Impact of oncology pharmacy services on the management of chemotherapy-induced nausea and vomiting: A systematic review and meta-analysis

Abstract

Purpose

To evaluate the effect of oncology services rendered by clinical pharmacists on reducing chemotherapy-induced nausea and vomiting (CINV) and improving overall treatment experiences.

Methods

A systematic review and meta-analysis were conducted using studies retrieved from PubMed, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), and Research Information Sharing Service (RISS). The incidence and severity of CINV were evaluated as primary outcomes. Secondary outcomes were patient adherence, patient satisfaction, quality of life (QoL), emergency department (ED) visits, hospitalizations, and costs.

Results

A total of 12 studies were selected for systematic review, with 8 studies eligible for meta-analysis. We found that clinical pharmacy services contributed to preventing and alleviating CINV as well as improving patient’s medication adherence, treatment satisfaction, and QoL, reducing hospital visits, and achieving cost savings. In the meta-analysis, pharmacists’ interventions were notably effective in reducing the incidence of nausea (odds ratio [OR], 1.917; 95% CI, 1.243-2.955; P = 0.003) and vomiting (OR, 2.491; 95% CI, 1.199-5.177; P = 0.014) during overall treatments periods relative to results in control groups. In addition, the impact of clinical pharmacy services on CINV control was greater during the delayed phase compared to the acute phase.

Conclusion

This study demonstrated the important role of clinical pharmacy services in controlling CINV and enhancing the overall treatment experience for patients with cancer. Further studies with standardized pharmacists’ services and outcome measures are needed to validate our findings.

Chemotherapy-induced nausea and vomiting (CINV) is one of the most common and troubling adverse effects for patients undergoing chemotherapy treatment.^1,2 Despite therapeutic advancement with antiemetic medications, uncontrolled CINV is associated with decreased treatment adherence, impaired functional status, and a negative impact on quality of life (QoL).^3-5 Poorly managed CINV can also increase the overall medical costs of anticancer treatments.^4,6 Therefore, proactive prevention and appropriate management of CINV still remain as important therapeutic goals for patients receiving chemotherapy.

CINV has been generally classified into 5 types: acute, delayed, breakthrough, refractory, and anticipatory nausea and vomiting (NV). Acute CINV is defined as the occurrence of NV within 24 hours after chemotherapy administration. Delayed CINV occurs 2 to 5 days following chemotherapy, and the peak intensity of the symptoms is typically known to occur in 48 to 72 hours post administration.^7,8 For managing different types of CINV, evidence-based clinical practice guidelines were developed by the National Comprehensive Cancer Network (NCCN), the American Society of Clinical Oncology (ASCO), and the Multinational Association of Supportive Care in Cancer (MASCC) and the European Society of Medical Oncology (ESMO).^9-11 These guidelines were structured considering the emetogenic risk of each antineoplastic regimen. Despite the availability of consensus recommendations, multiple studies have shown low adherence to the clinical guidelines for preventing CINV.^12,13 Gilmore et al¹² demonstrated that only half of the patients participating in their study adhered to guideline-consistent CINV prophylaxis, and results showed that the incidence of CINV was significantly higher in the guideline-inconsistent group compared to the consistent group. Furthermore, the occurrence and severity of CINV were closely related not only to the emetogenic potential of drugs but also to various patient-related risk factors such as age, gender, previous episodes of emesis, a history of motion sickness, and others.^8,14 Due to the complex interplay of these factors, healthcare professionals should consider guideline-based personalized antiemetic prophylaxis. This complexity has spurred numerous clinical pharmacy activities targeting patients receiving anticancer therapy.¹⁵

The oncology pharmacist is a clinical pharmacist with expertise in chemotherapy and profound knowledge of cancer prevention and treatment. Their roles have been recognized as increasingly essential with their experience and knowledge, maximizing the treatment effects on cancer patients while alleviating the adverse effects associated with cancer pharmacotherapy. Several systematic reviews have demonstrated the value of oncology pharmacists in the healthcare team.^16-20 Implementing clinical pharmacy services has significantly contributed to the effectiveness and safety of chemotherapy, patient education, and cost savings.^16,18,19 In the previous studies, clinical pharmacy services’ influence on patients’ clinical outcomes has been broadly discussed,^15-20 but few studies were focused on evaluating their effects on symptoms of NV. As studies systematically summarizing the impact of a pharmacists’ interventions in controlling CINV are scarce, a systematic and quantitative review study was constructed to describe the outcomes of pharmacists’ engagement in CINV management.

This study aimed to evaluate the effect of oncology services rendered by clinical pharmacists in reducing the incidence and severity of CINV and improving overall treatment experiences. To integrate the data quantitatively, we performed a meta-analysis using reported patient clinical outcomes of CINV following the initiation of chemotherapy.

Methods

Search strategy.

We performed a systematic review and meta-analysis in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (eAppendix).²¹ To identify relevant studies, we searched PubMed, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), and Research Information Sharing Service (RISS)—a Korean online database offering access to a wide range of academic materials—for studies published up to March 2024. Only articles related to humans and published in English were included. Additional manual searches to find relevant studies were also performed by reviewing the reference lists of the included studies. The following search keywords were used in combination: “cancer,” “pharmacist,” and “chemotherapy-induced nausea and vomiting.” The detailed search strategy is presented in Appendix  A.

Eligibility criteria and study selection.

According to the PICO (Patient, Intervention, Control, Outcome) framework,²² clinical studies meeting the following eligibility criteria were selected for the systematic review: (1) Patient, patients with cancer who received chemotherapy; (2) Intervention, pharmacist-led interventions; (3) Control, usual care or prior to the intervention implementation; and (4) Outcome, management of CINV.

There was no restriction on cancer types, and all patients receiving chemotherapy were included regardless of radiotherapy status. An intervention group (IG) is defined as the group that received an intervention led by a pharmacist or clinical pharmacy services. A control group (CG) is defined as encompassing both comparison and control groups, specifically referring to the group receiving the usual care. Studies were excluded if they (1) focused on pediatric or adolescent patients, (2) compared the effects of antiemetic drugs or evaluated the feasibility of implementing a new guideline, (3) were case reports, review papers, textbook, or not original research papers, or (4) were not available in full-text form. The literature screening and selection process were carried out independently by two reviewers (Y.S. and J.L.). Any inconsistency between reviewers during the literature selection process was resolved through discussions with a third researcher (S.S.).

Data extraction.

We extracted bibliographic information and study characteristics including country, study design, setting, number of patients, and type of cancer. We also collected information about the entity providing interventions and the method of outcome documentation. All activities led by pharmacists were extracted from each article’s full text and appendices, and lists of interventions types were sorted out as classified in the previously published studies.^16,20,23 The primary outcome was a comparison of NV symptoms between the IG and CG. The incidence rates of CINV during the acute (0 to 24 hours), delayed (2 to 5 days), and overall (0 to 5 days) periods during the anticancer therapy were collected. A comparison of patients’ symptom scores related to CINV was also included. Additionally, data on patient adherence, patient satisfaction, QoL, ED visits, hospitalizations, and medical costs were collected as secondary outcomes.

In the meta-analysis, a binary outcome of whether or not NV symptoms occurred during the follow-up period was evaluated. To calculate the effect size, we standardized the different formats of outcomes to incidence rates. If the result was presented as a bar graph, the WebPlotDigitizer version 4.6 software (Ankit Rohatgi, Pacifica, CA) was used to read the numerical values on the graph. Among various data extraction programs, we selected this program as the most suitable choice regarding reliability, validity, and usability as demonstrated in previous studies.^24,25

Statistical analysis and heterogeneity.

A pharmacist-led IG was compared with the CG to evaluate the effect of the clinical pharmacist interventions on improving CINV management. Effect sizes for pharmacist-led interventions were calculated as the odds ratio (OR) with 95% confidence interval (CI). Subgroup analyses were conducted to reduce the heterogeneity from different study designs. Results from the I² statistic method were reported to indicate the level of heterogeneity within the studies included in the meta-analysis. As the number of studies included in the analysis was small, we used the random-effects model to consider that each included study providing information about a different effect size was well represented in the summary estimate.²⁶ As the assessment of publication bias was considered difficult when the number of the included studies for meta-analysis was 10 or less, the potential for bias was not assessed in this study.²⁷ The meta-analysis was performed with Comprehensive Meta-Analysis version 4 (Biostat Inc., Englewood, NJ).

Quality assessment.

To evaluate the quality of the included studies for the systematic review, 3 different assessment tools were used considering the study design of the included studies: the Risk of Bias of Non-Randomized Intervention Studies (ROBINS-I) tool, the NIH Quality Assessment Tool for Before-after (Pre-Post) Studies with No Control Group, and the Risk of Bias of Randomized Control Trials (RoB) 2.0 tool.^28-30 Quasi-experimental studies with a control group were assessed by the ROBINS-I tool. The ROBINS-I tool is comprised in 7 domains: confounding variables, selection of participants, classification of interventions, deviations from intended intervention, blinding of outcome assessment, missing data, and selective reporting. Based on the ROBINS-I domains, all included studies were classified by risk of bias in one of the following categories: low, moderate, serious, critical, or “no information.”²⁸ For one-group pretest-posttest studies, the NIH assessment tool uses a scoring system to rate studies for risk of bias as follows: poor (rating of <50%), fair (rating of 50%-75%), and good (rating of 75%-100%).²⁹ The RoB 2.0 tool was used to assess 5 areas of study performance (randomization, deviations from intended intervention, missing data, measurement of outcome, and selection of the reported result).³⁰ Risk of bias in each area was evaluated as low, “some concerns,” high, or “no information.” Two researchers (Y.S. and H.R.) independently evaluated the quality of all studies, and any disagreement was resolved through discussion sessions with other researchers (S.S. and E.L.).

Results

Literature search.

A total of 2,414 studies were identified in the literature search using 4 electronic databases (Figure 1). Following the removal of duplicates, 2,282 studies remained for screening titles and abstracts, and of those, 1,843 were excluded. The majority of the excluded studies focused on the efficacy or metabolism of specific anticancer agents or were not related to chemotherapy. Studies published as guidelines or review articles were also excluded during the first screening process. The remaining 439 studies were assessed for eligibility by screening full-text articles according to the established inclusion and exclusion criteria. A total of 12 studies were selected for systematic review and qualitative analysis.^2,31-41 Of those, 8 studies were eligible for meta-analysis.^2,31-36,41

Study characteristics.

Overall characteristics of the included studies are summarized in Table 1. The interventional studies were conducted in various countries: United States (N = 4), Japan (N = 4), Germany (N = 2), Spain (N = 1), and India (N = 1). Among a total of 12 studies, 11 were nonrandomized studies, including cohort and pretest-posttest designs, and one study was a randomized controlled trial (RCT). Cancer types were diverse, including both solid and hematological malignancies. Patients in the CG received usual care with physician appointment scheduling. The IG was provided with individual clinical care by clinical pharmacists during cancer treatment.

Pharmacists’ interventions for patients receiving chemotherapy.

Descriptive summaries about pharmacist-led interventions that were implemented in various forms according to the clinical settings of different countries are presented in Table 2. Patients in the CG received usual care without any provision of clinical pharmacy services.^2,31-41 Pharmacists’ practices were limited to verifying prescription orders and mixing anticancer injections,³² and patients in the CG did not have regular appointments with a pharmacist.^31,36,41 For IG patients, services led by pharmacists targeting patients with cancer were classified by the following types of services: medication reconciliation, identification of medication-related problems (MRPs), management of adverse drug reactions (ADRs), patient counseling and education, establishment of guidelines on supportive care services, and educating other healthcare professionals. The most frequently observed intervention types were patient counseling and education (91.6%), followed by identification of MRPs (83.3%) and management of ADRs (66.6%).

At first visit, patients were asked for their medication history and current medication list and their medication-use patterns were assessed by pharmacists during the appointment.^37,38,41 Clinical pharmacists conducted medication chart reviews to check the appropriateness of the drug doses, dosage forms, frequencies, and drug-drug interactions (DDIs), including anticancer agents, antiemetic agents, and other medications. Through the medication reviews, clinical pharmacists identified individual patient’s MRPs.^{2,31-33,35-38,40,41} Pharmacists monitored patient’s conditions regularly and offered supportive care. Antiemetic medications were added, discontinued, or adjusted in dosage, duration, or frequency as clinically appropriate when CINV occurred.^{2,32,35,37-41} To monitor CINV symptoms after chemotherapy administration and provide persistent interventions, pharmacists conducted face-to-face interviews at every regular clinic visit, supplemented by consultation with a pharmacist at any time via telephone or email.^{2,32-36,38,41} Through these activities, clinical pharmacists identified patient’s current problems and made recommendations to physicians to resolve any potential problems. If the medication change was included in a pre-established protocol with the agreement of the interprofessional team, pharmacists were permitted to make direct modifications.^38-40 During counseling and education sessions, patients received information about their treatment plan, potential adverse effects, and preventative strategies to optimize medication adherence.^{2,31-36,38-41} Patients received brochures or fact sheets made by pharmacists, which included instructions regarding their chemotherapy and information about ADRs.^34,35,41 Pharmacists also intervened in medication reconciliation during a patient’s transitions of care.^37,38,41 For indirect patient care, pharmacists developed evidence-based treatment guidelines related to antiemetic prophylaxis.^31,34,38-40 In addition, pharmacists interacted with patients, other members of the healthcare team, and professional students to offer them a better understanding of medications, potential DDIs, and ADRs through drug information and education sessions.^32,36-38

Primary outcomes.

Patient clinical outcomes related to CINV were defined as the primary outcomes (Table 1). We quantitatively analyzed the incidence rate of CINV, comparing the IG and CG. Due to a limited number of studies and diverse outcome measurements, the severity of the symptoms was qualitatively assessed.

Incidence rate of CINV.

Among the 12 studies included, 10 studies reported the occurrence of NV as an outcome measure of the impact of pharmacists’ interventions.^{2,31-37,40,41} The incidence of CINV denoted that the prescribed antiemetic medication was inappropriate, insufficient, or not adhered to by patients, resulting in inadequate control. The data on clinical outcomes related to nausea and to vomiting were separately reported in all studies, excluding Suzuki et al.³⁷ In Suzuki et al,³⁷ the incidence rate of a grade 2 or higher level of NV was significantly reduced after the implementation of pharmacist interventions (from 8.0% to 0.9%, P < 0.01). Due to the inability to ascertain which patients achieved the outcomes, Quinn et al⁴⁰ was also excluded from the quantitative analysis. Quinn et al⁴⁰ showed that 89.1% of patients had an overall reduction in their nausea and vomiting from baseline after intervention. Therefore, we conducted meta-analyses for separate nausea and vomiting outcomes with 8 studies.^2,31-36,41

Meta-analysis of the overall period (days 0 to 5).

A total of 6 studies presented outcomes for CINV incidence rate for the overall period (days 0 to 5).^{2,31,32,35,36,41} Both nausea and vomiting during the overall period were well controlled in cancer patients who received pharmacists’ interventions such as medication review, management of ADRs, treatment education, and counseling. A meta-analysis showed that the odds of nausea events were almost twice as high in the CG compared to the IG (OR, 1.917; 95% CI, 1.243-2.955; P = 0.003) (Figure 2A). For chemotherapy-induced vomiting, a pooled analysis showed that the odds of vomiting events in the CG were 2.5 times higher than in the IG (OR, 2.491; 95% CI, 1.199-5.177; P = 0.014) (Figure 2B). Due to the difference of study design, subgroup analyses were performed with only non-RCT studies (Appendix  B). The odds of nausea events in the CG were 2.2 times higher in the CG than in the IG (OR, 2.218; 95% CI, 1.177-4.178; P = 0.014). The odds of vomiting were also 3 times as high in the CG compared to the IG (OR, 3.055; 95% CI, 1.398-6.677; P = 0.005).

Meta-analysis by each phase (acute or delayed).

In the overall period, it was clear that pharmacists’ interventions led to a significant improvement in the management of CINV based on the quantitative analyses. With the CINV classified based on the timing of occurrence, additional analyses were performed to determine at which point the effectiveness of the intervention was maximized: whether interventions had an equal effect during both the acute and delayed phases or if they prominently affected a specific phase. A total of 3 studies were included in the additional analyses (Figure 3).^32-34 Findings showed that the delayed phase was a key period associated with the impact of pharmacist-led interventions. In the delayed phase, the odds of nausea events were 1.8 times higher in the CG than in the IG (OR, 1.822; 95% CI, 1.114-2.979; P = 0.017). The odds of vomiting were also higher, although not significantly, in the CG compared to the IG among patients who received pharmacists’ interventions (OR, 2.292; 95% CI, 0.925-5.679; P = 0.073).

Severity of patients’ symptoms related to CINV.

Severity of patients’ CINV symptoms were evaluated in 3 studies.^31,38,39 In Liekweg et al,³¹ the median scores indicating the severity of nausea in the IG were reduced by 39.4% and 34.5% during the acute and delayed phases, respectively, compared to the CG. However, these reductions were not statistically significant. The frequency of vomiting in the IG was significantly decreased relative to the frequency in the CG in both the acute phase (P < 0.001) and delayed phase (P = 0.002). Also in Jackson et al,³⁹ the change in symptom scores for CINV assessed by 5 patients were presented. At baseline, the median scores were 8 points for acute nausea, 10 for delayed nausea, and 4 for delayed vomiting. After the implementation of pharmacists’ clinical interventions, all previously mentioned symptoms were reduced, with a rated median score of 0.

Secondary outcomes.

For secondary outcomes, a qualitative systematic review was conducted regarding patient adherence, treatment satisfaction, QoL, ED visits, hospitalizations, and medical costs (Table 3).

Patient adherence.

Patient adherence to treatment was reported as an outcome in 6 studies.^{2,34,35,38,40,41} In 3 studies presenting data on self-reported medication adherence,^34,35,41 2 studies showed significantly increased adherence rates in the IG compared to the CG.^35,41 Based on the claims data, Nhean et al² reported that adherence to capecitabine was significantly higher for patients in the oral chemotherapy management program than for those in the CG, with 175 patients in each group (97% vs 94%; P = 0.03). Quinn et al⁴⁰ measured the status of adherence to antiemetic therapy involving a total of 46 patients. At baseline, 23.9% of the patients were considered as nonadherent. However, after implementing a pharmacist-led program to manage the refractory CINV, all patients became adherent.

Patient satisfaction.

Two studies reported patient satisfaction with cancer treatment and pharmacist interventions.^31,38 Liekweg et al³¹ reported that patient satisfaction with cancer treatment was significantly improved upon implementation of clinical pharmacy services (P = 0.009). Providing patients with information on cancer therapy, adverse effects, and information sources greatly improved their satisfaction. Hansen et al³⁸ found that all 12 patients enrolled in the program had a favorable response to the quality of pharmacists’ services and 91% expressed interest in scheduling a return visit with the clinical pharmacists.

QoL outcomes.

Two studies demonstrated a difference in QoL between the IG and CG after chemotherapy.^36,41 In an RCT study, the mean symptom score was notably lower in the IG compared to the CG at week 4, as measured by the nausea/vomiting sections of the European Organization for Research and Treatment of Cancer QLQ-C30 tools (P = 0.036).⁴¹ In another study in patients with breast cancer, quality-adjusted life-years (QALY) were calculated with the EQ-5D-5L instrument in both groups.³⁶ After the completion of 6 cycles of chemotherapy, a significant improvement in QALY was observed in the IG (P < 0.05).

ED visits and hospitalizations.

As many patients were reported to have ED visits or hospitalizations due to uncontrolled adverse events during chemotherapy, the impact of the role of clinical pharmacists on the management of NV was observed in 3 studies.^2,35,37 In Todo et al,³⁵ 2 out of 13 patients (15%) in the CG visited the ED because of grade 3 nausea. Following the interventions, there were no cases of ED visits. Nhean et al² demonstrated that ED visits and the number of hospitalizations due to adverse events were decreased after the implementation of the intervention program. Particularly, the rate of ED visits due to nausea or vomiting was reduced by 12.5% (from 17.1% to 4.6%, P < 0.001) or 6.9% (from 10.3% to 3.4%, P = 0.01), respectively, in the IG. Similarly, the rates of hospitalizations due to nausea (P < 0.001) and vomiting (P = 0.003) were significantly decreased. In the study by Suzuki et al³⁷ involving 121 patients undergoing chemo-radiotherapy, the mean duration of hospitalization of patients showing improvement of grade 2 or higher ADRs after medical interventions was significantly shorter than among those who did not receive interventions (43.1 vs 67.7 days, P = 0.01).

Medical costs.

Two studies reported economic impact in terms of cost savings.^32,37 Iihara et al³² reported that appropriate use of antiemetics led to cost reduction of medications by 15.7% from $147.7 to $124.5 (P < 0.01) per course of chemotherapy. In Suzuki et al,³⁷ the estimated reduction of hospital stays through pharmacists’ interventions resulted in cost saving of $517,000 over 18 months.

Quality assessment.

For 7 quasi-experimental studies with a control group assessed for risk of bias by the ROBINS-I tool (Appendix  C), 3 studies showed a moderate level,^2,31,35 and 3 had a serious level due to deficiencies in the confounding domain.^32-34 Compared to a well-performed randomized trial, at least one known important confounding factor was detected in those studies. One study was judged to have no information because there was a lack of information on missing outcome data.³⁶ Using the NIH quality assessment tools for one-group pretest-posttest studies, all studies showed a fair level of risk of bias (Appendix  D).^37-40 Due to the nature of the intervention studies, it was not feasible to blind outcome assessors to participants’ interventions across all studies. Moreover, the majority of the studies did not achieve sufficient sample size to guarantee the robustness of the findings. One RCT study was rated as “some concerns” using the RoB 2.0 tool in the area of deviations from the intended intervention in the blinding process (Appendix  E).⁴¹

Discussion

For oncology practices, research specifically addressing CINV, ie, the most prevalent and challenging adverse effects in patients with cancer, has been notably lacking. This systematic review documented the value of clinical pharmacy services in management of CINV. The meta-analysis suggested that pharmacist intervention significantly reduced both nausea and vomiting in the 5 days immediately following chemotherapy administration. Additionally, pharmacist-led interventions were associated with improvements in medication adherence, treatment satisfaction, and QoL, as well as reductions in ED visits, hospitalizations, and costs.

It was especially noteworthy that clinical pharmacy services were associated with greater control of CINV in the delayed phase. This finding was important because controlling delayed CINV is generally more problematic than controlling CINV in the acute phase.^13,42-45 In a survey on challenges in preventing and managing CINV, “controlling nausea/vomiting in the delayed phase” (64%) ranked as the top concern.¹³ Studies showed that despite prophylactic antiemetic therapy, patients exhibited a higher incidence of CINV in the delayed phase compared to the acute phase.^42-44 To address uncontrolled symptoms, twice as many patients needed rescue antiemetics during the delayed phase compared to the acute phase.⁴² Moreover, in a retrospective cohort study with 19,139 patients, 99.3% of CINV-associated hospital visits occurred during the delayed phase.⁴⁵ Thus, our results suggest that the pharmacists made important positive professional contributions in reducing or mitigating CINV in patients with cancer, especially during the delayed phase.

The scope of the clinical pharmacist’s practice has been expanded, moving beyond verification of prescription orders to finding current problems, providing interventions for preventing ADRs, and solving DRPs. For studies included in the meta-analyses,^2,31-36 the major intervention types commonly observed were patient counseling sessions and identification of MRPs. Providing appropriate education and counseling may help to increase patient treatment adherence and control of CINV.^2,3,34,35 Additionally, pharmacists checked patients’ medication list and facilitated the implementation of prophylactic treatment with anticancer agents following evidence-based guidelines. Pharmacist-driven patient-centered care and comprehensive medication management services are tailored to patients’ needs and preferences considering their clinical conditions. This approach could ultimately lead to improved clinical outcomes, especially in preventing and managing CINV.

The implications of our study suggest that clinical pharmacy services can have an important role in controlling CINV and improving the overall treatment experience for patients with cancer. While we cannot definitively ascertain which components of and to what extent the pharmacist intervention exerted the greatest influence, it is reasonable to assume that a combination of pharmaceutical factors led to an additive or synergistic beneficial effect on CINV management.

This study had some limitations that should be considered. First, there was heterogeneity across the included studies. Pharmacist interventions provided to cancer patients were not standardized, and there was a wide range of cancer types and treatment regimens in each study. Previous descriptive reviews of clinical pharmacy services had also pointed out the heterogeneity of activities performed by pharmacists in practice settings.^23,46 As it was challenging to consider care setting, region, and countries due to the combination of clinical outcomes from various studies, a subgroup analysis by intervention types was not performed. Second, this systematic review was primarily comprised of quasi-experimental studies. Compared to a well-performed randomized trial, these studies were prone to selection bias and failed to maintain blinding of outcome measurement. Thus, there were constraints on the validity of the results obtained in the analysis. However, quasi-experimental studies gave information about clinical pharmacy models currently used in oncology practice and patient clinical outcomes in the real world. We confirmed the impact of pharmacist-led interventions on CINV control and overall treatment experience in actual clinical settings rather than an overly controlled environment.

To move forward and overcome the limitations, it is necessary to establish standardized forms of oncology pharmacy services and conduct well-designed intervention studies with large sample sizes. Further studies may be able to provide a high level of evidence for the impact of clinical pharmacists’ interventions on managing ADRs in oncology patients.

Conclusion

This systematic review and meta-analysis on clinical pharmacy services in oncology units demonstrated improvements in the intervention group regarding CINV control, treatment adherence, and patient satisfaction. Clinical pharmacists also contributed to enhancing the QoL for patients and reducing healthcare resource utilization and medical costs. Further randomized controlled trials with standardized pharmacists’ services and outcome measures are needed to validate our findings.

Data availability

The data underlying this article are available in the article and its online supplementary material.

Disclosures

This work was supported by the 4 phases of the Brain Korea 21 program in 2023. The authors have declared no potential conflicts of interest.

Appendix A—Search strategy

Appendix B—Results of meta-analysis of only non-RCT studies

Appendix C—Results of bias assessment using ROBINS-I tool

Appendix D—Assessment of risk of bias using NIH Quality Assessment Tools

Appendix E—Results of bias assessment using RoB 2.0 tools

References