Integrated Risk Information System (IRIS) response to "Assessing risk of bias in human environmental epidemiology studies using three tools: different conclusions from different tools"

Abstract

"Assessing risk of bias in human environmental epidemiology studies using three tools: different conclusions from different tools," a recent publication in this journal, applied the study evaluation approach developed by the U.S. Environmental Protection Agency's Integrated Risk Information System (IRIS), as well as other approaches, to a set of studies examining polybrominated diphenyl ethers (PBDEs) and neurodevelopment. They concluded that use of the IRIS approach resulted in exclusion of studies, which would lead to hazard conclusions based on an incomplete body of evidence. As scientists in the IRIS program, we support the comparison of approaches to improve systematic review methods for environmental exposures; however, we believe the IRIS approach was misrepresented. In this letter, we demonstrate that the ratings attributed to the IRIS approach were not consistent with our own application of the tool. We also clarify the use of studies rated as "low confidence" and the use of an overall study confidence rating in our systematic reviews. In conclusion, the IRIS study evaluation approach is a transparent method to inform certainty in our evidence synthesis decisions and ensures consistency in the development of IRIS health assessments.

Keywords: Environmental epidemiology; Risk assessment; Risk of bias; Study evaluation; Systematic review.

Fig. 1

Heat map of re-evaluations of studies of PBDE exposure and IQ using Integrated Risk Information System (IRIS) approach. Evaluation rationales are available at: https://hawcprd.epa.gov/summary/visual/100500646/. Hashed shading indicates that there are multiple judgments in this domain for the study, which analyzed associations with exposure metrics with different temporal relationships with the outcome