Summary of Main Findings
Interoperability was largely an unfamiliar concept to patients and caregivers. Once a brief definition was provided however, most participants recognised its necessity and potential benefits. Participants were able to highlight how the lack of EHR interoperability negatively impacted their care, predominantly as a source of notable inconvenience for themselves and their healthcare providers.
Limited interoperability has reportedly contributed to inaccurate, incomplete, or incoherent medical histories within EHR which were often difficult to amend, thereby leading to the unencumbered perpetuation of errors. Patients recounted that this has negatively impacted clinical decision-making. Consequences ranged from repeat history taking or investigations, to more serious ones such as misdiagnoses of potentially stigmatising conditions. Patients regularly had to serve as the main means to ensure information was shared appropriately. This often included carrying paper-based documentation when travelling between providers, verbally conveying care plans, and acting as the final backstop for providers to verify clinical information found within EHR.
Participants proposed several solutions or factors which should be considered in any future attempts to address poor EHR interoperability. Most participants conveyed a desire for ‘one centralised system’. As improving interoperability will result in greater amounts of clinical information being shared, participants stressed the importance of a corresponding increase in data and privacy protection measures. There was an overarching sentiment that efforts to improve provider-to-provider information sharing should also include parallel efforts to improve information sharing with patients. Patient participants perceived that increased interoperability with their involvement would attenuate the current risks of information sharing, namely sharing of inaccurate information between different providers. More robust training for NHS staff to foster better HIT competency and greater involvement of allied health services and social care in the upkeep of EHR content were areas in which participants believe further efforts should be made to improve overall EHR interoperability.
Strengths and Limitations
This study contributes to the growing body of evidence investigating how patient perspectives can be used to inform future directions for key healthcare challenges. The topic guide was informed by existing literature and was piloted with two patient research groups before use. The focus groups were conducted online, which allowed patients and caregivers from across the UK to participate and thus broadened the range of perspectives included.
However, our findings must also be taken in the context of certain limitations. Firstly, the relatively small number of participants, mostly of elderly age, limits the range of views and perspectives discussed in our focus groups. There is also some level of self-selection bias as participants are likely to be patients who feel more strongly about their views or are more able or inclined to participate.
The inclusion of only adult patients with chronic conditions also neglects exploration of the views of other patient groups, such as parents and children. The inclusion of only patients who were able to participate using video conferencing introduced selection bias for those who are more digitally literate, and likely from socio-economic backgrounds who could afford the technology necessary. Lastly, the restriction of participants to only those able to converse in English likely contributed to excluding those less proficient, or unable to speak the language.
Comparison with Prior Work
Few studies have investigated specifically EHR interoperability and its practical implications on patient care from their perspective (26).
Kelly et al., summarised how current EHR interoperability levels contributed to suboptimal care for dialysis patients in the US (27). Specifically, patient perceptions of HIE were found to be dependent on the perceived risks and benefits to their care resulting from increased clinical data sharing, which were based on bespoke lived experiences. For example, patients from ethnic minority backgrounds were found to have greater concerns regarding the sharing of their clinical information through HIEs. The authors also pointed to potential solutions to the patients identified to mitigate these concerns, many of which align with those reported in our own study including greater patient engagement and increasing opportunities for patient feedback longitudinally (27).
In 2016, Legler et al., conducted a survey study which explored patient satisfaction levels and their perception of their healthcare provider’s knowledge of their medical history after the introduction of a centralised portal (26). While the portal did not allow for the manipulation or transfer of patient health data stored across the US VA, Department of Defence, and community providers, the authors nonetheless found that merely using it to view its contents resulted in patients being 14% more likely to report higher levels of satisfaction with their healthcare providers as they appeared more knowledgeable about their medical history (26). This was especially noticeable for patients with long-standing relationships with their healthcare providers (26). However, the authors also noted that providers were more likely to use the portal with newer patients and thus be the ones more likely to reap the greatest clinical benefits compared to patients with existing relationships (26).
A recent publication by Hussein et al., investigated patient-generated health data (PGHD) interoperability from three distinct perspectives (technological, clinical, and users) concerning the proposed DH-Convenor platform, an initiative to support the collection, organisation, and interoperability of PGHD in the Austrian national health system (28). Many of the overarching opportunities and barriers identified mirror the views expressed by our study participants. For example, the authors found that patients were largely accepting of the clinical necessity for PGHD interoperability and its purported benefits. However, participants clearly opposed the sharing of socioeconomic data and any attempts to use patient data for commercial or political purposes (28).
A qualitative study by Sanyer et al., examined patient perspectives on clinical information sharing via EHR in team-based care (29). Though authors noted that patients recognised the value of centralising their clinical information, concerns were raised towards documentation inaccuracies and the associated difficulties in resolving them (29). Patients acknowledged that the ability to share health information amongst members of their care team was important, but also voiced apprehensions regarding data security, who should have access to their information, and the scope of information shared (29). This was especially poignant for sensitive clinical information such as mental health diagnoses. Aside from the sharing of ‘basic’ information (e.g., name, date of birth) with all members of the care team, patients were only willing to share more sensitive or personal information found in their EHR with their immediate healthcare provider or someone the patient already had a well-established relationship with (i.e., GP) (29).
Implications for Policy, Practice, and Further Research
The past two decades of EHR use have demonstrated the importance of greater data sharing and the necessity of interoperability in the HIT infrastructure of modern healthcare systems. As demonstrated in our study, the need for a more coordinated, interoperable EHR system is evident. Despite the priority often being placed on tailoring such systems to the requirements of healthcare providers, our recent experiences, especially throughout the pandemic, have made apparent the urgency with which future EHR systems and the policies surrounding their use, must better accommodate the needs and expectations of patients. Tackling the lack of interoperability will necessitate fundamental changes ranging from the types of data that are stored and how it is shared, to addressing more systemic questions such as aligning information governance with evolving patient preferences. A renewed attempt at realising some form of a centralised EHR system should be revisited, especially given that the recognition of its need is apparent even amongst patients.
As shown in our findings, the idea of increased EHR interoperability is synonymous with greater access to patient data for more stakeholders, including for patients themselves. HIT policies and procurement contracts must encourage vendors to incorporate this into their design of future EHR systems from the outset and ensure that as information sharing is enhanced for providers, it is also made more robust for patients. As illustrated by our participants and supported by findings from other similar studies, the greater sharing of clinical information possible through better interoperability is contingent upon efforts to strengthen data security and privacy protections (28–30).
Future research should seek to quantify any improvements through the implementation of new EHR interoperability interventions. Outcome measures of interest may include the accuracy of stored clinical information, patient safety incidents, user satisfaction, or time spent reconciling patient information. These efforts would help demonstrate the value of expanding EHR interoperability and serve to corroborate the findings found in patient-centric qualitative studies such as our own. Likewise, additional qualitative studies focussing on underrepresented ethnic minority groups, non-English speaking communities, and different target patient demographics (e.g., paediatrics), would also be useful in ensuring future efforts at improving EHR interoperability are inclusive of their needs.