The Centralized Open-Access Rehabilitation Database for Stroke (SCOAR) is a tool that stroke therapy researchers can use to better understand the relationships among variables, efficiently share data, generate hypotheses, and streamline clinical trial design. The purpose of this website is to bring people as close to the data as possible through effective and efficient visualizations. For more information about how to use the website, please see the video below. For a description of variable names, see the "Database" section on Clinical Trials.
The Google map shows which countries randomized clinical trials (RCTs) in the SCOAR database come from. Click on the paper symbol to see more information about that RCT. (Clicking on the author name will then transfer you to the URL of the related paper.) In order to zoom in/zoom out the map use +/- on the right bottom corner of the map.
The number of RCTs per continent in the SCOAR database is shown in this bar chart.
Trials in the SCOAR database came from 114 cities/states in 28 different countries. From this wide range of trial locations, SCOAR represents rehabilitation trials across the globe. Presenting the data geographically, we can see there are relatively few trials for regions such as South America, Africa, central Asia, and eastern Europe.
The evidence-base in rehabilitation is always expanding. Researchers, clinicians, and other stakeholders need effective tools for collecting, managing, analyzing, and visualizing data, but currently RCTs are published in a wide range of formats, across many different journals, and often unavailable (due to paywalls). As such, the goal of SCOAR is to improve methods for searching the literature, conducting power analyses, and place individual studies in their broader context. As SCOAR develops, we hope these data will help improve experiments in rehabilitation, and (ultimately) improve the quality of life for individuals with disabilities.
Our team developed SCOAR to ease data integration across studies and to increase the visibility of a comprehensive body of evidence. The initial SCOAR database was based on a systematic review and screening that left 215 RCTs in the database. From these RCTs there are 489 independent group representing 12,847 patients. SCOAR is a living database, however, and data are continuous extracted from publications and uploaded into the database.
Since there are large amounts of data with very different attributes in SCOAR, this website is interactive, specifically to allow researchers to explore various relationships among variables. Our goal is to allow researchers to efficiently ask their own research questions, generate hypotheses, see the relationships among the reported variables across hundreds of trials.
As new data are published, we also hope that researchers will use SCOAR as a platform to share their summary data with other researchers in a structured format. You can interact with the data in the SCOAR database through this website, but you can also download and work with the raw SCOAR data at any time. All of the data in SCOAR are public and require no permissions to access/use them. We are also happy to help answer your questions about the database at any time! If you do use data from SCOAR in your research, we ask that you cite our initial manuscript that introduced the database.
For this website, we have arranged the visualizations around three main research questions. Users can interact with these visualizations to modify them, filter the data, and link back to original studies to explore data points that are new or surprising! We want powerful visualizations that will get users as closer to the data as possible. (Please refer to Clinical Trials to explore the answer for these research questions)
By plotting summary statistics from rehabilitation RCTs, which can be further filtered by study characteristics, it is easy to look at publication bias across trials in stroke rehabilitation (or in sub-areas when filters are applied).
Cross-indexing common study parameters with experimental outcomes allows researchers to quickly and easily visualize relationships in the field. These visualizations are useful for summarizing past research but also inform future research. For instance, users could filter SCOAR data to see the range of effect sizes for different amounts of therapy at different times post-stroke. Having these experimental parameters available in a few clicks means that researchers can engage in more evidence-based plannng of their experiments, do more case-specific power analyses, and in general see what the "history" of the field has to say about a question (rather than just a few recent and/or seminal studies).
As above, cross-indexing common study parameters with outcome measures makes it easy to see important relationships in stroke rehabilitation. Using SCOAR, users can plot dose-response curves for therapy overall (using all outcome measures), for specific outcomes (e.g., upper extremity versus lower extremity), or for specific groups of participants (e.g., younger to older, severely impaired to minimally impaired).
In this video, Keith Lohse, PhD, talks about how to use each page of the SCOAR website. He introduces the SCOAR database, our research questions, and tips for efficiently using this website to interact with our visualizations.