Relative Source Contribution
When the Minnesota Department of Health (MDH) develops a drinking water guidance value for a contaminant, we must consider that water is not the only source of exposure. Contaminants can get into your body from several different sources, such as:
- drinking water,
- food and medications,
- contaminated dust and soil,
- products that are applied to the skin, such as sunscreen and lotion, and
- the air you breathe.
MDH uses a number called a Relative Source Contribution (RSC) factor to express what percentage of a person’s exposure may come from drinking water. However, it can be difficult to estimate how much of a contaminant gets into your body from each source. For years, our approach has been to use an RSC of 20 percent for most chemical exposures. An RSC of 20 percent means that we assume 80 percent of a person’s exposure to a chemical comes from a non-drinking water source.
This approach, based on recommendations from the U.S. Environmental Protection Agency, was created to assess traditional contaminants. We are now trying to assess newer, non-traditional contaminants known as contaminants of emerging concern (CECs). CECs often come to the attention of MDH because of new exposure information, including types of exposure never considered in the past. Because of this, it is possible that our historic approach for estimating exposure may not work well for them. We wanted to see if CEC exposures could be estimated using computer-based models, providing MDH with additional tools to estimate relative source contributions for CECs.
There were two phases to this project. In the first phase, we hired a contractor to review all available exposure models and determine which models would work best to estimate RSCs for CECs in Minnesota. At the end of this phase, the contractor identified ten models out of a total of 33 that seemed to work best for evaluating relative source contributions from CECs.
In the second phase, the contractor evaluated the ten identified models. They evaluated the models by applying them to a specific set of CECs that already had measured experimental exposure data. Doing this allowed comparisons to be made between the modeled values and the exposure values that already existed for this set of CECs. The evaluation showed areas where the models are less effective, and areas where we need more data to run the models effectively. In these test cases, the values we got from the models were sometimes very different from the measured exposure values. There are many possible reasons for this difference: the models are based on a simplified simulation of how exposures happen, and do not perfectly represent the real world. However, they are still a useful tool to turn to when measured data are not available.
One of the main challenges MDH faces is explaining to the public how they can be affected by the chemicals that fill our environment. Our use of computer models adds to our understanding of chemical exposure and helps us determine how to reduce exposure to harmful substances. Since this project was completed, we have used the models in a limited way to compare experimental exposure data to modeled results. The process of applying the models to real-world problems highlighted the importance of using good judgment in selecting model inputs, and interpreting the results. The models are still being incorporated into our process for evaluating CECs. While more testing on the models with different types of CECs needs to be done, these models may allow us to better understand chemical exposures in cases where real-world data are not available.