Web Interactive Presentation of EPC Reports: A Foray Into Interactive Reports [Internet]

Excerpt

Background: Systematic reviews are consumed by different types of users to address an ever-expanding set of questions and needs. It is unlikely that a single static report will efficiently satisfy the diverse needs of diverse users. It might be practical and satisfactory to give users the ability to generate their own report from the information produced by the systematic review.

Methods: We developed an open-source web-based interactive report presentation (IRP) of a systematic review on 14 nonsurgical treatments for Urinary Incontinence (UI) in women. We used data from a systematic review we conducted through the Agency for Healthcare Research and Quality Evidence-based Practice Center Program for the Patient Centered Outcomes Research Institute. We presented wireframes (mockups) of the IRP to a panel of three Duke Health System stakeholders in teleconferences, and incorporated their feedback in the final IR, which includes a visual representation of the evidence base, and allows users to access generic or detailed descriptive and analytic information through a point-and-click interface. Subsequently, the stakeholders piloted the IRP and evaluated its ease of use by answering open ended questions. We proposed a roadmap for scalable IRPs of systematic reviews starting from a thematic analysis of these suggestions.

Findings: The final tool (accessible at https://to1.infalliblekitty.com/#index) allows users to obtain descriptive and analytic results for a network of 14 treatment categories, various outcomes (cure, improvement, satisfaction, quality of life) and several subgroups (all women, older women, or those with stress or urge UI). Users can access descriptions overall or for specific sets of studies. The stakeholders were satisfied with the functionality of the tool, and commented on its ease of use. They proposed numerous augmentations, which we organized in themes of presentation (e.g., present information on numbers of trials on figures), analytic (e.g., allow on-the-fly subgroup analyses, explore tradeoffs between several outcomes) and information sharing (e.g., ability to import/export data from/to other software). A roadmap to satisfying these augmentations involves the following: (i) integrating analytic capabilities in the tool, (ii) incorporating ways to assess the tradeoffs between several distinct benefits and harms of interventions with minimal assumptions, (iii) defining an information technology standard for evidence synthesis objects (i.e., what information, in in what format, completely describes an evidence-base and its synthesis), and (iv) developing a rigorous representation of evidence synthesis objects as mathematical graphs that can be analyzed.

Conclusions: In sum, we described a foray into an alternative view of a systematic review that complements a static systematic review report. Health system representatives found it useful and practical, and requested analytical, expository, and information sharing augmentations to the tool.