Background: During systematic reviews, data abstraction can be inefficient and error-prone. We developed Data Abstraction Assistant (DAA), an open-access, open-source tool to help data abstractors mark sources of information by dropping 'flags' in articles (Figure). DAA is currently implemented in the Systematic Review Data Repository (SRDR), but is compatible across data systems. Using a randomised trial, we are evaluating the accuracy and efficiency of data abstraction comparing DAA with traditional approaches.

Objectives: To present opinions of users of the DAA tool (DAA Trial participants) regarding the user-friendliness of the tool, and to present preliminary results from the trial.

Methods: There are 52 participants in the DAA Trial. We formed 26 pairs of individuals, each pair comprising one less-experienced and one more-experienced abstractor. After data abstraction for the trial, we surveyed each abstractor using Qualtrics®.

Results: The 40 abstractors who had completed the DAA Trial as of 14 March 2017 completed the survey. 33/40 abstractors (83%) found using DAA to be either very or somewhat easy overall. When asked about future use during data abstraction, 30/40 abstractors (75%) said they are very or somewhat likely to use it themselves, and 24/40 (60%) stated that they are very or somewhat likely to recommend that others use it. When asked about their favourite DAA feature, 21/40 abstractors (52%) named the ability to click on flags marking information sources (Figure). At the Summit, we will present DAA Trial data pertaining to the possible effectiveness of DAA in improving the accuracy and efficiency of the data-abstraction process.

Conclusions: Most users found the DAA tool user-friendly, and most would use it and recommend that others use it for data abstraction in the future. The most popular feature of DAA appears to be the ability to click on existing flags to navigate to portions of text/figures/tables in the article that contain relevant data, a feature that could be useful when verifying abstracted data and during updates of systematic reviews. DAA was released for public use in April 2017.