A Precautionary Tradition
The procedure used by the FDA to establish a level of exposure from
pesticides and new food additives that will be considered safe by the U.S.
government was developed over 60 years ago. Even though it was and is
somewhat arbitrary, this relatively simple standardized process is used to define “safety” without resorting to personal judgments. This is achieved largely because many of the
arbitrary judgments are standardized.
In particular, the statistical evaluation of the way uncertainties will be
treated, and the use of standard safety or uncertainty factors to calculate a
“safe” exposure determine the extent to which the evaluation will “err on the
side of safety”. There also may still
be one more somewhat arbitrary decision, that will typically be made by an
expert or group of experts, about what constitutes an adverse effect, but that’s it.
A key characteristic of the Safety Assessment procedure is
that it is designed to be precautionary.
In addition to erring on the side of safety, it was designed to be used
for new chemicals that were previously not present in food. But since it was first developed in the 50’s,
procedures using safety factors have come to be used for many other regulatory
applications as well. In particular,
when pesticide regulation was moved from the FDA to the EPA in the early 70’s,
Safety Assessment was adopted by the EPA for more general purposes. However, it didn’t always work very
well. In fact, sometimes it can be quite
silly.
A Safe Level for Water
Hydrogen hydroxide, otherwise known as water, is a common
ingredient in many foods and beverages. Since
it was commonly used before regulations were instituted for new food additives,
it is clearly the sort of chemical that would be among the food ingredients
that are Generally Recognized As Safe had, and therefore it is would not be regulated
by a safety-factor based procedure - if anyone even considered it. Nonetheless, for the sake of comparison,
let us suppose hydrogen hydroxide was to be added to food or suddenly discovered
for the first time. What level of
exposure would be considered to be safe?
There are very few formal studies of the toxicology of
hydrogen hydroxide. That is probably because water is not very toxic
at all. The median lethal dose (the LD50)
in rats has been determined to be greater than 90 g/kg body weight (Lewis, 2014),
which means that even very large amounts of water (9% of total body weight)
will kill some, but fewer than half of the rats consuming that amount . The reason drinking too much water can be lethal is well understood. The kidney needs to maintain a proper balance
of electrolytes, and it does so by eliminating water as needed. But, if you drink too much at one time, it
can’t keep up. The hyponatremia that results
can be fatal.
Using the value of 90 g/kg body weight as a starting point,
a safe exposure to water may be calculated with the application of standard
safety factors:
- A factor of 10 because humans may be more sensitive than rats
- A factor of 10 to account for variability among humans
- A factor of 10 because a No Observed Effect Level was not identified
Dividing 90 g/kg bw by 1000 (10 x 10 x 10) yields a safe
dose (e.g. an ADI or RfD) of 0.09 g/kg bw, or 90 mg/kg bw. For an average adult (about 75 kg), that comes
out to 7 g per serving, about half a tablespoon. Any more than that wouldn’t be safe. Yes, that’s silly – but why?
Thinking Again
Perhaps the first objection should be that applying the
Safety Assessment procedure to water constitutes a use that it was not designed
for; not unlike hitting a screw with a sledge hammer. Perhaps Safety Assessment shouldn’t be used
for chemicals that have always been in food.
You could even say that maybe it shouldn’t be used for chemicals that
aren’t strictly optional. You could
still use Safety Assessment to identify already present chemicals that might be
a problem, but not automatically conclude that being above the level identified
as safe necessarily means that there is a significant problem. Under those circumstances, perhaps the issue
should be given more thought.
So, if we are going to get serious about the safety of
hydrogen hydroxide, what else needs to be considered?
- Perhaps we can whittle away at the safety factors a bit. While there are no epidemiology studies concerned with the toxic effects of water, there are enough case reports to suggest that the susceptibility of humans is about the same as rats. So, that should allow the animal-to-human factor to be dispensed with. We could also find that there are many known instances of adults consuming one or even two liters at a time with no obvious adverse effects. Those two adjustments will get our safe level of water to a far more practical 100-200 g (4 to 8 ounces) at a time. But still, how terrible would it be to drink the whole can or bottle?
- Water is hard to avoid. It’s in just about everything. It’s even in the air you breathe!! You would have to carry around a calculator and a table of the percentages of water in everything you eat to make sure the safe level is not exceeded. Is it worth it?
- Water is necessary. Generally speaking, the risk of having too little hydrogen hydroxide is far greater than having too much. It’s called dehydration. So, instead of avoiding the consumption of 200 g in one sitting, on occasion it may be far riskier to not drink 500 g. You can’t err on the side of safety when there are two sides to err upon. Perhaps even one safety factor is too many.
Running the Marathon
For marathon runners, getting the right amount of hydrogen
hydroxide is a serious matter. One of
the major obstacles to completing, or even surviving the race, is to get enough
water to replenish the amount lost while running. But, the amount needed varies, and it is also
important not to drink too much. At the
very least, it will depend on the weight on the runner and the weather
conditions. Simple formulaic advice may
be more dangerous than the more traditional approach of drinking when thirsty
(Hendricks, 2011). Therefore, any
direction to the contrary deserves careful scrutiny.
References
Ballantyne C (2007). Strange
but True: Drinking Too Much Water Can Kill. Scientific American, June 21, 2007.
Hendricks N (2011). Marathon
runners who drink too much water are at risk of a deadly condition. Washington Post, Oct 24, 2011.
Lewis RJ (2014). Sax's
Dangerous Properties of Industrial Materials, 12th Ed. John Wiley & Sons.
Official Post Soundtrack
Post Notes
Thesis Post #53. Serves as an example of the misuse of the Safety Assessment Paradigm.
