Feasibility and acceptability of a virtual clinical pharmacy service for elective orthopaedic inpatients in an Australian metropolitan hospital

Virtual healthcare services are usually provided from urban centres to outpatient clinics or underserved rural areas. This study utilises virtual pharmacy as an innovative model to provide services to a metropolitan hospital from a rural area.


INTRODUCTION
Medication-related incidents are frequently associated with hospital admissions and evidence suggests clinical prescribing error rates are between 1 and 1.5 per patient in Australian hospitals. 1 While it is recognised that clinical pharmacists play a vital role in ensuring best practice medication management, they are not deployed in all hospital settings. Research into Australian hospital pharmacies reveals that most pharmacists spend less than half of their time providing clinical services and very few hospitals provide comprehensive clinical pharmacy services to patients admitted overnight. 2 Virtual care offers a potential mechanism to improve access to clinical pharmacists with the aim of improving compliance with hospital accreditation standards for medication safety and to drive high-quality care. The implementation of electronic medical records (eMRs) and electronic medication management (eMeds) in New South Wales (NSW) health facilities has provided the digital infrastructure to support integrating virtual pharmacy into many health settings across the state. In 2020, Western New South Wales Local Health District (WNSWLHD), a public healthcare provider with 38 inpatient facilities servicing a population of 276 000 people in rural and remote NSW, expanded a pre-existing ad hoc telehealth pharmacy service to a comprehensive, proactive Virtual Clinical Pharmacy Service (VCPS). This service aimed to improve medication management and safety for inpatients across eight rural and remote hospitals that would otherwise not have had access to clinical pharmacy services. The VCPS is undergoing a formal evaluation across the range of clinical services it provides, including best possible medication history, medication reconciliation, medication review, medication lists, and patient education. 3 Virtual pharmacy is an emerging model utilised in several healthcare settings worldwide with a 2019 review reporting on 18 virtual clinical services. 4 Until recently most models have focused on remotely optimising medication in chronic disease management through outpatient clinics to increase access to underserved communities. [5][6][7] Recently, virtual pharmacy services have also been described internationally in metropolitan settings in response to the COVID-19 pandemic, 8 although there does not appear to be any feasibility or acceptability studies of such models conducted for inpatients in an Australian metropolitan hospital. A study of virtual pharmacy in rural and remote Australian hospitals demonstrated acceptability, with staff valuing access to a dedicated workforce, specialist medication advice, medication education, and improved communication which was facilitated by the electronic medical record. 9 Reliable and easy-to-use technology coupled with familiarity with the process and systems have been reported to improve staff acceptability of telehealth interventions. [10][11][12] There is limited evidence of patient acceptability of inpatient virtual pharmacy, but an Australian outpatient cardiology virtual pharmacy program, which included medication reconciliation, has reported high patient satisfaction. 13 WNSWLHD collaborated with Sydney Local Health District (SLHD) to investigate the feasibility and acceptability of the VCPS in a tertiary hospital ward that traditionally had limited clinical pharmacy input. Historically, most virtual healthcare delivery models have been delivered from tertiary metropolitan hospitals to regional communities. 14 This study is unique in that pharmacy staff experienced in the VCPS model from a rural hospital network provided services to a metropolitan public hospital ward where staff were largely unfamiliar with virtual care delivery. This study examines the feasibility of the VCPS in a tertiary hospital and the acceptability to patients and clinicians in a metropolitan environment.

Design, Setting, and Clinical Service
A non-randomised feasibility study commenced in April 2021 with implementation of a VCPS from WNSWLHD to SLHD. A 12-week period was chosen to embed the new virtual model and allow staff to attain a working knowledge of the service. The study site was a 33-bed elective orthopaedic ward in a major metropolitan teaching hospital in Sydney, Australia. The ward performed minor and major orthopaedic procedures for both local and out-of-area patients who did not have multiresistant organisms. The ward had access to hospital specialist services and provided comprehensive allied health services, and patients were either discharged home or to another facility for further rehabilitation. The ward used electronic medical records (Cerner [Oracle Corporation, Kansas City, Missouri, USA] eMR) and electronic medication (Cerner eMeds) systems and had minimal on-site clinical pharmacy services. All elective orthopaedic patients were eligible for the VCPS, but patients admitted for major surgeries such as elective knee and hip replacements or those taking high-risk medications were prioritised for pharmacy review in accordance with standards of practice for hospital pharmacies. 15 The pharmacist proactively identified patients for review using the eMR and local staff could refer patients via phone communication. It was anticipated approximately 300 patients would be admitted during the trial period.
Pharmacists from WNSWLHD experienced in virtual clinical pharmacy were credentialed and orientated to study site systems and processes to provide the virtual service 32 h per week from Monday to Friday. One pharmacist primarily provided clinical services but was backfilled for rostered days off. The VCPS provided all clinical services, including best possible medication history, medication reconciliation at transitions of care, medication review, medication lists, and patient education via videoconference. Most patients were only seen by the pharmacist once on admission for the provision of all clinical pharmacy services, which were primarily best possible medication history, education, and discharge counselling. Medications requested for supply were remotely verified by the virtual pharmacist and dispensed by on-site pharmacists. Discharge medications were supplied via the onsite pharmacy though existing processes.

Data Collection
The measures of feasibility included indicators of service utilisation (number of patients seen by a pharmacist, number of best possible medication histories, and medication reviews) and detection of medicationrelated issues which were recorded in the eMR. The virtual pharmacist recorded observations throughout the trial.
Patient acceptability was measured through a sevenquestion patient-reported experience measures (PREM) survey, which used six Likert-type responses, a free-text option, and demographic information ( Table 2). Seventythree patients out of 97 who had interacted with the pharmacist via videoconference were invited to complete the PREM, which was delivered via text message. The 24 participants who were not invited were aged <18 years, did not have access to a mobile phone, did not speak English, or had complex communication requirements and were excluded from the PREM.
Clinical staff perceptions of the issues, benefits, barriers, and overall acceptability of the VCPS were explored through a virtual focus group. The focus group was held at the end of the trial to provide adequate time for staff to be exposed to and use the service. Focus group participants were nurses and managers from the elective orthopaedic ward. The focus group took approximately 30 min, was audio-recorded, transcribed verbatim, and the participants provided informed consent prior to the session commencing.

Analysis
Descriptive statistics were performed on quantitative data. Qualitative data underwent thematic analysis. Transcripts were open coded by two investigators with qualitative research experience (JA and EW) and discussed with a third with clinical experience (BC). Codes were grouped to form themes. The analysis sought semantic themes in that the explicit meanings of what the interview participants said were used to identify the benefits and challenges of the VCPS to inform recommendations about the service's feasibility. 16 Ethical approved was granted by the Greater Western Human Research Ethics Committee prior to the study commencement (Reference No: 2021/ETH00097). This study was retrospectively registered with ANZCTR on 26/07/2021 (ACTRN12621000983808). Trial registration was retrospective as the authors did not initially plan to publish results.

RESULTS
The VCPS service was implemented and continued for 65 days from 27 April 2021 and concluded on 30 June 2021. The trial ended two weeks earlier than planned due to delays starting the trial and funding constraints.
During the study period 389 patients were admitted (Figure 1), with an average age of 57 years ( Table 1). The patient population was considered typical for an elective orthopaedic ward (Table 1). Most patients stayed in hospital for more than 24 h (74%) and most were discharged home after an average stay of 3.29 days (Table 1).
Measures from the eMR demonstrate that the VCPS provided a comprehensive service with high utilisation. The pharmacist performed 535 clinical and medication reviews for 225 patients (  Pharmacist medication reviews resulted in the detection of 151 medication-related issues or recommendations. The medication-related recommendations were categorised in the eMR by the pharmacist. The most common medication-related issues were omitted medications (30%), dosage changes (22%), cease medication (11%), and monitoring (11%) ( Table 2).
Patient-reported experience measures surveys were completed by 22 of 73 (30%) patients (10 females, 12 males). From an age perspective, respondents were evenly distributed (11 < 65, 11 > 65). Patients' responses to the PREM questions were positive (Table 2). Patients reported good audio and video quality (question 1), good communication with the pharmacist (question 2), more confident managing medications (question 3), and satisfaction with involvement in medication-related decisions (question 4). Overall patients rated their experience of care highly (question 5) and most would recommend the service to others (question 6). Seven free-text comments were provided, of which two expressed concerns related to patient privacy, four provided compliments, and one suggested using a remotecontrolled videoconference machine.
One focus group was conducted with six participants, including a nurse manager, nurses, and one pharmacist. One nurse left after 10 min but had not contributed to the focus group. Verbal consent was recorded after checking participants had information about the study and any questions were answered (no questions were asked). The focus group participants discussed the ways in which the VCPS interacted with staff and with patients. They perceived the VCPS to have benefits for both groups. Staff benefits included ease of access to specialist medication advice, confidence that patient medications were correct for their condition, and communication facilitated by the eMR and eMeds (Table 3).
Challenges included staff concerns about patient confidentiality in open wards, especially relating to sensitive conditions, lack of experience in using virtual healthcare, working a new process and equipment into usual tasks, and communication barriers because the equipment was not fit for purpose (Table 3).

DISCUSSION
This research describes an innovative approach to the provision of pharmacy services in a metropolitan hospital and introduces potential opportunities to expand or complement existing pharmacy service delivery. Clinical pharmacy provided virtually is a feasible and acceptable model in a tertiary elective orthopaedic ward. The results identified high service utilisation, detection of medication errors, and patient and clinician acceptability of the model. The focus group found strong support from nursing and pharmacy staff, who experienced easy access to clinical pharmacy and increased confidence with prescribed medications. Adoption of virtual technology has the potential to enhance the agility and sustainability of clinical services and could be a rapid response to workforce shortages or pandemic restrictions. More importantly, as a standard model of care it offers a way to bring skilled pharmacists to areas of need irrespective of their physical location. High service utilisation was expected and achieved through a proactive model where the pharmacist selfreferred patients for clinical pharmacy review based on identified risk factors. Even in a ward with high turnover almost 80% of patients staying over 24 h were seen by a pharmacist. Typically, at large metropolitan hospitals only 50% of overnight admitted patients are seen by a pharmacist. 2 This reflects this study's pharmacist-topatient ratio of 32 h per week to 33 patient beds, with included on-site support for medication supply.
The VCPS resulted in the detection of 151 medicationrelated issues and recommendations from the review of 225 patients. This detection rate is consistent with other studies of medication errors in Australian hospitals. 1 Due Table 3 Staff views on the benefits and challenges of virtual clinical pharmacy services on the orthopaedic ward Benefits Ease of access to specialist medication advice . . . you'd ring (pharmacist) of a morning and go, okay, this is my problem. . . where they are allergic to this, this, and this. We're trying to find other options for pain relief. What can we do that we are not exactly used to? She was a really good resource for us. If you did have a question of her, it was very easy to chat to her about it, really quickly.

Confidence about patient medications
Our doctors here, the ones that we directly work with, who prescribe, are junior medical officers and they are usually looking for maybe guidance from the nurses, or guidance from the pharmacist. I think they are happy to take guidance from anyone because they are just rotating here usually. They [patients] appreciated speaking to (pharmacist) a lot because we were able to get (a) hold of their drugs very quickly, and (b) we were able to sort out a lot of their stress very quickly because (pharmacist) was able to go, right, this is what you are on at home, this is what you'll get here, this is what we are going to do for you. They often have a lot of anxiety coming in about specific things to do with medications that she just took over and alleviated. Electronic medical record and eMeds . . . from a quality perspective as well and just a patient safety perspective, we actually had a number of phone calls where (pharmacist) would pick up the phone and go, hey just before I jump on virtual, just to let you know, I've been through the patient charts and I've [checked it] with doctors already and I've already had a discussion. We've picked up a couple of errors, or double-duplication orders or things like that. She's been through each patient's file already. . . you would not have that, if it was on a paper chart.

Patient confidentiality
We have the volume up all the way [so the patient can hear], and it can be really loud and everybody else in the rest of the bays could hear what was going on. I'm a little bit worried about the privacy aspect of it because you cannot have the volume up all the way. . . . for example, we had a patient with HIV, and they did notwe were not able to start the medications with that patient because they were in a four-bed bay and for confidentiality reasons, so they were not able to fulfil their whole job with that patient. New processes and equipment . . . we had an issue with the actual equipment. . . the computer died or something. . . if you do not have a second computer or a second way of communicating with the virtual pharmacist, you were out. She [pharmacist] would say, hey, can I talk to x-y-z, and then that person would have to saythey would not know how to log on or anything like that. They'd have to run around and try and find somebody who knew how to work the equipment. Communication barriers with devices not fit for purpose [Bulky]. It's hard to drive, hard to see. The patients, because they are a bit further away, they are struggling to [read] off you as well. They're struggling to hear from that distance as well, because the speaker is on a little desk underneath. We had a lot of troubles with that too. Improvements Better equipmentuse iPads and headphones.
to the nature of VCPS, prescribing errors are more likely to be detected than administration errors (other than omitted doses) as the administration process is not witnessed. As part of the implementation medical input was sought recognising doctors are required to make medication changes identified by the pharmacist. No grading for the severity or impact of the intervention or the prescriber uptake of the recommendations was available within the eMR, which limits the interpretation of the significance of the interventions. Patients provided positive feedback on audio and video quality, communication, confidence managing medications, and overall experience with the VCPS, indicating the service was acceptable to patients. There is limited evidence of acceptability of a VCPS in the inpatient setting. However, the results are comparable to a recent study of an outpatient telepharmacy cardiology clinic where patients reported increased confidence in managing medications and high overall satisfaction. 13 The low number of PREM surveys completed means the results need to be interpreted with caution, but patient suggestions about the need to address confidentiality and use dedicated videoconferencing equipment were consistent with staff feedback.
Staff valued the service and described benefits of access to specialist medication advice, quality and safety, confidence managing medications, and facilitation of communication through eMR. These findings are similar to those described in smaller rural and remote facilities with a virtual pharmacy service. 9 Critical to the success of the project was a local staff member who championed the project and supported the change management and implementation. Staff described challenges with new processes and technology, patient communication over videoconference and patient confidentiality. Interestingly, patient confidentiality was not raised as a challenge in smaller rural hospitals, perhaps due to patients being known to staff in smaller rural communities, fewer onsite allied health staff and visitors, different videoconferencing equipment being used, or smaller patient-to-room ratios. 9 As the only virtual service in operation on the ward it was expected that staff would find the adoption of new processes, systems, and technology challenging and this was in fact the case. Other studies have suggested observation, hands on training and practice with the process and technology may lead to more positive experiences for nurses. 10,11 This could be addressed in future trials through a hybrid face-to-face implementation phase where the pharmacist provides additional support with technology and creates an opportunity to build relationships with local teams.
Reliable and easy-to-use technology is frequently reported as a critical factor for acceptance of telehealth programs. 12 Only basic videoconferencing technology was employed due to the short duration of the trial at the site where telehealth was not routinely used. This limited some of the available functions and therefore, the user experience. Pharmacists were unable to dial directly into the videoconference machine or control the camera lens, which increased some administrative tasks for staff on the patient end. An unexpected finding was concern for patient privacy, with nurses reporting the speaker on the videoconference machine could be heard outside the patient's room. It is possible the same concerns could be raised during a face-to-face conversation between a pharmacist and a patient in the same environment. Despite this, addressing privacy and ensuring appropriate technology may further increase the acceptance of a VCPS.
A key strength of this trial is that it was embedded into routine clinical care over a relatively short study period, demonstrating the agility of the VCPS model. This study was able to go ahead despite broader environmental and health crises, including the COVID-19 pandemic and subsequent vaccination roll out, which diverted health resources. Key enablers of the study were pharmacists experienced in the delivery of a VCPS. Other considerations for successful implementation include streamlined referral processes, appropriate technology, patient privacy, and local change management and support. This study demonstrates the eMR, eMeds, videoconferencing, and communication processes are the minimum fundamental requirements to support a VCPS. In this study virtual pharmacists were located many hundreds of kilometres from the hospital site. The physical location of the pharmacists is not a factor in care delivery so long as the other implementation processes are in place.
This research is directly translatable to a tertiary elective orthopaedics or rehabilitation-type ward, with potential translation to other wards in hospitals without on-site clinical pharmacists. Several factors need to be considered when determining the generalisability of the findings and further trials may be required before expanding to other clinical areas. Typically, elective orthopaedic patients are younger, and have fewer medication changes and fewer ward transfers. It is unlikely that a VCPS will replace face-to-face services in a metropolitan context, but rather complement or expand existing services. Potential opportunities include extending clinical services though a centralised, after-hours service or to complement infection control in areas such as haematology, oncology, intensive care, or infectious diseases wards. The VCPS has the potential to provide sustainable inpatient pharmacy services where they are not currently available, as virtual pharmacists can work across sites, wards, and health conditions. Some limitations to this study include that it was conducted over a relatively short time period, and it is possible a longer trial could have yielded different results, particularly as there was an absence of an implementation period prior to data collection. It is also possible the low number of patient surveys means the sample is not representative of the broader hospital population. There is also a potential for selection bias with the surveys as a small number of patients were unable to participate as they did not have a smartphone device, or had complex communication needs or poor English language skills. Further research would be required to confirm patient acceptability and future studies could investigate the effectiveness of an inpatient virtual pharmacy program. 3 There were no harms or potential safety concerns reported throughout this trial.

CONCLUSION
The findings of this study demonstrate the VCPS is a suitable model for complementing on-site pharmacy services and demonstrates a VCPS is both feasible and acceptable in a metropolitan environment. It also provided an opportunity for collaboration, networking and shared learning across local health districts. The study provides evidence for clinicians, managers and health system planners that a VCPS is a potential solution that can be rapidly implemented to deliver clinical services when face-to-face is not practicable. This innovative and agile solution is potentially scalable within a metropolitan health network to enhance the sustainability of clinical service provision.