Knowledge Linkages: Augmenting Online Clinical Care Discussions with Published Literature



Samuel Stewart*, Dalhousie University, Halifax, Canada
Syed Sibte Raza Abidi, Dalhousie University, Halifax, Canada


Track: Research
Presentation Topic: Building virtual communities and social networking applications for health professionals
Presentation Type: Oral presentation
Submission Type: Single Presentation

Building: MECC
Room: 0.9 Athens
Date: 2010-11-30 01:00 PM – 02:30 PM
Last modified: 2010-09-21
qrcode

If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL).

Abstract


BACKGROUND: The emergence of online discussion forums has provided clinicians alternate knowledge dissemination mediums to converge and share their knowledge through problem/topic-specific discussions. Tyically, such discussions contain the experiential knowledge of clinicians—i.e. vital and pragmatic insights into what worked, what did not work and what are the best practices in specific clinical situations
OBJECTIVE: The objective is to augment health professional’s online communication by connecting it to evidence-based literature. The rationale is that practitioners need to reaffirm their practice-related online discussions with evidence reported in the medical literature so that the online recommendations can be applied safely.

METHODS: We have developed a Web 2.0 based knowledge management framework that (a) leverages Web 2.0 based communication methods to gather clinical experiential knowledge through online interactions between a virtual community of clinicians, (b) organizes the rather unstructured online discussions in terms of meaningful clinical topics/discussions—i.e. a discussion thread; (c) parses the content of the online discussions to derive medically salient keywords compliant to MESH terminology; (d) applies our specialized search strategy, using the derived keywords from the selected online discussion, to retrieve research articles from Pubmed; (e) presents the research articles to the user through a web-based interface. The data for the project pertains to the Pediatric Pain Mailing List (PPML)—an online forum for practicing clinicians to contact peers on the subject of pain in children. PPML brings together over 700 pediatric pain practitioners from around the world to share their clinical experiences and seek clinical advice. The PPML contains over 13,000 messages spanning more than 15 years.

RESULTS: A pilot study was conducted on a sample of the archives using all messages on the PPML in 2007 and 2008. One hundred threads were reviewed to determine (a) the accuracy of the message parsing, (b) the accuracy of the thread assignment, and (c) the accuracy of the papers returned.
The results of the pilot study were promising. The message parsing was successful on 74% of the messages. The organization of messages into threads was achieved accurately over 92% of the time. Our search strategy was compared against implementations of a Vector Space Model (VSM) strategy and an extended Boolean Information Retrieval model (eBIR). The literature searches based on an online discussion thread returns a ranked list of articles. Our results conclude that the precision and relative recall of our search strategy outperforms the VSM and eBIR methods.

CONCLUSIONS: As the prevalence of online communication between clinicians increases, it is vital that these communications containing experiential knowledge are aligned with published evidence. Our framework allows health professionals to (i) search through the PPML archives to find interesting discussion topics, (ii) select interesting discussions threads within the discussion topics, and finally (iii) retrieve a set of related research articles from Pubmed, through a ‘single-click’ evidence retrieval mechanism. This project has successfully augmented Web 2.0 based communications with published evidence through the process of knowledge linkage. The featured framework is generic in nature and can be applied to any specialized medical domain.




Medicine 2.0® is happy to support and promote other conferences and workshops in this area. Contact us to produce, disseminate and promote your conference or workshop under this label and in this event series. In addition, we are always looking for hosts of future World Congresses. Medicine 2.0® is a registered trademark of JMIR Publications Inc., the leading academic ehealth publisher.
Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.