On biodiversity and disease risk

This post was originally published on BioDiverse Perspectives – a research blog aimed at fostering communication about biodiversity.

Few studies have had as large of an impact on me as Charles Mitchell’s study of the impacts of plant species diversity on fungal diseases at the Cedar Creek grassland in Minnesota, USA.

Ok; quick caveat, Charles Mitchell is my advisor. But I’m not saying this to put my advisor on a pedestal.  This study is in large part the reason that I study what I do, and that I am a graduate student where I am.  By evaluating disease impact in an experiment that directly manipulated host species diversity, Mitchell was able to provide empirical evidence that decreased host diversity should increase the abundance of many diseases. Not only did it key in on the link between biodiversity loss and health risk, but the study showed me that such a complicated question could be approached in a way that was experimentally tractable.

But I don’t want to focus on Mitchell’s research here.  See, although his study provided evidence to support the diversity-disease hypothesis, I am highlighting it here because it led to the search for general mechanisms behind that phenomenon.  Instead I want to focus on a paper that I consider a true frontier in biodiversity science. This is a paper that took an often disjointed and complicated field, grounded it in a very simple theoretical model, and then generated some clear, testable hypotheses to move the field forward.

In their 2006 paper, Effects of species diversity on disease risk, Keesing, Holt, and Ostfeld provided a synthesis that would address the key question that underlies the diversity-disease hypothesis: What is the mechanism by which biodiversity influences disease risk?  By generating 5 discrete mechanisms from a litany of previous research, they provided what would hopefully become a roadmap for future research aiming to understand and possibly mitigate for the relationship between biodiversity loss and increased disease risk.

I’m not going to get into the nitty-gritty details of this paper. Rather, I want to highlight one really cool aspect of it that I think was truly innovative and inspirational: that they take something almost immeasurably complicated (the ecology of plant and animal hosts, and the epidemiology of specialist, generalist, and vector born pathogens) and reduce it to the simplest system possible (a simple epidemiological susceptible-infected model) to identify the specific mechanisms by which diversity can influence disease risk. From this simplified model, they are then able to scale up in complexity to explain patterns observed in far more complicated systems.

So obviously, this paper is important to disease ecologists and conservationists aiming to prevent the spread and emergence of infectious diseases (not a trivial thing in and of itself). But I think this paper has value to all biodiversity researchers.  It’s so easy to get bogged down in our own subfields and forget that we can often look to other disciplines or simple theory to synthesize our own research. Keesing, Holt, and Ostfeld used a simple epidemiological model to decompose nearly 100 years of research into 5 testable hypotheses. Biodiversity, with it’s multiple dimensions, drivers, results, and feedbacks, can often seem immeasurably complicated. Is there a simple, ecological theory that can unify this field as well?

Update: The PEGE Journal Club just posted a review of a recent empirical study of biodiversity and disease risk in a trematode parasite of amphibians that was published in Nature. Pieter Johnson’s lab at CU Boulder is doing a lot of really cool research in disease ecology, and this recent paper is a great example! Here, they argue that there’s an emergent property of host diversity that can decrease disease risk that acts independent of host density.

8 Replies to “On biodiversity and disease risk”

  1. Nice post on a great paper. Ostfeld and Keesing recently published an Annual Review hitting on many of the same key points and potential mechanisms (http://www.annualreviews.org/doi/abs/10.1146/annurev-ecolsys-102710-145022). (I don’t know how to embed links).

    I still wonder why an amplification effect isn’t seen more often in nature, though the recent article on Bd and amphibian diversity gets at that (most susceptible host is often the most common as well). With so many different views on the diversity-disease paradigm (seems overly general, no?) I’m excited to see some of the other disease systems in which a dilution effect exists…and a bit more excited to see those in which an amplification effect exists.


  2. There was a fantastic empirical study of diversity and disease risk published in Nature just a few weeks ago (http://www.nature.com/nature/journal/v494/n7436/full/nature11883.html). I’m hoping to find the time to write a follow-up on this one, since it’s such a great study. But my general impression was that species-poor communities tend to be dominated by the most competent hosts, thus amplifying disease risk. I think this is probably pretty consistent across host-pathogen systems. For example, in another fantastic study, this time in a plant-disease system (reviewed here: http://briecoblog.blogspot.com/2011/12/cronin-jp-et-al-ecology-letters-2010.html by a senior undergrad student), the most competent hosts tended to exhibit “quick-return phenotypes” characteristic of many dominant plants in species-poor habitats.


  3. The species you linked to is the Bd study I referenced. Disease was amplified as biodiversity decreased, which is the same as saying that diversity reduced disease. I believe they suggest the relationship between susceptibility and commonness is immune-mediated in the sense that common organisms will invest less in immunity and more in reproduction. It just leads to more thinking, especially along the lines of what systems will yield an amplification effect (i.e. what systems is the most common host not the most susceptible?). Thanks for the link to the plant-disease paper, I’ll have to check that out.

    I don’t think that enough work has been done on different systems to even suggest that the diversity buffers disease paradigm is consistent across host-pathogen systems. As Randolph and Dobson (I think…too lazy to look it up) pointed out in their recent “Pangloss revisited…” paper in Parasitology, diversity can buffer…but also can amplify…disease, depending on host ordering (I can build a lab system right now where diversity would enhance pathogen dynamics…this is not bragging), community evenness, transmission rates (both within hosts of the same species and cross-species transmission) etc. etc.

    Long story short: great paper, super interesting hypothesis, needs a better theoretical foundation (in my opinion) and some more studies in natural systems to fill it out (this is more of a fact…it’s really only been studied to my knowledge in Lyme disease, Bd, and the plant pathogen system you linked to).


      1. Also, completely wrong. The linked Nature letter wasn’t Bd, it was a Ribeiroia species . I’ll stop commenting and correcting myself now. I was thinking of Searle 2011 in PNAS for Bd, but the whole ‘commonness-susceptibility’ thing was Johnson.


  4. Haha. There seems to be a lot of amphibian disease research out there, so I understand how easy it can be to get mixed up.

    I haven’t read the Searle 2011 article, but it looks like a good one!

    I noticed a great post on the Johnson paper this morning. Worth taking a look at and joining the conversation! (http://pegejournalclub.wordpress.com/2013/03/04/biodiversity-decreases-disease-through-predictable-changes-in-host-community-competence/)


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s