Although the early detection of DDI-induced ADRs can contribute to affording safe drug therapy, there is limited information on drugs capable of increasing DDIs in clinical practice for utilization by community pharmacists. Therefore, in the current study, we retrospectively investigated the occurrence of potential DDIs in patients who received prescriptions from multiple clinical departments.

The average number of drugs taken was 7.2, which was 1.6-fold higher than the 4.5 drugs per prescription in the same region reported in our previous study [12]. Additionally, in the current study, the incidence of CYP3A4-related potential DDIs with prescriptions from multiple clinical departments (84.8–100.0%) was 1.6–1.9-fold higher than that of the subdivided control data (52.5%). These results indicate the importance of centralized drug management from the perspective of DDIs for patients who receive prescriptions from multiple clinical departments. Although 81.2% of patients visited two clinical departments in the current study, an analysis based on receipt data from more than 1 million patients in Tokyo has revealed that the sum of patients who visit three or four medical institutions does not differ from the proportion who visit two medical institutions among those aged ≥ 75 years [14]. This discrepancy may be attributed to differences in the methodology of this study, which was counted by the clinical department, not by a medical institution, or conducted in one local city area, the Kojima region, Okayama. However, the receipt-based study discussed above supports the significance of focusing on prescriptions from multiple clinical departments.

Previous study has reported a substantially higher incidence of ADRs in hospitalized patients taking ≥ 6 drugs when compared with those taking 1–3 drugs [15]. Thus, in the current study, eligible patients were divided into two groups for analysis: those taking ≥ 6 drugs and those taking ≤ 6 drugs. The incidence of CYP3A4-related potential DDIs due to two or more or three or more drugs was markedly elevated only in patients taking less than six drugs in the multiple clinical department group when compared with the subdivided control data. This finding may suggest that DDI-related ADRs require more attention in patients receiving prescriptions from multiple clinical departments than in those receiving prescriptions from a single department, even if the number of drugs taken is less than six. Additionally, given that metabolic mechanisms via CYP molecular species become more complex, DDIs due to three or more drugs may exert a greater impact on drug blood levels and the occurrence of ADRs than DDIs due to two drugs. Herein, the 2.6-fold higher incidence of CYP3A4-related potential DDIs due to three or more drugs in patients taking less than six drugs in multiple clinical department group, compared with that in the subdivided control data, emphasizes the importance of centralized management of multiple clinical department prescriptions. In contrast, when six or more drugs were taken, there was no difference in the frequency of potential DDIs between the subdivided control data and the prescription data from multiple clinical departments. Polypharmacy is a well-recognized risk factor for DDIs [16, 17]. Our ROC curve analysis also showed that the calculated threshold values of the number of drugs taken, which was associated with the existence of potential CYP3A4-related DDIs, were both six drugs in potential CYP3A4-related DDIs due to ≥ 2 or ≥ 3 drugs in the prescriptions from multiple clinical departments. Although these results support our decision to divide the data into more than and less than six drugs, the findings of this study highlight that attention should be paid to ADRs caused by CYP3A4-related potential DDIs in patients taking over six drugs, regardless of the number of prescribing clinical departments. One study has reported that the use of potentially inappropriate medications is not necessarily associated with the number of drugs prescribed [18]. Therefore, in patients with polypharmacy, in addition to reducing the number of drugs, it is crucial to focus on optimizing the content of prescriptions from the perspective of safety, including DDIs.

Multiple regression analysis identified the top three therapeutic categories with the highest impact on CYP3A4-related potential DDIs, i.e., cardiovascular agents, agents affecting central nervous system, and urogenital and anal organ agents. Although the number of prescriptions for urogenital and anal organ agents was lower than that for cardiovascular agents or agents affecting central nervous system, most of prescribed urogenital and anal organ agents were CYP3A4-related drugs. Considered in conjunction with the finding that more than 90% of urogenital and anal organ agents are prescribed in urology, whereas most cardiovascular agents and agents affecting central nervous systems are not prescribed in urology, urogenital and anal organ agents may increase the risk of ADRs due to potential CYP3A4-related DDIs that prescribing doctors are not always aware of in patients receiving prescriptions from multiple clinical departments. These findings would be useful in alerting community pharmacists to the existence of CYP3A4-related potential DDIs, and also highlight the importance of centralized management of prescription drugs by community pharmacists. Patients are more likely to experience potential DDIs and serious DDIs with increasing age [19, 20]. However, as 85.8% of the patients in this study were aged ≥ 65 years or older, “age” may have been overlooked as a factor contributing to the increased number of potential DDIs. In addition, the number of clinical departments was not extracted as a factor influencing the increase in potential DDIs by multiple regression analysis in this study. This may be because the number of patients who received prescriptions from three or four clinical departments was smaller than the number of patients who received prescriptions from two departments. A survey in community pharmacies in Croatia revealed that 84% of eligible patients experienced at least one clinically notable DDI [21]. A cross-sectional and cohort study in Denmark has reported that CYP3A4 is the most prominent CYP isoenzyme involved in mortality and readmission rates; the authors also found that combinations of clarithromycin with ticagrelor, tacrolimus, or everolimus were associated with substantially elevated readmission rates [22]. Thus, the optimization of prescriptions to reduce the potential DDIs is of great importance in reducing the risk of ADRs occurrence. Future research on the occurrence of ADRs due to CYP3A4-related DDI with a focus on combinations with high-risk drugs (e.g., antiplatelet agents and immunosuppressant drugs) may be needed, in addition to exploring the number of drugs involved in CYP3A4-related potential DDIs.

This study has several limitations. First, because we identified patients who were prescribed drugs from multiple clinical departments based on their prescription itself, drug history accumulated in the pharmacy, or medication notebook, all patients who were prescribed drugs from multiple clinical departments may not have been identified. Second, because this study was conducted in five pharmacies within the same region, the drugs included in the analysis may reflect the characteristics of the region. Third, we excluded the intensity affecting metabolism by CYP3A4. Although concomitant use of drugs metabolized by the same CYP molecular species (potential DDIs by more than only two “competition” drugs in this study) is often listed as a concomitant use precaution in the package insert, the actual impact on DDI has not been fully investigated. We also excluded the involvement of CYP3A4-inducing drugs. Of the 372 patients, 7 (1.9%) were prescribed some CYP3A4-inducing drugs, and actual DDIs may differ from our results. The same limitation would be applicable to metabolic pathways other than CYP3A4. Finally, we did not examine the presence of ADRs. Further clinical studies are required to determine the extent to which CYP3A4-related potential DDIs result in ADRs. Nevertheless, the results of the current study highlight the importance of centralized drug management for patients who receive prescriptions from multiple clinical departments to avoid DDIs. Furthermore, ADR monitoring utilizing the results of this study would contribute to the early detection of DDI-induced ADRs in an aging society.