Paper Plain : Making Medical Research Papers Approachable to Healthcare Consumers with Natural Language Processing

Abstract

When seeking information not covered in patient-friendly documents, healthcare consumers may turn to the research literature. Reading medical papers, however, can be a challenging experience. To improve access to medical papers, we explore four features enabled by natural language processing: definitions of unfamiliar terms, in-situ plain language section summaries, a collection of key questions that guides readers to answering passages, and plain language summaries of those passages. We embody these features into a prototype system, Paper Plain . We evaluate Paper Plain , finding that participants who used the prototype system had an easier time reading research papers without a loss in paper comprehension compared to those who used a typical PDF reader. Altogether, the study results suggest that guiding readers to relevant passages and providing plain language summaries alongside the original paper content can make reading medical papers easier and give readers more confidence to approach these papers.

Know Your Audience: The benefits and pitfalls of generating plain language summaries beyond the "general" audience

“…This joint variation is reflected in the guidelines for plain language summaries 3 and in the strategies science writers use to communicate with interested publics [7]. At the same time, these two dimensions have real-world constraints: there are situations in which technical words must be used to convey specific meaning, or where there is a desire to understand the majority of the details in the original scientific article, such as a patient reading a medical research paper or lab report [11,76]. In studies 1 and 2, we allow plainness and information content to vary based on the intended audience ( §4 & §5).…”

“…Adapting interfaces to different users has the potential to improve usability [13,87], aesthetic judgements [28,69], and trust [61,68]. Increasingly, language styles, such as community language norms [26], formality [9], and text complexity [11,95] have been the focus of adaptable user interfaces. Work has shown that language styles can impact behavior in online experiments [9], counseling conversations [4], online communities [26], and security interfaces [98].…”

“…Work has shown that language styles can impact behavior in online experiments [9], counseling conversations [4], online communities [26], and security interfaces [98]. This work has highlighted the benefits of adapting language to people with different backgrounds [11].…”

“…This is especially enticing in scholarly and scientific communication, where language styles (e.g., medical jargon) can present major communication barriers [85]. Research has explored using models to adapt scientific papers for non-experts (referred to as general audience readers in this paper) [11,44], and paid services like Elicit, 1 or Explainpaper 2 promise to make scientific language easier to read and understand.…”

“…Further, simplified summaries usually convey less information [7] and can unintentionally lead to people being overconfident in their understanding [90]. In contexts where details are important, it may be important to preserve all information, even at the cost of longer or more complex text (e.g., a medical research paper [11]). This gap in research is particularly important for developers of new interactive text interfaces [11,63] because it is currently not clear what the benefits and pitfalls of adaptive text are: when is simplifying necessary, what are the costs in doing so, and do these costs differ for readers with different background knowledge?…”