Article, Emergency Medicine

How reliable are patient-completed medication reconciliation forms compared with pharmacy lists?

Unlabelled imageAmerican Journal of Emergency Medicine (2012) 30, 1048-1054

Original Contribution

How reliable are patient-completed medication reconciliation forms compared with pharmacy lists?

Carolyn Meyer MD a, Michael Stern MD a, Wendy Woolley DO a,

Rebecca Jeanmonod MD b,?, Donald Jeanmonod MD b

aAlbany Medical Center, Albany, NY 12208, USA

bSt. Luke’s Hospital and Health Network, Bethlehem, PA 18015, USA

Received 19 April 2011; revised 27 June 2011; accepted 30 June 2011

Abstract

Objectives: Medication reconciliation is a Joint Commission for the Accreditation of Healthcare organizations requirement to reduce Medication errors. This study evaluated the reliability of patient- completed medication reconciliation forms (MRs) compared with pharmacy-generated lists and determined if there was a difference in concordance when patients completed the forms from memory compared with when they brought a separate list or pill bottles.

Methods: We prospectively enrolled patients with completed MRs. Research assistants contacted the patient’s pharmacy to determine medications filled in the prior 3 months, which was compared with the MR. Discrepancies and the method by which the patient completed the MR (memory, list, or pill bottles) were recorded.

Results: Three hundred fifteen patients were enrolled. Thirty-three percent made errors of omission (reported by pharmacy, but not on MR), 12.7% made errors of addition (reported on MR, but not by pharmacy), 18.1% made both types of errors, and 36.3% made no errors. Patients with errors were on

5.6 medications compared with 3.6 medications for those without errors (P b .0001). Those completing the MR from a list made 2.3 errors compared with 1.2 for those completing from memory and 1.8 for those completing from their pill bottles (P b .001). Of 390 medications omitted from patient lists, 16% were cardiac medications, 13% were neuropsychiatric agents, and 9.5% were narcotics.

Conclusions: Thirty-six percent of patients were able to provide a medication list that matched their pharmacy-prescribed drugs. More errors were noted from patients taking more medications and from those completing their MR from a separate list.

(C) 2012

Introduction

Medication reconciliation refers to the process of comparing a patient’s Medication orders to all of the medications that he or she has been taking. The 2005 National Patient Safety Goals (NPSG) included medication

* Corresponding author. Tel.: +1 610 954 1102.

E-mail address: [email protected] (R. Jeanmonod).

reconciliation as a Joint Commission for the Accreditation of Healthcare Organizations (JCAHO) requirement to reduce medication errors and adverse patient events in health care systems [1]. The JCAHO’s sentinel event database includes more than 350 medication errors resulting in death or Major injury. Effective medication reconciliation may have avoided more than half of the errors that were the result of communication breakdown [2]. Despite the implementation of medication reconciliation as an NPSG in 2005, medication

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.06.038

errors continue to be the fifth most frequent cause of preventable Patient harm [3].

The JCAHO recognizes that many institutions have struggled with the development of an efficient process to meet the goal of medication reconciliation, and it plans to continue to evaluate and further refine this NPSG [4]. As currently described, however, proper medication reconcili- ation is a 5-step process that includes developing a current list of medications being taken, a list of medications to be prescribed, comparison of the medications on the 2 lists with Clinical decisions made based on that comparison, and communication of any changes to the patient and subsequent caregivers [2]. Because the emergency department (ED) is often the point of entry for patients into the hospital, the medication history obtained in the ED becomes a vital part of the medication reconciliation process. There are many challenges to gathering an accurate medication history in this setting; and often, we depend on the information provided by the patients and/or their family.

In the study institution, all patients are asked to complete a medication reconciliation form (MR) when they arrive in the ED to complete the first step in the medication reconciliation process described above. In this study, we evaluated whether patient-provided MR forms agree with filled pharmacy prescriptions as reported by the patient’s pharmacy. We also sought to determine whether there is a difference in concordance to the pharmacy when patients report their medications from memory compared with those who use a medication list or bring their pill bottles from home. Our hypothesis was that the MR forms would be significantly discordant from the pharmacy-filled prescription list and that patients who bring a medication list or their pill bottles with them would be more accurate with their medications than those who completed the MR from memory.

Methods

Study design

This is a prospective observational cohort study that was reviewed and approved by the institutional review board.

Study setting and population

The study was performed in a tertiary care ED with an annual census of 75 000.

Study protocol

Researchers prospectively enrolled a convenience sample of patients who had completed their MR forms. All data were obtained by research assistants (RAs) who were trained in the approach to patient interview, pharmacy interview, and completion of the data collection form. After obtaining

informed consent, the RAs contacted the patients’ pharmacies for a list of medications filled in the preceding 3 months. The pharmacies that were contacted were the pharmacies listed by the patient as the location in which he or she filled prescriptions. The RAs transcribed the medications listed over the phone from the pharmacy and then compared this list with the completed MR. Any discrepancies were noted. It is usual practice in the department that patients complete their own MR when possible; and if not possible, this is completed by family or nursing staff. The choice of whether to fill out the MR from a list, bottles, or memory was the choice of the patient or family. Patients were excluded from the study if they did not complete an MR, if they did not speak English, or if their medical status precluded their ability to consent. In addition, patients for whom their pharmacies were unavail- able were excluded.

Key outcome measures

Discrepancies between the lists were classified as errors of “omission” if the pharmacy reported a medication that was not listed on the MR or “addition” if the MR contained drugs not reported by the pharmacy. We did not include any reported over-the-counter medications or herbal or homeo- pathic remedies as errors of addition, as there was no way to verify these medications with the pharmacy. The age and sex of the patient were recorded, as was the method by which the patient completed the MR (memory, list, or pill bottles). Researchers also collected the total number of medications reported and type of medications omitted.

Data analysis

Data were analyzed using descriptive statistics, t test, analysis of variance, and simple and Multivariate linear regression analyses.

Results

A total of 484 patients were screened for the study, of whom 315 were enrolled. Of those patients screened that we did not enroll, 31 patients declined to participate and 138 had incomplete data sets because their pharmacies were not available (Fig. 1). The distribution of days and times that data were collected is presented in Table 1.

Of the 315 patients enrolled, 213 (67.6%) completed their MR from memory, 65 (20.6%) completed their MR from a list, and 29 (9.2%) completed their MR from pill bottles brought to the ED with them. The remaining 8 (2.5%) used another unspecified method. One hundred four (33%) of 315 MRs completed contained errors of omission, 40 (12.7%) of

315 contained errors of addition, 57 (18.1%) of 315 contained both types of errors, and 114 (36.2%) of 315 contained no errors (Fig. 2). Patients with errors (Table 2)

Fig. 1 Patient population.

reported taking a mean of 5.6 medications compared with 3.6 medications for those who did not have errors (P b .0001). The patients completing the MR from memory reported taking a mean of 3.3 medications, and their MR had a mean of 1.2 errors per patient. Eighty-seven patients completing their MR from memory had no errors (41%). The patients completing the MR from a list reported taking a mean of 9.2 medications and made a mean of 2.3 errors per patient. Eighteen patients completing their MR from a list made no errors (28%). The patients completing the MR from their pill bottles reported taking a mean of 6.4 medications and made a mean of 1.8 errors per patient. Nine patients completing their MR from pill bottles made no errors (31%) (Table 3). The differences between errors based on method of completion were statistically different

(P = .0001, analysis of variance).

Multiple linear regression was used to assess for correlation between method of list completion, age, and number of reported medications taken with observed number of errors. The number of medications patients reported taking had a modest relationship with number of errors (r2 = 0.17; range of reported medications, 1-20; range of errors, 0-10). Method of list completion also had an independent relationship with number of errors (r2 = 0.17), with completion from a separate list resulting in more errors

Table 1 Data collection days and times

n

day of the week

%

Weekday Weekend Shift time

7:00 AM to 2:59 PM

3:00 PM to 10:59 PM

11:00 PM to 06:59 AM

Unrecorded

273

42

86.7

13.3

165

138

3

9

52.4

43.8

0.9

2.9

than completion from pill bottles, which resulted in more errors than lists created from memory. Once number of medications and method of list completion were accounted for, age had no further impact on number of errors.

One hundred sixty-one patients omitted a total of 390 medications from their MR. Sixteen percent of these were cardiac medications. Forty-three (11.1%) of the 390 omissions were Antihypertensive medications, and 5 (1.3%) of the 390 omissions were other cardiac medications such as antiarrhyth- mic medications. Six (1.5%) of the 390 omissions were Antiplatelet agents or anticoagulants such as aspirin, clopido- grel, warfarin, or heparin. Fifty-two (13%) of the 390 omissions were neuropsychiatric agents. Seventeen percent of the omitted medications were analgesics. Thirty-seven

Fig. 2 Types of error.

No. of patients (n) Male sex, %

Age, y, median (IQR)

No. of pharmacies used, % 1

2

3

List completion method, % Memory

List

Pill bottles Other

No. of medications reportedly taken by patient, mean

Difference in number of medications between

patient and pharmacy report, median (IQR) [range]

315

104

40

57

114

49.5

47.1

52.5

43.9

53.5

49 (38-61)

46 (38-61)

52 (40-63)

50 (41-63)

48 (32-60)

84.2

84.6

70

80.7

91.2

13.0

12.5

25

15.8

7.9

2.2

2.9

2.5

3.5

0.9

67.6

68.3

50

61.4

76.3

20.6

20.2

27.5

26.3

15.8

9.2

8.7

17.5

7

7.9

2.5

2.9

5

5.3

0

4.9

4.4

6.7

7.1

3.6

-2 (-3 to -1)

1.5 (1-3)

0 (-1 to 2)

[-9 to -1]

[1-10]

[-9 to 8]

IQR indicates interquartile range.

(9.5%) of the 390 omissions were narcotics. Twenty-three (6%) of the 390 omissions were nonsteroidal anti-inflamma- tory drugs, and 6 (1.5%) of the 390 omissions were Muscle relaxants.

Table 2 Patient characteristics associated with error type

Characteristic Total

Error type

Omission

Addition

Both

None

Ninety-eight patients added a total of 200 additions to their Medication lists. Twenty-five percent of these were cardiac medications. Thirty-three (17%) of the 200 additions were antihypertensive medications, and 16 (8%) of the 200 additions were cholesterol-lowering medications. Nineteen (11%) of the 200 additions were antiplatelet agents or anticoagulants such as aspirin, clopidogrel, warfarin, and heparin. Thirty (15%) of the 200 additions were neuropsychi- atric agents. Fifteen percent of the additions were analgesics.

Seventeen (9%) of the 200 additions were narcotics. Nine (5%) of the 200 additions were nonsteroidal anti-inflammatory drugs, and 2 (1%) of the 200 additions were muscle relaxants (Fig. 3).

Discussion

The current knowledge regarding the accuracy of medication orders as they compare to the medications the patient has been taking at home is consistent. One study done in the Baystate Medical Center Emergency Department found that 87% of the medication lists that they obtained

Table 3 Patient characteristics associated with list completion method

Characteristic

Total

Completion method

Memory

List

Pill bottles

No. of patients, n

315

213

65

29

Male sex, %

49.5

47.9

60

37.9

Age, y, median (IQR)

49 (38-61)

44 (31-56)

64 (53-70)

51 (39-56)

No. of pharmacies used, %

1

84.2

84.5

84.6

82.8

2

13.0

12.7

13.8

13.8

3

2.2

2.3

1.5

3.4

Error type, %

Omission

33

33.3

32.3

31

Addition

12.7

9.4

16.9

24.1

Both

18.1

16.4

23.1

13.8

None

36.2

40.8

35.7

31

No. of medications reportedly

4.9

3.3

9.2

6.8

taken by patient, mean

Fig. 3 Types of medications associated with errors.

from their patients older than 64 years had at least one error [5]. In addition, a recently published study done by Mazer et al [6] found a 37% discrepancy rate between the medication list generated during triage and an intentional medication history taken after triage. This study, however, used the patients’ verbal response as the standard for medications taken and did not verify the information with outpatient or pharmacy records as in our study or the Baystate study previously mentioned.

Medication errors on admission to the hospital can reduce the quality of treatment and can lead to adverse events [7].A study conducted in 2006 showed that of the Medication discrepancies encountered, 18% of them were potentially serious [8]. Another study showed that there is a significant disagreement between the primary care physicians’ medica- tion records and the information given by the patients at the time of hospitalization. The authors concluded that there is an urgent need for communication improvement between patients and physicians concerning medications being prescribed [9].

We know that the implementation of effective and efficient medication reconciliation has been complex and challenging. The information we gather about a patient’s medications is often incomplete, and the process of obtaining a full medication history results in an increased time demand on providers. Currently, physicians and nurses work together to compile medication orders that are consistent with a patient’s Home medications. It has been shown that pharmacists are underused in the medication reconciliation processes [10,11]. Green et al [12] found significant numbers of discrepancies when they compared medication lists brought by patients to other sources of information, particularly the primary care physicians’ records or their caretakers’ records. This emphasizes the importance of taking a thorough medical history that includes a medication history. It is clear that there is a need to improve the

information ascertained through medical reconciliation at hospital admission [13-16]. The medication history that is taken at the time of first presentation in the ED has been shown to be important for medical reconciliation. Often, the ED list is used as a baseline and continued on admission to the hospital; so its accuracy is critical to avoid errors.

Development of an efficient medication reconciliation process that includes all of the 5 steps defined by the Joint Commission for proper reconciliation is the best way to meet the intent of NPSG 8; however, it remains an ongoing struggle. Several solutions have been proposed and tested, but there has yet to be a method that can reliably obtain the correct information and reduce medication errors. Truitt et al

[17] suggest a Drug history method using a review of systems format to obtain a more accurate drug history. An algorithm that matches medications to the patient’s problem list has been tried [18,19]. The Walter Reed Army Medical Center in Washington, DC, uses a codified medication process that they have shown increases the number of complete drug entries that includes dosing information [20]. Others have tried a discharge survey that involved a review of all the anesthesia records, allergies, and home medications with the patients or their family before the patients were discharged [21]. Even a study using standardized wallet-sized medica- tion lists has been conducted to get patients more involved with their medical care and increase communication [22]. Interestingly, pharmacist-acquired medication histories are shown to have significantly fewer errors in documentation and have more documentation of patient allergies [11,23]. However, the use of pharmacists in the medication reconciliation process is not a standard often used in the ED. Fitzgerald [24] concludes that prescribers need to be educated in taking accurate medication histories and in clinical pharmacology, and it is easy to agree that education is vital for reducing errors [25].

It has been suggested that the appropriate medication reconciliation is a 3-part process [26]. The first process is establishing the database including all prescription, over-the- counter, herbal, and Alternative medications by reviewing medication lists and pill bottles or contacting pharmacies and primary care physicians. In addition, it is recommended that reviewing past admission MR forms and querying patients about medications associated with reported medical prob- lems will assist in formulating a complete list. The next step is validating the information obtained, where questions about dosing are answered and patients are specifically queried about commonly forgotten medications such as neuropsy- chiatric agents, aspirin, vitamins, inhalers, eye drops, insulin, and topical patches. Further validation includes reviewing allergies as to the nature of the allergic response. The final step includes documentation of the information in an unambiguous format for the patient’s medical record.

The degree of involvement that is required for appropriate medication reconciliation often necessitates a time commit- ment and medication knowledge that are not available to ED

triage or ward nurses with significant patient care responsibilities. Therefore, it is natural to assign the job to an individual with intimate knowledge of medications, whose responsibility is to perform medication reconcilia- tion. A number of studies have demonstrated improved medication reconciliation with the use of pharmacists in the ED. In their study of 100 patients studied pre- and postintervention (50 patients each group), Mills and McGuffie [27] found that the reported error rate decreased from 3.3 errors per patient to 0.04 error per patient after introduction of a pharmacist in the ED. The data from this study should be interpreted with caution given the introduction of a formal MR form at the same time as the introduction of the pharmacist, potentially accounting for some of the reported effect. Carter et al [28] found that approximately 25% of MRs completed by ED staff had errors of omission or, more commonly, dosing errors. These findings were also supported by Hayes et al [11], who found that their error rate dropped from 1.7 to 0.3 error per patient with the use of an ED pharmacist.

It is not realistic to think that a single pharmacist can accommodate the volume of patients that pass through the ED. Thomas [26] admits that it is not possible for a pharmacist to be involved in every patient’s medication reconciliation process. In his institution, they use the following criteria to prompt a pharmacist’s review of the medication reconciliation: age greater than 65 years, taking more than 5 medications, likely to be admitted to the hospital, and the patient’s ED nurse requests the medication review. Part of the burden may be alleviated by assigning medication reconciliation to trained pharmacy technicians. In a Canadian ED, trained pharmacy technicians were able to perform medication reconciliation with the accuracy of a pharmacist [29]. Despite best efforts of pharmacists, errors still continue. In the reviewed literature, between 0.04 and

0.3 error or discrepancy per patient is reported despite use of pharmacists [11,27,29]. Finally, in an era of cost containment in health care, one needs to consider the financial implications of having additional salaried professionals involved in patient care.

Our study sought to further define some of the discrepancies encountered in a patient-dependent medication reconciliation process in the ED setting. We found that the minority of patients are able to provide a medication list that matches those medications dispensed by the pharmacy. The medications that were omitted from the lists included agents from a number of different classes that could have a high rate of significant adverse events for a patient if omitted from patient care during a hospitalization. There was also omission of a number of Medication classes that could cause significant adverse events from Drug interactions with those agents that are commonly prescribed during a hospital stay. Interestingly, in our sample, there were more errors noted when the patients were completing their MR from a separate list than when completing the form from memory, even when controlling for number of medications listed. Not

surprisingly, the more medications a patient was taking, the more likely it was that errors would be found.

Limitations

The design of the study places limitations on the results. This study used convenience sampling to collect data. Convenience sampling is a type of nonprobability sampling that includes a sample of the population that is readily available; and because our RAs were available more often during days and evenings, most of our data were collected during those times. It is possible that patients presenting during overnight shifts may represent a different population in terms of the medications they take or the way in which they complete their MR forms, but we were not able to capture these data. In addition, at certain times of day, some pharmacies were less likely to be accessible, which resulted in exclusion of some patients in the late evening or overnight period due to pharmacy business hours.

A single institution conducted this research; and as such, results may not be extrapolated to other institutions with different populations and different systems of medical reconciliation. We based our study on a single system of medical reconciliation (ie, patient completion) used in our institution, and there are many different methods of medical reconciliation currently in use. The errors demonstrated in the current study may not be present to the same degree in systems with electronic medical records where a list of medications from the previous visit is compared with the list of medications on the current visit.

There were limitations of knowledge to this study, as it only compared the reliability of patient-generated informa- tion with pharmacy-distributed medications. We were unable to take into account medications provided as samples, over- the-counter medications, or herbal medications. There was no way to know if a health care provider with no documentation in the pharmacy record had recently verbally discontinued a patient’s medication.

Finally, there is no criterion standard for accuracy of a medication list. We do not have a set of criteria that describe an accurate medication list and, as such, based the accuracy of the patient-generated medication lists on the records held at the pharmacy. Although patients were queried about which pharmacies they used and all reported pharmacies were contacted, it could be that patients failed to disclose the name of a pharmacy that they used. Furthermore, we did not collect information on the dosage of medications; nor did we look for medication interactions within the patient list. We only recorded the medication itself and only defined errors as differences between the pharmacy and patient list. Certainly, had we considered dosing errors, our rates of errors would have been much higher than described. As well, there were occasions when patients’ reported medications could not be matched to any known generic or brand name medications because of spelling, thereby creating a reported error when one may not have existed.

Conclusions

This study provides further information on the types of errors encountered with patient-generated medication recon- ciliation. The minority of patients (36%) were able to provide a medication list that was concordant with the medications dispensed by pharmacies. Approximately half of the patients providing a medication list failed to list at least one medication that had been dispensed by their pharmacy, and these medications came from a wide variety of classes that can cause significant adverse events. This study also shows the need for the conduction of further investigation to find an optimal mode of gathering the information needed to complete proper medication reconciliation for all patients in the ED to reduce the number of preventable medication errors that occur.

References

  1. JCAHO 2005 national patient safety goal 8.
  2. Joint Commission sentinel event alert, issue 35. Accessed 10/27/2010 http:// www.jointaddition.org/sentinel_event_alert_issue_35_using_medication_ reconciliation_to_prevent_errors/.
  3. The Joint Commission. Sentinal event statistics (quarterly). November

30th, 2010 http://www.jointaddition.org/assets/1/18/Stats_with_ all_fields_hidden30September2010_(2).pdf (accessed Novemeber 30th, 2010).

  1. Medication reconciliation national patient safety goal to be reviewed, refined. http://www.jointaddition.org/PatientSafety/NationalPatient- SafetyGoals/npsg8_review.htm.
  2. Caglar S, Henneman PL, Blank FS, et al. Emergency department medication lists are not accurate. J Emerg Med 2008, Epub ahead of print. Available at: http://dx.doi.org/10.1016/j.jemermed.2008.02.060 Accessed 11/5/2010.
  3. Mazer M, Pharm D, DeRoos F, et al. Medication History taking in emergency department triage is inaccurate and incomplete. Acad Emerg Med 2011;18(1):102-4.
  4. Vira T, Colquhoun M, Etchells E. Reconcilable differeces: correcting medication errors at hospital admission and dscharge. Qual Saf Health Care 2006;15(2):122-6.
  5. Larsen MD, Nielsen LP, Jeffery L, et al. Medication errors on hospital admission. Ugeskr Laeger 2006;168(35):2887-90.
  6. Rabol R, Aroe GR, Folke F, et al. Disagreement between physicians’ medication records and information given by patients. Ugeskr Laeger 2006;168(13):1307-10.
  7. Clay BJ, Halasyamani L, Stucky ER, et al. Results of a medication reconciliation survey from the 2006 Society of Hospital Medicine national meeting. J Hosp Med 2008;3(6):465-72.
  8. Hayes BD, Donovan JL, Smith BS, et al. Pharmacist-conducted medication reconciliation in an emergnecy department. Am J Health Syst Pharm 2007;64(16):1720-3.
  9. Green CF, Burgul K, Armstrong DL. A study of the use of medicine lists in medicines recnciliation: please remember this, a list is just a list. Int J Pharm Pract 2010;18(2):116-21.
  10. Frei P, Huber PL, Bonani RW, et al. Insufficient medicaiton documentation at hospital admission of cardiac patients: a challenge for medicaiton reconciliation. J Cardiovasc Pharmacol 2009;54(6): 497-501.
  11. Cornish PL, Knowles SR, Marchesano R, et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med 2005;165(4):424-9.
  12. Lau HS, Florax C, Porsius AJ, et al. The completeness of medication histories in hospital medical records of patients admitted to general internal medicine wards. Br J Clin Pharmacol 2000;49(6): 597-603.
  13. Beers MH, Munekata M, Storrie M. The accuracy of medication histories in the hospital medical records of elderly persons. J Am Geriatr Soc 1990;8(11):1183-7.
  14. Truitt CA, Longe RL, Taylor AT. An evaluation of a medication history method. Drug Intell Clin Pharm 1982;16(7-8):592-6.
  15. Carpenter JD, Gorman PN. Using medication list-problem list mismatches as markers of potential error. Proc AMIA Symp 2002: 106-10.
  16. Slain D, Kincaid SE, Dunsworth TS. Discrepancies between home medications listed at hospital admission and reported medical conditions. Am J Geriatr Pharmacother 2008;6(3):161-6.
  17. Green DL, Boonstra JA, Bober MA. Use of a codified medicaiton process for documentation of home medications. J Am Med Inform Assoc 2010;17(5):608-12.
  18. Pronovost P, Weast B, Schwarz M, et al. Medication reconciliation: a practical tool to reduce the risk of medication errors. J Crit Care 2003;18(4):201-5.
  19. Chae SY, Chae MH, Isaacson N, et al. The patient medcation list: can we get patents more involved in their medical care? J Am Board Fam Med 2009;22(6):677-85.
  20. Murphy EM, Oxencis CJ, Klauck JA, et al. Medication reconciliation at an academic medical center: implementation of a comprehensive program from admission to discharge. Am J Health Syst Pharm 2009;66(23):2126-31.
  21. Fitzgerald RJ. Medication errors: the importanc of an accurate drug history. Br J Clin Pharmacol 2009;67(6):671-5.
  22. Chan AH, Garratt E, Lawrence B, et al. Effect of education on the recording of medicines on admission to hospital. J Gen Intern Med 2010;25(6):537-42.
  23. Thomas MC. A practical approach to medication reconciliation by pharmacists in a community hospital emergency department. Hosp Pharm 2010;45(1):41-4.
  24. Mills PR, McGuffie AC. Formal medicine reconciliation within the emergency department reduces the medication error rates for emergency admissions. Emerg Med J 2010;27:911-5.
  25. Carter MK, Allin DM, Scott LA, et al. Pharmacist-acquired medication histories in a university hospital emergency department. Am J Health- Syst Pharm 2006;63:2500-3.
  26. Johnston R, Saulnier L, Gould O. Best possible medication history in the emergency department: comparing pharmacy technicians and pharmacists. Can J Hosp Pharm 2010;63(5):359-65.

Leave a Reply

Your email address will not be published. Required fields are marked *