Whiskey’s a good medicine. It keeps your muscles tender.
—Richard Overton, 1906–2018
In his later years, Richard Overton liked to take the edge off his days with a shot of bourbon and a few puffs of a Tampa Sweet cigar, lit directly from the gas stove in his home in Austin, Texas. He insisted that he never inhaled—word to the wise. Mr. Overton, as he was known, was born during the Theodore Roosevelt administration and died in late 2018 at the age of 112.
Not to be outdone, British World War I veteran Henry Allingham attributed his own 113-year lifespan to “cigarettes, whiskey, and wild, wild women.” It’s a shame he never met the adventurous Frenchwoman Jeanne Calment, who once joked, “I’ve only ever had one wrinkle, and I’m sitting on it.” She rode her bicycle until she was 100 and kept smoking until the age of 117. Perhaps she shouldn’t have quit, because she died five years later at 122, making her the oldest person ever to have lived.
Mildred Bowers, at a comparatively youthful 106, preferred beer, cracking open a cold one every day at 4 p.m. sharp—it’s five o’clock somewhere, right? Theresa Rowley of Grand Rapids, Michigan, credited her daily Diet Coke for helping her live to the age of 104, while Ruth Benjamin of Illinois said the key to reaching her 109th birthday was her daily dose of bacon. “And potatoes, some way,” she added. They were all youngsters compared with Emma Morano of Italy, who consumed three eggs a day, two of them raw, up until her death at age 117.
If we were epidemiologists from Saturn and all we had to go on was articles about centenarians in publications like USA Today and Good Housekeeping, we might conclude that the secret to extreme longevity is the breakfast special at Denny’s, washed down with Jim Beam and a good cigar. And perhaps this is so. Another possibility is that these celebrity centenarians are messing with the rest of us. We cannot be certain, because the relevant experiment cannot be done, as much as I’d like to open JAMA and see the title “Do Cream-Filled Chocolate Doughnuts Extend Lifespan? A Randomized Clinical Trial.”
We yearn for there to be some sort of “secret” to living a longer, healthier, happier life. That desire drives our obsession with knowing the special habits and rituals of those who live longest. We are fascinated by people like Madame Calment, who seem to have escaped the gravitational pull of mortality, despite having smoked or done other naughty things throughout their lives. Was it the bike riding that saved her? Or was it something else, such as the pound of chocolate that she purportedly consumed each week?
More broadly, it’s worth asking: What do healthy centenarians actually have in common? And, more importantly, what can we learn from them—if anything? Do they really live longer because of their idiosyncratic behaviors, like drinking whiskey, or despite them? Is there some other common factor that explains their extreme longevity, or is it simply luck?
More rigorous research into large groups of centenarians has cast (further) doubt on the notion that “healthy” behaviors, which I can’t resist putting into scare quotes, are required to attain extreme longevity. According to results from a large study of Ashkenazi Jewish centenarians, run by Nir Barzilai at the Albert Einstein College of Medicine in the Bronx, centenarians are no more health-conscious than the rest of us. They may actually be worse: a large proportion of the nearly five hundred subjects in the Einstein study drank alcohol and smoked, in some cases for decades. If anything, the centenarian males in the study were less likely to have exercised regularly at age seventy than age-matched controls. And many were overweight. So much for healthy lifestyles.
Could the centenarians merely be lucky? Certainly, their age alone makes them extreme statistical outliers. As of 2021, there were just under 100,000 centenarians in the United States, according to the Census Bureau. And although their number has increased by nearly 50 percent in just two decades, the over-one-hundred age group still represents only about 0.03 percent of the population, or about 1 out of every 3,333 of us.
After ten decades of age, the air gets pretty thin, pretty quickly. Those who live to their 110th birthday qualify for the ultra-elite cadre of “supercentenarians,” the world’s smallest age group, with only about three hundred members worldwide at any given time (although the number fluctuates). Just to give you a sense of how exclusive this club is, for every supercentenarian in the world at this writing, there are about nine billionaires.
Yet nobody has come close to Madame Calment’s record. The next-longest-lived person ever recorded, Pennsylvania native Sarah Knauss, was a mere 119 when she died in 1999. Since then, the world’s oldest person has rarely exceeded the age of 117, and she is almost always female. While some individuals have claimed extremely lengthy lifespans, of 140 years or more, Calment remains the only person ever to be verified as having lived past 120, leading some researchers to speculate that that may represent the upper limit of human lifespan, programmed into our genes.
We are interested in a slightly different question: Why are some people able to just blow past the eighty-year mark, which represents the finish line for most of the rest of us? Could their exceptional longevity—and exceptional healthspan—be primarily a function of their genes?
Studies of Scandinavian twins have found that genes may be responsible for only about 20 to 30 percent of the overall variation in human lifespan. The catch is that the older you get, the more genes start to matter. For centenarians, they seem to matter a lot. Being the sister of a centenarian makes you eight times more likely to reach that age yourself, while brothers of centenarians are seventeen times as likely to celebrate their hundredth birthday, according to data from the one-thousand-subject New England Centenarian Study, which has been tracking extremely long-lived individuals since 1995 (although because these subjects grew up in the same families, with presumably similar lifestyles and habits, this finding could be due to some environmental factors as well). If you don’t happen to have centenarian siblings, the next best option is to choose long-lived parents.
This is part of why I place so much importance on taking a detailed family history from my patients: I need to know when your relatives died and why. What are your likely “icebergs,” genetically speaking? And if you do happen to have centenarians in your family tree, let me offer my congratulations. Such genes are, after all, a form of inherited luck. But in my family, you were doing well if you made it to retirement age. So if you’re like me, and most people reading this book, your genes aren’t likely to take you very far. Why should we even bother with this line of inquiry?
Because we are probing a more relevant question: Can we, through our behaviors, somehow reap the same benefits that centenarians get for “free” via their genes? Or to put it more technically, can we mimic the centenarians’ phenotype, the physical traits that enable them to resist disease and survive for so long, even if we lack their genotype? Is it possible to outlive our own life expectancy if we are smart and strategic and deliberate about it?
If the answer to this question is yes, as I believe it is, then understanding the inner workings of these actuarial lottery winners—how they achieve their extreme longevity—is a worthwhile endeavor that can inform our strategy.
When I first became interested in longevity, my greatest fear was that we would somehow figure out how to delay death without also extending the healthy period of people’s lives, à la Tithonus (and à la Medicine 2.0). My mistake was to assume that this was already the fate of the very long-lived, and that all of them are essentially condemned to spend their extra years in a nursing home or under other long-term care.
A deeper look at the data from multiple large centenarian studies worldwide reveals a more hopeful picture. It’s true that many centenarians exist in a somewhat fragile state: The overall mortality rate for Americans ages 100 and older is a staggering 36 percent, meaning that if Grandma is 101, she has about a one-in-three chance of dying in the next twelve months. Death is knocking at her door. Digging further, we find that many of the oldest old die from pneumonia and other opportunistic infections, and that a few centenarians, such as Madame Calment, really do die of what used to be called old age. But the vast majority still succumb to diseases of aging—the Horsemen—just like the rest of us.
The crucial distinction, the essential distinction, is that they tend to develop these diseases much later in life than the rest of us—if they develop them at all. We’re not talking about two or three or even five years later; we’re talking decades. According to research by Thomas Perls of Boston University and his colleagues, who run the New England Centenarian Study, one in five people in the general population will have received some type of cancer diagnosis by age seventy-two. Among centenarians, that one-in-five threshold is not reached until age one hundred, nearly three decades later. Similarly, one-quarter of the general population will have been diagnosed with clinically apparent cardiovascular disease by age seventy-five; among centenarians, that prevalence is reached only at age ninety-two. The same pattern holds for bone loss, or osteoporosis, which strikes centenarians sixteen years later than average, as well as for stroke, dementia, and hypertension: centenarians succumb to these conditions much later, if at all.
Their longevity is not merely a function of delaying disease. These people also often defy the stereotype of old age as a period of misery and decline. Perls, Barzilai, and other researchers have observed that centenarians tend to be in pretty good health overall—which, again, is not what most people expect. This doesn’t mean that everyone who lives that long will be playing golf and jumping out of airplanes, but Perls’s ninety-five-and-older study subjects scored very well on standard assessments of cognitive function and ability to perform those tasks of daily living we mentioned in chapter 3, such as cooking meals and clipping their own toenails, a seemingly simple job that becomes monumentally challenging in older age.
Curiously, despite the fact that female centenarians outnumber males by at least four to one, the men generally scored higher on both cognitive and functional tests. This might seem paradoxical at first, since women clearly live longer than men, on average. Perls believes there is a kind of selection process at work, because men are more susceptible to heart attacks and strokes beginning in middle age, while women delay their vulnerability by a decade or two and die less often from these conditions.
This tends to weed the frailer individuals out of the male population, so that only those men who are in relatively robust health even make it to their hundredth birthday, while women tend to be able to survive for longer with age-related disease and disability. Perls describes this as “a double-edged sword,” in that women live longer but tend to be in poorer health. “The men tend to be in better shape,” he has said. (The authors didn’t measure this, but my hunch is that it may have something to do with men having more muscle mass, on average, which is highly correlated to longer lifespan and better function, as we’ll discuss further in the chapters on exercise.)
But even if they are not in such great shape in their eleventh decade, these individuals have already enjoyed many extra years of healthy life compared with the rest of the population. Their healthspan, as well as their lifespan, has been extraordinarily long. What’s even more surprising is that Perls’s group has also found that the supercentenarians and the “semisupercentenarians” (ages 105 to 109) actually tend to be in even better health than garden-variety hundred-year-olds. These are the super survivors, and at those advanced ages, lifespan and healthspan are pretty much the same. As Perls and his colleagues put it in a paper title, “The Older You Get, the Healthier You Have Been.”
In mathematical terms, the centenarians’ genes have bought them a phase shift in time—that is, their entire lifespan and healthspan curve has been shifted a decade or two (or three!) to the right. Not only do they live longer, but these are people who have been healthier than their peers, and biologically younger than them, for virtually their entire lives. When they were sixty, their coronary arteries were as healthy as those of thirty-five-year-olds. At eighty-five, they likely looked and felt and functioned as if they were in their sixties. They seemed like people a generation younger than the age on their driver’s license. This is the effect that we are seeking to mimic.
Think back to the notion of the Marginal Decade and Bonus Decade that we introduced in chapter 3, and the graph of lifespan versus healthspan. Because Medicine 2.0 often drags out lifespan in the context of low healthspan, it lengthens the window of morbidity, the period of disease and disability at the end of life. People are sicker for longer before they die. Their Marginal Decade is spent largely as a patient. When centenarians die, in contrast, they have generally (though not always) been sick and/or disabled for a much shorter period of time than people who die two or three decades earlier. This is called compression of morbidity, and it basically means shrinking or shortening the period of decline at the end of life and lengthening the period of healthy life, or healthspan.
One goal of Medicine 3.0 is to help people live a life course more like the centenarians—only better. The centenarians not only live longer but live longer in a healthier state, meaning many of them get to enjoy one, or two, or even three Bonus Decades. They are often healthier at ninety than the average person in their sixties. And when they do decline, their decline is typically brief. This is what we want for ourselves: to live longer with good function and without chronic disease, and with a briefer period of morbidity at the end of our lives.
The difference is that while most centenarians seem to get their longevity and good health almost accidentally, thanks to genes and/or good luck, the rest of us must try to achieve this intentionally. Which brings us to our next two questions: How do centenarians delay or avoid chronic disease? And how can we do the same?
This is where the genes likely come in—the longevity genes that most of us don’t have because we failed to pick the right parents. But if we can identify specific genes that give centenarians their edge, perhaps we can reverse-engineer their phenotype, their effect.
This would seem to be a relatively straightforward task: sequence the genomes of a few thousand centenarians and see which individual genes or gene variants stand out as being more prevalent among this population than in the general population. Those would be your candidate genes. But when researchers did this, examining thousands of individuals via genome-wide association studies, they came up almost empty-handed. These individuals appeared to have very little in common with one another genetically. And their longevity may be due to dumb luck after all.
Why are longevity genes so elusive? And why are centenarians so rare in the first place? It comes down to natural selection.
Hold on a second, you might be saying. We’ve been taught all our lives that evolution and natural selection have relentlessly optimized us for a billion years, favoring beneficial genes and eliminating harmful ones—survival of the fittest, and all that. So why don’t we all share these pro-longevity centenarian genes, whatever they are? Why aren’t we all “fit” enough to live to be one hundred?
The short answer is that evolution doesn’t really care if we live that long. Natural selection has endowed us with genes that work beautifully to help us develop, reproduce, and then raise our offspring, and perhaps help raise our offspring’s offspring. Thus, most of us can coast into our fifth decade in relatively good shape. After that, however, things start to go sideways. The evolutionary reason for this is that after the age of reproduction, natural selection loses much of its force. Genes that prove unfavorable or even harmful in midlife and beyond are not weeded out because they have already been passed on. To pick one obvious example: the gene (or genes) responsible for male pattern baldness. When we are young, our hair is full and glorious, helping us attract mates. But natural selection does not really care whether a man in his fifties (or a woman, for that matter) has a full head of hair.
Hair loss is not terribly relevant to longevity, luckily for me. But this general phenomenon also explains why genes that might predispose someone to Alzheimer’s disease or some other illness, later in life, have not vanished from our gene pool. In short, natural selection doesn’t care if we develop Alzheimer’s disease (or baldness) in old age. It doesn’t affect our reproductive fitness. By the time dementia would appear, we have likely already handed down our genes. The same is true of genes that would accelerate our risk of heart disease or cancer in midlife. Most of us still carry these lousy genes—including some centenarians, by the way. Indeed, there is a chance that those same genes may have conferred some sort of advantage earlier in life, a phenomenon known as “antagonistic pleiotropy.”
One plausible theory holds that centenarians live so long because they also possess certain other genes that protect them from the flaws in our typical genome, by preventing or delaying cardiovascular disease and cancer and maintaining their cognitive function decades after others lose it. But even as natural selection allows harmful genes to flourish in older age, it does almost nothing to promote these more helpful longevity-promoting genes, for the reasons discussed above. Thus it appears that no two centenarians follow the exact same genetic path to reaching extreme old age. There are many ways to achieve longevity; not just one or two.
That said, a handful of potential longevity genes have emerged in various studies, and it turns out that some of them are possibly relevant to our strategy. One of the most potent individual genes yet discovered is related to cholesterol metabolism, glucose metabolism—and Alzheimer’s disease risk.
You may have heard of this gene, which is called APOE, because of its known effect on Alzheimer’s disease risk. It codes for a protein called APOE (apolipoprotein E) that is involved in cholesterol transport and processing, and it has three variants: e2, e3, and e4. Of these, e3 is the most common by far, but having one or two copies of the e4 variant seems to multiply one’s risk of developing Alzheimer’s disease by a factor of between two and twelve. This is why I test all my patients for their APOE genotype, as we’ll discuss in chapter 9.
The e2 variant of APOE, on the other hand, seems to protect its carriers against dementia—and it also turns out to be very highly associated with longevity. According to a large 2019 meta-analysis of seven separate longevity studies, with a total of nearly thirty thousand participants, people who carried at least one copy of APOE e2 (and no e4) were about 30 percent more likely to reach extreme old age (defined as ninety-seven for men, one hundred for women) than people with the standard e3/e3 combination. Meanwhile, those with two copies of e4, one from each parent, were 81 percent less likely to live that long, according to the analysis. That’s a pretty big swing.
We will explore the function of APOE in more detail in chapter 9, but it is likely relevant to our strategy on multiple levels. First and most obviously, it appears to play a role in delaying (or not delaying) the onset of Alzheimer’s disease, depending on the variant. This is likely not a coincidence, because as we’ll see, APOE plays an important role in shuttling cholesterol around the body, particularly in the brain; one’s APOE variant also has a large influence on glucose metabolism. Its potent correlation with longevity suggests that we should focus our efforts on cognitive health and pay special attention to issues around cholesterol and lipoproteins (the particles that carry cholesterol, which we’ll discuss in chapter 7), as well as glucose metabolism (chapter 6).
Researchers have identified two other cholesterol-related genes, known as CETP and APOC3, that are also correlated with extreme longevity (and may explain why centenarians rarely die from heart disease). But one individual gene, or even three dozen genes, is unlikely to be responsible for centenarians’ extreme longevity and healthspan. Broader genetic studies suggest that hundreds, if not thousands, of genes could be involved, each making its own small contribution—and that there is no such thing as a “perfect” centenarian genome.
This is actually good news for those of us without centenarians in our family tree, because it suggests that even on this genetic level there may be no magic bullet; even for centenarians, longevity may be a game of inches, where relatively small interventions, with cumulative effect, could help us replicate the centenarians’ longer lifespan and healthspan. Put another way, if we want to outlive our life expectancy and live better longer, we will have to work hard to earn it—through small, incremental changes.
One other possible longevity gene that has emerged, in multiple studies of centenarians worldwide, also provides some possible clues to inform our strategy. These are variants in a particular gene called FOXO3 that seem to be directly relevant to human longevity.
In 2008, Bradley Willcox of the University of Hawaii and colleagues reported that a genetic analysis of participants in a long-running study of health and longevity in Hawaiian men of Japanese ancestry had identified three SNPs (or variants) in FOXO3 that were strongly associated with healthy aging and longevity. Since then, several other studies have found that various other long-lived populations also appear to have FOXO3 mutations, including Californians, New Englanders, Danes, Germans, Italians, French, Chinese, and American Ashkenazi Jews—making FOXO3 one of the very few longevity-related genes to be found across multiple different ethnic groups and geographical locations.
FOXO3 belongs to a family of “transcription factors,” which regulate how other genes are expressed—meaning whether they are activated or “silenced.” I think of it as rather like the cellular maintenance department. Its responsibilities are vast, encompassing a variety of cellular repair tasks, regulating metabolism, caring for stem cells, and various other kinds of housekeeping, including helping with disposal of cellular waste or junk. But it doesn’t do the heavy lifting itself, like the mopping, the scrubbing, the minor drywall repairs, and so on. Rather, it delegates the work to other, more specialized genes—its subcontractors, if you will. When FOXO3 is activated, it in turn activates genes that generally keep our cells healthier. It seems to play an important role in preventing cells from becoming cancerous as well.
Here’s where we start to see some hope, because FOXO3 can be activated or suppressed by our own behaviors. For example, when we are slightly deprived of nutrients, or when we are exercising, FOXO3 tends to be more activated, which is what we want.
Beyond FOXO3, gene expression itself seems to play an important but still poorly understood role in longevity. A genetic analysis of Spanish centenarians found that they displayed extremely youthful patterns of gene expression, more closely resembling a control group of people in their twenties than an older control group of octogenarians. Precisely how these centenarians achieved this is not clear, but it may have something to do with FOXO3—or some other, as yet unknown, governor of gene expression.
We still have more questions than answers when it comes to the genetics behind extreme longevity, but this at least points in a more hopeful direction. While your genome is immutable, at least for the near future, gene expression can be influenced by your environment and your behaviors. For example, a 2007 study found that older people who were put on a regular exercise program shifted to a more youthful pattern of gene expression after six months. This suggests that genetics and environment both play a role in longevity and that it may be possible to implement interventions that replicate at least some of the centenarians’ good genetic luck.
I find it useful to think of centenarians as the results of a natural experiment that tells us something important about living longer and living better. Only in this case, Darwin and Mendel, the Russian geneticist, are the scientists. The experiment entails taking a random collection of human genomes and exposing them to a variety of environments and behaviors. The centenarians possess the correct combination of genome X required to survive in environment Y (perhaps with help from behaviors Z). The experiment is not simple; there are likely many pathways to longevity, genetic and otherwise.
Most of us, obviously, cannot expect to get away with some of the centenarians’ naughty behaviors, such as smoking and drinking for decades. But even if we don’t (and in many cases, shouldn’t) imitate their “tactics,” the centenarians can nevertheless help inform our strategy. Their superpower is their ability to resist or delay the onset of chronic disease by one or two or even three decades, while also maintaining relatively good healthspan.
It’s this phase shift that we want to emulate. But Medicine 2.0, which is almost solely focused on helping us live longer with disease, is not going to get us there. Its interventions almost always come too late, when disease is already established. We have to look at the other end of the time line, trying to slow or stop diseases before they start. We must focus on delaying the onset rather than extending the duration of disease—and not just one disease but all chronic diseases. Our goal is to live longer without disease.
This points to another flaw of Medicine 2.0, which is that it generally looks at these diseases as entirely separate from one another. We treat diabetes as if it were unrelated to cancer and Alzheimer’s, for example, even though it is a major risk factor for both. This disease-by-disease approach is reflected in the “silo” structure of the National Institutes of Health, with separate institutions dedicated to cancer, heart disease, and so on. We treat them as distinct when we should be looking for their commonalities.
“We’re trying to attack heart disease, cancer, stroke, and Alzheimer’s one disease at a time, as if somehow these diseases are all unrelated to each other,” says S. Jay Olshansky, who studies the demography of aging at the University of Illinois–Chicago, “when in fact the underlying risk factor for almost everything that goes wrong with us as we grow older, both in terms of diseases we experience, and of the frailty and disability associated with it, is related to the underlying biological process of aging.”
In the next chapter, we will look at one particular intervention, a drug that likely slows or delays that underlying biological process of aging at a mechanistic level. It may become relevant to our strategy as well, but for now this means pursuing two approaches in parallel. We need to think about very early disease-specific prevention, which we will explore in detail in the next few chapters dedicated to the Horsemen diseases. And we need to think about very early general prevention, targeting all the Horsemen at once, via common drivers and risk factors.
These approaches overlap, as we’ll see: reducing cardiovascular risk by targeting specific lipoproteins (cholesterol) may also reduce Alzheimer’s disease risk, for example, though not cancer. The steps we take to improve metabolic health and prevent type 2 diabetes almost certainly reduce the risk of cardiovascular disease, cancer, and Alzheimer’s simultaneously. Some types of exercise reduce risk for all chronic diseases, while others help maintain the physical and cognitive resilience that centenarians largely get via their genes. This level of prevention and intervention may seem excessive by the standards of Medicine 2.0, but I would argue that it is necessary.
In the end, I think that the centenarians’ secret comes down to one word: resilience. They are able to resist and avoid cancer and cardiovascular disease, even when they have smoked for decades. They are able to maintain ideal metabolic health, often despite a lousy diet. And they resist cognitive and physical decline long after their peers succumb. It is this resilience that we want to cultivate, just as Ali prepared himself to withstand and ultimately outlast Foreman. He prepared intelligently and thoroughly, he trained for a long time before the match, and he deployed his tactics from the opening bell. He could not have lasted forever, but he made it through enough rounds that he was able to fulfill his objective and win the fight.