Developing an intelligent, socially-oriented search recommendation service for electronic health records

This project researched the development of a search query recommendation service for health practitioners and researchers to help professionals retrieve information from electronic health records more easily and efficiently when using full-text search engines. This recommendation service would also allow users to develop and refine search queries for more effective searches in the future.

Start date: 9/27/2010
End date: 9/26/2011

Read More

In an era where the rapid growth of knowledge is distributed among multiple databases, management of these resources has grown far beyond what a single individual can master. As a result, clinicians and researchers face many challenges in information management. New services are needed to help organize all of this information so that it can be used by others in the future.

The goal behind this project was to understand how to give practitioners and researchers with little or no formal information retrieval training the ability to optimize their search queries with this recommendation service. When physicians can access and analyze electronic health records more efficiently, they can save time, improve results, and provide better levels of care.

A full-text search engine is critical to increase the reuse value of documents stored in electronic health records. However, many users lack the search expertise or domain knowledge needed to construct effective and inclusive search queries. This new recommendation system was designed to improve the quality of care that health practitioners can provide and promote collaboration among users so that search queries can be continually improved through social interaction. In this way, search expertise can be developed, preserved, and shared among people and across domains.

The research delineated the difference between electronic health record (EHR) searches and Web-based searches and found that EHR searches are more sophisticated and more challenging for users to formulate their information needs with appropriate queries. Also recognized was the effectiveness of collaborative search, encouraging further exploration of social-information foraging techniques to bring together and utilize the collective expertise of the users.

Kai Zheng from the U-M School of Public Health collaborated on this project.

Read the full project analysis here

Grants

Developing an Intelligent and Socially Oriented Search Query Recommendation Service for Facilitating Information Retrieval in Electronic Health Records, National Institutes of Health: $301,251

 

The National Institutes of Health is made up of 27 different  Institutes and Centers, each with its own specific research agenda. NIH’s mission is to seek fundamental knowledge about the nature and behavior of living systems and the application of that knowledge to enhance health, lengthen life, and reduce illness and disability.