Tuesday, April 26, 2011
Are Dirty Hands Good for You? – The Unwashed Evidence
I have received a number of comments on my previous post (Rally Around the Peanut?) on this blog and on other websites. My objective in my previous post was social commentary – that it is reasonable for the majority to accept a very minor inconvenience in order to enable one with a serious health condition to participate in important activities – like going to school.
I mentioned in that post that I was not specifically addressing the actual prevalence of allergies, rather discussing the situation pertaining to an individual with a known serious allergy. Some commenters brought up the prevalence issue however. In that regard it is worth citing one study (Yocum et al, Journal of Allergy and Clinical Immunology, 1999, Volume 104, 452-456) which investigated all cases of very severe allergic reactions (anaphylaxis) in Olmsted County Maryland over a period of time and reported a rate of 21 per 100,000 person-years which translates to approximately 65,000 cases per year in the US. Food allergies were responsible for 36% of these cases.
I mentioned in my previous post that one measure adopted in the Florida elementary school for the protection of the 8 year old child with a severe peanut allergy was the requirement that the other children wash their hands before entering the classroom. I suggested that this would also be beneficial for the other children - reducing transmission of infectious diseases among them. A number of commenters objected citing the “hygiene hypothesis” positing that being “too clean” is bad and leads to the development of allergies.
I think that this comment brings up an important point regarding what is the best way to make decisions. Does one rely upon hypotheses or does one rely upon evidence collected specifically to test the outcome of specific interventions? This was the topic of one of my earlier posts.
The “hygiene “hypothesis” postulates that reduced exposure of children to microorganisms and parasites increases the probability that they will develop immunologic disorders including allergic diseases. It has been used to explain the increased incidence of such diseases and the increase in asthma in developed countries compared to underdeveloped countries. There is some experimental evidence supporting this hypothesis. However, the epidemiological data are not uniformly consistent with this hypothesis. A recent Australian study (Ponsonby et al, International Journal of Epidemiology, 2008, 37, 559–569) showed a reduction in the prevalence of asthma and hay fever without evidence for a decrease in hygiene. Asthma prevalence has also been dropping in other developed countries. In addition, asthma is more prevalent in poor inner city neighborhoods in the US and these areas are unlikely to be more hygienic than the more affluent areas. In addition, improved hygiene is not the only environmental difference between developed and underdeveloped more rural countries. For example, in more developed countries people tend to live in tight buildings which are fabricated from and contain artificial materials which emit chemicals that could possibly facilitate the development of allergies.
While the hygiene hypothesis has attracted much scientific attention and is the active subject of investigation, the exact role that it plays in the development of immunological disorders is not yet established. In any case, even if it is found to play a significant role that doesn’t necessarily imply that we should abandon hand-washing.
Biology and medicine are extremely complex. The behavior of even the simplest physical systems is difficult to predict using the basic laws of physics. Human beings and microbes are fantastically more complex and we simply cannot reliably predict with certainty what the impact of a new intervention might be. Common sense, understanding of basic mechanisms, and “clinical experience” may be the basis for formulating a hypothesis of how to proceed but are far from a sufficient basis for setting policy. Whenever possible a prospective experimental test, or clinical trial, is required for that.
Taken to an extreme, based on the hygiene hypothesis one could advocate for not only eliminating hand-washing but also eliminating chlorination of the public water supply or perhaps intentionally contaminating the water supply with human excrement. This would be ridiculous. It is in fact well established that poor sanitation practices contribute to high infant and child mortality rates in underdeveloped countries.
With regard to hand-washing there have in fact been many studies that established its value. Interventional studies (for one review see, Curtis and Cairncross, Lancet Infectious Disease, 2003, Volume 3, 275–81) have shown that introduction of hand-washing reduces incidence of diarrhea by approximately 50% and thus would be projected to save approximately one million lives per year worldwide. Other studies show that introduction of hand-washing is associated with similar large reductions in respiratory and skin infections.
These observation do not necessarily mean that the hygiene hypothesis is wrong, only that the world is complex. A decrease in hand-washing increases the incidence of infectious diseases which may more than outweigh the benefit of a possible reduction in immune disorders. The hygiene hypothesis may suggest other interventions that do not increase the incidence of serious infectious diseases. For example, it has been suggested that there is benefit to acquiring a family pet in order to introduce into the home microbes which do not cause human disease.
The point here is that, whenever possible, interventions need to be prospectively tested experimentally, not just introduced on the basis of - even what appear to be very sensible - hypotheses. The world is too complex for that.
In the mean time, if you don’t wash your hands after using the restroom, please do not come shake my hand. Just wave – I will understand and appreciate the gesture.