Using the 4 NPT constructs, we review and interpret the findings of our study in turn:
Coherence: ‘Making sense of new electronic systems’
It is clear that considerable effort has been put into policy building and dissemination of information both locally and nationally in relation to the universal switch to a new primary care EMR in March 2012. The GPASS system was until relatively recently used by around 80% of practices in Scotland
[40, 41]. However, many GPs increasingly felt that the system was no longer meeting their needs. The Scottish Local Medical Committee Conference (2006) called for GPASS to be replaced by alternative systems
. EMR systems have been in wide-spread use in Scotland for many years and their adoption is now – to the best of our knowledge – almost, if not entirely universal
. A majority of GPs interviewed considered that their EMR system was to some degree beneficial to their work practices. Most stakeholders were clear about the need for change and this has facilitated the development of coherence, that is, a shared view of the purpose of these initiatives, with individuals able to grasp potential benefits and has facilitated normalisation of these new technologies. The key lesson here is that the successful adoption of new technology therefore needs to be seen as the result of a sustained effort to communicate the rationale for change and sustained efforts to promote changes in practices, culture and IT use within NHSScotland over a prolonged period.
Cognitive participation: ‘Achieving buy-in’
Although the work of engaging with users is central to the successful implementation of any new technology, work aimed at actively involving GPs in the take-up of new EMRs was barely mentioned in the interviews. While many GPs felt that the previous GPASS system was no longer fit for purpose, most had been using it for years. Many felt that it was – although perhaps not optimum – a system that they had grown familiar with and felt confident using. Several GPs felt that they had received insufficient training before having to switch to the new EMR systems within their practices. However, both INPS and EMIS vendors have provided individualised progressive migration calendars to primary care practices, including training sessions during systems transition and several of the GPs we interviewed also admitted that they too had some responsibility towards making the effort required to improving their skills with the new system. However, they often cited a lack of time as a barrier to do so. The considerable time and effort required to adopt new electronic primary care systems has also been reported in other studies, which suggested that dedicated time for training as well as basic knowledge of ICT were important factors in the successful adoption of these systems
A substantial incentive for the use of practice EMR was audit-related tasks to implement the QOF and this was a key feature. Also, the visible benefits, for example, in terms of improved access to patient information was clearly a positive driver to uptake. However, it is clear that – although there may have been deficiencies in some aspects of the system functionalities – the presence of financial incentives and other system benefits outweighed the barriers to the uptake and adoption of the new systems. Performance-related financial incentives were also identified as important drivers of EMR adoption in a systematic review of the impact of EMR systems in primary care practices
Collective action: ‘Operationalising new systems’
The emphasis of collective action involves the work performed by individuals, groups of professionals or organisations in operationalising a new technology in practice and socio-technical issues, such as how e-health systems affected the everyday work of individuals, organizational structures and goals
. The impacts of practice EMR in that respect are substantial. Overall it is clear that the uptake, adoption and normalisation of these new systems have been possible because, to a large extent, they make the completion of clinical tasks easier.
While GPs will usually work alone during the patient consultation and interact individually with the EMR, an electronic repository of clinical records will facilitate the sharing of patient data with other health professionals within the practice (i.e. nurses and other GPs) and within the health-boards (i.e. with district nurses), as well as enabling electronic transfer of patient information to secondary care services through electronic referrals
. This was considered by a majority of GPs as an important step towards an integrated patient care pathway within the NHS
Several GPs considered that the EMR was therefore a facilitator of shared and continuity of care. Many GPs work part-time and the EMR enables the treating GP to have immediate access to a patient record which may have been accessed and modified by another member of staff within the practice. It also allows for GPs, nurses and healthcare assistants to have concurrent access to the patient medical record. Within the practice, the EMR integrates with an electronic document repository (Docman), allowing to store patient laboratory results and clinical letters such as hospital discharge information. The EMR records are also used to transfer information to the local health-boards electronic data repositories (SCI Store). This information is used among other purposes to populate the patient Emergency Care Summary, available in secondary care hospitals in case of clinical emergencies
Impact on workflows
Many GPs reported perceived usability issues with their EMR and several attributed this to a lack of understanding of their work by system developers. However, the EMR is a complex artefact and it is not entirely clear how individual tasks and functionalities could be further simplified in future. In addition, it is likely that some of the perceived difficulties GPs have with their systems could actually be resolved through additional training and familiarisation with the systems. Indeed, our results suggest an increased overall satisfaction with the EMR systems according to the length of use, which has also been reported in other studies
[20, 22]. The use of EMR also had a substantial impact within the broader practice, in terms of space and storage. As a consequence, administrative support tasks within the practice are now heavily reliant on the use of ICT: for booking patient appointments, record-keeping, quality assurance of clinical coding and completing electronic referral letters on behalf of GPs
Recurrent usability issues during the course of the consultation, such as ‘multiple clicks’ – often perceived by GPs as frustrating and unnecessary – have frequently been reported. The format of our study can not ascertain whether these were legitimate usability issues or else, embedded checks and safety features which were not perceived as such by GPs. In any case, it appears that this potential distinction was not clear to end-users. Furthermore, this also suggests that the use of ‘multiple clicks’ as an error prevention mechanism can be perceived as a blunt instrument for avoiding clinical errors in EMR systems, particularly if this feature is recurrent throughout the system. While it might make sense from the system developers’ point-of-view to introduce double-checks at key decision points – as a typical consultation will usually last approximately 10 minutes on average – the frequency of this type of system interaction can be very high (i.e dozens or even hundreds of times a day), therefore becoming disproportionally frustrating for GPs in the course of the consultation.
Roles, responsibilities and training
The routine use of EMR has an impact on medical training as recently qualified GPs had all trained with one or several emr systems and consequently appeared more comfortable in using or switching from one system to another. Yet, even recently qualified GPs had some difficulties and reservations when using their practice EMR which raises the question of whether further ICT training would be a useful addition to their medical training?
The lack of ICT skills among GPs has been identified as a safety concern in other studies. A previous study by Morris et al. suggested that – although GPs in primary care trusts in England ranked patient safety highly – they often had insufficient knowledge and training to make optimum use of embedded clinical decision support features of their computer systems
. Shojania et al. suggested in their systematic review on the impact of computer decision support systems (CDSS) on doctors’ behaviour that computer reminders only provided modest improvements on clinical processes and guideline adherence
. Avery et al. conducted semi-structured interviews with a range of key stakeholders of GP computer systems in order to identify features which could lead to patient safety improvement, particularly in the area of medication prescribing and decision support alerting
. The authors suggested that a concerted effort from a range of stakeholders would be needed to promote increased safety in the use of ICT in primary care. This would include: additional training of primary care practitioners in the effective use of ICT systems, incentives for systems developers to improve the safety features of their systems and the importance of change management to promote an increased use of ICT for safety purposes. Short et al. identified a number of barriers to the use of CDSS in general practice consultations, including: limited time and consequently the potentially infrequent use of a CDSS, GPs’ limited skills in ICT, a lack of understanding and the risk assessment functionalities, algorithms and results, the reluctance of GPs to rely on a third-party system for risk assessment, the potential concerns of patients with a CDSS and the possible lack of patients’ understanding of risk results