About the SA Blog Network

Lab Rat

Lab Rat

Exploring the life and times of bacteria
Lab Rat Home

Global Handwashing Day – why you probably have poo on your phone.

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Apparently yesterday was Global Handwashing Day! This is a day designed to encourage awareness of proper handwashing procedures and in the spirit of this I thought I’d take a look at a paper that came out recently about the amount of faecal matter found on mobile phones in the UK.

A cluster of E. coli

A cluster of the faecal bacteria E. coli. Photo by Eric Erbe, digital colorization by Christopher Pooley via wikimedia commons.

Carried out by the London School of Hygiene and Tropical Medicine the report measured the bacterial load of people’s hands and phones, including the amount of the faecal bacteria E. coli. E. coli is a strange little bacteria because it has two separate incarnations, good and evil. Good E. coli are the harmless little bugs that are used for general laboratory research, but some E. coli, particularly the strains naturally found in the human digestive tract, can be very dangerous if they stray from the part of the body they are meant to be found in.

The research focused on sampling from major cities, and the data for the average number of bacteria found on phones looks like an advertisement by the Southern Snob tourist board. London, Brighton, Cardiff and Exeter have the lowest, while Glasgow, Edinburgh and Newcastle show the highest bacterial load. This is the load of all bacteria though, many of which will be completely harmless, and the data for the presence of E. coli looks surprisingly different. Birmingham has the dubious distinction of winning, with 41% of sampled phones containing faecal E. coli while Glasgow (5%) and Newcastle (9%) come off relatively lightly [link to all data below the article].

Surprisingly enough the city with the lowest percentage of E. coli on phones is Carlisle, which also has a relatively low bacterial load. For those readers who don’t live in the UK, or are geographically challenged, there’s a map below:

map of the UK

Map from wikimedia, captions by me. The dots might not be in *exactly* the right places, but they are approximately correct.

Although the first response to this research is to shrink away in disgust, this isn’t really that surprising. Hands are often not washed correctly, with soap and water, nearly enough times during the day. And as humans in general are bacteria-attracting machines it is hardly surprising that 95% of people were found with bacteria on their hands. I’d be worried about the other 5% myself.  These studies always bring up the numbers, which are changed into nice headlines for the press releases, but the one question never addressed is “is this worrying.”

The press release tells us that “Every year, 3.5m children under the age of five are killed by pneumonia and diarrhoeal diseases” but I’m pretty sure those aren’t children from Glasgow. Also I know that nowadays mobile phones are all in, but how many children under five own one? Washing hands is important, vitally important if you’re a doctor, nurse, or schoolteacher, but the number of people who get E. coli infections from their mobile phones in negligible. The largest E. coli scare of recent times happened because of bean-sprouts. That would not have been prevented by anyone washing their hands (although washing the bean-sprouts would, admittedly, have helped a bit).

I’m also a bit saddened by the sample size here. The researcher’s went to “12 cities and took 390 samples” which by my mathematics works out as 32.5 samples per city. That means that not only did some poor guy travel all the way from London to Glasgow just to collect around 30 samples, it means that unless someone was carrying half a phone they didn’t even sample the same numbers per city. And thirty is not a large number. Sampling fourty random street commuters in London and twenty-five people who’ve just rushed out the loo in the train station in Birmingham would be enough to skew the odds significantly.

Washing hands with soap and water is an incredibly effective way of getting rid of bacteria, far more effective and convenient than those little alcohol gels at the hospital, or packets of tissues. On the other hand, while studies like this are good to boost awareness every now and again, they probably aren’t as panic-worthy as they sound. Yes there are bacteria on your phone. And your hands, your doorknobs, your skin, your sofas cushions and basically anything you come into contact with. But it’s all far less likely to hurt you than crossing the road, or eating undercooked meat.


Or bean-sprouts! Image from wikimedia.

The data and contact information for this research can be found here.

S.E. Gould About the Author: A biochemist with a love of microbiology, the Lab Rat enjoys exploring, reading about and writing about bacteria. Having finally managed to tear herself away from university, she now works for a small company in Cambridge where she turns data into manageable words and awesome graphs. Follow on Twitter @labratting.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 2 Comments

Add Comment
  1. 1. JDahiya 7:46 am 10/17/2011

    Please don’t sneer at the sampling. 30 samples per ‘cell’ is generally considered to be enough, since it allows you to use Z-stats rather than the more complex t-stats. All the market research you read about, for example, is done with this sample size (or, if not, it ought to, to make sense). To ensure you get at least 30 per city/cell, you would expect a slight over-sampling, particularly if it was done by more than one person. And why would you throw away an oversample if you got one? There are also well-established processes to make sure the sample is as random as possible. Hopefully, they weren’t all sampled from the sewers or, alternatively, the clean-rooms for disease control. So, it shouldn’t be all that skewed as you fear. Of course, you could always check out their raw data, if you can get at it.

    Link to this
  2. 2. S.E. Gould in reply to S.E. Gould 8:21 am 10/17/2011

    Thanks for the comment! I’m used to working with bacteria myself, so a sample size of thirty seemed bafflingly small. I also understand the explanation for the oversampling. I couldn’t find the raw data linked anywhere, but I’m sure enquiring to one of the contacts would yield helpful results.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article