Prescription errors are clinically meaningful errors that occur when prescribing decisions or formulary processes, resulting in an unintentional and significant decrease in the probability of timely and effective treatment or an increased risk of harm compared to generally accepted practice (Abdel-Qader et al., 2010) and can occur in daily clinical practice (Bates et al., 1995; Noguchi et al., 2016). Therefore, preventing prescription errors is essential for hospitals, and hospital pharmacists play a vital role in comprehensively intervening in patient care to prevent adverse drug events (ADEs) (Nester and Hale, 2002; Bond and Raehl, 2007; Abu-Naser, 2021).

The number of medications prescribed to patients increases with age and comorbidities (Park et al., 2016). Polypharmacy, defined as the inappropriate use of multiple medicines in patients in the 1960s (Canadian Medical Association, 1966), is associated with various problems, such as drug interactions, ADEs, increased medical expenses, and decreased medication adherence (Hersh et al., 2017). Studies have found that there is a dose-dependent relationship between polypharmacy and mortality, and excessive polypharmacy, such as regular use of ten or more medications, can lead to death (Leelakanok et al., 2017). However, increasing the number of drugs essential for a patient’s health can result in polypharmacy. Several studies have defined polypharmacy as the use of more drugs than is clinically indicated (Fulton and Allen, 2005). Recent research suggests that deprescribing, a process of identifying and discontinuing inappropriate medications (IMs), can reduce inappropriate polypharmacy in older patients (Reeve et al., 2015). However, it remains unclear whether deprescribing can improve clinical outcomes (Reeve et al., 2014; Scott et al., 2015). The use of multiple drugs in various healthcare settings is associated with potentially inappropriate medication (Nothelle et al., 2017; Nothelle et al., 2019).

According to Japan’s guidelines for the appropriate use of drugs for older patients, polypharmacy is a condition in which various problems occur due to the use of multiple drugs (Ooi, 2019). In Japan, there have been reports on the management of polypharmacy in a specific area of a single hospital (Hashimoto and Tensho, 2016; Horii and Atsuda, 2020). However, it resulted from medical staff with high-level knowledge, and it was difficult to generalize the data. As such, there is insufficient evidence of pharmacists’ contributions to addressing this problem in daily practice. To fill this gap, the Japanese Society for Pharmaceutical Palliative Care and Sciences (JSPPCS) Research Committee previously conducted a questionnaire survey on polypharmacy for members who worked as hospital pharmacists (Uchida et al., 2019) and community pharmacists (Suzuki et al., 2019). These results show the prevalence of polypharmacy and the benefits of pharmacy interventions for drug-related problems (Suzuki et al., 2019; Uchida et al., 2019).

However, since the findings were obtained from a retrospective observational study, the data lacked information on the patient background (age, disease, etc.), the actual number of cases of pharmacist recommendations and acceptance, detailed information such as discontinuation of IMs or reduction of drug doses following pharmacist recommendations, and information linking pharmacist interventions to each patient to improve the symptoms of ADEs. Furthermore, to date, no multicenter prospective studies have been conducted on deprescribing interventions by pharmacists. Therefore, to bridge this gap, the JSPPCS conducted a multicenter, prospective study to clarify the benefits of pharmacy interventions in reducing IMs in Japan. In our previous study, the questionnaire included analyses based on drug use status and professional certification. In this prospective study, we utilized the data obtained to perform a subanalysis, examining the impact of pharmacist intervention based on these factors (Suzuki et al., 2019; Uchida et al., 2019).

This was the first multicenter prospective observational study conducted in Japan to demonstrate how hospital pharmacists contribute to eliminating polypharmacy among inpatients. Of the 1,415 patients involved in the study, the pharmacists recommended discontinuing or reducing the dosage for 347 patients (24.5%). A total of 391 patients received recommendations from pharmacists’ to reduce IMs use, of which 368 (94.1%) were accepted. A total of 523 drugs were discontinued following the pharmacist’s recommendations. Pharmacist intervention improved hypnotic sedation, delirium, and hypotension. Although the trends in IMs were similar to those of a previous questionnaire survey (Uchida et al., 2019), we were able to reveal the results of medication reduction that the questionnaire survey could not clarify. The average number of 8.0 regular medications may reflect the actual situation of patients with cancer in clinical practice. This also reveals polypharmacy and the fact that pharmacists can reduce the number of medications. The pattern of deprescribing due to pharmacy intervention in the questionnaire survey (Uchida et al., 2019) differed from that in the current prospective observational study, indicating a difference between responses based on general knowledge and cases encountered in actual clinical practice. The main reasons for drug reduction were “less meaningful drug use” and “ADEs,” which were apparently associated with patient-related disadvantages. This result also differs from that of a questionnaire survey (Uchida et al., 2019). However, our prospective study of community pharmacists confirmed a similar trend (Uchida et al., 2022). Compared with prospective studies of interventions by community pharmacists, the rate of prescribing interventions by hospital pharmacists was clearly higher and more diverse. This observation could be attributed to the closer working relationships between hospital pharmacists and prescribing doctors. The proximity and collaboration between hospital pharmacists and physicians may have facilitated prescribing interventions, making it easier for hospital pharmacists to intervene compared to pharmacists working in community pharmacies.

This study provided a case report form through which hospital pharmacists could document whether the symptoms of ADEs improved after the intervention. Although the symptoms of ADEs varied widely, approximately 40% of the patients’ adverse symptoms improved after the intervention. Interventions aimed at reducing ADEs are the mainstay of IMs use. A multicenter study in Japan reported that 29 of 100 hospitalized patients had ADEs, of which 4.9% were severe and 1.6% were life-threatening (Morimoto et al., 2011). Patients who experience drug reactions have higher mortality rates and longer hospital stays than those who do not (Bond and Raehl, 2006). Half of the adverse drug reactions are preventable (Chan et al., 2001; Leendertse et al., 2008; Zed et al., 2008), and this intervention can improve patient quality of life, healthcare costs, and treatment. Especially in older individuals, renal dysfunction is a cause of adverse drug reactions due to unintentional overdose caused by delayed elimination of many drugs, such as water-soluble antibacterials, diuretics, and non-steroidal anti-inflammatory drugs (Mangoni and Jackson, 2004). A prospective observational study reported that the most common types of drugs causing adverse drug reactions were anti-infectives, steroids, anticoagulants, non-steroidal anti-inflammatory drugs, and diuretics (Geer et al., 2016). In this study, the reasons for drug reduction varied, leading to a corresponding reduction in the number of medications prescribed to patients. The diversity of reasons contributed to the decision to decrease the usage or dosage of specific drugs in order to address various concerns or optimize the patients’ medication regimens. ADEs were also diverse; however, because many of the patients were in the palliative medicine field, such as oncology, symptom improvement in hypnotic sedation, hypotension, and delirium were achieved in many cases.

Although the number of drug-drug interactions in the breakdown of IMs was small, the acceptance rate was high. Drug interactions are a potential concern for most medications, but it is important to note that the majority of interactions are not absolute contraindications. Instead, they often serve as guidelines for cautious administration. The use of medications with cautious administration can indeed present challenges, and even with knowledge and expertise, it can be difficult for pharmacists to determine in actual practice whether a patient’s condition is adversely affected by a drug interaction or whether to continue a medication that has been identified as causing a potential drug interaction. Although the number of interactions identified by pharmacists in this study was low, the acceptance rate by doctors was high because the interactions identified by pharmacists were clearly unfavorable to the patient. A Swedish drug registry study reported a strong relationship between the number of drugs and drug-drug interactions (Hajjar et al., 2007; Johnell and Klarin, 2007). Since the average number of regularly used drugs in the study was 8.2, it is expected that the number of drug-sensory interactions was high and consequently the associated potential for patient disadvantage was also high. In polypharmacy, drug interactions are more potentially concerning, and education or training on how to deal with them is necessary. Therefore, in Japan, continuing education to detect drug-drug interactions in routine clinical practice is necessary. As suggested by Scott et al., incorporating education on polypharmacy and deprescribing into pharmacists’ training would be advantageous for the success of future pharmacists (Scott et al., 2023).

Pharmacists reduced the number of drugs used in various pharmacological categories. The most common drugs discontinued following pharmacists’ recommendations were gastrointestinal medications (14.0%), followed by analgesics (13.2%), antipsychotics (8.2%), and hypnotic sedatives (8.0%). This was due to long-term administration of irresponsible or aimless medications, adverse drug reactions, duplication of the same type of drug, and indiscriminate prescription. Sleepiness or sedation, cognitive function, and other symptoms improved after discontinuation of the medications. These interventions resulted in the resolution of the ADEs. Our data suggest that the clinical services provided by pharmacists can contribute to reducing the disadvantages faced by outpatients.

Although most patients targeted for pharmacist intervention in this study were cancer patients, it was clear that those using opioids had a higher number of medications and were more likely to be able to reduce their medications. Since the JSPPCS conducted the study, half of the pharmacists who participated in the study had BCPPP, and most of the interventions were for patients with cancer. Comparing patients with and without opioid use showed that patients prescribed opioids use more medications, contributed more to medication reduction, and had a greater rate of improvement in adverse event symptoms. Although pharmacists must consider the use of opioids, the results clearly demonstrate their significance. When comparing pharmacists’ interventions with and without the BCPPP, there was no statistical difference in the number of drugs used between the two groups and no difference in the mean value of medication reduction. There was no statistical difference in the acceptance rate of interventions for IM, but there was a twofold difference in the number of suggestions made by the 27 pharmacists with BCPPP and the 23 without, 252 and 139, respectively. Furthermore, the rate of ADEs improvement was considerably higher in the BCPPP group than in the non-BCPPP group. Hence, the training and certification in BCPPP provided by the JSPPCS holds remarkable value. However, since these results were based on a univariate sub-analysis, a carefully designed prospective intervention study with or without BCPPP is needed to confirm this important clinical question.

Several definitions of inappropriate prescribing exist (O’Connor et al., 2012; O’Mahony et al., 2015; Panel et al., 2015; Levy, 2017; Chun et al., 2018), including the American Geriatrics Society Beers Criteria (Panel et al., 2015) and the Screening Tool of Older People’s Prescriptions (STOPP) (O’Mahony et al., 2015) which are well-known criteria that address multiple elements to reduce polypharmacy. However, to alleviate the burden on hospital pharmacists in their daily practice, we did not use international criteria to detect IMs in this prospective study. Hamilton et al. reported that potential inappropriate medications, as defined by the STOPP criteria, were extensively associated with avoidable drug reactions in older patients (Hamilton et al., 2011). Although the Beers criteria (Panel et al., 2015), the STOPP/START criteria (O’Mahony et al., 2015), and the Medication Appropriateness Index (Hanlon and Schmader, 2013) have clear evidence and should be used for evaluation, only a few facilities in Japan currently implement these criteria in their daily clinical practice. In addition, the use of these criteria requires diagnostic studies and imaging evaluations that pharmacists cannot perform regularly. Therefore, we did not use these criteria in this study. This is further because a Dutch study on outpatients reported that most drug-related problems were not associated with the STOPP/START criteria (Verdoorn et al., 2015); those criteria are not absolute. These criteria should be carefully selected according to the purpose of the study. As the purpose of our study was to clarify pharmacists’ drug reduction activities in actual clinical practice, the study was successful without evaluation using international standard criteria. Polypharmacy is often defined in numbers (Levy, 2017; Masnoon et al., 2017; Wastesson et al., 2018), but there is no unanimous definition of what constitutes polypharmacy (Masnoon et al., 2017). In our study, we define polypharmacy as the use of more drugs than clinically indicated rather than by a specific number (Fulton and Allen, 2005). ADEs cause some symptoms related to the central nervous system and the gastrointestinal tract, often resulting in the prescription of additional drugs to control these symptoms (Budnitz et al., 2011; Carroll and Hassanin, 2017; Wastesson et al., 2018), which is a necessary treatment for patients. Therefore, pharmacists must propose the use of necessary drugs based on the successive assessment of the patient’s treatment and not just the number of drugs. Prescribing multiple medications can negatively impact patient adherence and health-related quality of life (Schenker et al., 2019); it has been reported that medication adherence decreases with the number of medications prescribed (Pasina et al., 2014), making the involvement of pharmacists’ rather crucial.

This prospective study proved that pharmacists could contribute to improving the use of IMs in clinical practice in Japan. This study, however, has few limitations, including the possibility of bias in that most registered pharmacists were JSPPCS members, the small number of participating pharmacists, and the level of interest in polypharmacy among the participating pharmacists. Nonetheless, the results obtained in this study were not influenced by these limitations as they provide a background similar to the current situation in Japan, where there are many older patients suffering from cancer. While it is important to strive for minimizing bias in study design, it is worth noting that complete elimination of bias may not be feasible in certain prospective studies that involve real-world clinical data. We designed this study to minimize the burden of clinical trials by considering the work of pharmacists. In the data analysis, since the pharmacist intervention in actual clinical practice was summarized, its detailed analysis was not designed before the study, and the data obtained could not be analyzed in detail due to the study design, which is similar to a fact-finding survey. Therefore, subsequent studies are required to collect more detailed data. Alternatively, it is desirable to use a study design that clearly demonstrates the effectiveness of pharmacist intervention, such as a two-arm comparison with and without pharmacist intervention, or a pre- and post-pharmacist intervention comparison, rather than a single-arm observational study design. However, pharmacist intervention is a standard practice within Japanese medical care, and it is considered unethical for medical institutions to intentionally create an intervention group without including pharmacist intervention. Unfortunately, the more we aim to collect detailed data, the greater becomes the burden, drifting us away from actual clinical results. In addition, as there are reports of improvement in IMs by multidisciplinary teams led by nurses overseas (Garland et al., 2021), such clinical work should be conducted together with physicians, nurses, and other multidisciplinary teams (Liu, 2014); in terms of collaborative work, the role of clinical pharmacists in Japan is still in the process of development and may not be considered mature. In addition, because of the importance of working with patients to resolve polypharmacy and IMs, Woodward suggested deprescribing principles by reviewing all current medications, identifying medications for discontinuation, planning a deprescribing regimen, and working with patients and caregivers (Woodward, 2003). Therefore, a comprehensive approach is required.

This multicenter prospective observational study conducted in Japan is the first to demonstrate the effectiveness of hospital pharmacist interventions in promoting appropriate prescriptions. As a result, there was a significant reduction in the number of medications used and a decrease in polypharmacy. The study also highlighted the active role of hospital pharmacists in addressing polypharmacy by discontinuing inappropriate drugs or reducing the dosage of regularly prescribed medications, which contributed to the mitigation of ADEs.