What does ePrescribing mean for patients? A case study of the perspectives of hospital renal patients

Background Hospital ePrescribing systems are expected to improve quality of care for patients, yet the perspectives of patients themselves have seldom been explored in the context of ePrescribing deployments. Objective We sought to understand the significance of ePrescribing for patients through a case study of renal in-patients on a hospital ward, before and after the introduction of an ePrescribing system. Methods Three data sources were drawn on as part of the case study: interviews with representatives from national patient groups (n = 10), in-patients on a renal ward (n = 11 pre-implementation; n = 12 post-implementation) and fieldnotes (n = 25) of observations made on the case study ward. Data were analysed thematically focusing on: (1) perceived benefits of ePrescribing; (2) patient awareness and understanding of the medications prescribed and (3) patient views on medicines reconciliation at admission and discharge. Results While ePrescribing was viewed positively overall, its implementation in the case study site failed to address the lack of patient involvement in the prescribing process and poor medication counselling upon discharge. Importantly, the limited impact of the ePrescribing system in these particular areas appeared to be the result of institutional and cultural practices rather than solely technological factors. Conclusions The introduction of ePrescribing systems offers new opportunities to improve sharing of knowledge and communication with all those involved in the patient’s care pathways, including patients, carers and healthcare professionals across diverse care settings. Achieving this will, first and foremost, require significant cultural and policy shifts in how the patient’s role is perceived by clinicians in relation to medicines management.


INTRODUCTION
The introduction of hospital ePrescribing systems is seen as a critical step in improving quality of care for patients as a result of anticipated increased safety 1-4 as well as improved efficiency and communication between healthcare team members. [5][6][7][8][9] There are still considerable sociotechnical issues and challenges 10- 19 to be overcome before these systems become commonplace across hospitals in England and the vision for a paperless National Health Service (NHS) 20 can be achieved. However, concerns about the negative impact of using clinical computer systems on the quality and amount of face-to-face interactions between patients and healthcare professionals (HCPs) 21, 22 underline how benefits realisation in one area may reduce quality of care in another. Indeed, even in countries like Denmark, which are seen as having more mature digital healthcare environments than in the UK, 23 ePrescribing systems still leave scope for important medication errors. 24 There is some evidence that patients have broadly positive attitudes towards the digitisation of healthcare. [25][26][27][28] More specifically, they recognise, on the whole, the potential of electronic systems in enabling a paradigm shift in the patient-HCP relationship 29 through improved communication, knowledge, data and information sharing. Yet technology still needs to be used appropriately in order to deliver benefits such as improved communication, 30 and there is an urgent need, therefore, to better understand what ePrescribing in hospitals means for patients and what their priorities are when it comes to medicines management. In order to investigate this, we undertook a qualitative case study, designed to explore in detail patient perspective on ePrescribing. We report here on the main findings.

Study design
The study was designed as part of a wider programme of work investigating the implementation, adoption and use of ePrescribing systems that offer varying degrees of functionality in the supply, administration, recording and prescribing of medication. 31 The aim of the case study was to focus specifically on the patient perspective of medicines management and ePrescribing. The approach was both inductive and deductive, drawing on themes initially identified in the wider programme of work 31 and through expert interviews with patient organisations (see topic guide 1, Appendix) and then explored through interviews of in-patients on a renal ward of a hospital studied in the programme. 31 The patient organisation interviews were also used to refine the patient topic guide (see topic guide 2, Appendix). We chose to focus on the renal ward at the hospital for both strategic and practical reasons. First, renal patients were identified as one of the groups of patients at increased risk of experiencing events related to the introduction and use of ePrescribing systems due to the complex regimen of their medication and the co-morbidities they may suffer from. Second, renal patients as frequent users of in-patient facilities were seen as most likely to notice changes in service delivery over time. As such, while these patients' prescribing experiences may not have been unique, they were seen as illustrative of the types of issues that may ensue in such complex cases. Third, the renal ward had been earmarked as an early adopter ward, where both preand post-implementation data could be collected within the timeframe of the study. The two-stage data collection design was used to allow changes in quality of care to be captured within the datasets. The case study was not intended to be representative of all patients' experiences and was instead seen as providing an in-depth exploration of localised experiences on a single ward, which could be combined with the broader perspective of the expert communities representing the needs of patients nationally to offer wider relevance. 32

Sampling and recruitment
Two categories of participants were recruited for the study to provide both breadth and depth of data.
Category 1: We drew up a sample of 13 patient organisations, whose expertise could be used to inform the findings from the interviews with renal patients. This included groups representing renal patients and other conditions these patients may concurrently suffer from.
Category 2: Individual patients were self-selected following a verbal invitation from nursing staff on the renal ward of a large urban hospital in England both before and after the implementation of an ePrescribing system. This was a specialist renal ward caring for patients with renal problems. Therefore, the main reason for the patients' presence on the ward was to receive treatment for renal conditions, although this did not preclude continuation of medication for other existing long-term health issues (e.g. diabetes). Only inpatients were included. Eligible patients were only included if available and if judged well enough by staff. Initial acceptance was taken first verbally by the nursing staff who supplied the patient with an information sheet. Written consent to take part was obtained by the researcher from each patient prior to each interview and any queries answered. In total, 15 patients were approached by the nursing staff before the introduction of the ePrescribing system, of whom 11 were interviewed. After implementation, a total of 14 patients were invited to participate in the study, of whom 11 were interviewed. There were across the two data collection

Perceived benefits of ePrescribing
The patients and patient organisations valued the potential of ePrescribing systems which they believed could improve speed, efficiency, accountability, clarity of information and support decision making. In particular, in a paper-based environment, errors relating to drug names and dosing appeared to be a commonly occurring problem across conditions and treatments. It was also pointed out that as well as making such errors less likely, the precision and clarity of information in an ePrescribing system (such as food intake for correct dosing of insulin) offered real potential for improved treatments. ePrescribing was further seen as enabling better compliance and use of available patient data. For instance, weight recorded on admission could be used to automatically calculate the appropriate dose rather than relying on estimates at the time of prescribing.
For many groups of patients, knowledge that such improvements would be happening in the hospital they are being cared for would be extremely reassuring as … "a lot of kidney patients have a fear of just going to an ordinary district and general hospital" (NKF Head of Advocacy, Interview #40). It was also suggested that widespread use of these systems could lead to more effective information sharing by harnessing the opportunities offered by patients' growing smartphone use, or through data mining for example to identify the side effects of medication.
The patients interviewed on the renal ward at the case study site echoed this positive perception and referred to the ubiquity of computer technologies in everyday life. Indeed, for most, the thought that prescribing at the hospital would be or had gone electronic was seen as something that should be expected in this day and age. For others, ePrescribing was simply seen as reassuring and a good way to reduce the likelihood of prescribing errors being made.

Medication awareness and patient communication
The prescribing process was virtually invisible to most of the in-patients due to the complexity of drug regimens as well as the sometimes urgent circumstances of treatments. Overall patients … "assume that they're going to be looked after properly and that errors will be minimised if not zero" (BHF Medical Director Interview #50). While many patients were described as quite happy for the doctor to do whatever (DUK Interview #49), others, especially those with long-term conditions such as diabetes, had clearly become experts of their own medication. It was also explained how self-management was important because there was often a personal preference of how the patient wanted to feel, for instance, after taking insulin. The introduction of an ePrescribing system appeared unable on its own to either tap into patient expertise or to facilitate self-management. Such developments would depend above all on local prescribing policies and practices. The problem phases, four patients who declined to take part and three who were repeatedly unavailable at the times when data collection took place. Although most participants had been in-patients on the ward before ePrescribing was introduced, the timeframe of the study and the variability of the patient's admission schedule did not allow for the same patients to be interviewed twice. The number of interviews was determined by saturation, i.e. when no new themes were emerging from the data.

Data collection
Interview data (see topic guides, Appendix) were collected by the researcher (LL) between January 2014 and February 2015 and resulted in three interview data subsets. The first subset was intended to provide a general overview which could then be explored through patient interviews before ePrescribing was used on the renal ward (subset 2) as well as after (subset 3). Subset 1 involved eight telephonebased interviews and two face-to-face interviews with patient organisations. In subsets 2 and 3, 11 patient interviews were carried out before the ePrescribing system was introduced on the ward and another 11 additional interviews 4-5 months later. All patient interviews were carried out in a private space at the patient's bedside on the ward, either in a private single room or behind a screen. The collection of data for each subset lasted 4-8 weeks. Notes were also made of relevant observations on the ward resulting in 25 fieldnotes.

Data analysis
The interviews were audio recorded with the participant's consent, transcribed, fully anonymised and then analysed in NVivo 10 (QSR International Inc., Melbourne, Australia) to facilitate a systematic and integrated approach following initial coding by key areas of interest. 25-30 The same thematic approach was used for both categories of interviewees. The coding framework was applied to the observational notes, although these notes were linked primarily to individual interviews and therefore used to provide contextual clarifications to help guide the analysis, elucidating details where appropriate. The analytical approach and study results were reviewed by the Programme's Patient and Public Involvement (PPI) Group in order to improve the robustness of the findings and the inclusion of the patient perspective throughout the study. PPI members provided feedback on the design of the study, topic guides and findings. This included a discussion of findings, during which, the results were found to resonate with perceptions of this group of patient representatives.

Ethical Considerations
Participation in the study was voluntary and informed consent obtained from each participant prior to the interview. Records of individual names were removed from all datasets prior to analysis. The study was approved by the NHS National Research Ethics Service Committee. which would represent a step back. Furthermore, it was not clear to many how ePrescribing systems could address the lack of communication which was seemingly deeply engrained in hospital prescribing practices. Indeed, it appeared that the ePrescribing system, by removing the physical presence of the paper chart, had diminished a patient's ability to input to discussions about their medication: It was also explained that this lack of communication meant that both patients and carers had a poor understanding of the medication regimens post-discharge, with potentially detrimental consequences for drug adherence levels.

Medicines reconciliation and discharge
Medicines reconciliation is a fundamental aspect of patient safety since it involves sharing, verifying and correcting information regarding a patient's medication at admission, transfer and discharge. 33 The potential of computer-based systems to facilitate this process is seen as considerable. 34 The introduction of ePrescribing on the ward had made little difference, however, in improving the medicine reconciliation process, as it could not completely prevent human error. Patients felt frustrated that errors were being made with their regular medication, even though they had provided the correct information to the HCP.
Patients on the renal ward reported both before and after the introduction of the ePrescribing system considerable problems with "absolutely total lack of information" (Patient Renal Ward, Interview #97) between primary and secondary care. Discharge was also described in negative terms, the result of poorly designed processes where tracking progress on orders and checking drug availability was difficult. Surprisingly, the extended wait patients had to endure at discharge did not improve following the introduction of the ePrescribing system and were a continuing and visible matter of concern on the ward (Fieldnotes #62-65 Furthermore, medicine reconciliation issues at admission were not corrected at the point of discharge. These problems were compounded by sparse detailed information on the discharge summary, poor communication with the patient's GP and the lack of medicine counselling given to patients pointed out earlier. There was a sense that a more personalised approach to prescribing at the point of discharge was urgently needed to ensure patient safety. was thus seen as cultural rather than simply a technical matter: "Historically a cultural problem, you went in hospital, you stopped whatever you were taking and we'll decide what you should now be on so it was very much we know best" (NA Trustee, Interview #41). This lack of communication with patients was also described by one patient on the ward who explained how he had experienced two prescribing errors at the hospital in the past and that this had led him to showing greater assertiveness when prescribed medication: They didn't bother to ask me what I'd been taking for 10 years without any conflict they just decided to stop and then when I asked the pharmacist basically I'd been given the wrong type … It means any time someone says something about medication I want to know why, when, how … I'm the idiot that takes them. (Renal Patient, Interview #66) The patient organisations interviewed further described the need for improved dialogue between HCPs and the patient, as well as a better understanding of the role of carers in the conversation: This lack of information, awareness and understanding highlighted according to some interviewees the issue of consent, especially when a medication was seen as 'routine'.
There were concerns among some of the patient groups that ePrescribing could lead to a reduction in self-management,

Quality of care and safety
Some respondents expressed concern that the use of an ePrescribing system on the ward may reduce quality of care, especially if the system's potential to replace clinical expertise was overstated or if the system allowed increased delegating of tasks. Patients themselves described how the introduction of the system had, immediately after rollout on the ward, impacted negatively on the patient-HCP interaction: I was in here late last year for an operation and because they weren't so adept at it they seemed more zombie-like … they were afraid of making a mistake. It definitely was because they were so concentrated on getting it right on the computer that sometimes they forgot that you were there. (Patient Renal Ward, Interview #86) While familiarity with the system did seem to improve with time, the quality of interactions between patients and HCPs and the speed of prescribing and administration had in some cases been reduced following the introduction of the system: … before they'd get the drug chart out, it's all been signed and if a patient queries something she'd go straight to the doctor and get it signed there and then and come back but this time he's got to put it in the system, he's got to wait for it to come through the system (Patient Renal Ward, Interview #82) Lack of hardware on the ward was mentioned by some patients as contributing to the reduced speed of prescribing, with queues forming at times to use the computer. This type of focus on the computer rather than the patient was noted also in the fieldnotes during data collection (Fieldnotes 80-85; 86; 95-98).
A nurse complains that some information has gone missing from the screen. Another nurse joins to help out, as they talk, the patient appears to have no place in the interaction and subsequent delay. (Fieldnote #86) Some patients voiced their concerns (without prompts) regarding the security of the information stored and the procedures in the event of the system crashing.

DISCUSSION AND CONCLUSIONS
A central theme running through the patient perspective of the prescribing process in hospital is a lack of patient involvement which may become especially significant in an ePrescribing environment. The ePrescribing system did not in this case allow patients to access further information on their medication. By removing the paper chart at the end of their bed, patients had less access to information about their medication than before. It is clear also that an ePrescribing system does not in itself facilitate best practice in self-management; rather, what is needed is a care redesign.
There is a pressing need to address the lack of inpatients' involvement in prescribing and decision making, 35 yet the introduction of an ePrescribing system may be doing little to change such cultural norms. Those implementing ePrescribing in hospitals should be mindful that patients are more concerned with knowing that an opportunity exists to ask questions and input into decisions, rather than emphasising the existence of the new ePrescribing system itself.
The situation here may be the result of temporal and localised circumstances, reflecting in part the limitations of the study. Indeed, patient interviews were collected on one ward only and in a relatively immature ePrescribing context. 10, 11 That said, perspectives from the patient groups helped redress this by providing broad ranging perspectives in terms of medical conditions, geographical locations and time. What is more, lack of patient engagement is strongly embedded in organisational cultures that often ignore the complexity of the prescribing process. 36 Using new ePrescribing technologies to support this engagement will need to be accompanied by a cultural shift in how patients' existing knowledge or conversely lack of understanding of their medication can be addressed within the prescribing process. By so doing, safety benefits anticipated from ePrescribing systems in hospitals are more likely to be achieved, and drug compliance upon discharge improved, thereby reducing readmissions. 37