Adding exposure information to inventories

Part of the challenge in making LCAs more useful as predictive tools has been incomplete and unreliable life-cycle inventories (LCIs). An LCI describes and quantifies the flows of matter and energy into and out of the systems being assessed. This includes matter entering by water and air flows, as well as releases to air, land, water, and biota. LCIs describe and quantify the flows of matter and energy into and out of the systems being assessed. This includes natural resources used or consumed throughout the life cycle, as well as releases to air, land, water, and biota. LCIs exist in many different formats, often distributed through commercial software (such as GaBi and SimaPro). Recently, open-source software is making data more accessible to a larger audience and improving transferability of data.

There are a number of ways that exposure and toxicity can be combined in LCAs, and this integration is continuing to develop. In the past, the U.S. EPA's Exposure Factors Handbook has been incorporated into risk assessment spreadsheets such as CALTOX and developed into LCA characterization factors such as TRACI's original human health characterization factors.

Currently, other tools are available with several being evaluated as part of the U.S. EPA's Systematic Empirical Evaluation of Models (SEEM) framework. SEEM applies models to extrapolate exposure and internal dose for multiple-exposure scenarios, routes, and pathways. These high throughput predictions can be made by using data and models in a forward manner, or be inferred in a reverse manner, such as from biomarkers. However, these tools have not yet been utilized within LCA.

Recently it has been recognized that microenvironments (μE) may have a significant impact on human health, but have not been included in LCAs. The initial focus to estimate exposure to chemicals in products used in μE encountered in their daily habits and routines necessitates a “systems” model to delineate data needs arising from numerous knowledge bases to integrate product formulations, purchasing and use activities, and human activities. This will indicate products likely to be in different μEs and the potential for human contact with chemicals in these products. Modeling consumer product exposures requires a wide array of data, including (1) product ingredients, (2) pharmacokinetic factors, (3) consumer product category–specific “exposure factor surrogates,” and (4) time/activity estimates (human factors). These different data streams must then be integrated within an interface such that different exposure scenarios for individual, population, or occupational time-use profiles can be interchanged and quantitatively explored to screen tailored chemicals for potential exposure, and ultimately dose. This allows estimates of multichemical signatures of exposure, internalized dose (uptake), remaining dose or body burden, and elimination.

The resulting exposure data can then be mapped onto toxicity data. For example, EPA's ToxCast system predicts and prioritizes chemicals based on their potential toxicity by using advanced science tools to help to understand how normal human body processes are impacted by exposures to chemicals and to help determine which exposures are most likely to lead to adverse health effects. Exposure information can be combined with toxicity information in various ways ranging from qualitative to quantitative (see illustration). Although the approaches described in the illustration were developed for human exposure, similar approaches may be used for integrating exposure and ecotoxicity. For example, potential exposures to sentinel or endangered species may not differ substantially from human exposures since the target is a single species. Thus, the LCA may account for a chemical agent in a product that is intentionally applied (for example, a pesticide) or one that migrates to a habitat (for example, released from a stack or outfall structure and then transported to the habitat). Thereafter, the activities of the species (such as predator-prey, migration, and bioaccumulation factors) and pharmacokinetic factors within the organism can be modeled and estimated.

More comprehensive exposure assessment within LCA

Throughout the history of LCA, there has been a search for the appropriate level of sophistication and comprehensiveness concerning how to incorporate human health. The evolution from the exclusive use of toxicity information to the current practice acknowledges that most LCAs continue to lack incorporation of important environmental and human health risks, since they do not include all of the exposure pathways of agents, especially the thousands of chemical compounds in the marketplace. In addition to the exposure pathways found in the Exposure Factors Handbook and the pathways traditionally included in LCA, current and future research is focusing on consumer product exposures (for example, cosmetics), indoor air releases (such as off-gassing of building materials), and occupational exposures. This greater comprehensiveness in human and environmental exposure information should improve the utility and predictive ability of LCAs and make them more valuable risk management tools.