Clinical pharmacist interventions during clinical rounding in a psychiatric hospital: Positive evidence for pharmacists in interdisciplinary rounding at psychiatric hospital


 Clinical pharmacists have not yet become an integral part of interdisciplinary ward rounds in most psychiatric hospitals across the European Union. This retrospective study examined the impact of the recommendations from a clinical pharmacist during hospital rounds as part of an interdisciplinary medical team in a psychiatric hospital. This retrospective observational pre-post study included all patients in a Slovenian psychiatric hospital who were hospitalized in 2019–2020. The clinical pharmacist made 315 recommendations for a total of 244 participants (average age M = 59.4, median = 56). Psychiatrists accepted 295 (93.7%) of the recommendations. After the recommendations, the number of expressed DRPs decreased in 166 interventions (93.8%) and the number of potential DRPs in 129 interventions (93.8%). Three months after discharge, 222 accepted recommendations were continued (70.5%). The most common recommendations were related to antipsychotics (19.4%, N = 61) followed by antidepressants (16.8%, N = 53). Including a clinical pharmacist in the interdisciplinary ward rounds in a psychiatric hospital reduced the number of expressed and potential DRPs with a very high recommendation acceptance rate. These results are the first in the European Union to show the benefits of including a clinical pharmacist in ward rounding.


Introduction
Mental disorders pose a high disease burden and health costs, which continue to increase with signi cant health, social, ethical and economic consequences in all countries. Mental disorders account for 13% of the global disease burden, and major depression alone is expected to be the largest contributor by 2030 [1]. Wittchen et al. report that the most frequent disorders in the European Union are anxiety disorders (14.0%), insomnia (7.0%), major depression (6.9%), somatoform disorders (6.3%), alcohol and drug dependence (> 4%), ADHD among youth (5%), and dementia (1-30%, depending on age) [2]. Though there are nonpharmacological treatment options (e.g., psychotherapy), pharmacotherapy is often necessary. Patients with mental disorders are often treated with several medications (i.e., polypharmacy) and for several diseases. This opens the risk of unnecessary polypharmacy, potentially inappropriate medications (PIMs) and irrational medication combinations. In spite of lacking evidence, patients are often treated with psychotropics to manage several conditions (e.g., insomnia, depression, and behavioral symptoms of dementia treatment) [3,4]. Consequently, there is a need for well-designed studies with strong ecological validity in clinical practice to minimize medication-related problems and plan appropriate strategies.
Inpatients in psychiatric hospitals often have treatment-resistant conditions, are concurrently treated for other diseases (e.g., diabetes, infectious diseases, seizures), and receive medication combinations that are not always evidence-based. This creates a risk of inappropriate polypharmacy, which can result in inappropriate treatment, adverse events, and treatment failures [3]. Psychotropics have a particularly high risk of potential drug-drug interactions (pDDIs), adverse events (e.g., weight gain) and PIM prescribing. A retrospective descriptive chart study in a Slovene psychiatric hospital found that 47% and 22% of the prescriptions were concomitant prescriptions of two and three antipsychotics, respectively. It also found many inappropriate combinations, including antipsychotic polypharmacy (APP) [4]. Clozapine prescriptions in psychiatric hospitals are less frequent than expected in patients with two unsuccessful treatments and many patients are treated with APP before clozapine initiation [5].
A promising approach to address these drawbacks for patients with mental health disorders is collaborative care that involves a clinical pharmacist, which rst became more widespread in the 1990s in the US, where it demonstrated improved outcomes and medication adherence in primary care settings [6,7]. Outside the US, Stuhec M et al. report a decrease in medication-related problems (number of PIMs and pDDIs) and an improvement in the patients' quality of life following a clinical pharmacist's medication review service in a Slovenian nursing home [8]. There is limited research on whether these bene ts also apply to psychiatric hospitals, in which patients often require complex treatments, and most studies on the topic have been conducted in the US [6,7,8,9]. A retrospective analysis in a Californian acute care psychiatric hospital reported a positive impact of clinical pharmacist interventions during ward rounds. The interventions had a 92.5% acceptance rate. Most frequently, they were medication discontinuations (38.5%), laboratory monitoring (26%), and medication order modi cation (13.5%). The 200 clinical pharmacist interventions were also associated with signi cant cost savings and cost avoidance [9]. A 2002 study of US hospitals (N = 1081) found a signi cant inverse relationship between medical errors and pharmacist participation on medical rounds (slope = 0.6974303, p < 0.001), as well as with pharmacist-provided admission histories (slope = -1.6021493, p < 0.001) [10]. While not obtained from a randomized study, these results suggest that pharmacist participation on ward rounds may be an important way to optimize pharmacotherapy in psychiatric hospitals. Because clinical pharmacists are usually not part of interdisciplinary medical teams in psychiatric settings in the European Union, this paper sets out to examine the impact of clinical pharmacist interventions during hospital rounds in a Slovenian psychiatric hospital.

Study setting
This study was conducted in the Ormoz Psychiatric Hospital (Psihiatrična bolnišnica Ormož), which covers approximately 300,000 inhabitants in Northeastern Slovenia. It has 120 hospital beds and provides treatment for all mental disorders, a community psychiatric service and daily hospital and outpatient service. It receives approximately 1,000 inpatients annually with an average length of stay of 28 days. It has six wards: a psychogeriatric ward, a daily hospital center, an addiction treatment center, two closed wards, and an open ward for prolonged treatment. Daily hospital activities include interdisciplinary team activities (e.g., ward rounding) and individual activities (e.g., daily conversation between patients and various healthcare professionals, occupational therapy activities, psychotherapy).
During the study period, ward rounding was done 1-2 times per week in the morning for each ward and included all interdisciplinary team members: a psychiatrist, nurse, clinical pharmacist, psychologist, occupational therapist, and a social worker. Additional ward staff members may also be involved in roundings as well as residents (e.g., psychiatrists, psychologists and clinical pharmacists) and students (e.g., nursing, pharmacy, psychology and occupational therapy). Ward roundings covered 10-20 patients in 60-90 minutes, included by a discussion and potential decisions within the interdisciplinary team, such as changes in patient care including pharmacotherapy. Hospital employs one board-certi ed clinical pharmacist specialist. Clinical pharmacists may suggest interventions to ward psychiatrists during the roundings who need to con rm it. Each intervention is recorded in electronic patient records. Clinical pharmacists monitor all interventions and adjust them at subsequent ward roundings if necessary. He plays an important role in the team by contributing his expertise through medication recommendations and by designing and monitoring treatment plans. He does not measure clinical outcomes, but does have access to patient health data. The inclusion of a clinical pharmacist in the interdisciplinary team is a care standard since 2018 (as per the Slovenian Pharmacy Act), but not practiced in all psychiatric hospitals.

Study design
We designed a retrospective observational pre-post study that included all inpatients at the Ormoz Psychiatric Hospital between November 2019 and December 2020 for whom the clinical pharmacist provided recommendations. There was no additional selection criteria or a control group. The patients' data were retrieved from their electronic health records or hospital databases and were analyzed from when the time the clinical pharmacist provided recommendations to three months after hospital discharge. Clinical outcomes were not measured.
Drug-related problems (DRPs) were classi ed according to the Slovenian classi cation of drug-related problems (DRP-SLO-V1) with some adjustments. DRPs were identi ed as potential or expressed [11]. Potential DRPs were described in terms of risk factors for the DRPs to arise in practice. We examined if any potential DRPs occurred by 3 months after discharge.
We identi ed problems related to treatment effectiveness, adverse events (treatment safety), and unnecessary drug treatment problems (e.g., no indication). The interventions were categorized into drug discontinuation, drug initiation, and drug regimen adjustment. The clinical pharmacist interventions were considered effective if risk factors for DRPs were not present at 3 months after discharge, based on patients' medical documentation. A similar methodology was used in our previous paper on a nursing home setting [8]. Outcomes The primary outcome measure was the acceptance rate (%) of interventions. The secondary outcome measures were the continuation rate 3 months after discharge (%) and the difference in the total numbers of DRPs for each patient at the discharge. We only included clinical pharmacist interventions related to ward rounding.

Data collection and analysis
The patient data were collected by Valentina Tement (MPharm student) and Matej Stuhec ( rst author), who has been working in a psychiatric hospital setting for over a decade. We calculated the baseline characteristics of patients as the mean ± standard deviation as well as the differences before and after the interventions. SPSS Statistics 22.0 was used for statistical analysis. This study used the STROBE Statement checklist for observational studies [12] Ethical declaration All methods were carried out in accordance with relevant guidelines and regulations. Informed consent was not necessary from all subjects in Slovenia, because this was a retrospective chart study which was approved by the National Medical Ethics Committee of the Republic of Slovenia in 2021 (Number 0120 − 18/2021/3; approved 11.3.2021) and the Committee waived the requirement of informed consent.
The clinical pharmacist attended 75 ward roundings during the study period and gave 315 recommendations (4.2 per ward rounding), excluding 9 recommendations with missing data which were not included in the nal analysis.
Out of the total 315 DRPs, 56.2% (N = 177) were expressed and 43.8% were potential DRPs (N = 138). After the clinical pharmacist's recommendations, the number of both expressed and potential DRPs decreased by 93.8% to 11 and 9 DRPs, respectively.
Three months after discharge, 70.5%b of all proposed recommendations were maintained, 28.6% were not continued and 1.0% had missing data. The DRPs were most frequently related to the treatment of mental disorders, followed by the treatment of cardiovascular diseases. In terms of medication type, recommendations were most commonly given for antipsychotics (19.4%, N = 61) followed by antidepressants (16.8%, N = 53), antihypertensive medications (18.8%, N = 59), anxiolytics (8.2%, N = 26), antibiotics (7.0%, N = 22), hypnotics (4.1%, N = 13), analgesics, antiepileptics and beta blockers (3.8% each, N = 12 each) and other medications (14.3%, N = 45). b Antibiotics were included in 22 different interventions (7.0%). See Table 1 for detailed information on treatment guidelines adherence and clinical pharmacist recommendations. All antibiotic treatments were effective and patients did not require additional treatment. The psychiatrists accepted all proposed recommendations.

Discussion
This study is the rst in Central Europe to examine clinical pharmacist interventions in interdisciplinary ward rounding at a psychiatric hospital. The results from Slovenia are widely applicable to countries with comparable healthcare systems, such as those in Central, Southeastern, and Eastern Europe. Our main nding is that the clinical pharmacist recommendations had a high acceptance rate and were maintained at follow-up.
The rst important result is the high acceptance rate compared to medication review services. Pharmacist participation in an interdisciplinary ward rounding was shown to be signi cantly associated with decreased medication errors/occupied bed/year in a study of over 1,000 US hospitals [10]. The acceptance rate in ward rounding at 93.7% in this study is notably higher than in our two previous studies on a primary setting medication review service with an acceptance rate of 48.6% and 42.8% in which suggestions were con rmed or rejected by general practitioners (GPs) [13,14]. A higher acceptance rate of 88.0% was found in a 2014 study by Stuhec on a medication review service in a Slovenian psychiatric hospital [15]. While a medication review service alone may improve the quality of care received by psychogeriatric patients through reducing the number of medications, pDDIs, PIMs and improving treatment guideline adherence [14], the differences in acceptance rates suggest that the bene ts from clinical pharmacists may be increased if they are provided with greater access to psychiatrists, who are frequently the prescribers of psychotropics. This can be achieved either through closer collaboration in a hospital (as opposed to a primary care setting) or through direct inclusion in ward rounding. Most accepted recommendations were maintained 3 months after discharge, so the recommendations were positively accepted by psychiatrists and patients, although outcomes were not measured.
Our second most important nding is that pharmacist recommendations minimize DRPs. The pharmacist's recommendations in this study addressed 56% of the expressed DRPs. The most frequent recommendations in our study were dose adjustments in 37.8% (N = 119) of cases, followed by medication initiation 24.4% (N = 77), medication discontinuation 24.4% (N = 77) and other interventions (13.3%, N = 42). A 2019 study by Stuhec et al. had a similar distribution of intervention types and overall found that the clinical pharmacist identi ed several expressed DRPs, which had a positive impact on the patients' quality of life, although the acceptance rate was lower (29.2%) than in our current study [8]. These differences may be due to how a clinical pharmacist is included in the treatment process in ward rounding as opposed to a medication review. During ward rounding, a clinical pharmacist can communicate with psychiatrists, patients, and nurses and can raise recommendations several times. Getting from recommendations to actions is lengthier in medication reviews, as the recommendations are sent out to the GP, who may discuss them with a psychiatrist.
Most interventions (40.6%) were provided in the psychogeriatric ward, where patients were treated with excessive polypharmacy. This is similar to studies in primary care settings that show most clinical pharmacist recommendations were given for psychogeriatric patients [8,13]. Interestingly, interventions at the addiction department constituted 26.0% of the interventions and were mostly benzodiazepine discontinuations or related to non-psychiatric diseases (e.g., hypertension). The difference in interventions between departments shows that a clinical pharmacist in the interdisciplinary medical team can provide different speci c bene ts for each particular psychiatric hospital department.
The third important nding is that the clinical pharmacist in our study provided recommendations on pharmacotherapy in general and not exclusively on mental health disorders. While such recommendations were the most frequent, recommendations on cardiovascular, pain-related, and other conditions were also provided. This shows that a clinical pharmacist can be a versatile addition to the interdisciplinary medical team. For example, all 22 antibiotic-related recommendations in our study were accepted. The clinical pharmacist recommended the discontinuation of quinolones (e.g., cipro oxacin, levo oxacin, moxi oxacin) and cefuroxime and nitrofurantoin were added for lower urinary tract infection L(UTI). These recommendations are in line with treatment guidelines, because quinolones are discouraged as a rst-line treatment, in particular for patients with mental disorders because of possible drug-drug interactions and adverse events [16,17]. Some combinations (e.g., with duloxetine) are even contraindicated with cipro oxacin. Our ndings on antibiotics are similar to a Slovenian prospective study in a psychiatric hospital, that found the most commonly prescribed antibiotics were co-amoxiclav (36.5% of cases) and co-trimoxazole (25.7%), cipro oxacin (12.2%) and nitrofurantoin (6.80%), and detected important DDIs in 17 out of 74 patients (23% or 4.48/100 admissions), which were most frequently the combination of co-trimoxazole and quetiapine, and of cipro oxacin and olanzapine [18]. The antibiotic-related pharmacist recommendations in our study were evidence-based, as the rst-line treatment for psychogeriatric patients was cefuroxime (few DDIs and few adverse events) and nitrofurantoin (e cacy, few DDIs). The safest drugs for patients with mental disorders are penicillins, cephalosporins, and nitrofurantoin; macrolides and quinolones should be used with great caution [16,18].
The recommendations were also effective insofar as infections did not reoccur 3 months after discharge.
The fourth important nding is that the recommendations in our study were most frequently related to antipsychotics. The acceptance rate was much higher than in previous studies in a primary care setting, which may be due to a different working environment [14]. Interestingly, many interventions were connected with APP that guidelines discourage [19,20]. Our study found APP combinations of aripiprazole, levomepromazine and zuclopentixol, of olanzapine with aripiprazole or haloperidol, and of quetiapine with haloperidol, sulpiride, clozapine or risperidone. The clinical pharmacist recommendations were evidence-based as APP should be only considered after clozapine treatment when possible [20,21,22]. The results show that the clinical pharmacist reduced the use of APP and often suggested antipsychotic discontinuation in cases of excessive APP, in line with the results by Suzuki et al. that APP may be replaced with antipsychotic monotherapy in most cases [23]. A Finnish nationwide cohort study of schizophrenia patients found that the hazard ratio of psychiatric rehospitalization was 7% lower during any polypharmacy than any monotherapy period (HR, 0.93; 95% CI, 0.91-0.95; p < 0.001), but that the difference may be clinically insigni cant. Aripiprazole with clozapine was associated with the lowest risk of rehospitalization, so polypharmacy may be a feasible treatment option for schizophrenia (e.g., clozapine + aripiprazole, any long-acting antipsychotic + olanzapine and clozapine + olanzapine). However, the results for clozapine monotherapy were not substantially worse than those of the best polypharmacy option [24]. Many interventions were focused on the indication for antipsychotic and insomnia treatment. These results are in line with the clinical guidelines for antipsychotic use and insomnia treatment, according to which quetiapine and olanzapine are not a rst-line treatment, especially for insomnia, because the evidence is very weak and adverse events could be problematic [25,26]. Many patients treated with antipsychotics experience various adverse events associated with their use [27]. Adverse events are even more important in elderly patients, since they often have multiple diseases, including cardiovascular diseases. The results of our study show that the clinical pharmacist suggested many different recommendations including olanzapine discontinuation in APP (or for cases without indication), which can reduce the risk of important cardiovascular adverse events in a long-term treatment (e.g., weigh gain, QT prolongation and diabetes). These interventions were also maintained 3 months after discharge, suggesting they were effective and well tolerated.
This study also has many important limitations which should be discussed, mostly due to the selection criteria, minimum exclusion criteria and non-randomized conditions. The study is retrospective and nonrandomized, so causal relationships cannot be established. Patients were not monitored over a longer period of time (e.g., eight months), limiting the scope of our results. Another limitation is the population heterogeneity that increases intervariability as we included patients with various mental disorders.
Another source of selection bias stems from the inclusion of patients with polypharmacy and comorbidities. The sample size used in this study was small and not pre-calculated. We also did not measure clinical outcomes and other interventions (e.g., medication review). These limitations could be overcome with prospective studies in real clinical settings. Despite the limitations, our study is the rst to the best of our knowledge in the EU to examine the impact of clinical pharmacists in interdisciplinary team rounding in a psychiatric hospital.

Conclusions
Clinical pharmacist participation in interdisciplinary team rounds in a psychiatric hospital led to fewer DRPs with a very high acceptance rate. Their recommendations were mostly evidence-based and maintained at a 3-month follow-up. More countries across the EU could adopt this practice to improve outcomes for patients with mental disorders in psychiatric hospitals. Additional research with larger samples or a prospective design would be needed to replicate the positive results.