Patient Recruitment
Given that this was a feasibility study, the sample size is often a pragmatic balance between information needs and resources.8 Typically, up to 40 patients are recruited for a feasibility study.9 Recruitment of Drugs4Delivery patients was achieved via purposive sampling of those who were prescribed inhalers. In order to avoid the risk of not capturing eligible patients, all patients on inhalers were asked their diagnosis via telephone before inviting them to the study. Future studies could use a similar recruitment strategy by using their Patient Medication Record filter to identify all the pharmacy patients that use inhalers.
The most notable challenge of this research study was recruiting the required number of patients from a single community pharmacy’s patient base. A drawback of this study was that just 25 patients could be recruited via this method. This was overcome by adapting the approach half way through, via advertising more widely in local newspapers. Thus, the recruitment target was met. The advertising strategy mitigated against the limitations of the initial recruitment strategy and this approach could be incorporated from the outset in any larger, future studies. In addition, those participants that did not turn up to their baseline consultation, attended after they were re-scheduled.
In this study, participants recruited via advertising were either keen to be involved in research or were keen to consult a SRP and learn whether there were other treatment options available to them. The lack of referrals from general practice could potentially be circumvented by presenting the study proposal, in-person, to all the local surgeries and involving clinicians at an earlier stage in the research design. Future studies should be mindful that this could result in selection bias as participation in studies is usually accepted by motivated patients.
Patient Retention
A high retention rate was achieved in the study. A source of bias for prospective studies is the potential for dropouts, which was low in this study, as significant efforts were made to avoid this. These included, negotiating treatment decisions with the patient and conducting the clinics in a professional setup as advocated by Ali Murshid and Mohaidin.10 Other factors that may have contributed to the low number of participants being lost to follow-up could have been existing rapport with Drugs4Delivery patients and informing participants that they would be reassessed in 6-months’ time in order to determine any change in their condition. This latter information may have motivated them to find out about whether they had gained any benefits from the interventions after 6-months.
These strategies could be used to inform the design of future multi-centered studies in order to minimize attrition, although recruiting via the pharmacy’s patients should be interpreted with caution, as a larger study involving multi-centers may vary from this single-site study. Although, the multi-site approach improves recruitment and retention potential compared to single site, factors such as managing research, ensuring consistent processes and high-quality data, and having clear communication processes between researchers/clinicians, become crucial. These issues require a significant amount of planning and meetings for each research site, which involves going through protocol, standard operating procedures and processes.11
Acceptability of Tele-Consultations
An anonymized feedback questionnaire was administered to each participant (in both groups) in order to offer their opinions of the clinic. From the questionnaire used, it was determined that participants in the TC group were more receptive to using TC after having experienced its convenience during the study (see Table 1). It was thought that the older participants would be resistant to TC. However, this study found they became more amenable once they had been supported through the process of using it. From the study, it was evident that patients generally felt more comfortable talking from within the comfort of their home than in-person within a consultation room. Patients commented on the value of having a TC, especially during an exacerbation or for immobile patients. This was supported by Vitacca, Montini and Comini12 who reported that the best outcomes came from TCs dedicated to more severe cases, frequent exacerbators with multimorbidity and limited community support. Thus, selecting these ideal candidates could be the key to optimizing the use of TC in future RCTs.
The research literature on TCs is sparse but has begun to accumulate.13 Sorknaes et al.14 reported high patient satisfaction, which was also reported in other earlier studies.15–17 Although these studies had a similar demographic, Rahimpour et al.16 and Whitten and Mickus17 also included patients with congestive heart failure. Tatari18 reported that 12% of patients experiencing TCs for the first time thought they would benefit more from an IPC than a TC and 9.5% did not consent to having a TC, as they preferred to have an IPC; however, this was not specific to COPD patients, and participants were from a younger demographic on average.
Of the 185 patients screened during this feasibility study, only 8 (4.5%) specifically gave the reason of not wanting to use TC, however, this could have been more since 35 patients did not give a reason for declining to participate. Offering a tablet device to those without Skype access may have led to a higher acceptance rate. Participants in a study by Gund et al.19 also reported positive experiences with TCs, and patients found the technology easy to use and better than ordinary phone calls, although, this was not always the case with motivating staff.
There were some notable differences experienced between the two groups; most notably, during an IPC when the patient smelt of smoke, it was likely that they were either a smoker or they experienced passive smoking. However, this could not be picked up when conducting TCs. Furthermore, for TC to be successful, a clear uninterrupted transmission is required, a significant limitation of this medium. Losing transmission means the consultation cannot take place and must be rescheduled. From the pharmacist’s perspective, organizing the IPC was a great deal easier than the TC, especially if the participant did not have TC access or did not know how to work it. IPCs are the norm for patients when seeing their healthcare professional, whereas TC seemed overwhelming if they were new to the concept.
Practicalities of study implementation
This study sent recommendations via postal mail for both groups and lack of uptake (approximately a quarter were actually implemented) brings into question whether some recommendations actually reached the prescriber. Until read-write access to patient notes is implemented, future studies could try alternative communication channels such as phoning the practice pharmacist or emailing via NHSmail (this option was not available at the time of the study).
In a bid to improve uptake of recommendations, there were occasions where the SRP empowered patients to take the recommendations forward with their GP/RNS. This appeared to improve uptake of interventions in several instances, however, communicating via patients is not the recommended pathway. In these scenarios, the SRP felt frustrated for the patient, and despite having a prescribing qualification, there is no current pathway to prescribe an alternative inhaler device on the NHS in this setting. Future studies should focus on building stronger relationships with local surgeries and obtaining read-write access for the pharmacist to inform their recommendations.
During the study, similar numbers of patients from the TC group quit tobacco smoking compared to the IPC group, which suggests that TC was not a barrier in this regard. Only one person was recommended oxygen therapy by the pharmacist due to their oximetry readings and later went on to commence therapy. Pulmonary rehabilitation is a key non-pharmacological component that shows improvement in QoL in patients with COPD.20 There were referrals to pulmonary rehabilitation either directly or via the GP. Directly referring patients achieved a 100% success rate in attendance, whereas only a third attended via the GP.
During the PC of this current study, the appropriateness of patient’s inhaler device(s) was assessed using the In-Check DIAL meter. This device helps to establish the suitability of the inhaler device based on the patient’s inspiratory force capabilities. If their inhaler device was inappropriate, an alternative inhaler device was suggested to their GP. A barrier to TCs was in the assessment of patients’ inspiratory force using this device; from a practical perspective, this measurement had to take place in-person for all participants, as the pharmacist needed to instruct the patient on its use.
The results in this study provide some assurance that TCs are not a barrier for pharmacists to making interventions/recommendations when compared to IPCs and give an indication of what can be expected in future studies. These findings are of importance, particularly when reviewing immobile patients, who are harder-to-reach and often deprived of pharmaceutical advice. To enable an all-inclusive NHS, pharmacists who may otherwise only access these patients via telephone or a rare domiciliary visit, could use TCs with its visual advantage over telephone communication and bearing in mind the time and costs of travel that domiciliary visits inevitably incur (see Fig. 2).
As a SRP, the researcher conducted the consultations and mitigated against potential sources of bias by concealing the randomization of participants, as well as using validated and objective tools wherever possible. The process of randomization into the respective groups proved acceptable to patients, thus will be feasible to conduct in a future definitive trial. Employing an independent SRP to conduct these clinics or analyse the data would be ideal if funding was not a barrier. This would also reduce the risk of ‘vested interest bias’ that could be attributed to the researcher. It is crucial to mitigate against the risks that ensue in having an insider researcher. Also, it was not possible to blind the SRP or the participant to which group they were allocated (open-labelled trial). This issue was mitigated against by keeping systematic differences between the two groups to a minimum (performance bias); this was achieved by observation by another pharmacist and standardizing the consultations of the two groups, where the only major variability was conducting the consultation via IPC or via TC.