Medical Device Link.

Originally Published MX January/February 2004

INFORMATION TECHNOLOGIES

Numerous methods to evaluate the quality of scientific literature have been described, a few of which have endeavored to identify and categorize the messages in the documents. One such process is the Astrolabe Message Mapping System (AMMS). This system reflects the current state of knowledge of scientific message mapping.

AMMS utilizes software algorithms and relevancy filters to cull a large mass of scientific literature to a subset of publications that represents the information most likely to influence health professionals' knowledge and treatment habits in the area of interest. Specialized human evaluators then read and rate the identified publications using a series of sophisticated criteria involving the relevancy, accuracy, and presentation of research data. This results in a source score, a numerical value assigned to each individual published source of information.

The evaluators next identify and categorize the key messages in each source document by means of other decision-tree algorithms. How well a message relates to and is supported by the study findings is the measure of its strength. The strength of each message is determined by an algorithm incorporating the value of the message, the value of the source, the value of the literature type, and a few other criteria. The resulting message score, related to the source score, enables a user of this system to review several dozen relationships among factors such as publication target audience, clinical investigators, publication lag time, sponsor influence, and message frequency and age. Data reports and graphic displays of this information are generated from analyses of these relationships.

Reports based on analyzed data help clients to develop a SWOT (strengths, weaknesses, opportunities, and threats) analysis of their products and the competition. A good SWOT analysis enables a company to improve the effectiveness of product development and marketing strategies. Knowledge of publication lag times helps in choosing a publication vehicle that will publish an article relatively soon after it is accepted. Knowing who sponsored a study can be important in understanding the source of negative information about a product. Reports addressing the frequency and age of messages can assist the client in identifying messages it thinks are important but that may not be making much of an impact on target markets. They also bring to light older messages that may need updating—or simply repeating.

AMMS uses algorithms that enable evaluators to interpret information found in the scientific literature consistently. The system presents resulting interpretations in a way that reliably replicates healthcare professionals' perception of the published knowledge, by mimicking readers' extraction processes in rating messages by value.

Copyright ©2004 MX