- And when we hear that sperm counts are going down, are they going down to the point where fertility is impacted? That's really the one of the, I think, functional questions. - And when you have 45 to 50 million per milliliter and below, it matters a lot what your sperm count is.

I mean, people say, "Does it matter?" Yeah, if you get in this range where the probability of conception is dropping off really rapidly, it matters a lot. And then around 45 to 50, it starts to level off. And then after that, after certainly after a hundred, probably 75, it doesn't matter at all.

So when people say, "Does sperm count matter for fertility?" Yes, it matters a lot if it's low. And no, it doesn't matter at all if it's high. So we just have too many sperm. I mean, I don't, but humans, you know, and there's- - Nature runs a probability game, overproduce sperm.

- Right. - And sperm counts range anywhere from, you know, it could be low eight, nine, 10 million per milliliter in the very low situation. - It could be zero. - It could be zero in some people, right? All the way up to 400 million. There's a huge range.

- That's right. - And that's a function of age. It's a function of genetics. It's a function of presumably phthalate exposure. - I was asked to join a committee of the National Academy of Sciences. And that committee was assembled to look at the question of whether hormonally active chemicals, endocrine disrupting chemicals in the environment posed a threat to human health.

Because at that time it was like, well, yeah, we hear about this, but should we care, right? And so that committee wanted to consider a study that had come out of Denmark a few years earlier, which claimed that sperm count had dropped 50% in 50 years. - Wow, that's a huge drop.

- That's what, we're seeing worse than that by the way now. So, okay. They said to me, I was the only statistician on the panel, would you look at this and see if we need to consider this in our deliberations? And as I mentioned, I'm skeptical and I looked at it and I thought, eh, I don't think so.

That was my initial reaction. And that was because, first of all, I didn't know who had written this, I just saw it in a journal and it was not very big and not very many figures, not very much data. And I thought of it and I thought, that's a big claim for a little paper.

But I'll look at it 'cause it's important. And so, what I did then was to think about all the factors that we epidemiologists call confounders, things that might have caused that decline if it wasn't real biologically, right? And so, we could think of some of them together, maybe the method of counting sperm had changed so that later methods counted fewer sperm in the same sample.

That's certainly possible, right? But it turned out that wasn't the case 'cause they actually had all used the same method. And maybe the men had changed. So, maybe you can't get a sperm count at random, you have to get somebody to volunteer, right? So, who were these men? Or were they very different in the early part of the study and the late part of the study in a way that maybe in the late part of the study, there were men with lower sperm count and they were more concerned?

Maybe they were more obese. That's pretty plausible. Obesity is related to sperm count, fertility. And maybe they smoked more and maybe, and so on and so forth, right? So, what I did was to get the 61 studies, go through them and try to extract information on all the factors that could explain the decline.

So, I created a multivariable model and ran that model. And to my astonishment, when I was done, the slope of the decline was exactly the same to the first decimal place. It had not explained anything. I was like, oh my God, this looks like it might be real. - This is really important to, because I think what we're talking about here in parallel to the main conversation is how to do really great science.

- So, when I saw that and actually did another study to select my own studies and not accept her 61 studies that had been published, that Elizabeth Carlson had published. So, new studies came up to more recent times, went back further, did it again, found exactly the same thing.

Okay, so there were three looks at that. And I thought, okay, I'm gonna accept this now. This is, sperm count is declining. And why? I turned to the why. Okay, because up and down now, we hadn't said anything about why. We just said, is it doing that? Yes. Okay, now we believe it is declining.

Why? And so then I thought quite a lot and talked to people and ruled out genetics because it was too fast. It's two generations, it's too fast. 50 years, two generations. So, if it's not genetics, then it's environment. And so what is it about the environment that could do this?

So, I asked, okay, in the environment, there could be things that are making sperm decline. So, if you think about how you might look at that, you might design the study that I designed next, which was another study. And by the way, this preceded the AGD. So, we had four cities in the United States that we picked with different environments.

And then we got men to come in and we used the same equipment at each place. We used the same method of selecting the men. The technicians were trained centrally at UC Davis. We had very good quality control. So, samples were sent around every quarter to make sure that everybody was measuring things the same way.

We didn't want drift, right? And then we got their urine. And that's how I had those urine samples. So, if you wanted to do this study and you wanted to get a representative sample of men, where would you go? Because you can't, I can't ask a guy in the street to give me a semen sample, right?

I mean, it's not something you'd get very, you know. So, I thought, how can I get a representative sample, and which would teach me something about a larger population called the parent population. So, here's a sample, it should represent the parent. So, how do I ensure that? And what I decided was to sample partners of pregnant women.

Because pregnant women all come to medical care, almost all. And if their partners will give a semen sample, then we have a representative sample. And we know what we're looking at. So, that's what we did. So, this is a, the semen study is the study of partners of pregnant women.

And of course, they'll have slightly higher semen quality 'cause they got their partner pregnant, but. And so, we had their urine, we had their blood, and we looked at their semen quality. And then we decided to look at pesticides. And the reason we look at pesticides was because there was a lot of gradation across our four centers in pesticide use.

And what we found was really extraordinary that men who were living in central Missouri, where I was living at the time, who were in the middle of a agricultural belt where there was spraying all the time for soybeans and so on. Those men had half as many moving sperm as men in Minneapolis.

- Whoa. - Whoa. Huge, right? And then we went one step further. And within Missouri, we looked at a sample of men who had very high sperm parameters and very low sperm parameters and showed that five pesticides were significantly higher in the men with the low sperm parameters. That include motility, morphology, you know.

- So, these are pesticides that are being sprayed in the air on crops. You mentioned soybeans, what other types of crops? - I don't know. I don't remember. - So, plant and fruit crops, presumably. - Yeah, whatever they were growing in Columbia, Missouri at that time. - And just to make sure I understand, it's not that the men need- - Soybeans, corn and soybeans.

- Corn and soybeans. But we're not talking about eating corn and soybeans. We're talking about living in an area where pesticides are being used by, what is it called? Is it still called dust crop? - Yeah, we didn't go into how they got these. We just looked in their urine and there were the metabolites.

The metabolites don't get in their urine unless they were exposed. - Exposed through the air or exposed by eating corn and soybeans? We don't know. - We don't know. - Okay. - We don't know. But this was not a particularly, you know, we didn't sample farmers only or anything like that.

So, whoever came into the, you remember how we got these men? Their wives were pregnant. They were having prenatal care at the University of Missouri. So, that's where we got them. Whoever happened to come in to the prenatal clinic and agreed to be in our study, their, the male, you know, urine, male's urine was measured for these pesticides.

- I'm sure a number of people, including myself, are wondering in what other products are these five pesticides present? Are these commonly used pesticides or is it something about- - They're called the triazine pesticides. Atrazine is, was the most widely used and it's a huge use around the world.

I mean, it's highly, you know, one of the most, the largest commercial pesticides. So, these were very big players in the pesticide field. (upbeat music) (upbeat music) you