Medical Terminology Management in Evidence Synthesis: Early Versus - - PowerPoint PPT Presentation

medical terminology management in evidence synthesis
SMART_READER_LITE
LIVE PREVIEW

Medical Terminology Management in Evidence Synthesis: Early Versus - - PowerPoint PPT Presentation

Medical Terminology Management in Evidence Synthesis: Early Versus Late Binding Craig Whittington, PhD, Sandra Zelman Lewis, PhD, Tobias Sayre, and Robert Battista, MBA Disclosures All 3 of us are employed by Doctor Evidence . 2 Outline


slide-1
SLIDE 1

Medical Terminology Management in Evidence Synthesis: Early Versus Late Binding

Craig Whittington, PhD, Sandra Zelman Lewis, PhD, Tobias Sayre, and Robert Battista, MBA

slide-2
SLIDE 2

Disclosures

All 3 of us are employed by Doctor Evidence.

2

slide-3
SLIDE 3

Outline

The context Ontology management Why late binding is superior How researchers can benefit

3

slide-4
SLIDE 4

Doctor Evidence is a global technology and services company, providing dynamically-updatable systematic evidence reviews

Digitizing evidence from published studies, clinical trials, meeting abstracts, and other forms of clinically-relevant data permits:

  • Sophisticated analytic capabilities

– Direct (pairwise) meta-analyses – Network meta analyses (Bayesian, frequentist) – Cohort analyses

  • Graphical visualizations of a body of evidence
  • Easily customized evidence tables

4

slide-5
SLIDE 5

Ontology Management

Involves application of poorly-defined term ontologies,

  • r arbitrary

decisions to group terms Different variables representing what are thought to be the same construct are

  • ften bound

together using subjective and poorly-defined criteria These bindings are not easily restorable Does not allow for transparent review of authors' source language and context.

One obstacle in evidence synthesis: primary research studies use various terminologies based on geography, therapeutic area, time, researcher, etc.

CURRENT STANDARD PRACTICE OF DATA CAPTURE

5

slide-6
SLIDE 6

Ontology Management

Term, taxonomy, and ontology management system literature Terms are bound after these extractions: “late-binding” Terms can be unbound

  • r rebound.

Learning system based

  • n previous bindings

DOCTOR EVIDENCE’S APPROACH:

Enables data to be captured, stored, and indexed "as is" in the literature

  • preserving the original

terms and definitions,

  • efficiently accommodating

synonyms, and

  • utilizing a variety of

medical ontologies

6

suggests options selected before As appropriate for specific analyses

slide-7
SLIDE 7

Advantages of Late Binding

ADVANTAGES LATE BINDING

7

slide-8
SLIDE 8

DEMO

8

slide-9
SLIDE 9

Value to Healthcare Providers

Specificity is captured. Thoughtful and relevant term binding has been incorporated into the appropriate analyses. Clinical relevance has been maintained.

Healthcare providers can trust evidence reviews created with this term management approach

VALUE TO HEALTHCARE PROVIDERS

9

slide-10
SLIDE 10

Value to Researchers

High quality data extractions preserving

  • riginal

terms and definitions Granularity

  • f concepts

Healthcare providers can trust evidence reviews created with this term management approach

VALUE TO RESEARCHERS

10

Ability to quickly review

  • riginal

terms and definitions in context Filtered and bound

  • utcomes &

variables, as appropriate for each analysis subgroup analysis Comparative reviews of various

  • ntologies or

creation of new

  • ntologies
slide-11
SLIDE 11

Contact Us for a Complete Demo

Sandra Zelman Lewis, PhD Chief Guidelines Officer slewis@doctorevidence.com Craig Whittington, PhD Global Head of Quality, Methodology, and Innovations cwhittington@doctorevidence.com

11