Intravenous compounding robots in pharmacy intravenous admixture services: a health technology assessment

This study was performed to conduct a health technology assessment of intravenous compounding robots and provide the currently available best evidence for hospital decision-makers. A comprehensive search of relevant professional health technology assessment websites and electronic databases (Embase, PubMed, The Cochrane Library, CNKI, VIP, CBM, and Wanfang Database) was conducted from inception to 11 August 2020. Two reviewers independently screened literature according to the inclusion and exclusion criteria and extracted data. The results were reported by qualitative description because of heterogeneity in the characteristics of the data in the included studies. Forty studies were included: 2 health technology assessments, 24 control studies (6 randomized controlled trials, 13 non-randomized contemporaneous controlled trials, and 6 non-randomized historical controlled trials), 12 non-controlled studies (11 single-arm studies, 1 investigation report), 1 qualitative study (in-depth interview), and 1 economics research. Effectiveness: The robots improved the production eciency compared with usual/manual preparation; however, the intravenous preparation process requires further optimization. Additionally, robots reduced the incidence of medicine residues, preparation errors, and preparation failures. The accuracy and solution properties of intravenous admixture medicines were satisfactory, and the robots also contributed to error recognition. Safety: The robots reduced product pollution and environmental pollution, but vigilance is still required to ensure that pollution stays low. The robots also reduced the incidence of health damage to technicians. Economy: The robots reduced material costs in these studies; however, whether they can reduce labor costs remains unclear. Social suitability: Technicians had a high degree of satisfaction with the robots, but few relevant studies focused on this aspect.


Abstract Background
This study was performed to conduct a health technology assessment of intravenous compounding robots and provide the currently available best evidence for hospital decision-makers.

Methods
A comprehensive search of relevant professional health technology assessment websites and electronic databases (Embase, PubMed, The Cochrane Library, CNKI, VIP, CBM, and Wanfang Database) was conducted from inception to 11 August 2020. Two reviewers independently screened literature according to the inclusion and exclusion criteria and extracted data. The results were reported by qualitative description because of heterogeneity in the characteristics of the data in the included studies.

Results
Forty studies were included: 2 health technology assessments, 24 control studies (6 randomized controlled trials, 13 non-randomized contemporaneous controlled trials, and 6 non-randomized historical controlled trials), 12 non-controlled studies (11 single-arm studies, 1 investigation report), 1 qualitative study (in-depth interview), and 1 economics research. Effectiveness: The robots improved the production e ciency compared with usual/manual preparation; however, the intravenous preparation process requires further optimization. Additionally, robots reduced the incidence of medicine residues, preparation errors, and preparation failures. The accuracy and solution properties of intravenous admixture medicines were satisfactory, and the robots also contributed to error recognition. Safety: The robots reduced product pollution and environmental pollution, but vigilance is still required to ensure that pollution stays low. The robots also reduced the incidence of health damage to technicians. Economy: The robots reduced material costs in these studies; however, whether they can reduce labor costs remains unclear. Social suitability: Technicians had a high degree of satisfaction with the robots, but few relevant studies focused on this aspect.

Conclusion
Intravenous compounding robots have certain advantages in terms of safety, effectiveness, economy, and social adaptability. High-quality and large-sample randomized controlled trials or well-designed observational studies are still needed to evaluate such robots, especially in terms of economic and social suitability.

Background
Intravenous infusion therapy is one of the main routes of administering medication in the clinical setting. The pharmacy intravenous admixture service (PIVAS), an essential department involved in the centralized preparation and supply of intravenous admixture medicines in medical institutions, plays an important role in the quality control of intravenous admixture medicines [1]. The working mode of PIVAS is characterized by a large workload, tight timeline, multiple links, and high requirements. With the increasing demand for intravenous admixture medicines, completion of the preparation by manual labor alone will result in problems such as low e ciency, numerous errors, and high risks. One systematic review showed that the incidence of an incorrect dose, incorrect concentration, and inadequate aseptic technique in the preparation of intravenous admixture medicines in medical institutions were 32.6%, 88.6%, and 92.7%, respectively, bringing potential harm to patients [2]. A multicenter study showed that the concentration of a biomarker of DNA oxidative damage (8-hydroxy-2deoxyguanosine) was signi cantly higher in PIVAS staff exposed to anti-tumor drugs than in the control group, suggesting that PIVAS staff face severe health risks due to the manual preparation of anti-tumor drugs [3]. Another multicenter survey showed that PIVAS nurses were under great pressure and that the main sources of pressure were the fear of medical accidents and occupational injuries, insu cient sleep, and fatigue [4].
With the rapid development of electronic information technology, increasingly more intelligent software and equipment are being applied to PIVAS, assisting PIVAS staff in completing various tasks [5]. Automated technology is an effective way to reduce preparation errors and relieve the working pressure and occupational injuries of PIVAS staff. According to data released by the American College of Hospital Pharmacists, 0.3% of hospitals have introduced intravenous compounding robots [6]. In 2019, Amodeo et al. [7] found that the application of intravenous compounding robots increases precision, improves safety, decreases costs, and saves time, which may lead to a reduction in medication errors and improvement of patient and family care. In 2014, Urbine et al.
[8] established a pathway model based on the published literature to estimate the nancial bene t of robotic preparation compared with traditional manual preparation. The results showed that the use of the robotic device prevented 5,420 medication errors and resulted in an associated savings of $288,350 per year, showing excellent preparation accuracy and economic bene ts. However, in 2015, Nurgat et al. [9] reported that robots were used to prepare only 13.79% of anti-tumor drugs in the third year after installation, 15% of intravenous admixture medicines failed to meet the dose accuracy requirement, and robot throughput relative to the manual compounding process was low and associated with substantial medication waste. In 2013, Chen et al. [10] assessed the human resource cost after applying robot technology and found that robots were unable to improve the full-time equivalents of pharmacists/technicians, and an additional on-site engineer was even required in some cases to supervise and resolve possible machine problems.
Before introducing medical equipment, hospitals need to perform a comprehensive evaluation and ensure that the medical equipment can bene t the patients, medical staff, and hospital. Considering the signi cant differences in the evaluations of intravenous compounding robots in China and abroad, it is necessary to conduct a health technology assessment (HTA) to evaluate the effectiveness, safety, economy, and social suitability of intravenous compounding robots.
Therefore, based on the attention recently being given intravenous compounding robots, we performed the present HTA to evaluate the application of intravenous compounding robots and thus provide evidence for the introduction of intravenous compounding robots in PIVAS.

Inclusion and exclusion criteria
The inclusion criteria for this assessment were as follows. (1) Study types: comprehensive HTA reports, systematic reviews/meta-analyses, randomized controlled trials (RCTs), non-randomized controlled trials (NRCT), prospective/retrospective observational studies, and economic studies. (2) Study objects: Intravenous admixture medicines that were prepared in the hospital pharmacy. (3) Interventions/comparisons: An intravenous compounding robot/automatic dispensing equipment (ADE) was applied in the intervention group, and conventional/traditional manual preparation or other types of intravenous compounding robots/ADE were applied in the control group; alternatively, the study had no control. (4) Outcomes: The effectiveness outcomes were production e ciency, medicine residues, preparation accuracy, preparation errors, preparation failures, error recognition, and solution properties. The safety outcomes were product pollution, environmental pollution, and health damage to technical personnel. The economic outcomes were labor costs and material costs. The social suitability outcome was the satisfaction of personnel.
The exclusion criteria were as follows: (1) literature without available full texts and conference abstracts; (2) news, subjective views, and editorials; and (3) repeated published literature. follows: intravenous, infusion, robot, automation, intelligence, admixture, dispense, compound, preparation, and deployment. The Chinese search terms were translations and modi cations of the English search terms used in the Chinese databases. Appendix 1 shows the detailed search strategy. The retrieval time was from database inception to 11 August 2020. Finally, we manually searched the references of the included studies as supplements.
Literature screening and data extraction Two researchers independently screened the literature and extracted the data. First, the titles and abstracts of the literature were read, and then the full texts were read if they were relevant. The inclusion of all literature was decided jointly by the two researchers. In case of disagreements, the literature in question was further discussed or its inclusion/exclusion was decided by a third researcher. The data were extracted according to a predesigned data extraction table and included basic information such as the study design, sample size, intervention, effectiveness outcomes, safety outcomes, and economic outcomes.
Disagreements were resolved by discussions or a decision by the third researcher. The necessary information with uncertainty was obtained by contacting the authors of the primary studies.

Quality assessment of literature
The quality of the included individual studies were assessed independently by two researchers. Currently, there is no HTA quality assessment method in the world, so HTA Checklist was used to evaluate its quality. The primary studies selected different scales (Cochrane risk of bias tool, Newcastle-Ottawa Scale, ROBINS or JBI Qualitative Assessment and Review Instrument) for quality assessment according to study types. The quality of pharmacoeconomic studies was assessed using Quality of Health Economic Studies (QHES).

Data analysis
Descriptive analysis was used because of heterogeneity in the characteristics of the data in the included studies.

Result of literature search
In total, 165, 52, 153, and 198 studies were obtained from CNKI, Wanfang Database, VIP, and CBM, respectively; 463, 2744, and 154 studies were obtained from PubMed, Embase, and The Cochrane Library, respectively; and 28 studies were obtained through the HTA professional websites and the references of the included studies. All studies were screened strictly in accordance with the inclusion and exclusion criteria. Finally, 40 studies [7-46] were included (2 HTAs, 26 controlled studies, 11 non-controlled studies, and 1 economic study), among which 25 studies were in foreign languages and 15 studies were in Chinese. The literature inclusion process and results are shown in Figure 1.

Results of quality assessment
The quality assessment of HTAs showed the quantity of evidence was small and the quality was low. The quality assessment of the economic study showed the the quality score (89 points) was relatively high. The description of methodology in the qualitative study was not clear and it was di cult to assess the quality. In addition, a survey report didn't have an appropriate scale for quality assessment. Details of quality assessments for other studies can be found in Appendix 2.
Results of bibliometric analysis ( gure 2) Since publication of the rst study, the number of studies published in this eld has increased each year. After more than 10 years of development, the application of intravenous compounding robots in China and abroad has been gradually increasing, and the number of relevant studies has reached its peak in the past three years.

Results of comprehensive analysis
HTAs Two HTAs [11,12] were included (Table 1); both were published by the CADTH in 2013 and 2016. The results of these two HTAs showed that automation for the preparation of intravenous admixture medicines had certain security and economy. However, because the number and quality of the included original studies were low, it is necessary to update the currently available HTAs or re-evaluate the available data. International (INAHTA, HTAi, ICES, ISPOR), American (AHRQ), European (EUnetHTA), British (NIHR HTA Programme), Canadian (CADTH), Swiss (SBU), and Australian (AGDHHTA) HTA databases do not yet contain any evaluations of intravenous compounding robots.
System (Baxter Healthcare) as well as robots independently developed in China, such as WEINAS series robots and Angel compounding robots. The sample size of each study ranged from 10 to 11,865, and the observation time ranged from 2 days to 3 years. The main types of intravenous admixture medicines were anti-tumor drugs (14 studies, 36%), unclassi ed medicines (14 studies, 36%), antibiotics (2 studies, 5%), and total nutrient admixture (2 studies, 5%).
Effectiveness: Thirty studies evaluated the effectiveness (Table 3), including 21 controlled studies and 8 non-controlled studies. The evaluation indicators were production e ciency, medicine residues, preparation accuracy, preparation errors, preparation failures, error recognition, and solution properties.
However, one controlled study [34] showed no signi cant difference in the preparation time before and after introducing the robot, and another controlled study [26] showed that the average preparation time in the robot group increased by 47% compared with that in the robot group but that the pharmacists' working time decreased by 76%. Yet another controlled study [9] showed that the average number of preparations by the robot in 7 hours was comparable with that by the trained and experienced pharmacy staff in 2 to 3 hours; thus, the robot produced limited improvement in production e ciency in practice. Another non-controlled study [41] showed that although robots can signi cantly improve production e ciency, the manual pre-processing and post-processing steps were time-consuming and had to be reorganized.

Medicine residues
Nine studies [13][14][15]17,19,20,23,40,46] evaluated medicine residues, including 7 controlled studies, 1 non-controlled study, and 1 qualitative study. Six controlled studies [13][14][15]17,19,23] conducted a quantitative analysis of medicine residues. The data could not be merged because of the different medicine types in each study, but the results of all studies showed that the rates of residues in the robot groups (0.00%-4.50%) were signi cantly lower than those in the manual groups (3.67%-50.00%). One non-controlled study [40] also showed that the use of robots reduced medicine residues, and another in-depth interview [46] showed that the robots reduced the residual amount of some insoluble medicines and alerted the technicians through an alarm when the residual amount was large. Only one controlled study [20] showed that the amount of medicine residues in the robot group (compound coenzyme: 0.11 ± 0.01, sodium carbamate: 0.12 ± 0.01) was slightly higher than that in the manual group (compound coenzyme: 0.09 ± 0.02, sodium carbamate: 0.08 ± 0.02); however, the amount in both groups was lower than the hospital inner quality standard.

Preparation accuracy
Ten studies [28,9,23,7,30,44,41,38,35,46] evaluated the preparation accuracy, including ve controlled studies, four non-controlled studies, and one qualitative study. Seven controlled studies [28,23,9,7,30,44,41] conducted a quantitative analysis on the preparation accuracy (deviation of less than ±5%). The data could not be merged because of the different medicine types in each study, but the results of six studies [28,9,7,30,44,41] showed that the rates of preparation accuracy in the robot groups (0%-23%) were signi cantly higher than those in the manual groups (5%-53%). One controlled study [23] showed that robotic preparations were more accurate and precise (mean absolute dose error of 0.83 for uorouracil and 0.52 for cyclophosphamide) than manual preparations (1.20 and 1.70, respectively). In an in-depth interview [46], the vast majority of respondents indicated that compared with manual preparation, the robot improved the accuracy by adjusting the dose by itself. Two non-controlled studies [38,35] showed that the preparation accuracy was related to the viscosity of the liquid and the minimum volume of the dose. The minimum volume of the non-viscous solution and viscous solution that could be accurately prepared (deviation of less than ±10%) was 2 ml and 5 ml, respectively [38].

Preparation errors
Eleven studies [13][14][15]17,19,26,31,32,37,40,46] evaluated the incidence of preparation errors, including eight controlled studies, two non-controlled studies, and one qualitative study. Seven controlled studies [13][14][15]17,19,26,31] conducted a quantitative analysis on the incidence of preparation errors in the robot groups and manual groups. The data could not be merged because of the different medicine types in each study, but the results of all studies showed that the incidence of preparation errors in the robot groups ranged from 0.0% to 0.9%, which was signi cantly lower than that in the manual group (0.013%-12.5%) [13][14][15]17,19,26,31]. One study showed that the number of daily preparation errors was reduced from 0.26 ± 0.78 to 0.06 ± 0.13 [32]. One non-controlled study [40] and one in-depth interview [46] also showed that the robots could reduce the incidence of preparation errors by warning, without interference by factors such as visual fatigue. Finally, a 3-year real-world study [37] showed that the percentage of actual failed preparations was <1% (0.21%, 0.15%, and 0.18% in 2015, 2016, and 2017, respectively).

Preparation failures
Three studies [9,10,42] evaluated the incidence of preparation failures, including one controlled study and two non-controlled studies. The model used in two of the studies was the CytoCare, and the incidence of failures was 12% (n = 1028) [10] and 19% (n = 4509) [9], respectively. The model used in another study was the APOTECAChemo, and the incidence of failures was 2% (n = 11,642) [42]. The reasons for preparation failures included robot shutdowns, mechanical failures, re-calibration, and other practical problems [9]. However, as the application time increases, the rate of preparation success might gradually improve. For example, one study [10] showed that the rate of preparation success increased from 76.8% in week 1 to 95.3% in week 10.

Solution properties
Two RCTs [22,25] evaluated the properties of the solutions prepared by the robots. In one study [22], measurement of the size, pH value, and osmotic concentration of lipid particles in the nutrient solutions showed that the nutrient solutions prepared by the automated compounding device were superior to those prepared by gravity infusion in terms of stability and compatibility. By measuring antibody aggregation, another study [25] concluded that robotic compounding of monoclonal antibodies was feasible and that the robot could be used to achieve reproducible high-quality compounding for delicate formulations.

Error recognition
One non-controlled study [42] evaluated the robots' role in error recognition. The study showed that the robot recognized 1.12% (n = 133) of errors that would have caused harm to patients.
Safety: Twenty-two studies evaluated safety (Table 3), including 13 controlled studies, 8 non-controlled studies, and 1 qualitative study. The evaluation indicators were product pollution, environmental pollution, and health damage to technical personnel.

Product pollution
Eleven studies [27,21,39,30,45,43,41,38,36,35,46] evaluated product pollution, including three controlled studies, seven non-controlled studies, and one qualitative study. Five studies [39,30,41,38,36,35] did not detect microbial pollution, and one non-controlled study [43] showed that non-contaminated bags were not contaminated after preparation, revealing that the robot enabled the preparations with low levels of product contamination. One controlled study [21] showed that the positive rate of bacterial cultures in the piston decreased from 26.67% to 13.33%, revealing that the robot could signi cantly reduce the possibility of product pollution. One in-depth interview [46] revealed that the robot was cleaner than the biological safety cabinet because it had an air channel cleaning system and a closed negative-pressure environment. However, another study [27] showed that the microbial culture positive rate was 7.9% (n = 152), which was related to product pollution. Another investigation showed that microbial contamination might originate from the washing station, which was easily be ignored [45]. Finally, the technicians' gloves were also key sites of microbial contamination [35].

Environmental pollution
Eight studies [24,23,39,18,43,38,36,35] evaluated environmental pollution, including three controlled studies and ve non-controlled studies. Two controlled studies [24,18] showed that the external pollution of the robots was relatively low through monitoring gloves, infusion bags, and other equipment, indicating that environmental pollution could be improved. One non-controlled study [38] detected only one type of contamination associated with pulling the needle out of the syringe, which was very small (spots of <3 mm) but relatively frequent (11%). Pollution was mainly observed inside the robot [43], especially in the area of the dosing device [24,23,39]. The robot was able to perform automatic microbial decontamination by ultraviolet radiation [36]. One controlled study and one non-controlled study showed that the settling plate/contact plate met the recommended limits for the class A area of the clean room [39] and that the surface and air samples complied with an ISO 5 class environment [35], indicating that the robots had well-controlled programs.

Health damage
Eleven studies [26,17,23,15,16,14,13,19,43,40,46] assessed health damage to the technical personnel, including eight controlled studies, two non-controlled studies, and one qualitative study. Seven controlled studies [26,17,15,16,14,13,19] performed a quantitative analysis of the incidence of health damage to technical personnel in both the robot groups (chapped ngers: n = 0, ampoule scratches: n = 0, fatigue: n = 0-1, syringe stabs: n = 0-3) and manual groups (chapped ngers: n = 1-29, ampoule scratches: n = 2-38, fatigue: n = 3-19, syringe stabs: n = 1-28). The results showed that the incidence of health damage to technical personnel was signi cantly lower in the robot groups (0.0%-2.9%) than in the manual groups (5.1%-40.0%), and the somatic pain score was also signi cantly lower in the robot groups than manual groups (2.65 ± 0.47 vs. 5.76 ± 0.03, respectively). One non-controlled study [40] also showed that the robot could provide effective occupational protection for nurses. In an in-depth interview [46], all interviewees were nurses in the oncology ward and daytime chemotherapy center, and they had strong experience with occupational protection provided by the robots. Two studies [23,43] showed that no contaminant exposure was found in the gloves or on the hands of the technicians by applying the robots.
Economy: Six studies [34,26,10,7,30,8] conducted an economic evaluation, including four controlled studies, one non-controlled study, and one economic study. The evaluation indicators were labor costs and material costs. Through 1000 simulations, an economic study [8] showed that robots could reduce healthcare costs by preventing medication errors and reducing adverse drug events, saving an average of $288,350 per year. Three controlled studies [26,7,30] showed that robots could reduce costs by 8% to 66% by saving materials, and the cost savings might continue to rise with the increase in preparations [7]. One controlled study [34] showed that a robot could reduce the labor costs of three pharmacists, but two other studies [26,10] showed that the robots could not reduce labor costs, could not improve the full-time equivalent of a hospital general pharmacist/technician, and even needed additional on-site engineers to help resolve possible breakdown.
Social suitability: Only one study [37] investigated technicians' satisfaction with robots. The results showed that responders agreed that the overall impression of robots was "very good," and the safety features of robots had a median score of being "very safe".

Findings and clinical value of this study
Based on the currently available evidence, robots can improve production e ciency over the usual manual preparation, but the intravenous preparation process still requires further optimization. Robots can also reduce the incidence of medicine residues, preparation errors, and preparation failures. The studies evaluated herein showed that the accuracy and solution properties of intravenous admixture medicines were satisfactory and that robots had a role in error recognition. With respect to safety, robots can reduce product pollution and environmental pollution; however, vigilance is still required to ensure that pollution stays low. Additionally, robots can reduce the incidence of health damages to technicians. With respect to economic factors, robots can reduce material costs, but whether they can reduce labor costs remains to be further studied. Finally, in terms of social suitability, technicians had a high degree of satisfaction with the robots. However, there are relatively few relevant studies on this aspect at present.
Only two HTAs of intravenous compounding robots have been performed to date, and the number and quality of the original studies included are low. Therefore, in the present study, we re-evaluated intravenous compounding robots. We included the original studies in all respects and selected evaluation indicators covering various aspects to comprehensively analyze the effectiveness, safety, economy, and social suitability of intravenous compounding robots, providing a reference for future introduction and application of intravenous compounding robots in medical institutions.

Limitations of this study
This study has three main limitations. First, because of our limited access to data, we were unable to obtain the full texts of some studies that met the inclusion criteria, and only studies from which we could extract data were included. This might have introduced a risk of bias that affected the conclusion. Second, the original studies differed in terms of the robot models, types of intravenous admixture medicines, and technical personnel's familiarity with robots.
There was great heterogeneity among the studies, preventing the data from being merged; thus, only qualitative analyses could be conducted. Third, most of the original studies involved evaluations of effectiveness and safety and lacked evidence of economy and social adaptability. This made it impossible to carry out a more comprehensive HTA for intravenous compounding robots.

Future research directions
This analysis showed that most of the currently available studies of intravenous compounding robots have a high risk of bias and small sample size. Therefore, high-quality, large-scale, multicenter RCTs or well-designed observational studies are needed for further evaluation. Additionally, only ve studies in this analysis included economic evaluations, and only one study included a social suitability evaluation. More attention should be paid to the economy and social suitability of intravenous compounding robots in future studies to provide more evidence for medical institutions. In addition, PIVAS in medical institutions should cooperate with the manufacturers of intravenous compounding robots to jointly determine how to further optimize intravenous compounding robots, reduce preparation errors and safety events, and improve the service quality of intravenous compounding robots.

Conclusion
Intravenous compounding robots have certain advantages in terms of safety, effectiveness, economy, and social adaptability. High-quality and large-sample RCTs or well-designed observational studies are still needed to further evaluate these robots, especially in terms of economic and social suitability. Ethics approval and consent to participate: As this study does not involve human or animal experiments, this section is not applicable.
Consent for publication: As this manuscript contains no individual personal data, this section is not applicable.
Availability of data and materials: The datasets generated and/or analyzed during the current study are not publicly available because they are subject to the West China Second University Hospital, Sichuan University. However, the data and materials are available from the corresponding author on reasonable request. The robot guaranteed the accuracy and safety of infusion, and reduced the error rate of disposition.
The robot was bene cial to strengthen the occupational protection of personnel. The bene ts of robot included increased precision in drug preparation, improved safety, and saved time.
--The bene ts of robot included improved safety. Automatic system improved the work e ciency and reduce the labor intensity of the staff. - The dispenser could improve work e ciency and had high precision.
The bacteriology examination of product dispensed by the dispenser was satisfactory.
The dispenser could save consumables and reduce costs.