Saturday, August 11, 2012

What you don't know can hurt you

In 1972, the U.S. passed the Clean Water Act, despite a presidential veto by Richard Nixon. Did this act also end an era of unusually high estrogen levels in the environment? (cartoon source)

There has been much concern over the presence of “environmental estrogens” in our drinking water and elsewhere in the environment. These are man-made chemicals, like DDT, PCBs, and dioxins, that mimic the effect of natural estrogens. Among other things, they’re blamed for declining sperm counts and rising male infertility.

Yet estrogen also enters our environment from a source that excites much less concern. This is the estrogen that women excrete every day in their urine. Shouldn’t that source also be cause for worry?

At first thought, no. People have been urinating for a very long time. And other animals for even longer. During that time, microorganisms have evolved to break down and feed on whatever is present in urine. Larger organisms have likewise had plenty of time to adapt to this aspect of their environment. Urine is so ubiquitous and unchanging that it could not possibly pose a danger. Or could it?

Actually, two things have changed in recent times. One is that humans have become much more numerous, with the result that much more urine is being discharged into the environment. Another is the way it is discharged.

Before the late 19th century, urine entered the environment via privies, cesspools, and ditch sewers (Rockefeller, 1996). It was thus discharged into a warm stagnant medium rich in organic matter—ideal conditions for rapid breakdown of the estrogen molecule by nitrifying bacteria (Vader et al., 2000). These same conditions, however, increasingly became a threat to public health, particularly in the ever larger and more numerous urban centers.

And so a new disposal system was developed. Human waste was now expedited via sewers to a central facility where the liquid component would be separated and rapidly discharged into the nearest cold body of water—which often doubled as the city’s source of drinking water. It was a perfect system for discharging urinary estrogen into the environment with as little biodegradation as possible … and then bringing it back into the human organism. As for urinary androgen, it was also present in wastewater but at much lower levels because of its lower solubility in water (Tabak et al., 1981).

That system survived until the late 1960s and early 1970s, when concern about pollution brought an upgrading of almost all sewage treatment facilities. If we look at the total number of Americans who produce untreated wastewater, this number peaked at 70 million in 1960 before falling to 2 million after passage of the Clean Water Act in 1972 (Copeland, 1993; US Council on Environment Quality, 1984). Primary treatment removes 35‑55% of estrogen from wastewater, and this proportion rises to 50‑70% for secondary treatment (Tabak et al., 1981). Today, tertiary treatment removes 90% of all natural and synthetic estrogenic compounds (LeQuire, 1999).

But what about the 100-year period when most wastewater went untreated? During that time, the main sources of drinking water must have been highly contaminated with estrogen. What were the effects? The most obvious ones would have been a decline in sperm counts and a rise in male infertility. But were there others? And should we soon see a reversal of these trends?

I tried to publish a paper on this subject, but the reviewers were skeptical. I was indulging in speculation that could never be proved one way or another. There simply are no records of estrogen levels in the environment for the period in question.

So I published my speculations on my blog (here, here, and here) and let the matter lie … while keeping an eye on the literature. Recently, three findings have come to my attention:

Primary treatment less effective than previously thought

A recent paper confirms that tertiary treatment of wastewater removes over 90% of all estrogen. On the other hand, primary treatment is less effective than previously thought, the removal rate being only 10% (Limpiyakorn et al., 2011).

Estrogen content of wastewater higher than previously thought

In the past, estrogen levels were measured only for the three most common kinds of estrogen: estrone (E1), estradiol (E2), and estriol (E3). Other natural estrogens, however, are present in urine. The total estrogen level in wastewater is thus 18-34% higher than previously thought:

[…] the total excretion rates of EEQ [estrogen equivalent] by estrone (E1), 17β-estradiol (E2), and estriol (E3) only accounted for 66–82% of the total excretion rate of EEQ among four different groups, and the other corresponding natural estrogens contributed 18–34%, which meant that some of the other natural estrogens may also exist in wastewater with high estrogenic activities. (Liu et al., 2009)

River and ocean sediments reveal formerly high estrogen levels in the environment

It is possible to look into the past by taking cores of sediments from the bottoms of lakes, rivers, and coastal waters. A study of the River Thames has found that estrogen levels are higher in river sediments deposited before the 1960s:

There is an indication of higher concentrations of E1 and E3 in samples deemed to be deposited before the mid-1960s, prior to the introduction of biological treatment at STWs [sewage treatment works] discharging to the estuary. […] This provides indirect evidence that historical improvements to wastewater treatment have resulted in a decrease in the concentrations of steroids in the effluent, as observed for PCBs and DDT from the same core (Gomes et al., 2011)

A similar finding comes from a study of sediment cores from Japanese coastal areas:

The concentration of natural estrogens such as 17β-estradiol (E2) and estrone (E1) in the sediment of coastal areas in Japan was determined […] Core samples were sliced every 2cm from the surface to 20cm deep for the measurement of estrogen. Although the concentrations of estrogens decreased with the depth of core samples, fairly high levels of estrogens were again noticed at the layer deeper than 16cm. (Matsuoka et al., 2005)


The most promising line of research seems to be the use of sediment cores to estimate past levels of estrogen in the environment. One problem will be calibration of sediment dating. Gomes et al. (2011) have pointed to a possible solution by noting that the mid-1960s correspond to the first sediment layer that contains synthetic estrogen, i.e., from birth control pills. Another problem is that estrogen seems to degrade gradually over time, even in river or ocean sediments.


Copeland, C. (1993). Wastewater Treatment: Overview and Background [93-138 ENR] Washington, D.C.: Congressional Research Service.

Frost, P. (2009). The urinary estrogen theory. Part I, Evo and Proud, March 11

Frost, P. (2009). The urinary estrogen theory. Part II, Evo and Proud, March 18

Frost, P. (2009). The urinary estrogen theory. Part III, Evo and Proud, March 26

Gomes, R.L., M.D. Scrimshaw, E. Cartmell, & J.N. Lester.  (2011). The fate of steroid estrogens: partitioning during wastewater treatment and onto river sediments, Environmental Monitoring and Assessment, 175, 431–441.

LeQuire, E. (1999). Something in the Water. InSites, 7(1),

Limpiyakorn, T., S. Homklin, & S.K. Ong. (2011). Fate of estrogens and estrogenic potentials in sewerage systems, Critical Reviews in Environmental Science and Technology, 41(13), 1231-1270.

Liu, Z., Y. Kanjo, S. Mizutani. (2009). Urinary excretion rates of natural estrogens and androgens from humans, and their occurrence and fate in the environment: A review, Science of the Total Environment, 407, 4975–4985

Matsuoka, S., R. Sakakura, M. Takiishi, Y. Kurokawa, S. Kawai, & N. Miyazaki. (2005). Determination of natural estrogens in the sediment of coastal area in Japan, Coastal Marine Science, 29(2), 141-146.

Rockefeller, A.A. (1996). Civilization and sludge: Notes on the history of the management of human excreta. Current World Leaders, 39, 99‑113.

Tabak, H.H., R.N. Bloomhuff, & R.L. Bunch. (1981). Steroid hormones as water pollutants II. Studies on the persistence and stability of natural urinary and synthetic ovulation‑inhibiting hormones in untreated and treated wastewaters. Developments in Industrial Microbiology, 22, 497‑519.

U.S. Council on Environment Quality. (1984). Annual Report. Washington D.C.

Vader, J.S., C.G. van Ginkel, F.M.G.M. Sperling, J. de Jong, W. de Boer, J.S. de Graaf, M. van der Most, & P.G.W. Stokman. (2000). Degradation of ethinyl estradiol by nitrifying activated sludge. Chemosphere, 41, 1239‑1243. 


Beyond Anon said...

What does it matter?

Can we detect any effect on fertility? Have males become less reproductive?

Sean said...

A lot of the research on sperm counts is from Denmark and might be misleading.

Schizophrenia onset at adolescence indicates there is a connection to sex hormones. Schizophrenia was rare before 1800. The prevalence of schizophrenia began to decline in Britain around 1960.

sykes.1 said...

As a civil/sanitary engineer, I can testify that back in the 70s around 70% of the American people were connected to sewerage systems that discharged to sewage treatment plants then to surface waters. The percentage is not much larger today.

In the 70s a significant fraction of sewage treatment plants were settling only, which is what "primary treatment" means. The effect of the Clean Water Act was convert these facilities to biological treatment and to upgrade the degree of biological treatment provided.

Pharmacological discharges were going on even in the 60s, especially estrogens, but have undoubtedly increased per caput since then as psychology has largely become drug oriented.

Some drugs are removed by biological treatment, but others pass through. Engineers and scientists are studying the problem, but alarm does not yet appear warranted. No doubt more developments can be expected.

Schizophrenia was not a recognized or defined illness in the 19th Century (there were no defined psychological illnesses then), and the modern definition is quite recent. I doubt it is more than 20 years old. So, any trends regarding this disease are utterly fictitious.

Sean said...

I wonder if sex ratios have changed since the 60's

Doctors could recognize schizophrenia by the middle of the 20th Century, They knew the age of onset and course of the disease is different in the sexes . The prevalence of schizophrenia is less than it was around 1960. There are precious few catatonics about now.

It wouldn't be all that surprising if the overlooked metabolites of estrogen had subtle effects on the brain.

As for the metabolites of xenoestrogens - God knows!

Jason Malloy said...

Israel sperm banks find quality is plummeting:

"Though declining sperm quality is an international phenomenon, the change in Israel is occurring at nearly twice the pace as other developed countries...

Some scientists fear that Israelis are being overexposed to female hormones.

... Laurence Shore, a retired hormone and toxicology researcher at the Kimron Veterinary Institute ... has found Israeli milk and associated products such as butter and cheese can contain 10 times as much estrogen as products from other countries because of Israel's aggressive milk-production practices.

... Haimov-Kochman is looking into water quality... The water, she says, has been found to contain traces of ethinyl estradiol, a synthetic estrogen used in birth control pills, which gets into the water through the urine of women taking the pills.

"You can't clean this from the water," she said."

Anonymous said...

""You can't clean this from the water," she said."

same was said by a woman in a news blurb about male fish turning to female in american rivers. You can find the clip on youtube.
There was a 1990 article about the same in britain that I had come across while searching about this phenomenon.