A method for assessing the potential for confounding applied to ionic strength in central Appalachian streams



Causal relationships derived from field data are potentially confounded by variables that are correlated with both the cause and its effect. The present study presents a method for assessing the potential for confounding and applies it to the relationship between ionic strength and impairment of benthic invertebrate assemblages in central Appalachian streams. The method weighs all available evidence for and against confounding by each potential confounder. It identifies 10 types of evidence for confounding, presents a qualitative scoring system, and provides rules for applying the scores. Twelve potential confounders were evaluated: habitat, organic enrichment, nutrients, deposited sediments, pH, selenium, temperature, lack of headwaters, catchment area, settling ponds, dissolved oxygen, and metals. One potential confounder, low pH, was found to be biologically significant and eliminated by removing sites with pH < 6. Other potential confounders were eliminated based on the weight of evidence. This method was found to be useful and defensible. It could be applied to other environmental assessments that use field data to develop causal relationships, including contaminated site remediation or management of natural resources. Environ. Toxicol. Chem. 2013;32:288–295. © 2012 SETAC