Discussion
This scoping review identified 25 articles assessing clinical e-consenting applications. We sought to understand the dimensions by which these publications characterised or evaluated clinical e-consent. We found that publications most commonly focused on the impact of electronic processes on consent data quality and patients’ experiences, with lesser attention paid to providers’ experiences, adoption, efficiency and workflow, and care quality, access and equity.
A plurality of studies addressed the quality of clinical consent data. Overall, data completeness, legibility and accuracy were consistently improved by electronic processes as compared with paper-based ones. This is an important and hoped-for benefit, with implications for healthcare quality and safety as well as organisational efficacy.14–17 These critical findings should assure provider institutions that there is immediate and concrete value to automating clinical consent processes.
Many studies examined dimensions of patient experience in a variety of settings. Almost uniformly, e-consenting was acceptable to patients and did not result in a loss of comprehension. These encouraging findings are consistent with results from the literature on research e-consent, in which patients similarly find electronic consenting systems comparable, or even preferable, to paper-based ones.18 19 A notable exception among the articles we examined was a study of consent for health information exchange, where an attempt to communicate risks and benefits entirely via computer (as opposed to paper plus human interaction) resulted in a decreased comprehension by some patients. This suggests that complex consents, whether electronic or paper-based, may always require human interaction. Another important consideration is the continued emergence of new e-consent technologies (eg, videos, apps). While early evidence in the research e-consent literature suggests patients generally find these, too, acceptable, even greater technological changes may emerge over time.18 20–22 It will be vital to ensure that these developments do not compromise patients’ comprehension or experience of the consent process.
Similar to patients, providers were generally positively inclined towards clinical e-consenting and reported multiple benefits. Again, however, more research is needed, as we found very few articles addressing this topic. Assuring provider satisfaction will be a critical enabler of widespread adoption.23–25
Notably, several studies explored enablers, barriers and other adoption issues, with mixed results. To promote greater uptake among providers and patients, more research around these vital topics is needed. Better evidence could also spur wider implementation of clinical e-consent by healthcare institutions, which need assurance that technology development costs will yield returns on investment.26 27
Our scoping review found that additional—and particularly crucial—aspects of clinical e-consent remain largely unexplored. Measures of efficiency and workflow, for instance, are understudied. While one would expect improvements in these areas to be among the foremost benefits of automation, evidence either way remains scant. This may be because these issues can be difficult to measure, or there may be a lack of interest or opportunity among researchers. In any case, efficiency and workflow will be important topics for future study. For institutions to prioritise clinical e-consenting initiatives, these benefits must be well-established.
Similarly, we found almost no data on the impact of clinical e-consent on care quality, access or equity. Further research in these areas will be crucial to ensure that potential harms to patients are avoided and existing disparities by age, race and other key characteristics are mitigated rather than exacerbated.28–30
On the whole, the literature on clinical e-consent is nascent. While the findings thus far are nearly all positive, the number of research publications is relatively low and reflects a limited number of geographic locations and practice settings. More, several publications came from the same setting, so the range of locations and settings is even smaller than the number of publications. Thus, the data at present are insufficient to guide institutional efforts around clinical e-consent. To save provider organisations from needing to reinvent the wheel or rely on vendors’ marketing claims, additional studies should explore a larger variety of settings, particularly in locations beyond North America and the UK, with an eye for advancing more generalisable knowledge and best practices. Similarly, only one study examined consents for health information exchange, and few studies examined paediatric or proxy consents. More research is needed to explore these vital topics. It is also notable that fewer than half of the studies included information about funding or conflicts of interest, so undisclosed bias in the reported results is possible.
Our study has some limitations. In our literature search, we may have missed relevant articles. However, we developed a comprehensive search strategy with a professional librarian (KM), reviewed additional reference lists from key articles and had two reviewers screen each article for fit. There also is potential subjectivity in review and data abstraction practices; however, we had two reviewers use explicit criteria for all procedures.
