, Redmond, WA). Descriptive analyses were conducted to characterize the types and rates of medication reconciliation discrepancies between the inpatient and
Am J Health-Syst Pharm—Vol 71 May 15, 2014
outpatient settings and the health literacy of CTS patients. Univariate Poisson regression models were used to determine variables that would be retained for testing in the final model, and an a priori level of significance of 0.2 was chosen in order to prevent the elimination of variables too soon. A multivariate Poisson regression model with an a priori level of significance of 0.05 was used as the final model to determine patient-specific predictors of MRPs. Results Baseline characteristics. Baseline characteristics of patients included in the evaluated inpatient and outpatient CTS samples are listed in Table 1. Inpatient CTS sample. A total of 191 patients received admission medication reconciliation (CTS phase 1) over the five-month inpatient evaluation period; their mean age was 61 years, and 48% were male. Approximately 50% of patients were of Hispanic/Latino ethnicity, which is similar to the ethnic mix of the general population in New Mexico. Ninety-two patients (48.2%) received discharge medication review and hospital-to-community pharmacist handoff (CTS phase 2), and 46 patients (24.1%) received a follow-up phone call (CTS phase 3). Outpatient CTS sample. Thirty patients were eligible for phase 4 of the CTS during the two-month pilot testing period. Seven of the 30 patients failed to attend their scheduled postdischarge appointment. Three patients were ineligible for the service because they were unable to independently answer questions. One patient deferred the service. Missed opportunity, due to external factors (e.g., unavailability of the pharmacist to perform a medication reconciliation), occurred in 3 patients. Therefore, a total of 16 patients received outpatient CTS services and were included in the analysis. Those patients had a mean age of 57.8 years, 56%
PRACTICE REPORT Care transitions service
were male, and 50% were Hispanic/ Latino. Medication reconciliation sources. Inpatient CTS sample. One hundred forty-one patients (73%) had their medication list reconciled via three sources (EMR, patient interview, and outpatient pharmacy data). Thirtynine patients (20%) had their medication list reconciled via two sources (EMR and outpatient pharmacy data). The remainder of the patients had only one source available to use for reconciliation. The mean time to complete the phase 1 admission medication reconciliation was 45 minutes. Outpatient CTS sample. Adherence barriers to therapy (cognitive, physical, financial, or transportation factors or patient beliefs) were identified in 62.5% of patients. Approximately 32% of patients in the CTS phase 4 evaluation were receiving a high-alert medication; no patients were taking medications that required monitoring. Summary data on the results of inpatient and outpatient medication reconciliation by the CTS team are shown in Table 2. Identified MRPs. Inpatient CTS sample. A total of 1140 MRPs were identified during admission medication reconciliation (CTS phase 1). The mean number of MRPs identified per patient was 6. Forty-nine percent of MRPs were attributable to patient variables and patient nonadherence, with 25%, 21%, and 5% attributable to miscellaneous, safety, and provider variables (issues of appropriateness or effectiveness), respectively. The medication class most commonly associated with MRPs was cardiovascular agents (22%), followed by gastrointestinal–genitourinary, central nervous system (CNS), analgesic, and endocrine–metabolic agents. These classes comprised nearly 70% of all medications associated with MRPs. Outpatient CTS sample. Overall, the number of MRPs identified, per
Table 1.
Baseline Patient Characteristicsa Characteristic
No. (%)b,c
Inpatients (n = 191) Mean age, yr Male Languaged English Spanish Ethnicity White, non-Hispanic Hispanic/Latino African American Native American Other Health insurance Private Public County indigent None Marital status Single Married Divorced Widowed Living with partner Source of admission Home Caregiver’s home LTCF/SNF Other Median ± S.D. Charlson Comorbidity Index score Outpatients (n = 16) Mean age, yr Male Language English Spanish Ethnicity White, non-Hispanic Hispanic/Latino African American Native American Health insurance Private Public County indigent None
61.0 92 (48) 168 (88) 21 (11) 61 (32) 95 (50) 10 (5) 15 (8) 10 (5) 39 (20) 71 (37) 64 (33) 17 (9) 79 (41) 54 (28) 28 (15) 28 (15) 4 (2) 175 (92) 4 (2) 7 (4) 5 (3) 2.0 ± 1.8 57.8 9 (56) 14 (87) 2 (12) 6 (37) 8 (50) 1 (6) 1 (6) 5 (31) 4 (25) 6 (37) 1 (6)
LTCF = long-term care facility, SNF = skilled nursing facility. All data are no. (%) unless specified otherwise. c For some characteristics, percentages do not sum to 100 due to rounding. d Not reported by two patients. a
b
Am J Health-Syst Pharm—Vol 71 May 15, 2014
805
PRACTICE REPORT Care transitions service
patient, from admission medication reconciliation (CTS phase 1) to postdischarge medication reconciliation (phase 4) decreased. A total of 28 MRPs were identified in the phase 4 pilot (range, 0–3 per patient), with an average of 1.56 per patient. The maximum number of MRPs for a single patient was 3 (this occurred in 37.5% of patients), whereas 18.8% of patients had no identified MRP. Approximately 75% of the MRPs identified were related to patient variables and nonadherence. In over 80% of patients, fewer MRPs were identified at the follow-up appointment than had been identified during their admission to the hospital (Figure 2). The specific MRPs were tracked through the continuum of care. Eighty-six percent of all postdischarge MRPs were newly identified MRPs, while 14% had persisted through the continuum of care. Pharmacy interventions and recommendations. A pharmacy intervention or recommendation was documented in response to each MRP. Eight hundred seven
MRPs (70%) were resolved through a pharmacy intervention performed independently of the provider under pharmacy-driven protocols; 43% of pharmacy interventions involved the removal of drugs from the medication list, and 29% involved the addition of drugs to the medication list. The remaining roughly 30% of MRPs required provider follow-up; the majority of these recommendations included recommendations to clarify the patient’s medication regimen (25%), consider removing a drug from the medication list (22%), and add an indication for a prescribed medication (18%). A breakdown of pharmacy interventions per protocol and pharmacy recommendations can be found in Figures 3 and 4, respectively. Predictors of inpatient MRPs. Certain marital status, insurance coverage, and age variables, as well as the presence of medications requiring monitoring and the use of a higher number of medications before CTS phase 1 medication reconciliation, were patient-specific predictors of
Table 2.
Medication Reconciliation Findings Finding Inpatients (n = 191) Adherence barriersa ≥1 None Presence of medication(s) requiring monitoring Yes No Presence of high-alert medication(s) Yes No Mean no. medications prior to medication reconciliation Mean no. medications after medication reconciliation Outpatients (n = 16) Adherence barriersa ≥1 None Presence of high-alert medication(s) Yes No Includes cognitive, physical, transportation, financial, and patient-belief barriers.
a
806
Am J Health-Syst Pharm—Vol 71 May 15, 2014
No. (%)
111 (58.1) 80 (41.9) 28 (14.7) 163 (85.3) 69 (36.1) 122 (63.9) 12 11
10 (62.5) 6 (37.5) 5 (31.2) 11 (68.8)
MRPs (Table 3). Patients with medications requiring drug monitoring had a 35% higher incidence rate of MRPs than patients not prescribed such medications (incidence rate ratio [IRR], 1.35; 95% confidence interval [CI], 1.15–1.159). For each additional medication above three on a patient’s medication list, there was a 6% increase in the incidence rate of MRPs (IRR, 1.06; 95% CI, 1.05–1.07). Patients having no insurance (IRR, 1.48; 95% CI, 1.18–1.85) or private insurance (IRR, 1.20; 95% CI, 1.03–1.41) had 48% and 20% higher MRP incidence rates, respectively, than patients enrolled in the UNM Care indigent-care program. There was a 1% decrease in the MRP incidence rate for every one-year increase in age (IRR, 0.99; 95% CI, 0.98–0.99). Health literacy. The majority of patients (69%) who completed the NVS health literacy assessment during the phase 4 outpatient service evaluation were found to have a high likelihood of limited health literacy (defined as an NVS score of 0 or 1). Thirty-one percent of patients were found to have adequate health literacy (an NVS score or 4 or 5). MRPs were identified in patients with limited health literacy as well as those with adequate health literacy. Discussion This single-center descriptive study of an ongoing pharmacydriven inpatient service and pilot outpatient service demonstrated the need for medication reconciliation at each transition of care. The 1140 MRPs identified in CTS phase 1 were found after the admitting team had already conducted medication reconciliation as part of routine care, indicating the need for a pharmacist at the initial admission medication reconciliation. Several factors contributing to MRPs identified during CTS phase 1 were identified, which in the future may help to identify the patients most in need of a care tran-
PRACTICE REPORT Care transitions service
sitions service. Our study found that for each additional medication above three on a patient’s medication list, there was a 6% increase in the MRP rate, suggesting the total number of medications can be used as a means to identify patients at highest risk.15,16 We found a 35% higher rate of MRPs in patients receiving medications that required monitoring, indicating another group with a greater need for CTS-type services. In addition, patients with either no insurance or private insurance were found to have 48% and 20% higher rates of MRPs, respectively, relative to patients in the UNM Care program, which provides discounted medications to New Mexico indigent patients. Because the UNM Care program is specific to UNMH, these findings could be explained by
prescriber awareness of and familiarity with the UNM Care formulary as opposed to differing formularies of private insurance carriers. We also found that age appeared to have a protective effect against MRPs, which was not expected and differed from the findings in some published studies. 16-18 Pippens et al. 19 and Kripalani et al.,18 however, found that elderly patients were at lower risk for potential adverse drug events identified during both inpatient and postdischarge medication reconciliation; this effect persisted when adjusted for multiple patient characteristics and could be explained by the wide age range (21–99 years) of patients included in the studies or, perhaps, the greater diligence of outpatient providers in maintaining accurate medication lists in older patients. In
our study, medication classes such as cardiovascular, gastrointestinal– genitourinary, and CNS agents were frequently involved in MRPs, which was consistent with the findings in other studies.20-22 There were a number of important limitations to our study. First, patients readmitted to the hospital within 14 days did not receive inpatient CTS services because they received those services during the prior hospitalization; thus, we were unable to discern if any of these readmissions were due to an MRP. Second, CTS services are offered only to patients on the UNMH inpatient family medicine service, so these results only represent a small subgroup of the patient population within a large academic teaching hospital. Third, the CTS is an ongoing service staffed
Figure 2. Number of medication-related problems (MRPs) per Care Transitions Service (CTS) patient (n = 16) at admission medication reconciliation (green bars) and at the first postdischarge follow-up visit (black bars) during a two-month evaluation of CTS outpatient services.
Patient No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 No. 8 No. 9 No. 10 No. 11 No. 12 No. 13 No. 14 No. 15 No. 16 0
2
4
6
8
10
12
14
16
18
20
No. MRPs
Am J Health-Syst Pharm—Vol 71 May 15, 2014
807
PRACTICE REPORT Care transitions service
Figure 3. Pharmacy interventions per protocol conducted by the Care Transitions Service (CTS) team, by type. A total of 1140 medicationrelated problems among 191 patients were identified during the five-month evaluation of CTS inpatient services. DME = durable medical equipment. 400
350
No. Interventions
350 300 231
250 200 150 100
62
50
39
50
28
22
13
12
n re
r/t ea
ct
,a
lle
pr
rg
ov i
y/
de
in ue nt co
ad
d
C
dr
on
ug
ta
D
y/
D
la
rif
ov e
C
R
R
ac tio
m
ug dr
lis t n is
ed m m
M
M
E
od
fro
io
ify
na
do
li
se
nt
/fr
er
eq
ic at io
ue
th e
ve n
O
tio
n
lis t n at io at Ed
em
em
ov e
Ad
d
dr
ug
dr
uc
ug
fro
to
m
m
m
ed
ed
ic
ic
at io
n
lis t
r
nc y
0
Intervention Type
Figure 4. Pharmacy recommendations, by type, made by the Care Transitions Service (CTS) team during the five-month evaluation of CTS inpatient services. The team made a total of 333 recommendations targeting 1140 identified medication-related problems. 90
82 73
No. Recommendations
80 70
60
60
47
50 40 30
20
20
19 11
10
8
6
4
3
Re
m
ov e
dr u
Cl
g
ar
ify
do s
e do /fre q s ag ue fro e nc m fo y/ m rm ed ica Ad tio n d lis in di t ca Ad tio n d fo dr rd ug ru to g m ed ica tio n lis Di M t od sc on ify tin re ue gi m re en ta il p re sc rip tio Re n vie w in te ra So ct io cia n lw or k O re rd fe er rra la l bs /le M od ve ls/ ify te do st sa s ge fo rm /ro Do ut sa e ge ad ju st m en t
0
808
Recommendation Type
Am J Health-Syst Pharm—Vol 71 May 15, 2014
PRACTICE REPORT Care transitions service
by multiple pharmacists, which can lead to inconsistencies in data collection. However, a standardized database was used and all CTS pharmacists received training on MRP classification in the database. Among other study limitations, patient and caregiver recall and reporting of medications during the interview could have introduced bias into our results; however, the majority of medication lists (73%) were reconciled using three sources, minimizing the impact of this potential bias. Due to the short time frame of this analysis, we were unable to classify MRPs based on their potential to cause harm (i.e., clinically significant discrepancies) or to assess the clinical impact of the identified MRPs on endpoints such as readmission rates, patient outcomes, or health care utilization. This is an important area of future research through the continuum of care. It must also be emphasized that the outpatient service (CTS phase 4) was a small pilot extension of the ongoing inpatient service. Because only patients discharged to a predetermined family medicine clinic during the two-month pilot testing period were eligible for the outpatient service, the sample size was very small, which limited the statistical analyses; however, we still found a clinically significant number of MRPs in the outpatient setting at the first postdischarge appointment. Moreover, it is important to note that the interventions made by the CTS team among inpatients were associated with an overall reduction in the number of MRPs identified in the outpatient setting (Figure 2). Despite this, new MRPs that occurred during the time between discharge and the postdischarge appointment were identified. The presence of these newly identified MRPs revealed the need for a pharmacist to complete the final medication reconciliation in the outpatient clinic setting and thus enable improved MRP identification
and resolution across the continuum of care. Additionally, only patients who had received inpatient CTS services were eligible to receive outpatient CTS services at the clinic. A health literacy assessment was conducted for all patients targeted for CTS phase 4 services; however, due to the small sample size, we were unable to identify an association between health literacy scores and the presence of MRPs. We identified MRPs in patients with limited health literacy as well as those with adequate health literacy. Future studies are needed to evaluate associations of health literacy and MRPs. Lastly, it is important to note that some patients also received CTS inpatient phase 2 services (discharge
medication review and hospital-tocommunity pharmacist handoff ) and phase 3 services (postdischarge follow-up phone contact), which may have also reduced the number of MRPs identified, further supporting the need for medication reconciliation at several points throughout the continuum of care. The majority of MRPs identified in both the inpatient and outpatient settings involved patient variables and patient nonadherence. Seventy percent of inpatient MRPs were resolved independently by the CTS team as a result of pharmacy-driven protocols. Despite identification and resolution of MRPs during hospital admission, new MRPs developed between the time of discharge and the first follow-up appointment. How-
Table 3.
Predictors of Inpatient Medication-Related Problemsa Characteristic Marital status Widowed Living with partner Divorced Married Singleb Health insurance None Private Public (Medicare/Medicaid) Indigent (UNM Care)b Adherence barriers ≥1 Noneb Presence of high-alert medication(s) on admission reconciliation list Yes Nob Age Presence of medication(s) requiring monitoring on admission reconciliation list Yes Nob No. medications prior to medication reconciliation
Incident Rate Ratio (95% CI) 1.17 (0.95–1.43) 0.94 (0.63–1.40) 0.74 (0.60–0.92)c 1.28 (1.11–1.48)c . . .d 1.48 (1.18–1.85)c 1.20 (1.03–1.41)c 0.98 (0.84–1.14) . . . 1.09 (0.96–1.24) . . .
0.90 (0.79–1.03) . . . 0.99 (0.98–0.99)c
1.35 (1.15–1.159)c . . . 1.06 (1.05–1.07)c
UNM = University of New Mexico. Reference condition. c Statistically significant (p ≤ 0.05). d Not applicable. a
b
Am J Health-Syst Pharm—Vol 71 May 15, 2014
809
PRACTICE REPORT Care transitions service
ever, in the majority of patients, the admission medication reconciliation (phase 1) was associated with a lower number of MRPs after transition to the outpatient setting. Conclusion MRPs were identified through the continuum of care. The majority of MRPs identified in both the inpatient and outpatient settings involved patient variables and patient nonadherence. Seventy percent of inpatient MRPs were resolved independently by the CTS team under pharmacydriven protocols.
iation patient centered, clinically relevant and implementable: a consensus statement on key principles and necessary first steps. J Hosp Med. 2010; 5:477-85. Institute for Safe Medication Practices. ISMP’s list of high-alert medications. www.ismp.org (accessed 2012 Jun 15). Chen S, for Healthcare Communities. Medication therapy intervention and safety documentation form (version 8, 9/12/11). Available with registration at http://healthcarecommunities.org/ Comm_DocsDetails.aspx?groupid= 2&id=12345&tid=81321 (accessed 2011 Oct 12). Matsen Picone D, Titler MG, Dochterman J et al. Predictors of medication errors among elderly hospitalized patients. Am J Med Qual. 2008; 23:115-27. Tschantz Unroe K, Pfeiffenberger T, Riegelhaupt S et al. Inpatient medication reconciliation at admission and discharge: a retrospective cohort study of age and other risk factors for medication discrepancies. Am J Geriatr Pharmacother. 2010; 8:115-26. Bedell SE, Jabbour S, Goldberg R et al. Discrepancies in the use of medications, their extent and predictors in an outpatient practice. Arch Intern Med. 2000; 160:2129-34. Kripalani S, Price M, Vigil V et al. Frequency and predictors of prescriptionrelated issues after hospital discharge. J Hosp Med. 2008; 3:12-9. Pippens JR, Gandhi TK, Hamann C et al. Classifying and predicting errors of inpatient medication reconciliation. J Gen Intern Med. 2008; 23:1414-22. Steurbaut S, Leemans L, Leysen T et al. Medication history reconciliation by clinical pharmacists in elderly inpatients admitted from home or a nursing home. Ann Pharmacother. 2010; 44:1596-603. Gleason KM, McDaniel MR, Feinglass J et al. Results of the Medication At Transitions and Clinical Handoffs (MATCH) study: an analysis of medication reconciliation errors and risk factors at hospital admission. J Gen Intern Med. 2010; 25:441-7.
13. 14.
15.
16.
References 1. National Quality Forum. Preventing medication errors. www.qualityforum.org/ N P P / . . . / P r e v e n t i n g _ Me d i c a t i o n _ Errors (accessed 2012 Jun 15). 2. Santell JP. Reconciliation failures lead to medication errors. Jt Comm J Qual Patient Saf. 2006; 32:225-9. 3. Cornish PL, Knowles SR, Marchesano R et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005; 165:424-9. 4. Haynes KT, Oberne A, Cawthon C et al. Pharmacists’ recommendations to improve care transitions. Ann Pharmacother. 2012; 46:1152-9. 5. Davis TC, Wolf MS, Bass PF III et al. Literacy and misunderstanding prescription drug labels. Ann Intern Med. 2006; 145:887-94. 6. Institute of Medicine. Health literacy: a prescription to end confusion. In Nielsen-Bohlman L, Panzer A, Kindig DA, eds. Washington, DC: National Academies Press; 2004:1-17. 7. Weiss BD, Maysm MZ, Martz W et al. Quick assessment of literacy in primary care: the Newest Vital Sign. Ann Fam Med. 2005; 3:514-22. 8. Kwan JL, Lo L, Sampson M et al. Medication reconciliation during transitions of care as a patient safety strategy: a systematic review. Ann Intern Med. 2013; 158:397-403. 9. Mueller SK, Cunningham Sponsler K, Kripalani S et al. Hospital-based medication reconciliation practices: a systematic review. Arch Intern Med. 2012; 172:105769. 10. Reeder TA, Mutnick A. Pharmacistversus physician-obtained medication histories. Am J Health-Syst Pharm. 2008; 65:857-60. 11. Nester TM, Hale LS. Effectiveness of a pharmacist-acquired medication history in promoting patient safety. Am J HealthSyst Pharm. 2002; 59:402-5. 12. Greenwald JL, Halasyamani L, Greene J et al. Making inpatient medication reconcil-
810
17.
18.
19.
20.
21.
22. Quélennec B, Beretz L, Paya D et al. Potential clinical impact of medication discrepancies at hospital admission. Eur J Intern Med. 2013; 24:530-5.
Appendix A—Medication-related problem classifications14 Appropriateness/effectiveness Inadequate dosing regimen Monitoring standards not followed Untreated medical problem Home medication not resumed Safety Allergy/food/drug interaction Contraindication Dosage adjustment needed (renal or hepatic) Drug dosing excessive for treatment Incorrect dose/frequency Intended use differs from actual use Laboratory test indicated, not ordered Missing dose/frequency No indication for drug prescribed Therapy duplication Patient variables/nonadherence Adherence divergence Dosage form/route inappropriate Medication continued despite discontinuation orders Patient self-discontinuation despite continuation orders Patient not taking medication due to financial reasons Medication on medication list, patient no longer taking Medication not on medication list, patient taking Miscellaneous Durable medical equipment more than two years old and patient no longer using Medication more than one year old and patient no longer taking Medication allergy/reaction unclear Hospital/retail medication dose/frequency mismatch Over-the-counter/herbal medication missing from medication list Other
Appendix B—Pharmacy interventions and recommendationsa,b Intervention per Protocol Add drug to medication list Remove drug from medication list Contact provider/team Modify dose/frequency Discontinue drug Clarify/add drug allergy/reaction Educational intervention Remove DME from medication list Other
Recommendation Add drug to medication list Remove drug from medication list Add indication for drug Clarify dose/frequency/dosage form Discontinue retail prescription Dosage adjustment (renal or hepatic) Modify dose/frequency/duration Modify dosage form/route of administration Order laboratory tests/drug levels/diagnostic tests Review drug/allergy/food interactions Social work referral
DME = durable medical equipment. Interventions per protocol were performed independently by the pharmacist. Recommendations were made to the admitting team. a
b
Am J Health-Syst Pharm—Vol 71 May 15, 2014
Copyright of American Journal of Health-System Pharmacy is the property of American Society of Health System Pharmacists 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.
![[Medication reconciliation in primary care after hospital discharge].](/assets/img/hold.png)
