Understanding chemical exposures associated with consumer products starts with a basic question: Which chemicals are present? Yet ingredient lists and traditional targeted testing may not capture the full range of chemicals found in the products people use every day. 

Non-targeted analysis, or NTA, offers researchers another way to investigate that question. Using high-resolution mass spectrometry, NTA can identify chemical signals beyond a predefined list of substances. As the number of NTA studies has grown, so has the challenge of comparing their findings. Results are dispersed across the scientific literature and are not always reported consistently, making it difficult to draw connections or apply existing data to new exposure questions.  

At the 2026 International Society of Exposure Science Annual Meeting, Chemical Insights researchers presented a standardized dataset designed to make existing consumer product NTA data more accessible to the scientific community.  

To build the dataset, the team reviewed 80 studies that directly analyzed consumer products or product extracts. They compiled reported chemicals and key study information, checked reported chemical identities against established data sources, and standardized the records for comparison across studies. 

Together, the 80 studies yielded 11,706 chemical reports from a wide range of consumer products, such as plastic and paper food packaging, apparel, and personal care products. After reviewing the identities and reconciling overlapping records, the team identified 6,876 unique, verified chemical structures. 

The analysis also brought gaps in the available chemical information into focus. Among the verified structures, 809 were absent from both the U.S. Environmental Protection Agency’s CompTox Chemicals Dashboard and CAS SciFinder. Another 538 reported chemicals could not be linked to a verified structure because there was insufficient information to establish their identity.  

The team found further gaps in available hazard and exposure information. In total, 2,541 chemicals, or 37%, were not represented in the evaluated resources, identifying where knowledge remains limited and future research may be needed. 

Computational methods for evaluating exposure, toxicity, and toxicokinetics can help researchers put existing data into context, particularly when measured data are limited, and identify chemicals that may warrant further study. The team plans to continue developing the dataset by adding information on analytical methods and measured concentrations, while also expanding it to include NTA studies involving biomonitoring and environmental samples.  

As the dataset continues to grow, it can deepen the insights available from existing consumer product data and support further investigation into potential everyday exposures.