Format

Send to

Choose Destination
Ann Intern Med. 2007 Dec 18;147(12):871-5.

Update on the methods of the U.S. Preventive Services Task Force: estimating certainty and magnitude of net benefit.

Author information

1
Department of Obstetrics, Gynecology and Reproductive Science, Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, California 94143-0856, USA.

Abstract

The major goal of the U.S. Preventive Services Task Force (USPSTF) is to provide a reliable and accurate source of evidence-based recommendations on a wide range of preventive services. In this article, the USPSTF updates and reviews the process by which it evaluates evidence, determines the certainty and magnitude of net benefit, and gives a final letter grade to recommendations. Because direct evidence about prevention is often unavailable, the Task Force usually considers indirect evidence. To guide its selection of indirect evidence, a "chain of evidence" is constructed within an analytic framework. The Task Force examines evidence of various research designs that addresses the key questions within the framework. New terms have been added to describe the USPSTF's judgment about the evidence for each key question: "convincing," "adequate," or "inadequate." For increased clarity, the USPSTF has changed its description of overall evidence of net benefit for the preventive service from "good," "fair," or "poor" quality to "high," "moderate," or "low" certainty. This rating considers the extent to which an uninterrupted chain of evidence exists across the analytic framework. Individual studies will continue to be judged as being of "good," "fair," or "poor" quality. Using outcomes tables, the USPSTF estimates the magnitude of benefits and the magnitude of harms, and synthesizes them into an estimate of the magnitude of net benefit. Although some judgment is required at all steps, the USPSTF strives to make the process as explicit and transparent as possible. The USPSTF anticipates that its methods for making evidence-based recommendations will continue to evolve.

PMID:
18087058
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Silverchair Information Systems
Loading ...
Support Center