A retrospective study comparing interventions by oncology and non‐oncology pharmacists in outpatient chemotherapy

Abstract Background The differences in the clinical pharmacy services (CPS) provided by oncology and non‐oncology pharmacists have not been sufficiently explained. Aim This study aimed to demonstrate the differences in direct CPS provided by oncology and non‐oncology pharmacists for patients and physicians, and to assess the potential impact of these services on medical costs. Methods We retrospectively examined CPS provided by oncology and non‐oncology pharmacists for outpatients who underwent chemotherapy between January and December 2016. Results In total, 1177 and 1050 CPS provided by oncology and non‐oncology pharmacists, respectively, were investigated. The rates of interventions performed by oncology and non‐oncology pharmacists for physicians‐determined treatment were 18.5% and 11.3%, respectively (p < .001). The rates of oncology and non‐oncology pharmacist interventions accepted by physicians were 84.6 and 78.8%, respectively (p = .12). Level 4 and Level 5 interventions accounted for 64.6% of all oncology pharmacist interventions and 53.0% of all non‐oncology pharmacist interventions (p = .03). The rates of improvement in symptoms from adverse drug reactions among patients resulting from interventions by oncology and non‐oncology pharmacists were 89.4 and 72.1%, respectively (p = .02). Conservative assessments of medical cost impact showed that a single intervention by an oncology and by a non‐oncology pharmacist saved ¥6355 and ¥3604, respectively. Conclusion The results of the present study suggested that CPS by oncology pharmacists enable safer and more effective therapy for patients with cancer and indirectly contribute to reducing health care fees.


| INTRODUCTION
Novel antineoplastic agents are being developed in large numbers, thus increasing options for therapy. At the same time, this proliferation of antineoplastic agents brings about a more complicated process in the use of chemotherapy, and heightens the importance of measures for adverse drug reactions (ADRs) other than hematotoxicity.
Consequently, pharmacists engaged in cancer therapy, are expected to serve many other roles. [1][2][3] Many studies have reported on the contributions of pharmacists to cancer therapy, including many Japanese studies. [4][5][6][7] Due to the high level of expert knowledge and skills required to fulfill these roles, Japan has created a system to train and certify oncology pharmacists (OP). For a pharmacist to be qualified for OP, the following items must be implemented. the services performed by OP. 8,9 One major Japanese study reported that the participation by OP in drug therapy can reduce medical costs and make therapy safer. 10 Another recent study has shown that intervention by pharmacists resulted in more suitable pharmacotherapy. 11 However, to the best of our knowledge, no study has ever compared the CPS conducted by OP with that conducted by non-oncology pharmacists (non-OP). Therefore, we aimed to demonstrate the differences between direct CPS provided by OP and non-OP for patients and physicians and to assess the potential impact of these services on medical costs.

| Intravenous chemotherapy practice at Shizuoka General Hospital
The present study is a retrospective comparison of services provided by pharmacists for outpatients undergoing chemotherapy at Shizuoka General Hospital, which is designated by the Japanese Ministry of Health, Labour, and Welfare as a Regional Core Hospital. This hospital provides high-quality medical care and plays a primary role in cancer care in central Shizuoka Prefecture. The Chemotherapy Center has 40 beds and provides outpatient chemotherapy on weekdays. To administer intravenous chemotherapy at this hospital, a physician must select a regimen that is previously registered in the electronic medical records. These regimens are based on evidence such as cancer therapy guidelines and clinical trials. Once a physician selects a regimen, the doses of the antineoplastic agents are automatically calculated based on the individual patient's background characteristics.
Also registered with these regimens are the recommended doses, routes, and rates of administration of infusions and supportive therapy injections necessary to administer the neoplastic agents. Therefore, by selecting the necessary regimen, physicians can provide consistent therapy. However, the physician can adjust the dosage based on the patient's condition. Before receiving antineoplastic agents, the patient is examined by a physician; if deemed eligible, the patient visits the Chemotherapy Center to commence chemotherapy.
At the Chemotherapy Center, CPS, which the management of ADRs and education for individual patients are conducted on weekdays by one OP and one non-OP. CPS for physicians includes pharmaceutical intervention; for example, if any problems arise in therapy, the pharmacist along with the physician intervenes to improve therapy. In Japan, pharmacists are not authorized to make the final deci-

| Study design
We retrospectively compared CPS provided by OP and non-OP for patients who underwent outpatient intravenous chemotherapy. All CPS were abstracted from the medical records and all endpoints were evaluated from the CPS with the type of pharmacist blinded. After evaluation, CPS were classified as either OP or non-OP service, and compared.

| Endpoints
The primary endpoint is the rate of pharmaceutical interventions con- According to several studies, the mean continuous pharmaceutical intervention rates by pharmacists that led to the prevention of major ADRs or of exacerbation of ADRs was 5.21%. [13][14][15] Hamblin et al.
believed that therapy using a Web-based system halved the risk of ADRs and also assessed the interventions by pharmacists. 16

| Statistical analysis
The sample size was calculated based on the number of CPS at Shizuoka General Hospital and past studies. The numbers of CPS for outpatients conducted by OP and non-OP at Shizuoka General Hospital are similar.
Sudou et al. reported that the rate of interventions by pharmacists before the examination by a physician is 17.6%. 17 Based on this figure, we predicted a mean baseline pharmaceutical intervention rate of 15.0% and hypothesized that the rate for OP would be 5% higher than the rate for non-OP. Using a two-sided test (α = .05, β = .10), we calculated that we would need to analyze a total of 2220 CPS. We began registering cases in January 2016 and continued until the month when the number of registered cases exceeded 2220. As a result, we analyzed CPS provided between January and December 2016. Because this was a retrospective study, it was difficult to obtain informed consent directly. Therefore, the study was conducted with a public document stating the study outline and the option for patient refusal to participate in the study. Percentages were compared using Fisher's exact test. All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (the R Foundation for Statistical Computing, Vienna, Austria).
More precisely, it is a modified version of R commander, designed to add the statistical functions frequently used in biostatistics. 18 The differences were considered statistically significant at p < .05.

| Patients characteristics
Patient characteristics is shown in Table 1. During the study period, a total of 2227 CPS were performed by five OP (1177 services) and six non-OP (1050 services). Neither group showed any evident imbalance in patient characteristics.

| Clinical pharmacy services
Specific CPS are shown in Table 2. During the study period, OP con-  We subsequently compared the quality of OP vs non-OP interventions which were based on Level 4-5 intervention rates. Significantly higher rates were found among OP than among non-OP, and these were 64.6% (Level 5, 5.6%; Level 4, 59.0%) and 53.0% (Level 5, 1.5%; Level 4, 51.5%), respectively (p = .03).
We then compared OP and non-OP in terms of symptom improvement rates where interventions for improving ADRs were accepted by the physician, and thereafter, the patient's symptoms were assessed. For both OP and non-OP, the most common ADR targeted for intervention was nausea/vomiting, followed by constipation/diarrhea. A significantly higher rate of ADR improvement was found among OP than among non-OP which was 89.4 and 72.1%, respectively (p = .02).  Table 3 and was then estimated to be ¥6355 for OP and ¥3604 for non-OP.

| DISCUSSION
The present study is a highly novel retrospective study in which we assessed and directly compared a sufficient number of CPS provided by OP and non-OP (Table 1). The study period included more CPS provided by OP than by non-OP, but this difference likely did not have a major effect on our analysis. The intervention rate among OP was significantly higher than that among non-OP. Our hospital provides chemotherapy according to a highly-regulated regimen system. Therefore, incorrect administration routes, rates of administration, or ADR measures that deviate from the guidelines are uncommon at the commencement of therapy. Thus, the number of interventions addressing these issues was fairly low. From the results in Table 2, the rate of Level 4-5 interventions and improvement of ADR after performed by OP was significantly higher than that performed by the non-OP, it is considered that OP has the higher skill to intervene in complex problems and better ADR management than non-OP. In a study by Sudou et al., when CPS were performed before a patient was examined by a physician, the intervention rate was 17.6%. 17 In the present study, all CPS were performed after examination by a physician. Although the study by Sudou et al. and the present study cannot be so easily compared, the fact that physicians try to optimize pharmacotherapy for individual patients by examining them suggests that rates of intervention by pharmacists would be lower when performed after examination than before examination. When this assumption is considered, the intervention rate among OP in the present study was sufficiently higher than the intervention rate reported

| CONCLUSIONS
The present study demonstrated the differences in the CPS provided by OP and non-OP and the impact of their interventions on medical costs. While CPS provided by both OP and non-OP greatly contributed to cancer therapy and reduction in medical costs, CPS by OP was, particularly of higher quality. The impact of CPS by both OP and non-OP may exceed the medical fees currently being charged for their services. Therefore, CPS for outpatients who undergo chemotherapy may not only provide better clinical management for patients but also reduce medical costs. Overall, the CPS performed by the OP has a greater effect than the CPS performed by the non-OP. Therefore, nurturing OPs to support more treatments is also an important issue.

ACKNOWLEDGMENTS
We would like to acknowledge the generous contributions and