In the early 1970s, Michael Lappin fell in love with a dog Americans had steadily embraced over the 20^th century: the golden retriever. As a young veterinarian working in small-town Massachusetts, he saw many of the fair-coated dogs come through the practice, and like so many others, he couldn’t resist them. There’s a reason they’re now one of the most popular breeds in the country.

“They’re always in love mode; they always do things to please you,” Lappin told me. In his early days as a vet, golden retrievers were also, he remembered, notably long-lived for large dogs: He’d see them thrive well into their teens, up to 17 years of age. They could be with families for nearly a generation at a time.

But somewhere along Lappin’s long career, he said something changed: Goldens were not living as long. He started seeing many of his golden retriever patients die of cancer before they hit 13. Many succumbed to the disease when they were even younger.

Years of anecdotal reports from other golden lovers as well as scattershot studies seemed to support the idea that something was wrong: Were the big, sweet dogs now perishing earlier than their forebears? Why?

Today, there is a consensus among veterinarians that golden retrievers have some of the highest rates of cancer of any dog breed. Perhaps, according to data spanning from the ’80s into the 2000s, the highest. But Lappin’s other observation—that golden retrievers’ lifespans have collectively and perhaps dramatically dipped—remains more contentious, years after he first started voicing his belief on a bigger stage. Across the country, veterinarians and researchers are puzzling over the question of how long these dogs live and why they die the way they do. Multiple long-term and retrospective research studies are now devoted to finding answers, including one led by the owner of a golden retriever who lived into her late teens. Lappin, now known to many as “the golden retriever guy,” has entered his own goldens into one study that has invested millions into the cause.

At stake in understanding if—and why—these dogs are dying younger is more than the health of just one beloved breed. It turns out researching the lifespan of golden retrievers can tell us a lot about our complicated relationship with dogs in general. What’s really happening may unlock a different future in how we think about our canine partners and their lives.

The first golden retrievers weren’t “fur babies.” They were hunting dogs, bred to fetch ducks and other waterfowl for the British social elite of the late 1800s. An affinity for water and enthusiasm for pleasing their owners made them particularly good at such tasks. Then, after the first World War and its border-crossing influence, the dogs began booming in popularity, and their mellow, sweet disposition was written into the “breed standard”—a set of criteria upon which judges at dog shows evaluate contestants. Later in the century, movies and shows like Homeward Bound, Air Bud, and Full House encouraged their popularity. Golden retrievers assumed their status as a member of the family.

Quality of life improved for dogs in general, said Audrey Ruple, a canine epidemiologist at the Virginia–Maryland College of Veterinary Medicine at Virginia Tech. Dogs moved from dog houses, where they were vulnerable to the elements, to the indoors. Preventative veterinary care, from vaccines to flea-and-tick medication, became the norm. Diagnostic care improved. “We now use the same equipment at veterinary hospitals as human hospitals,” Ruple said, which would have been unthinkable even a few decades ago. Today, you can buy your pooch health insurance, microchip them so they don’t get lost, or even outfit them with a doggy Fitbit. (It is unclear, let’s say, whether that last one provides measurable health benefits.)

Given these changes, Ruple is skeptical that golden retrievers are dying younger than they once were, though she has heard the claim time and again. “I say, ‘Show me the money, because I don’t believe that one tiny, eensy, little bit,’ ” she said.

Indeed, scant data exists on how long most dogs live. “There’s no canine census,” said Adam Boyko, a canine population geneticist at Cornell University. One study, published in the American Journal of Veterinary Research in 1982, does seem to suggest that Lappin is looking at the past with rose-colored glasses. Its authors analyzed the lifespans of 2,002 dogs treated between 1962 and 1976 in Boston at a referral hospital, which is where a vet sends a dog that needs additional or specialized care. Out of 33 golden retrievers in the study population, the average age at which death occurred was a mere 6.7 years of age. There’s a caveat: Referral hospitals, which are where most of the longevity data on dogs comes from, tend to treat the sickest of pets.

But unlike Ruple, Boyko does think it’s possible that golden retrievers are living shorter lives—even if those lives are comparatively plush in contrast to those of the dogs of yore. No study that he’s aware of has compared changes in the breed’s longevity over time, but declines have been documented in other breeds, like Irish wolfhounds and Doberman pinschers. According to one biologist’s analysis of owner-reported data on Doberman longevity in Russia, this breed appears to have dropped in lifespan since the early 1980s, from an average of 14 years to less than 10 years. And Boyko has an idea of what might be going on with those dogs, and by extension, golden retrievers.

To understand how dogs, with all their modern comforts and access to health care, could possibly be dying younger, it’s important to understand why dogs are prone to illnesses like cancer in the first place.

Golden retrievers emerged around the same time as the practice of modern dog breeding took hold. Humans had been shaping dog genetics since the first wolves joined us by the fireside—by raising and breeding only the most formidable pups, or those with the keenest intellect, or simply the cutest faces, we’d created different general dispositions of dogs: guard dogs, hunting dogs, lap dogs. Then, in the mid-19^th century, Darwin’s On the Origin of Species was published in Victorian England, where preoccupation with social class and “good breeding” was already a societal fixture among humans. The tome inspired the idea that breeds of dogs could be “perfected.”

Dog breeding became a fashionable hobby among Britain’s aristocracy, with dog enthusiasts gathering at shows to have their progeny evaluated for their looks, skills, and temperament. (At the same time, the dog owners were tacitly evaluated for their own social standing.) Kennel clubs formed to establish rules and regulations for these dog shows, including breed standards.

For golden retrievers, the defining characteristic was their beautiful coats (which went with the gorgeous ensembles worn by the hunting elite). The first litter of golden puppies was born after a wealthy banker’s son, Dudley Coutts Marjoribanks, First Baron Tweedmouth, came across a mutt with long, yellow fur. He bred the dog, named Nous, with his own russet-colored spaniel, Belle. He gifted the puppies to other members of the aristocracy, who continued tinkering with the new line of golden dogs. One dog was bred to an Irish setter, and its pups were bred to their canine aunts, uncles, and cousins. Later, a few dogs were sent over to Canada and the United States, and those dogs were, once again, bred with one another. This process, called line-breeding (which is really a nice way of saying “inbreeding”), ensured that the future generations of puppies all had the same distinctive characteristics.

Line-breeding dogs is very common, and it carries hazards. All living beings carry genes with harmful mutations, which they pass to their offspring. Most of the time, the descendant will inherit a working copy of the same gene from the other parent; that working gene takes over so that the harmful mutation never presents itself. But when two closely related individuals are bred together, their offspring are likelier to inherit two copies of the same mutation—say, a mutation that predisposes them to cancer—leaving them with no functional gene to step in. With selective breeding, in which dogs sharing desirable features are paired up, genetics gets even more complicated. Some genes come attached to one another, even though they code for completely different systems in the body. A boxy head, big brown eyes, or a long golden coat may be sneakily attached to a gene that regulates some aspect of cell growth. If two dogs that share the same physical trait mate, they may each be sending the same harmful mutation along for the ride in a process known as genetic hitchhiking.

When scientists study inbreeding, they use a statistic called the inbreeding coefficient, which measures the likelihood that the same variant of a particular gene—for instance, a gene that increases vulnerability to cancer—will be inherited from both sets of parents. Siblings have an inbreeding coefficient of 25 percent; this is why it would be disastrous, genetically, for siblings to have kids together. Inbreeding coefficients are commonly used by biologists to assess the health of an entire population of creatures. In human populations, an average inbreeding coefficient of 3 to 5 percent is considered unhealthy.

Studies suggest that in golden retrievers, that value, on average, hovers around 8 percent—not great. When Boyko and an international team of researchers analyzed the effects of inbreeding on longevity in golden retrievers, they found that dogs whose parents shared identical copies of the same genes lived shorter lives, on average, than those whose parents’ genes included less overlap.

The genetic mutations that erode dog lifespans can pop up seemingly out of nowhere, then spread rapidly through a population, like a spark exploding into a wildfire. Bernese mountain dogs, for instance, are plagued by a form of blood cancer called histiocytosis, said Ruple, the canine epidemiologist. In both humans and dogs, this cancer is associated with a mutation on one particular gene. While this cancer is incredibly rare in humans, 1 in 7 of these dogs dies of it. That wasn’t always the case: These gentle giants have existed for thousands of years, but it wasn’t until the 1970s that the first case of histiocytosis was described in a Bernese mountain dog. According to Ruple, it’s likely that the mutation happened in just one dog, was passed down to all of its puppies, then began causing cancer once those dogs were bred to one another.

In dogs, that can happen quickly. The average breeding-purebred male dog, called a sire, will father more than 100 puppies. That number can be much higher for particularly prolific sires—for instance, a male dog that wins a show. The tendency of one sire to spread a harmful mutation among its descendants even has a name, “the popular sire effect.” As it turns out, golden retrievers have the highest proportion of popular sires of any dog breed.

If many, many, many of the puppies are fathered by a relatively small concentration of male parents, then there is potential for a faulty gene to spread rapidly into a generation of offspring. Golden retrievers are “genetically lined up like a series of dominoes,” Lappin said.

So that’s the mechanism that could explain a precipitous drop in how long your cuddliest family member could live. But it’s not proof that the time the average golden will spend on this planet really has gotten shorter. Lappin and a cadre of canine study subjects are working on that part.

When it comes to human health, there are certain tenets that we take for granted: smoking causes lung cancer, high blood pressure puts strain on our hearts, regular exercise helps prevent a myriad of ailments. But these observations weren’t always common knowledge. In fact, we can trace them back to one decadeslong research endeavor: the Framingham Heart Study. This study, which is still ongoing, got its start in 1948 when it enrolled more than 5,000 adults in a small Massachusetts town, then followed them throughout their lives. The researchers regularly asked the participants questions about their lifestyles and tracked which participants went on to develop heart disease and which did not. Later, the study enrolled adult children of the original cohort.

Around the same time the grandchildren of the original Framingham cohort were enrolling in the study, a group of scientists and veterinarians at the Morris Animal Foundation had an idea: Why not run a similar study for dogs? They would follow thousands of pooches throughout their lives, gathering a wealth of data along the way: on their genes, on the residues of toxins in their urine, on the toys they chewed on and bowls they ate from, and even GPS data to track where dogs go.

They chose to focus on a single breed: golden retrievers. The relatively homogenous population would allow the scientists to more easily isolate lifestyle factors from genetic ones.

Starting in 2012, the Golden Retriever Lifetime Study (GRLS), as it came to be called, eventually enrolled more than 3,000 dogs. Kelly Diehl, a former veterinarian who is now a science communications officer at the Morris Animal Foundation, helped with enrollment. Again and again, devoted golden owners told her the same thing: They swore their dogs were dying sooner than they used to. “Their perception is a big driver for them to put their dogs into the study,” Diehl said. “Maybe they’re on to something, but we don’t know. That’s why we need research.”

The GRLS wouldn’t be possible without a contingent of enthusiastic veterinarians across the country who helped find and enroll participants and continue to diligently collect data. Lappin is one of those vets. When he caught wind of the GRLS, he immediately got on board, signing up 17 of his patients, including a golden retriever of his own, Isaac. (In total, Lappin owns four golden retrievers: Isaac, Emma, Lucy, and Otis.)

Each year, the owners of participating dogs each complete a questionnaire which digs into every possible factor in a pet’s health. For example: the foods the dogs eat on a regular basis (carrots are surprisingly popular), nearby sources of pollution such as highways and landfills, where dogs like to swim, sun exposure, and even how often they play fetch. Veterinarians collect samples of various sorts and send them to a lab. Both vets and owners submit reports on the pets’ behavioral and physical health, from new diagnoses to instances of aggression.

And, of course, the study records how long the pets live. While two-thirds of dogs in the study are still happily paddling around in ponds and snuggling with their owners, 804 have died. Some of those dogs were very young—the youngest only 9 months. Some died of fire, heat stroke, and being hit by cars. Others died of heart disease and infectious illness. But the overwhelming majority passed away due to cancer: a total of 600 participants.

Veterinarians and scientists collaborating on the study hope that by collecting as much data as possible on participants, researchers will discover risk factors for cancer and early death, and maybe even factors associated with longevity. So far, some trends have emerged. Preliminary data suggests there’s a link between exposure to pollution sources and certain types of lymphoma. However, it will be hard to assess patterns in longevity until all the participants have died. Then we’ll have some real answers. When the study launched, its estimated end date was 2024. The remaining participants are doing so well, it could take longer than that—potentially close to 20 years, Diehl said.

A lot of golden retrievers do die of cancer—that is clear. And that’s the primary concern of Robert Rebhun, a professor at the University of California, Davis, School of Veterinary Medicine and a cancer biologist. Rebhun was raised around goldens growing up, and lost two of them—Bourbon and Rum—to the disease when they were each around 10 years old. When he started a family of his own, it included a golden named Jessica. “Our four girls, she was their childhood dog,” Rebhun said. “She was very mischievous in her early years, loved to dig up any and all plants, counter-surf for food, and sneak out for an unsupervised run through the neighborhood when given the chance.”

He had a sense of what to expect with her lifespan—it would, he thought, be relatively short, if very sweet, the bargain most of us make when we adopt pets. But at 16, Jessica was still lively and cheerful, if a little bit more of a couch potato than she’d once been. Watching Jessica age with such pizazz made Rebhun wonder: Why do some goldens succumb to cancer so young, while others, like Jessica, continue to go on walks and chase tennis balls well into their teens? So when Jessica was 14 years old, Rebhun looked at her DNA, along with that of 300 other goldens, half of which lived more than 14 years, and half of which died before they reached 12.

His results, which are preliminary and haven’t yet been published, offer a peek into what the future of dog breeding may look like, after the tools of DNA analysis are coupled with the population-level data offered by the big studies. Rebhun said his findings suggest there’s one section of DNA that’s associated with longevity. That section has three different gene variants: “One variant appears to be associated with a longer lifespan and one associated with a shorter lifespan,” Rebhun said. Dogs that had two copies of the “bad” variant—one from each parent—were the most likely to die at a younger age. Dogs that had two copies of the “good” variant were the most likely to live into their teens. Rebhun, who told me he doesn’t fall one way or another on the question of whether goldens are really dying younger, said his next step is to replicate his work with genetic data from the Golden Retriever Lifetime Study.

Last year, at nearly 17, Jessica finally died of cancer, likely multiple different types. (Rebhun didn’t do much testing—he just recognized it was time.) Her decline was rapid, and though the cancer had been slowly growing for years, it didn’t impact her quality of life until the very end. As it turned out, she was one of the lucky goldens—she’d inherited two of the “good” variants of the gene Rebhun and his colleagues discovered.

Alas, it’s unlikely that in the future breeders will be able to produce pups that live long lives simply by selecting dogs that have the “good” variant of this gene. Longevity is much more complicated than that, Rebhun said. It’s a combination of many different genes, environmental exposures, and lifestyle factors like weight.

But Boyko and Rebhun are both hopeful that the plethora of new data on dog longevity could push dog breeding in a healthier direction. Right now, the best breeders play a kind of genetic “whack-a-mole,” Boyko said: While they do test for known genetic problems, unknown harmful genes invariably slip in whenever we select for those desirable superficial characteristics. In the future, Boyko expects that health will become a higher priority, thanks to advances in dog science. In 2018 came the Dog Aging Project, a groundbreaking study funded by the National Institutes of Health. Researchers on the project are gathering data, including full genetic sequences, from nearly 30,000 large-breed dogs. “I feel like it’s important to participate in these studies,” said Diehl, who enrolled her own 10-year-old Labrador retriever. “It makes you think about what you’re doing with your dog.”

And maybe the results will help breeders become less laser-focused on the perfect boxy head, big brown eyes, or fluffy coat. Instead, armed with data, they may start opening up their genetic pool to other breeds or dogs that have less archetypal features—a golden retriever with a deep red coat or long narrow snout, for instance—in order to water down those harmful genes. They may even be able to select for genes associated with longevity, and against genes associated with certain kinds of cancer—hardly a guarantee of a certain outcome for a given dog, but stacking the deck in their favor.

Imagine this: You bring home your golden retriever puppy. She doesn’t look exactly like the dogs that once won prizes in the ring. Maybe she has a slight curl in her tail and a large white patch on her chest. But she’s positively adorable, and you have assurances from the breeder that she has tested negative for a long list of genetic variants associated with canine cancer—knowledge we possess, in this future scenario, thanks to the studies in longevity that are currently underway. You carefully pick her food, toys, and bedding based on new veterinary recommendations from the research. As she grows older, you follow a set of exercise guidelines shown to improve life expectancy. When she gets old and develops a heart issue, your vet confidently lays out your treatment options and how likely each one is to improve her lifespan and quality of life.

She might not look exactly like the dogs your parents or grandparents had, the perfect golden hue or velvety tail. But that’s a tradeoff that recent generations of anguished dog lovers should not—and hopefully will not—hesitate to make.