Medication Discrepancies at Outpatient Departments for Mood and Anxiety Disorders in the Netherlands: Risks and Clinical Relevance

Medication Discrepancies at Outpatient Departments for Mood and Anxiety Disorders in the Netherlands:

Risks and Clinical Relevance

Mirjam Simoons, PharmDa,c,d; Hans Mulder, PharmD, PhDa,*; Arne J. Risselada, PharmD, PhDa; Frederik W. Wilmink, MD, PhDb; Robert Schoevers, MD, PhDc; Henricus G. Ruhé, MD, PhDc,’ ¡; and Eric N. van Roon, PharmD, PhDd,e,’ ¡

Psychiatric patients commonly use combinations of psychiatric and general medical drugs for their mental illness and the frequently occurring somatic comorbidities or side effects of psychiatric medication.1,2 Various prescribers from different health care institutions, including general practitioners, prescribe these drugs. In addition, patients may use nonprescription drugs, mostly unadvised and unsupervised by health care professionals.

To correctly evaluate a patient’s clinical status and allow adequate adjustment of pharmacotherapeutic treatment, clinicians need to have a complete medication overview. This overview is obtained by a process called medication reconciliation in which the actual medication use is determined. Despite reliable community pharmacy records in the Netherlands, previous research has shown that patient counseling is a crucial part of medication reconciliation to create a complete overview of the actual medication use by the patient.3,4 Medication reconciliation through combination of pharmacy records and patient counseling results in an up-to-date and complete medication overview including current medication use and all medication allergies or intolerances.

Most research on quality of medication reconciliation has been conducted in hospital settings and reported discrepant medication overviews in 34%-95% of patients.5,6 An incomplete or erroneous medication overview may lead to failure to detect cause and consequence of side effects and somatic complications, prescribing errors, and iatrogenic harm. However, little is known about the clinical importance of medication discrepancies. One systematic review5 examining the clinical relevance of such errors after hospital admission showed that approximately 11%-59% of the medication discrepancies were clinically important. However, in patients admitted to a geriatric psychiatric clinic, 82% of all discrepancies were clinically relevant.6 To our knowledge, there are no studies reporting medication discrepancies and their clinical relevance in psychiatric outpatients.

Therefore, we addressed 2 issues. First, we examined whether psychiatrists have an up-to-date medication overview available for treatment evaluation when their patients visit them. Second, we assessed whether incomplete and erroneous medication overviews at psychiatric outpatient clinics are clinically relevant. We investigated outpatients with mood and anxiety disorders in the northern part of the Netherlands, aiming to identify discrepancies between the medication overview available at psychiatry outpatient departments and the actual drug use as well as to investigate their clinical relevance. For comparison, we also assessed discrepancies in the medication overviews from the general practitioners and community pharmacies of the same patients.

METHODS

Design and Setting

We used a cross-sectional design to assess discrepancies between the reconciled medication use and the medication overview from different health care providers. The study was conducted at 4 outpatient departments for mood and anxiety disorders in the northern part of the Netherlands: 3 from 2 large secondary mental health care institutions and 1 from an academic hospital.

In the Netherlands, it is mandatory to have a complete and up-to-date medication overview (including drugs prescribed by other physicians) available for clinical decision making whenever a patient contacts a prescriber. The prescriber is responsible for updating this information through reconciliation with the patient. This information is recorded in the electronic medical record (EMR) of the patient.

• Psychiatrists at both outpatient academic and community settings may not have up-to-date medication information available during their consult with a patient. Absence of this information implies a clinically relevant risk for harm in 49.3% of patients.
• Systematic medication reconciliation with the combination of pharmacy records and patient counseling may prevent harm to patients by solving medication discrepancies.

Study Population

We included patients 18 years or older who had visited the participating outpatient department at least once. The latter criterion ensured that the treating mental health care provider had had the opportunity to certify information regarding medication use after the first visit.

We consecutively recruited patients when visiting the outpatient departments between March and November 2014. We obtained written informed consent after complete verbal and written description of the study. An independent medical ethics committee (rTPO Leeuwarden, the Netherlands) waived formal review and approval of the study protocol since participants were not subject to interventions nor were they required to follow rules of behavior for this study.

We used the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), criteria for psychiatric classification of participants.

Outcomes

Primary outcomes were the percentage of patients with at least 1 discrepancy and the number and type of discrepancies between the reconciled medication use and the medication use according to the medication overview from the mental health care institution, general practitioner, and community pharmacy on the day of inclusion. Discrepancies provide information regarding the actual medication use by the patient but not necessarily regarding the correct medication use in a pharmacologic sense.

Secondary outcomes were the clinical relevance of the discrepancies and the need for intervention as a consequence of the discrepancies.

Discrepancy Assessment and Classification

We determined actual medication use on the inclusion date by medication reconciliation with the patient immediately after the consult with a mental health care provider. In concordance with other studies,3,4 we considered medication reconciliation combining recorded (pharmacy records) and patient-reported information the gold standard for determining the actual medication use by the patient. If the patient-reported medication use differed from the pharmacy records, we used the patient-reported information to assess the actual drug use by the patient. This actual medication use might be discrepant from the drug use the psychiatrist expected.

We defined a discrepancy as any difference between the reconciled medication use or allergies/intolerances and the medication overview from the EMR at the outpatient department, the general practitioner, or the community pharmacy.

We considered all drugs approved by the Dutch Medicines Evaluation Board or European Medicines Evaluation Authority. Drugs with a unique active ingredient, strength, or route of administration were considered separate drugs. To prevent overestimation of discrepancies, we counted a maximum of 1 discrepancy per actual drug or allergy/intolerance instead of all differences (eg, total daily dose and route of administration). In addition, we used a margin of 28 days around the theoretical starting and end dates of a drug on the medication overviews from the health care providers in which period the drug was considered to be still in use.7 For example, when a patient’s lithium refill had theoretically ended 14 days before the inclusion date, but the patient reported to still use lithium, we recorded no discrepancy if the daily dose and route of administration matched as well. However, we would have recorded a discrepancy if the refill had ended more than 28 days before the inclusion date.

We subsequently classified discrepancies as 1 of 5 types: extra drug, omitted drug, difference in total daily dose, difference in route of administration, or difference in allergy or intolerance. Whenever we identified differences in both dose and route of administration for 1 drug, we classified the discrepancy as a difference in total daily dose.

Discrepancies do not necessarily reflect clinically relevant issues. To overcome this limitation to our design, we evaluated the clinical relevance of discrepancies. An expert panel consisting of a hospital pharmacist and clinical pharmacologist (A.J.R.) and a psychiatrist (F.W.W.) independently classified each discrepancy in 1 of 3 classes for its potential to cause patient harm, as adopted from Cornish et al.8 Class 1 discrepancies are those unlikely to cause patient discomfort or clinical deterioration, while class 2 and 3 discrepancies could potentially result in respectively moderate and severe discomfort or clinical deterioration. In addition, only for discrepancies at the mental health care institutions, the expert panel evaluated the need for intervention in order to prevent or alleviate any possible harm as a consequence of the discrepancy. The expert panel classified suggested interventions in 6 categories: consider measuring a somatic parameter, prescribing an extra drug, stopping a current drug, changing a current drug (without altering the active ingredient [eg, a change in dose or route of administration]), replacing a current drug (by another drug within the same therapeutic class [eg, 1 antidepressant by another antidepressant or 1 β-blocking agent by another β-blocking agent]), or reevaluating total medication use.

The two experts resolved all disagreements in classifications by discussion.

Statistical Analysis

Descriptive and statistical analysis was completed using Excel 2013 (Microsoft, Redmond, Washington) and IBM SPSS (version 20 for Windows; IBM Corp, Armonk, New York).

We investigated the number of discrepancies and the percentage of clinically relevant discrepancies (classes 2 and 3) for differences per setting (both the 4 outpatient departments separately and the academic/community departments) in univariate linear regression models, using a P < .05 significance level.

Interrater reliability of the expert panel members for judging the clinical relevance and the need for intervention was analyzed using a weighted κ score with squared weights9 and a Cohen κ score, respectively.

RESULTS

Participants

We asked 495 consecutively eligible patients to participate in the study at the 4 locations. Of these patients, 370 gave written informed consent (104, 103, 102, and 61 out of 131, 142, 143, and 79 patients at the 4 locations, respectively). Reasons for not participating included “no time” and “privacy.” We subsequently excluded 2 patients from analysis because they had withdrawn consent and 1 because medication reconciliation could not be achieved due to “no show” of the patient. Since outcomes did not statistically differ per setting, we present patient characteristics for the total population (N = 367) in Table 1. As expected for mood and anxiety disorders, female participants were overrepresented (62.1%). Participants mostly had a low level of education and used a mean of 4.6 drugs.

Number and Type of Discrepancies

We found at least 1 discrepancy in the medication overview of the outpatient departments for mood and anxiety disorders in 348 patients (94.8%), with a mean ± SD of 3.9 ± 2.8 discrepancies per patient (Figure 1). Discrepancy numbers did not differ significantly between outpatient departments (P = .362) or between academic and community settings (P = .773). In the medication overviews of the general practitioners and pharmacies, we found at least 1 discrepancy in 90.2% and 85.8% of the patients, respectively, with corresponding mean ± SD values of 2.9 ± 2.1 and 2.2 ± 1.7 discrepancies per patient.

Figure 1 shows numbers and types of discrepancies per patient per health care provider. Most discrepancies for each health care provider were omitted drugs, ie, drugs the patient was regularly taking but that were absent from the health care provider’s records (74.5%, 65.4%, and 63.4% for the outpatient departments, general practitioners, and pharmacies, respectively). Discrepancies regarding medication (allergies or intolerances excluded) mostly concerned acetaminophen (Anatomical Therapeutic Chemical [ATC] code N02BE; 15.1%), anxiolytic benzodiazepine derivatives (N05BA; 6.4%), and proton pump inhibitors (A02BC; 4.9%) at the outpatient departments; acetaminophen (20.8%), anxiolytic benzodiazepine derivatives (7.3%), and propionic acid derivatives (M01AE, eg, ibuprofen; 6.1%) at the general practitioners; and acetaminophen (25.9%), propionic acid derivatives (7.5%), and anxiolytic benzodiazepine derivatives (6.5%) at the pharmacies.

Clinical Relevance of Discrepancies

Figure 2 shows the classification of the discrepancies for their potential to cause patient harm, as assessed by the expert panel. The interrater reliability for judging the clinical relevance (classes 1-3) was moderate (weighted κ = 0.58; 95% CI, 0.53-0.63). Of the discrepancies at the outpatient departments, 77.2% were unlikely to cause harm (class 1), while 19.9% and 2.8% of the discrepancies were found to potentially cause moderate (class 2) or severe (class 3) discomfort or clinical deterioration, respectively. Table 2 shows a few examples of discrepancies and their classification. The clinically relevant discrepancies (class 2 or 3) affected 49.3% of all patients. The percentage of clinically relevant discrepancies did not significantly differ between outpatient departments (P = .440) or between academic and community settings (P = .379).

In comparison, 80.5%, 15.6%, and 4.2% of the medication discrepancies from general practitioners and 86.9%, 9.9%, and 3.2% of those from pharmacies were categorized as class 1, 2, and 3 discrepancies, respectively. Class 2 and 3 discrepancies from general practitioners and pharmacies were present in 33.2% and 22.3% of patients, respectively.

In 35.4% of all patients, the expert panel considered intervention clinically necessary as a result of a discrepancy at the outpatient department. Initial agreement for judging the intervention necessity was limited (Cohen κ = 0.13; 95% CI, 0.06-0.20), but this was solved by consensus in all cases. The expert panel suggested a mean ± SD of 0.5 ± 0.7 interventions per patient, with “consider measuring a somatic parameter” (38.4%) and “reevaluate total medication use” (37.8%) most frequently suggested.

DISCUSSION

This study indicates that medication reconciliation processes at outpatient departments for mood and anxiety disorders are potentially harmful and in need of improvement. Patients had a mean of 3.9 discrepancies. Moreover, almost 23% of all discrepancies had the potential to cause moderate to severe discomfort or clinical deterioration, affecting almost half of all patients. These figures were lower for general practitioners and community pharmacies.

To the best of our knowledge, this is the first study worldwide to evaluate whether mental health care providers are aware of the drugs used by their outpatients when they visit them. Previous studies4,10-14 in nonpsychiatric outpatient departments, such as those for hemodialysis or internal medicine, found on average 0.97 to 3.4 discrepancies per patient. A single study6 investigating psychiatric inpatients after admission to a geriatric psychiatric clinic showed discrepancies in 78% of 50 patients, with a median of 2 discrepancies per patient. While these results are in line with the numbers reported for general hospital inpatients,5 we found fairly higher discrepancy frequencies in psychiatric outpatients, which indicates that this issue may especially be unknown and problematic in psychiatric outpatient settings. In addition, we observed no differences in the discrepancy risk or associated patient harm between the participating outpatient departments or between academic and community settings. Moreover, we found high numbers and clinical relevance of discrepancies, despite the generally high quality of health care in the Netherlands and in particular a guideline demanding a complete and up-to-date medication overview with every contact between patient and prescriber. Therefore, although replication is warranted, we believe our results apply to psychiatry outpatient departments in general.

There are several potential explanations for the higher discrepancy frequencies in psychiatric outpatients. First, these patients often have more than one health care provider (eg, psychiatrist and general medical physician). This makes it difficult to keep track of changes in drug regimens made by different prescribers. Indeed, the number of prescribing physicians has been shown to increase medication discrepancies in outpatients.15 We could not evaluate whether there was a difference in number of discrepancies in prescriptions from the outpatient department and outside doctors, as we did not assess prescribers. Second, higher discrepancy numbers may reflect the often-reported suboptimal treatment of somatic conditions in psychiatric outpatients compared to nonpsychiatric individuals.16 In addition to receiving a lower quality of medical care, psychiatric patients receive fewer prescriptions for several common drugs for existing medical disorders than individuals without mental illness.16

In contrast to previous studies in nonpsychiatric outpatients, we also determined a measure of potential patient harm due to discrepancies. Assessing clinical relevance is essential to determine the impact of discrepancies. Almost 23% of all discrepancies had the potential to cause moderate to severe patient harm, which, importantly, affected almost half of all patients. Since there is currently no valid and reliable method to preidentify patients at risk for discrepancies, attention should not be limited to specific subsets of patients when implementing medication reconciliation.

Our results at general practitioners and pharmacies are in line with several studies15,17-21 in population-based samples visiting primary care physicians. Since general practitioners and community pharmacists at least in some countries have the role of gatekeepers with the responsibility of having an adequate overview of the medication information about their patients, the numbers of discrepancies are still surprisingly high.

In our opinion, our results demonstrate the need for implementation of a structured medication reconciliation process in clinical practice at psychiatry outpatient departments in order to minimize iatrogenic harm to outpatients. In different hospital settings, implementation of medication reconciliation with patient counseling substantially diminished discrepancies upon both admission and discharge in various countries.5,22-24 In addition, medication reconciliation upon hospital discharge resulted in higher benefits than costs related to the net time investment.25 Furthermore, prescribing safely and conducting adequate somatic monitoring of psychiatric patients as recommended by guidelines are impossible without a complete and up-to-date medication overview.26,27 We therefore developed an innovative care path called Monitoring Outcomes of Psychiatric Pharmacotherapy (MOPHAR), which is currently being implemented. In this care path, a nurse conducts medication reconciliation with each patient at every visit to a prescriber. In case of relevant medication discrepancies, MOPHAR will notify the treating psychiatrist. After reconciliation, recommended (somatic) monitoring is performed according to prespecified protocols per drug used. This information is immediately available in the electronic medical record in summarized form, thus instantly providing mental health care providers with up-to-date information on medication use and monitoring parameters. We will investigate the impact of this integrated care model regarding the benefits for psychiatric patients.28,29

Strengths of this study are the large population and the conduct of assessments at 4 different locations. However, a few limitations need to be considered. First, our study might suffer from performance bias, as collaboration of outpatient departments may have been selective for well-organized settings. This may have resulted in an underestimation of discrepancies, meaning true practice is more alarming still. Second, medication reconciliation involving psychiatric patients may not be as reliable as in other patient populations. However, medication reconciliation through combination of pharmacy records and patient counseling is currently considered the gold standard for determining the actual medication use. It is important to remark that this patient-reported medication use may not reflect the intended or correct use in a pharmacotherapeutic sense. Third, some medication overviews contain theoretical starting and end dates for medication refills that may not correspond with actual use by patients. However, we assume we covered most unintentional discontinuation periods by the 28-day permissible gap for medication refills. Fourth, we did not distinguish between different sources or reasons for discrepancies, such as clinical misunderstandings, clinical errors, or administrative errors. In clinical practice, it is important to make this distinction to resolve the discrepancy accordingly. Fifth, the classification method used to assess clinical relevance is, strictly speaking, unvalidated. However, this procedure has been used in previous studies on medication discrepancies and errors.6,8,30 Finally, our measure of clinical relevance concerned potential harm. Because of the cross-sectional design of this study, we were not able to collect evidence for actual adverse effects as a result of the discrepancies.

In conclusion, this study shows that outpatients with mood and anxiety disorders may be at substantial risk of medication discrepancies that may be clinically relevant in almost half of the patients. We consider this risk a potentially general problem in the treatment of psychiatric outpatients, for which we suggest that medication reconciliation processes be improved to increase medication safety in psychiatric outpatients.

Submitted: September 8, 2015; accepted March 24, 2016.

Drug names: atenolol (Tenormin and others), citalopram (Celexa and others), ciclesonide (Alvesco, Zetonna, and others), duloxetine (Cymbalta and others), esomeprazole (Nexium and others), fluticasone (Flonase and others), formoterol (Foradil, Perforomist), hydrochlorothiazide (Microzide and others), levothyroxine (Tirosint, Levo-T, and others), lithium (Lithobid and others), lisinopril (Prinivil, Zestril, and others), metformin (Glucophage and others), metoprolol (Lopressor and others), mirtazapine (Remeron and others), montelukast (Singulair and others), omeprazole (Prilosec and others), paroxetine (Paxil, Pexeva, and others), rosuvastatin (Crestor and others), salmeterol-fluticasone propionate inhaler (Advair), simvastatin (Zocor and others), temazepam (Restoril and others), topiramate (Topamax and others), valproic acid (Depakene and others), venlafaxine (Effexor and others).

Disclosure of off-label usage: The authors have determined that, to the best of their knowledge, no investigational information about pharmaceutical agents that is outside US Food and Drug Administration-approved labeling has been presented in this article.

Potential conflicts of interest: Drs Simoons, Mulder, Risselada, Wilmink, Schoevers, Ruhé, and van Roon have no personal affiliations or financial relationships with any commercial interest to disclose relative to the article.

Funding/support: Dr Ruhé is supported by an NWO/ZonMW Veni grant #016.126.059.

Role of the sponsor: The funding agency had no role in the conduct or publication of the study.

Previous presentation: Data from this study were presented as a poster at the 28th European College of Neuropsychopharmacology Congress; August 29-September 1, 2015; Amsterdam, the Netherlands.

Acknowledgments: The authors thank Casper van der Hoeven, PharmD; Eline Hemelt, BSc; Rabab Tarasse, BSc; and Ikrame Mouch, BSc, all from the Department of Clinical Pharmacy, Wilhelmina Hospital Assen, Assen, the Netherlands, for their assistance with data collection. None of these individuals have received specific financial support for their assistance nor did any of them have any potential conflicts of interest relevant to the subject of this study.

REFERENCES

1. Abdullah-Koolmees H, Gardarsdottir H, Stoker LJ, et al. Prevalence of medication use for somatic disease in institutionalized psychiatric patients. Pharmacopsychiatry. 2013;46(7):274-280. PubMed doi:10.1055/s-0033-1354405

2. De Hert M, Correll CU, Bobes J, et al. Physical illness in patients with severe mental disorders, I: prevalence, impact of medications and disparities in health care. World Psychiatry. 2011;10(1):52-77. PubMed doi:10.1002/j.2051-5545.2011.tb00014.x

3. Karapinar-Carkit F, Borgsteede SD, Zoer J, et al. Effect of medication reconciliation with and without patient counseling on the number of pharmaceutical interventions among patients discharged from the hospital. Ann Pharmacother. 2009;43(6):1001-1010. PubMed doi:10.1345/aph.1L597

4. 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(14):2129-2134. PubMed doi:10.1001/archinte.160.14.2129

5. Tam VC, Knowles SR, Cornish PL, et al. Frequency, type and clinical importance of medication history errors at admission to hospital: a systematic review. CMAJ. 2005;173(5):510-515. PubMed doi:10.1503/cmaj.045311

6. Prins MC, Drenth-van Maanen AC, Kok RM, et al. Use of a structured medication history to establish medication use at admission to an old age psychiatric clinic: a prospective observational study. CNS Drugs. 2013;27(11):963-969. PubMed doi:10.1007/s40263-013-0103-9

7. Caetano PA, Lam JM, Morgan SG. Toward a standard definition and measurement of persistence with drug therapy: examples from research on statin and antihypertensive utilization. Clin Ther. 2006;28(9):1411-1424, discussion 1410. PubMed doi:10.1016/j.clinthera.2006.09.021

8. Cornish PL, Knowles SR, Marchesano R, et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med. 2005;165(4):424-429. PubMed doi:10.1001/archinte.165.4.424

9. Roberts C, McNamee R. Assessing the reliability of ordered categorical scales using kappa-type statistics. Stat Methods Med Res. 2005;14(5):493-514. PubMed doi:10.1191/0962280205sm413oa

10. Manley HJ, Drayer DK, McClaran M, et al. Drug record discrepancies in an outpatient electronic medical record: frequency, type, and potential impact on patient care at a hemodialysis center. Pharmacotherapy. 2003;23(2):231-239. PubMed doi:10.1592/phco.23.2.231.32079

11. Wagner MM, Hogan WR. The accuracy of medication data in an outpatient electronic medical record. J Am Med Inform Assoc. 1996;3(3):234-244. PubMed doi:10.1136/jamia.1996.96310637

12. Peyton L, Ramser K, Hamann G, et al. Evaluation of medication reconciliation in an ambulatory setting before and after pharmacist intervention. J Am Pharm Assoc (2003). 2010;50(4):490-495. PubMed doi:10.1331/JAPhA.2010.09055

13. Ernst ME, Brown GL, Klepser TB, et al. Medication discrepancies in an outpatient electronic medical record. Am J Health Syst Pharm. 2001;58(21):2072-2075. PubMed

14. Chan WW, Mahalingam G, Richardson RM, et al. A formal medication reconciliation programme in a haemodialysis unit can identify medication discrepancies and potentially prevent adverse drug events. J Ren Care. 2015;41(2):104-109. PubMed doi:10.1111/jorc.12107

15. Tulner LR, Kuper IM, Frankfort SV, et al. Discrepancies in reported drug use in geriatric outpatients: relevance to adverse events and drug-drug interactions. Am J Geriatr Pharmacother. 2009;7(2):93-104. PubMed doi:10.1016/j.amjopharm.2009.04.006

16. Mitchell AJ, Lord O, Malone D. Differences in the prescribing of medication for physical disorders in individuals with v. without mental illness: meta-analysis. Br J Psychiatry. 2012;201(6):435-443. PubMed doi:10.1192/bjp.bp.111.094532

17. Andrus MR. Student pharmacist initiated medication reconciliation in the outpatient setting. Pharm Pract (Granada). 2012;10(2):78-82. PubMed doi:10.4321/S1886-36552012000200003

18. Andrus MR, Anderson AD. A retrospective review of student pharmacist medication reconciliation activities in an outpatient family medicine center. Pharm Pract (Granada). 2015;13(1):518. PubMed doi:10.18549/PharmPract.2015.01.518

19. Balon J, Thomas SA. Comparison of hospital admission medication lists with primary care physician and outpatient pharmacy lists. J Nurs Scholarsh. 2011;43(3):292-300. PubMed

20. Orrico KB. Sources and types of discrepancies between electronic medical records and actual outpatient medication use. J Manag Care Pharm. 2008;14(7):626-631. PubMed

21. Varkey P, Cunningham J, Bisping DS. Improving medication reconciliation in the outpatient setting. Jt Comm J Qual Patient Saf. 2007;33(5):286-292. PubMed

22. Vira T, Colquhoun M, Etchells E. Reconcilable differences: correcting medication errors at hospital admission and discharge. Qual Saf Health Care. 2006;15(2):122-126. PubMed doi:10.1136/qshc.2005.015347

23. Bond CA, Raehl CL, Franke T. Clinical pharmacy services, hospital pharmacy staffing, and medication errors in United States hospitals. Pharmacotherapy. 2002;22(2):134-147. PubMed doi:10.1592/phco.22.3.134.33551

24. Gleason KM, Groszek JM, Sullivan C, et al. Reconciliation of discrepancies in medication histories and admission orders of newly hospitalized patients. Am J Health Syst Pharm. 2004;61(16):1689-1695. PubMed

25. Karapinar-Çarkit F, Borgsteede SD, Zoer J, et al. Effect of medication reconciliation on medication costs after hospital discharge in relation to hospital pharmacy labor costs. Ann Pharmacother. 2012;46(3):329-338. PubMed doi:10.1345/aph.1Q520

26. Keck PE Jr. Monitoring pharmacotherapy response, safety, and tolerability to enhance adherence in bipolar disorder. J Clin Psychiatry. 2014;75(5):e12. PubMed doi:10.4088/JCP.13010tx4c

27. Dodd S, Malhi GS, Tiller J, et al. A consensus statement for safety monitoring guidelines of treatments for major depressive disorder. Aust N Z J Psychiatry. 2011;45(9):712-725. PubMed doi:10.3109/00048674.2011.595686

28. Bradford DW, Cunningham NT, Slubicki MN, et al. An evidence synthesis of care models to improve general medical outcomes for individuals with serious mental illness: a systematic review. J Clin Psychiatry. 2013;74(8):e754-e764. PubMed doi:10.4088/JCP.12r07666

29. Culpepper L. Improving patient outcomes in depression through guideline-concordant, measurement-based care. J Clin Psychiatry. 2013;74(4):e07. PubMed doi:10.4088/JCP.12075tx1c

30. Kwan Y, Fernandes OA, Nagge JJ, et al. Pharmacist medication assessments in a surgical preadmission clinic. Arch Intern Med. 2007;167(10):1034-1040. PubMed doi:10.1001/archinte.167.10.1034