Developing Criteria for Evaluating the Quality of Exposure Measurements in Epidemiological Studies during Systematic Review

Abstract

During a systematic review of epidemiological evidence on health risk from exposure to environmental chemicals, it may be necessary to develop criteria for evaluating the quality of exposure measurements that account for specific factors relevant to the chemical of interest. We describe approaches for developing chemical-specific criteria through three case studies for exposure to phthalates, methylmercury, and hexavalent chromium.Each case study began with a set of prompting questions aimed at answering the core question: Does the exposure measure reliably distinguish between levels of exposure in a time window considered most relevant for a causal effect with respect to the outcome? We then developed additional follow-up questions specific to each chemical (or class of chemical). We used a range of approaches to address the questions and draft chemical-specific criteria for rating exposure measurements as Good, Adequate, Poor, or Critically Deficient (Good indicating the least potential for bias in effect estimates due to exposure measurement error).The approaches and key chemical-specific criteria were as follows: For phthalates, based on a comprehensive review of the literature, we identified urine as the preferred biological matrix and emphasized repeat sample collection. For methylmercury we consulted multiple independent researchers with expertise in mercury biomonitoring and found agreement on the importance of standard reference materials and quality control related to mercury adhesion to analytical equipment. For hexavalent chromium, we used existing guidance to develop criteria that take into account uncertainty due to the reduction of hexavalent chromium to trivalent chromium in air and biological samples.Further development of chemical-specific evaluation criteria can enhance the use of standard systematic review tools. The views expressed in this abstract are those of the author and do not necessarily represent the views or policies of the U.S. EPA.