Spider bites are extraordinarily rare (and more unlikely than car crashes)

22 April 2014 by Christopher Buddle, posted in Spiders

It does not seem like this spider is a frequent biter and the reports of alleged bites in the media and from the general public greatly outnumber the reality of the situation”

That’s a quote from a recent paper published by McKeown et al., titled Verified spider bites in Oregon (USA) with the intent to assess hobo spider venom toxicity. An important word in that title is ‘intent’. The author’s original goal was to figure out, over a 3 year period, what spiders may actually bite people in Oregon, with the actual intent of assessing whether or not the hobo spider (Eratigena agrestis) is actually toxic.  The problem is that hobo spider bites were extremely rare, occurring in only 1 case out of 33 reported bites, and in that case, the symptoms were mild (pain and redness, and was resolved within 12 hours). So, although the total sample size was small, there is no evidence of hobo spiders being particularly dangerous to humans, despite a lot of mass hysteria about this species.
The (non-threatening) hobo spider. Photo by Sean McCann, reproduced here with permission)

The (non-threatening) hobo spider. Photo by Sean McCann, reproduced here with permission)

Wait a second: there were only 33 reported, verified spider bites in an entire state over a three year period!  Holy heck, the lack of data is itself an important part of the story!

Let’s think about risk a little bit: over a three year period in Oregon there were 134,417 car crashes (using data from 2009, 2010 and 2011, available here).  This clearly supports the assertion that it’s far more common to be in a car crash than it is to be bitten by a spider in Oregon, and the risk of being bitten by a hobo spider is even more remote.  To push the math a little further, on a per-person and per-day basis for Oregon, car crashes and spider bites are relatively rare events, yet it’s still about 4,000 times more likely to be in a car crash on any given day than it is to be bitten by a spider. To me, it’s astounding that we spend ANY time worrying about spider bites yet we often drive our cars feeling invincible, despite the risk.

What is especially useful with McKeown et al’s paper is that every bite incident is verified, meaning the spider was 100% implicated, and the spider itself was kept and later identified by an arachnologist. This means we can believe the data. In many cases, spiders are blamed for a ‘bite’ but it is possible and probable that the symptoms were caused by something else, hence the importance of capturing the spider in the act (see Vetter et al. for an excellent discussion of this issue).  So, if you are bitten by a spider, don’t crush it! Instead, collect the specimen and help Arachnologists and medical specialists figure out what species might bite people, and help us understand the full suite of symptoms.
Another beautiful shot of the hobo spider, by Sean McCann.

Another beautiful shot of the hobo spider, by Sean McCann.

The symptoms of the bites reported from Oregon were relatively mild, consisting in most cases of redness, pain and swelling, with more systemic symptoms occurring infrequently. The locations of the bites were, well, interesting: "The body part that was envenomated most often was the hand with nine incidents followed by the arm, leg and neck (four each), the back, finger and foot (two each) and single episodes of bites to the abdomen, ear, eyelid, face, lip and scrotum”.  The symptoms lasted, on average, about a week. However, there was much variation, many people reported their symptoms were gone in about a day, whereas some reported symptoms lasting many weeks.
Screen Shot 2014-04-22 at 11.58.50 AM

To summarize: some spider species can bite humans, but verified spiders bites are rare and spider bites can certainly be considered ‘low risk’, and even if bites do occur, the symptoms are relatively mild and go away quite quickly. I’ve written before about the extraordinary low probability of being bitten by a spider, and I’m delighted that the scientific evidence (or lack thereof!) continues to support this claim.


McKeown, Vetter & Hendrickson (2014). Verified spider bites in Oregon (USA) with the intent to assess hobo spider venom toxicity. ToxiconVetter et al. (2003). Diagnoses of brown recluse spider bites (loxoscelism) greatly outnumber actual verifications of the spider in four western American states. Toxicon 42(4): 413-418.

3 Responses to “Spider bites are extraordinarily rare (and more unlikely than car crashes)”

  1. Stephen Reply | Permalink

    Very interesting. It seems to be rooted in our minds that spiders equal danger and I wonder if there was some sort of psychological placebo effect, i.e. people experiencing worse symptoms or exaggerating their symptoms because they expected the spider to be toxic? Surprising statistics with the relative frequency of car crashes and spider bites there, but as you point out you don't often get people saying they won't go near a car in case they are involved in a car crash.

    All that being said, I feel sorry for the one bloke who got bit on the scrotum... now how would the spider have gotten there? Perhaps best not to dwell on that one two long, I think?

    • Christopher Buddle Reply | Permalink

      Stephen - thanks so much for your comment. Yes, there are deep roots around the 'danger' associated with spiders. I'm so curious about the gradient of this fear - i.e., there exists true arachnophobia, and there also exists fear because of societal (or parental?) influences. I think education can play a role in lessening the fear related to the latter, but how far along that gradient is is possible to go? Yes, OUCH re: the poor fellow who was bitten on the scrotum.

  2. Stephen Reply | Permalink

    Urmmm embarrassing typo there. I meant *too at the end, not two

Leave a Reply

− 1 = seven