A Disease Ecologist and His Discontents

If we save the animals, do we, in the end, save ourselves?

There is so much nature can do for us. It can clean our water. Absorb our carbon. Inspire us. Each of those benefits can be quantified, in the language of modern conservation, as an “ecosystem service,” as I describe this week in The Chronicle Review.

The stories behind some services are more alluring than others: Few people become environmentalists out of love for the lowly dung beetle and its waste management. But what if nature, at its most diverse, did something wholly unexpected for humanity? Something that appealed beyond the REI set; that touched people who rarely see more greenery than a weedy park. What if, for example, it protected us from disease?

For more than a decade, that idea has been developed by Richard S. Ostfeld, a disease ecologist at the Cary Institute of Ecosystem Studies. An expert on Lyme disease, Ostfeld has intensively studied the forested landscapes of the Northeast, where the bacterium that causes Lyme runs rampant, hosted in Ixodes scapularis, or, as it is commonly and misleadingly known, the deer tick.

Once, it was assumed that the number of Lyme-carrying ticks in a forest depended on the sheer amount of skin they could jam their saw-like mouths into; as more deer returned to the forests, ticks would thrive. But deer are unable to pass along the Lyme bacteria to ticks—in disease-ecology terms, they’re “incompetent” hosts.

Instead, Ostfeld saw that white-footed mice seemed particularly effective at spreading the Lyme bug to new ticks. Those mice, in turn, thrived in forests deprived of competitors, like squirrels, weasels, and foxes. The lower the biological diversity, the better for the mice and the disease, it seemed.

But add to the mix some chipmunks, which are less-competent Lyme carriers, and the disease’s incidence in ticks will go down, Ostfeld theorized. He called it the “dilution effect.”

Through various early field studies, it seemed as if the dilution effect held true for the Lyme system. (This work is described in David Quammen’s recent book, Spillover.) Ostfeld and his collaborators had shown that you can’t focus on just one animal as a disease source, he said. “We were among the first to impose a community perspective.”

The idea of biodiversity as a disease buffer soon got hot. Scientists gained fame for finding the dilution effect in their ecosystems; evidence emerged for West Nile and hantavirus, among others diseases. It started to seem like an ecological truism, a win-win for conservation and public health. Last year it hit the front page of The New York Times’s Sunday Review.

The dilution effect may not be universal, Ostfeld and others wrote in an influential Nature review, but biodiversity “should generally reduce the prevalence of infectious diseases.”

“A few years ago, we hit a phase where it seemed to become almost mainstream,” Ostfeld said. “And mainstream ideas become a target.”

Challenging the Truism

That’s been the case over the past two years, as several scientists have argued that the tick studies are flawed and have questioned the dilution effect. Certainly it exists in some regions, but it is rash to argue, from limited data, that it can be expected in most ecosystems.

“There’s no generality in this idea,” said Sarah E. Randolph, a retired professor of parasite ecology at the University of Oxford who spent her last working summer writing a long critique of the dilution effect, with the provocative title of “Pangloss Revisited.”

“The evidence is severely lacking,” Randolph said. Dilution may hold true in certain simple systems, but as soon as you broaden it to other tick-borne diseases, let alone other systems, “the evidence is no better or contradictory.” The mix of an ecological community is important, but it’s far from clear that biodiversity governs that mix.

Other papers followed. The first meta-analysis looking at dilution-effect data found limited evidence for the theory, though the analysis also threw out studies that depended on proxies for biodiversity, like forest-fragment size. (That included much of Ostfeld’s work.) It concluded that disease risk is more likely a local phenomenon than an ecological pattern governed by species biodiversity.

A young marine biologist, Chelsea L. Wood, came to the topic from studying disease-causing parasite systems in Pacific atolls. Some parasites depend on multiple hosts through their life cycles; a decline in species diversity hits them hard. She knew of other, similar systems. She was shocked when she talked with people at conservation groups who were starting to think of disease prevention as a reason for biodiversity.

The dilution effect, she said, “didn’t jibe at all with what I knew.”

With another author, she wrote a critique of Ostfeld’s work in Trends in Ecology and Evolution, and a tense back and forth of letters ensued. She ultimately concluded that it’s a “myth” that disease protection is a strong and general service of biodiversity.

She has followed that study with another paper, accepted for publication in Ecology, where she lays out hypotheses, based on natural history, for whether the dilution effect should exist in 69 diseases shared by human beings and animals. She expects the service to exist in 12 percent of the diseases, with higher biodiversity increasing disease rates in 38 percent of the cases. That all remains to be tested, of course.

The closest-hitting critique made its debut this week, when Maria Ana Diuk-Wasser, an epidemiologist at Yale University, released initial results from a Lyme study she has run on Block Island, off Rhode Island’s coast. The island has a low diversity of mammal hosts—essentially, white-footed mice, voles, and deer. Unlike some other studies, this one went through the painstaking work of counting species richness rather than using a proxy for it. If the dilution effect existed, Diuk-Wasser said, she expected her test area to have higher tick-infection rates than four control sites in Connecticut.

There was no difference between the sites, she said.

Tense Exchanges

Over the past two years, one thing that certainly hasn’t been diluted are the bruised feelings and tense exchanges caused by the debate.

Some of the critiques are ignorant of the literature, said Ostfeld, and create a caricature of his work. For example, years ago his team coined the term “amplification effect” for occasions when diversity might exacerbate a disease, and the term “rescue effect” for the notion that, once species richness gets low enough, some diseases, like those caused by Wood’s marine parasites, might blank out.

“The idea has never been simplistic. And it’s never been proposed to be universal,” Ostfeld said. It’s disappointing, he said, as he does welcome debate on how general the dilution effect is. So far, he added, the evidence from plant diseases has been especially strong, while animal diseases have been a bit better than mixed.

As far as anyone knows, there’s never been a conservation action that has been solely aimed at preventing disease, and that’s an appropriate caution. Even if biodiversity does generally present a disease buffer, that rule of thumb should not guide action, he said.

“If that is found to be generally true, that doesn’t by itself suggest a specific policy answer to a specific environmental problem,” Ostfeld said. “It tells you what to look for.”

The most vital work untangling the dilution effect’s future will be investigations into whether less-diverse ecosystems, as a rule, favor disease-favorable hosts like the white-footed mouse. Intricate research tracking the incidence of one disease in California frogs, out of the lab of Pieter T.J. Johnson, an associate professor of ecology and evolutionary biology at the University of Colorado at Boulder, found such a relationship this year, the disease-favorable frogs dominating as diversity declined.

As it happens, despite their differing views on how generalizable the dilution effect may be, Wood is now a postdoc in Johnson’s lab. She is planning an experiment to manipulate the abundance of birds in California systems prone to West Nile virus. It’s the type of rigorous experimental evidence the field needs.

“It would be awesome, if true, that biodiversity protected us from disease,” Wood said. “But nature is complicated.”

It now seems unlikely that a conservation group would build its actions on disease prevention as an ecosystem service. And that’s probably a good thing. Conservation science will get a thorough examination this next decade, like climate science has before it. The United Nations is now putting together the Intergovernmental Platform on Biodiversity and Ecosystem Services, modeled explicitly on its influential climate panel. Its first meeting occurred this year.

The platform will present conservationists with an unparalleled chance to influence policy. But it also presents risks, as several scientists noted last year in an editorial in Conservation Biology. “We tell compelling stories,” they wrote, but care is needed in which stories the science chooses to tell. Some, they warned, “will not stand up to careful scientific scrutiny.”

The idea that biodiversity can prevent disease is a compelling story, and it may be true for Lyme disease. It may even be more broadly true. But it’s not a story to be told just yet.