You are here:
Using study evaluation to inform evidence integration: Application in a systematic review of hexavalent chromium male reproductive outcomes
Citation:
Yost, E., X. Arzuaga Andino, A. Sasso, AND C. Gibbons. Using study evaluation to inform evidence integration: Application in a systematic review of hexavalent chromium male reproductive outcomes. NAS Workshop: Evidence Integration in Chemical Assessments, Washington, DC, June 03 - 04, 2019.
Impact/Purpose:
This poster uses an example from a systematic review of hexavalent chromium to demonstrate how study evaluations for risk of bias/sensitivity can facilitate evidence integration, particularly when studies find divergent results.
Description:
Study evaluation is used in systematic reviews to identify the strengths and weaknesses of the evidence base in a consistent and transparent manner. This can be used to inform evidence integration by identifying factors that may affect the reliability and interpretability of the results. Here, we describe how this principle was applied in a systematic review of the male reproductive effects of hexavalent chromium [Cr(VI)]. Twenty-three animal toxicology studies that examined effects of Cr(VI) exposure on the male reproductive system were evaluated by at least two independent reviewers for reporting quality, risk of bias, and sensitivity using a domain-based approach, and were rated as high confidence, medium confidence, low confidence, or uninformative. Of these, eight were considered uninformative due to serious flaws and were excluded from consideration. Four studies had no notable concerns and were considered high confidence, and eleven had significant concerns across multiple study evaluation domains and were considered low confidence. Whereas the high confidence studies found that the male reproductive system appeared unaffected by Cr(VI) exposure, the low confidence studies found a range of effects including decreased sperm quality and quantity and alterations in hormone levels. It was concluded that the evidence for Cr(VI)-induced male reproductive effects in animal models was slight, as the reliability of the observed effects was compromised by risk of bias and sensitivity concerns.