Designing a Study Without a Control Group
Figuring out how to set up a control group for the study was another challenge because most of the centenarians’ peers had been dead for decades. The first centenarians in our early studies were born between 1895 and 1910, when the average life expectancy was only about forty years because so many people died from childhood diseases. The people who made it to forty had a life expectancy of a little over sixty. So the centenarians lived forty years longer than their friends, who would have been the real control group. For a controlled study, we’d need to compare the DNA of centenarians with the DNA of unrelated people who were born and shared the environment at the same time as they had but had died. Grave digging wasn’t an option, so we had to find another way.
And while we looked for it, we also grappled with what we should measure in the centenarians’ blood samples. My line of thinking was that analyzing anything but their DNA at that age might be misleading. For example, assume that we’re measuring something in the blood of centenarians and we find that they have significantly higher levels than the normal value. On one hand, it could have contributed to their longevity, but on the other hand, since any given centenarian has almost a 30 percent chance of dying in the next twelve months, the high levels could also have been predictors of death.
With those considerations, we decided to recruit not only centenarians but also one offspring of each, because the offspring have half of the longevity genes and the phenotype of their centenarian parent. So if something measures high in centenarians and it’s also high in their offspring, it probably has something to do with their longevity. Another benefit of including the offspring is that we can isolate the longevity mutations and track them as they’re passed down through the generations. But the most important reason to recruit the offspring is that while we could not create a control group for their parents, we could create one for them.
Initially, the control group was composed of the spouses of the offspring if they were Ashkenazi Jews with all four grandparents being AJs. None of the spouses in the control group had grandparents who had lived longer than age eighty-five, so we knew they didn’t have longevity in their families. Since the spouses were part of the homogeneous population, lived in the same house and community as the offspring, and had similar health habits, we determined that they were a valid control group.
Our next step was to submit our plans for the study to the Institutional Review Board (IRB), which protects the rights and welfare of people who participate in research studies. The IRB is responsible for reviewing all proposed research studies that include human participants, and it has the right to approve, disapprove, oversee, and require changes in all research that is under its jurisdiction. So we sent our study proposal to the IRB, and I explained that we would be looking at centenarians and their children. The next day, the proposal was back on my desk with a note: “For children, you need different forms.” But of course, we were talking about eighty-year-old children, and we decided that the term offspring would probably cause less confusion.
Today, the control group consists mostly of AJs who are neighbors of the centenarians’ offspring rather than spouses. We included these people in the control group because the reviewers of our grants were concerned about assortative mating, which refers to a pattern of choosing mates in which people with similar traits and habits mate more frequently than we’d expect (under a random mating pattern). An obese person is likelier to marry an obese person, a vegetarian is likelier to mate with a vegetarian, and a smoker is likelier to marry a smoker. This tendency carries over into income and education levels, too. In the United States and to a lesser extent in other countries, when you marry, you choose a lot of the health issues that you may have later. We thought mates made a better control group than neighbors, because mates share the same environment and tend to have similar habits and diets (although, granted, one could certainly eat significantly more than the other). And we argued that spouses of the centenarians’ offspring didn’t marry them based on the longevity of their parents. For one thing, their parents weren’t centenarians when the offspring married, so they didn’t have that information at the time. But we rolled with the punches and recruited neighbors in addition to the mates, and our data shows that the two control subgroups are genetically similar.
Meeting Our First AJ Centenarians and Their Offspring
I met with the first several centenarians myself, and one of the questions I asked them was whether exceptional longevity ran in their families. It turned out that it usually did, with many of them saying they had family members who had lived to be one hundred or older. This supported our theory that exceptional longevity is primarily based on genetics, and so did the centenarians themselves. When asked why they think they live longer, their number-one response was genetics. None of this surprised us, though, because Tom Perls, director of the New England Centenarian Study at Boston University, Paola Sebastiani, genetics professor, and other investigators had already shown that exceptional longevity is often inherited.
Besides talking at length with the first centenarians, I examined them and took their blood. By the time the study was officially under way, I had used that information to create a questionnaire adapted from one that had been validated in large studies, but trying to figure out what to ask before that questionnaire was in place provided some challenges. For example, colleagues and friends had all sorts of ideas about what I should ask, but an overwhelming number of them thought there was a connection between longevity and napping. So I added napping to my list of questions, and the next time I interviewed a centenarian, I asked him if he napped.
“I nap every afternoon,” he said.
Wow, I thought, we might be onto something. “Did you nap every day last year, too?”
He thought about it. “No.”
“How about the year before that?”
He shook his head. “I don’t think so. I don’t remember. But I remember taking naps the year I retired.”
If he had retired a few years before this conversation, that might have been an indication of a pattern, but he’d retired more than twenty years earlier. So while centenarians may have once had habits that contributed to their longevity, we cannot rely on their memories to be accurate, and their habits may have changed from year to year. But we can rely on our lab results and the health histories that we take when we meet each centenarian and one offspring.
Meeting the centenarians quickly became a highlight of my work. I could easily sit with them all day to hear their stories, insights, and wisdom, carried from my grandparents’ generation. Many members of that generation were Holocaust survivors, including my uncle Ervin, who suffered in six different concentration camps before World War II ended. Ever since I was a young boy, I was impressed by stories about my uncle Ervin, as well as the stories he told me about my family that my mother—another survivor—wouldn’t talk about. Now here I was listening to the stories of other people who were part of the same generation and had experienced some of the same events. They all made indelible impressions on me, but the commonalities I shared with Benjamin were especially moving. At 104, Benjamin was alert, charming, and thoughtful. He had been born in 1898 in a tiny settlement in Israel called Rishon LeZion. Settled by Russian Jews, it was the second Jewish farm settlement in Israel. When I met him in 2002 in New Jersey, he said the settlement was located not far from Sarafand al-Amar, where the British army had built the biggest transportation base in the Middle East.