Top: The quality of the numbers you put into the risk assessment dictate how good of an answer you are going to get. Feature image by Doug Pinkerton
Sally Brown
There is nothing magical or mysterious about risk assessment. Granted it can often seem like some Wizard of Oz type exercise involving lots of hand waving and pulling numbers out of thin air. In reality, it is more like filling in a tax return using a software package. Filling in your IRS forms can at times involve some creativity, but it is a far stretch to call that magic.
It is also like filling out your tax return because the quality of the numbers you put into it dictate how good of an answer you are going to get. Maybe you have all of your receipts, maybe you don’t. Without those receipts, the numbers you enter are much less likely to have a strong bearing on your income or to reality. The same is true for risk assessment. Think of this column as a user’s guide to a risk analysis.
Two Initial Assessments
The first thing you have to do is decide what you are worried about. Yes, I know we have many, many things to worry about these days. Here you are focused though on how one might die. First will it be fast (acute), over a longer but still relatively short time frame (intermediate), or will it be over something that eventually wears you down (chronic)? These days almost all of these risk assessments focus on chronic toxicity — decades of exposure. Both PFAS and microplastics are things that many people in compost are concerned about now. For both, the focus is on chronic toxicity.
The next step is to decide whether you are worried about some type of bodily malfunction or if it is the big C (cancer)? If it is something other than cancer you start the process by calculating the Hazard Quotient (HQ). If it is cancer, you calculate the cancer risk factor. For both, you have to know how much of a particular compound you can be exposed to without anything out of the ordinary happening. Exposure can be by touching (dermal), breathing (inhalation) or eating (ingestion).
Three different sources of information for the HQ are typically used. EPA has Minimal Risk Levels (MRLs) and Reference Doses (RfDs) and the Agency for Toxic Substances and Disease Registry (ASTDR) has Minimal Risk Levels (MRLs). Each of these represents the quantity of the chemical that you can be exposed to on a regular basis without your body having any bad reaction to it. These are typically developed for ingestion and inhalation — eating or breathing. Poison ivy hasn’t figured in here. But take my word for it, dermal exposure would count as acute and nontoxic. For cancer, you would look at either the Cancer Slope Factor (CSF) for oral exposure (i.e., eating or swallowing something) or the Inhalation Unit Risk (IUR) for breathing in something.
PFAS compounds are in many household and consumer products. The risk concern with PFAS exposure is chronic toxicity. Photos by Sally Brown
Non-Cancer Vs. Cancer Risk Calculations
To assess the risk, take the estimated exposure that you are modeling, divide it by the modeled no impact level and you get a number.
For non-cancer impacts that number is referred to as the Hazard Quotient (HQ). Here’s the equation:
HQ = D (dose)/ MRL
If you get a number bigger than one, take notice. Something may be going on. For numbers less than one, go on with your day and figure out the next contaminant to question.
Cancer risk (CR) is also calculated by using an equation:
CR= D (dose in mg kg per day) * the CSF x EED (expected exposure duration/ lifetime)
Here the risk factor number is anything greater than one in a million. That means that if your exposure over a lifetime to the contaminant increases your chance of getting cancer by more than a one in a million shot, there may be a problem. From a regulatory perspective, If the CR is more than one in a million pay some attention. If not, go back to avoiding poison ivy.
This is the very simple version of the spreadsheet. The actual spreadsheet (and there is a program to do this with embedded values in a spreadsheet) has much more nuance. The model automatically accounts for the fact that a toddler will eat and drink less than a teenager, who in turn will eat and drink more than someone in the later years. Age-related consumption factors are built in. Something else that you don’t have to worry about.
Layers of Uncertainty — Some PFAS Math
This all sounds great and simple. And it is an amazing tool to have. However, as I said early on, it is only as good as the data that you use to fill in the columns. The first layer of uncertainty comes with the assumptions on how dangerous or potentially carcinogenic different compounds are. Take PFAS for example. I’m using this as an example because it is all over the news and according to the news, it is the deadliest thing since (you fill in the blank). The focus on PFAS has primarily been on two legacy compounds: PFOA and PFOS. The guidance to the risk spreadsheet (actually known as the Public Health Assessment Guidance Manual or PHAGM) is hosted by ATSDR, part of the Centers for Disease Control. To understand the potential for PFAS to be carcinogenic, they refer you to several sites including the International Agency for Research on Cancer, sponsored by the World Health Organization. Here is summary text from their recently issued guidance on PFAS:
The Working Group evaluated PFOA as carcinogenic to humans (Group 1) on the basis of sufficient evidence for cancer in experimental animals and strong mechanistic evidence that PFOA exhibits key characteristics of carcinogens in exposed humans. There was limited evidence in humans for cancer of the testis and for renal cell carcinoma. PFOS was evaluated as possibly carcinogenic to humans (Group 2B) on the basis of strong mechanistic evidence. There was limited evidence for cancer in experimental animals and inadequate evidence for cancer in humans for PFOS.
What that translates to is that they are pretty confident that PFOA causes cancer in people but that, while there are some indications for PFOS, the jury is still out.
ATSDR also refers you to the EPA, which is more convinced that PFOA is a problem, stating that kidney and testicular cancer have been observed in people. They set a reference dose of 3 x 10-8 mg kg day and a cancer slope factor of 0.0293 mg kg body weight day-1. They are also certain that PFOS is a problem with a reference dose of 1 x 10-7 mg kg body weight day and a cancer slope factor of 39.5 mg kg body weight day (U.S. EPA PFOA and PFOS).
One comment about what those numbers mean. They refer to how much is safe based on how much you weigh (kg) and on the assumption that you will be exposed every day. Two general comments — one is that it reminds me of the great debate on whether red wine is heart healthy or not (verdict lately is not) and that poison ivy is still dangerous. Here are a few sample calculations to translate that reference dose into some numbers:
Let’s say that you weigh 60 kg. That is the same as 132 lbs
140/ 2.2 kg per lb
For PFOA that means you can eat or drink 0.0000018 mg/day and not worry about it.
For PFOS you can eat or drink 0.000006 mg/day and not worry about it.
PFAS has now been phased out of both pizza boxes and microwave popcorn bags. But just for fun and now that we know we can go back to eating both without worry of PFAS exposure, let’s say that each contained 100 mg per kg of either PFOA or PFOS.
If the concentration of each of those compounds in the box or bag (box/bag) was 100 mg kg you could eat:
0.018 mg of the box/bag a day to reach your PFOA limit
0.06 mg of the box/bag each day to reach your PFOS limit
Those numbers are so small, that it basically means that a hint of the paper is all you need to reach your limit for each of these compounds.
Exposure Potential
The second fact that can be problematic is quantifying the exposure potential. After all, you don’t sit down on purpose every day to a plate of a known carcinogenic compound. Or make a deliberate effort to drink a certain quantity per day. You may do that with your steps (10,000 is a great goal), but it is highly unlikely you’d do it with something that gives you no joy and just might kill you. Ingestion and inhalation exposure for risk models is primarily done by accident (cigarettes aside). To be an accurate model, you have to include data on how much of a particular contaminant ends up in food and water that people do consume every day.
Using your taxes as a way to think about risk exposure potential is to model the estimated cost (single point) of two weeks away on a luxury vacation (left) versus a weekend getaway camping trip (right). Photos courtesy of Sally Brown
Typically, modelers turn to published research and data to fill in the holes. If there are a lot of holes, modelers often use the data that is available. Using limited data or data that may not reflect the real world can lead to questionable models. The model will still spit out numbers, but those numbers likely have little to do with reality. If there is a lot of data from highly contaminated sites and minimal amounts from other sites, the model may make predictions that don’t apply to mildly or even moderately contaminated sites. Models like lines, in particular lines that go through the zero point in an axis. That means that modelers will often take limited data and extrapolate to cover the range of values that the model requires.
A way to think about this is again, going back to your taxes. Say you took that luxury vacation of your dreams for two weeks and deducted it as a business trip (no judgement here). The cost of the vacation came to $5,000 per day for two weeks of bliss. In the absence of other data or other examples, the modelers would assume that all vacations cost $5,000 per day. That would put the modeled cost for a weekend getaway camping trip at $10,000. That would count as glamping to the extreme. The converse is also true. Use that two day camping trip as the single point for an estimate and you’ll find yourself in debt for decades when you book the cruise.
Hopefully this sheds some light on the risk assessment process and helps to take away some of the magic. Use this when you get told that something is hazardous as well as when something is safe. Remember that the goal here is to build something that closely resembles reality. Whether you use the same approach filling out your taxes is your own business.
Sally Brown, BioCycle Senior Advisor, is a Research Professor at the University of Washington in the College of the Environment.
