Historical approaches that search, compile, and manually evaluate relevant information would be very time and resource intensive to implement for all ~33,000 chemical substances in the non-confidential, active TSCA inventory. The EPA developed the PICS approach to integrate information from a variety of sources to better understand the landscape of publicly available information for these chemical substances. The PICS approach uses a large information management and technology infrastructure to synthesize traditional and NAM information in key scientific domains including human health hazard-to-exposure ratio, ecological hazard, carcinogenicity, genotoxicity, exposure to susceptible populations, persistence/bioaccumulation, and skin sensitization and skin/eye irritation. The output is a display of the chemical substances from the TSCA active inventory that reflects the overall degree of potential concern related to human health and the environment and the relative coverage of potentially relevant human health and ecological toxicity and exposure information. Behind this visual display is a quantitative summary of the individual domain metrics. This information could aid in determining the level of effort and resources that may be needed to evaluate specific chemical substances together with flags to identify potential information needs.

A proof-of-concept case study was performed by applying the PICS approach to a subset of the TSCA active inventory. The design of the scientific domain workflows was an iterative process using the results for chemical substances of known/expected hazard or exposure. For example, the results of the analyses for chemical substances with previous genotoxicity assessments helped to refine the genotoxicity domain workflow and determine where and why the workflow may vary from past assessments. The results of this case study showed that the overall SDM was generally correlated with the IAM, suggesting potential testing or reporting bias. However, the PICS approach was able to segregate the recently released TSCA high- and low priority candidate chemical substances, with some differences related to important aspects of expert review. Expert review would include data and study quality analysis, which may lead to removal of some studies or endpoints included in the PICS approach. Further, expert review would include a weight-of-evidence analysis and take into account the breadth of the available data unlike the PICS approach that focuses on selecting the more conservative result in order to limit the number of false negatives.

Apart from the TSCA high and low priority candidate chemical substances, most of the remaining chemical substances from the 2014 TSCA Work Plan were juxtaposed with TSCA high priority candidates. The chemical substances from the 2014 TSCA Work Plan were expected to have both a high SDM and IAM due to the rigorous selection process lead up to the Work Plan. However, a small subset had limited information availability suggesting that the Type 1 data sources may not capture all of the information sources utilized in the selection process. The POC238 also included chemical substances from the SCIL and intentional food ingredients lists. The PICS approach generally resulted in these chemical substances having a lower SDM, with some exceptions related to the conservative approach that may be addressed during a more systematic review (e.g., study quality). However, the chemical substances from the SCIL and intentional food ingredients lists also had lower than expected IAMs suggesting either missing information sources or the information collected on these chemicals may be targeted towards the specific uses and exposures.

As described above, the PICS approach has caveats and limitations. To accelerate the process of integrating publicly available data for a large number of chemical substances, the evaluation of the Scientific Domain and IAMs are performed using an automated process that may not account for all potential exceptions or contexts that may occur for a specific chemical substance or chemical substance group. The PICS approach relies on a large database of chemical substance properties, hazard, exposure, persistence, and bioaccumulation information that have been integrated from multiple publicly available sources and models. As the databases and methodologies are updated, the PICS approach can be applied again to update the results based on the latest available information. Although efforts have been taken to ensure the accuracy of the information, the database may contain errors propagated from the source databases. The cleaning and curation of the information will be an ongoing process and require significant resource investment to iteratively improve and develop new systems that avoid regeneration of legacy data. In many cases, data used in this analysis were not able to be verified back to primary source information. Data points that were verified from authoritative secondary sources were flagged with an information gathering flag and individual study quality was not considered. The quality control effort relied on the acceptance of data and information from authoritative sources. Finally, the domain workflows were designed to select the more conservative options unless otherwise stated; this likely results in a higher incidence of false positives, activities reported at lower doses, and exposures reported at higher doses. This was done to create the most comprehensive group of potential candidates for prioritization with the potential false positives identified in the subsequent expert review phase.