The first question is always whether the alleged phenomenon is real

A couple of days ago, I had a post, A rolling disaster about the possibility that mRNA vaccines might have temporarily or permanently reduced fertility rates.   While I accept this as both a theoretical possibility and the existence of some early studies suggesting that is what is happening, I stated my concern that the data volume and integrity is still too small and too weak to be confident in the hypothesis.

We should pay attention but probably don't need to panic.  Yet.

This morning there is a post from the often excellent Scott Alexander, Declining Sperm Count: Much More Than You Wanted To Know.  Where el gato malo was looking at the recent impact of mRNA on fertility rates around the world, Alexander is looking at the older and long standing debate about whether there is a danger of declining global sperm count and its possible impact on fertility.

The generic empirical response to any alleged phenomenon/problem is to ask the following questions:

1) Is the phenomenon real in a measurable sense?

2) What are the possible causal mechanisms?

3) Which among the causal mechanisms are most demonstrable and

4) Among those, which have the highest effect size?

5) Given the causes and effect size, what are the possible interventions, costs, and

probability of success?

Alexander focuses on the first question and lightly covers the second.

How Sure Are We That This Is Even Real?

Not too sure.

The authors of these studies are well-respected scientists - yes, even the one who wrote the book about imperiling the future of the human race - and they seem to be doing good statistics.

But an argument against might start with this graphic:

Source: Figure 2 here.

Each circle is an individual study examined in Levine’s first meta-analysis. I notice two things:

Yes, okay, that line is pointing very slightly down, and apparently this is statistically significant.

But also, the data are very noisy. Some studies from 2005 show higher sperm counts than most studies from the 1970s. The biggest pre-1980 study shows sperm counts very similar to today’s. 

He goes into a lot of detail.  His investigation is not glib.  

His conclusion is that there is data to support the hypothesis of global reduction in sperm count but that it is weak for many clearly understood reasons.  There is enough data to suggestion that possibly the phenomenon is real but that the evidence is demonstrably weak.  Again, enough evidence to suggest we should keep an eye on the issue but not enough to warrant any sort of five-alarm fire response.

He identifies five popularly ascribed reasons meant to explain the phenomenon which we are uncertain is real:

Plastics

Pesticides

Sunlight and circadian rhythm

Diet and obesity

Porn

IF the phenomenon is real, I would advance a different hypothesis - urbanization.  Two hypotheses in fact.

In 2020, 56% of the world is urbanized.  In 1900 only 14% of the world was urbanized.  

In the fifty or seventy year we are looking at the world has become markedly more urbanized.  Some of the countries with apparently the greatest decline in sperm count are also countries which have urbanized the fastest (e.g. China).  The causal mechanism I would argue is that urban living is stressful and that stress is a known inhibitor of fertility.  

All the other reasons remain candidates but I would suspect the transition from rural living to urban living might be the greatest impact.  If the phenomenon is even real in the first place.  

The second hypothesis is that if there is a real reduction in sperm counts, it might also be explained by demographic aging.  Most of the developed world has been aging for several decades but there are many developing middle income countries such as China which are also aging very rapidly.  

Again, sperm counts fall with age so if you population is aging, you would of course expect a decline in sperm count.  Median age in the world in 1900 was 20 years.  In 2020, it is 31 years.  And again, the countries which have aged the fastest are also the ones with the most notable declines in sperm count.

If there is a real decline in global sperm counts, I would suspect that the great bulk of the explanation lies with urbanization and with demographic aging.  But only if the phenomenon is real.