HORNUNG: I would be the guy that would say what is the risk of lung cancer if you smoke so many packs of cigarettes a day for so many years of your life. That's usually where I stop because beyond that, they probably lose interest. [LAUGHTER]

GRABER: It’s a simple explanation of what seems to be a complicated job – a biostatistician. Hornung began his training with the study of math, and followed with statistics. He received a doctorate in biostatistics which brings science into the mix to pull it all together. A biostatistician, he says, uses these skills in the initial stages of the research. At that point, the aim is to help scientists choose the type of study that would best tease out the answers to their questions.

HORNUNG: There’s, oh, dozens of different types of designs that could be considered. And so, my job is to try to steer the group to a design that’s not only doable but will sort of yield the maximum amount of information regarding the hypotheses that they want to test.

GRABER: For example, for the study of lead’s effects on children's IQ, the researchers chose to conduct a longitudinal study. This means they’ll follow their subjects over a relatively long-term period, in this case, five years. Hornung also helped the scientists figure out how large the study group would need to be to ensure that a drop in IQ is actually from lead and not from some other factor in the subject’s lives – in other words, the number of study subjects that would ensure a statistically significant result. So he asked researchers what they would consider a significant drop in IQ. He took this number and the length of the study into consideration.

HORNUNG: Given those things then, there are mathematical functions that involve probabilities and things of that nature that you can use to say, well, in order to do that, we will need a certain amount of subjects in, let's say, the low and high exposed area in order to be able to detect that difference.

GRABER: As a study progresses, Hornung does preliminary analyses to check how well the data is being collected. He’s at that stage now with the study on the effects of childhood lead exposure on adult anti-social behavior.

HORNUNG We have, oh, something – more than half of the data collected. And we kind of want to take a peek and see how things are going and give us some hints as to how we should analyze the data when it's completely collected.

GRABER: At the end of the study, Hornung will look at all the information about each subject – including factors such as income and education level – and tease out the effects of lead.

HORNUNG: Then it's the job of the statistician to use mathematical or statistical models to isolate the effect that you're looking for, in this case, let's say lead on, in the case of delinquency, lead on criminal conduct and juvenile delinquency and things of that nature, and correct for socioeconomic status, for example...

GRABER (to Hornung): That sounds really hard to me.

HORNUNG: That's why they pay us a lot. [LAUGHTER] I don't want to sound like I'm bragging about our profession, but it is difficult. It takes years of training, a lot of experience, to do these sorts of things. Thankfully now with the age of incredible computer power, a lot of the analyses that we can do now, just 20-30 year ago just couldn't be done.

GRABER: Some people might think that scientists do their own analyses of the data in their research. And the truth is, most scientists do have statistical training. But multi-faceted studies that involve all the complications of human life call for the type of skill that Hornung brings to the team. Douglas Ris is the neuropsychologist who’s working with Hornung on the study of lead’s effects on anti-social behavior.

RIS: Most scientists don't have that kind of expertise that they need to handle these large data sets, these complicated data sets. The biostatistician often works behind the scenes, but their critical role is very much appreciated by the rest of the investigators, and we all know that we can't get along without it.

GRABER: And Hornung’s grasp of numbers, and of details, has served him well in other aspects of his life. He’s a good poker player. He’s also a top-notch vacation planner. He tells a story of a two-week boat trip through America’s southern riverways that eventually brought him to a dock in Florida on the Gulf of Mexico. He had to figure out when to meet up with the owner. The owner suggested that Hornung call when he got close.

HORNUNG: And I said that's okay, I'll be there at noon on a certain day, and he said yeah, right. And I pulled up to the slip at like one minute after 12 on that day [LAUGHTER]. This guy was totally amazed. [LAUGHTER]

GRABER: But Hornung acknowledges that most people wouldn’t understand why he finds his work so exciting.

HORNUNG: I say this to people who are not statisticians and they look at me like I'm crazy, but I say I'm having a lot of fun doing this. And they think fun, how could this be fun. But if you work on a lot of different studies, then you know that some simply are more interesting or more intriguing or more challenging than others. And some, indeed, answer, in my view anyway, much more important questions for our society than others.

GRABER: And right now, Hornung says, figuring out how and at what levels lead affects us is one of those particularly challenging, intriguing, and important questions that he can help answer. For Living on Earth, I’m Cynthia Graber.

CURWOOD: You’re listening to Living on Earth.

ANNOUNCER: Funding for Living on Earth comes from the World Media Foundation Environmental Information Fund. Major contributors include the Oak Foundation, supporting coverage of marine issues, and the Geraldine R. Dodge Foundation. Support also comes from NPR member stations and Bob Williams and Meg Caldwell, honoring NPR's coverage of environmental and natural resource issues, and in support of the NPR president's council. And Paul and Marcia Ginsburg, in support of excellence in public radio.