Charles Ebikeme interviews Dr. Bahi Takkouche of the University of Santiago de Compostela and reviews his research on smoking and allergies, which was published earlier this year in PLOS Medicine.
Dr. Takkouche’s paper is included in the PLOS Clinical Immunology Collection, which is organized into broad categories in response to the commonly articulated request from our users that we provide more structured and efficient access to papers of interest in the PLOS corpus. The Collection has been updated today with two new sections on Immunodeficiency and Autoimmune Diseases.
Smoking is one of the most common factors when you think of allergens. Its link to a whole host of diseases within the atopic march – the cascade of allergies that starts with allergic dermatitis (such as eczema) during infancy, progresses to allergic rhinitis (for example hay fever) and food allergies, and finally exhibits with asthma – has been heavily researched. But unfortunately, no clear consensus exists as to the causes.
Research published in PLOS Medicine in March of 2014 took a wide look at the data from all published research on the theme of tobacco smoking and its relation to allergies. The sheer volume of amassed and analysed knowledge often makes it difficult for good recommendations to be made to practitioners on the risks involved with – let alone other researchers in the field wanting to see where their work fits in the landscape. A meta-analysis like this allows all studies to be assessed together, and gives a clearer picture of the research landscape.
Determining the exact role of smoking as a risk factor for allergies was the straight forward question Dr Bahi and his team set out to answer in this meta-analysis. What they were looking for was quite specific. Many studies were excluded from the analysis due to the fact that they did not provide any outcome measure or the outcome was allergy at large.
The authors identified 1457 papers from various database searches using key terms. Articles and papers published from 1966 to 2013 formed the basis of the literature search across Medline, Embase, regional bibliographies of the WHO and even personal contact with clinical researchers to highlight other key texts. The researchers identified 196 observational studies, published in 139 articles across 51 different countries that investigated the association between smoke exposure and allergic rhinitis, dermatitis, and food allergies. (Total includes 97 studies on allergic rhinitis, 91 on allergic dermatitis, and 8 on food allergies).
Link between Active Smoking and Allergic Rhinitis and Dermatitis in Children
One of the main conclusions of this meta-analysis is that in countries where smoking is a common occurrence, a significant proportion of allergic diseases can be attributed to smoking.
“14% and 13% of allergic rhinitis and allergic dermatitis, respectively, among children may be attributable to active smoking. Thus, the elimination of active smoking among children and adolescents could prevent one in seven cases of allergic rhinitis and one in eight cases of allergic dermatitis in such countries.”
Globally, smoking, especially active smoking, increases the risk of allergic diseases; slightly in adults and moderately in children and adolescents. Overall, the results of the large systematic review and analysis points to the fact that active and passive smoking are associated with a modest increase in risk for some allergic diseases.
- Active smoking is associated with increased risk of allergic dermatitis but not allergic rhinitis
- Passive smoking associated with modest increase in risk for both allergic dermatitis and rhinitis
Among children and adolescents they observed significant associations between both active and passive smoking and both rhinitis and dermatitis, while passive smoking was associated with an increased risk of food allergies.
Out of 97 studies, 34 were on active smoking and 63 centered on passive smoking. What they found was that there was no association between allergic rhinitis and active smoking when all studies were considered, although there was a slight association with second-hand smoke and passive smoking (using random effects analysis and subgroup analyses). From the data, researchers found that there is an indication of a potential lack of studies that favour a positive association of the disease with smoking — indicating a potential publication bias.
33 studies on active smoking and 58 on passive smoking revealed active smoking was significantly associated with an increased risk of allergic dermatitis. The same was true for passive smoking.
Only one study looked at the association between active smoking and food allergies, and 6 studies looked at passive smoking. Active smoking showed no significant association whereas passive smoking did.
Despite the analysis, limitations still exist. The accuracy of the studies themselves, being cross-sectional, do not allow for causal inference and can sometimes overestimate relative risks given.
Tell us a little about yourself and your research
I am a Professor at the Department of Preventive Medicine of the University of Santiago de Compostela (Northwest Spain), one of the oldest universities in Europe with more than 500 years of history. I am an Epidemiologist of Algerian origin, trained, among other places, at Harvard and a former Medical Officer at the World Health Organisation. My main interest is the epidemiologic methodology, especially that related to meta-analysis, as well as the epidemiology of “minor” diseases such as common cold, headaches, mouth ulcers, dysmenorrhea, and allergic rhinitis among others, diseases that have in common their extremely high frequency, their virtually nonexistent mortality and their high cost and impact on the quality of life.
My co-authors include Jurgita Saulyte, a PhD candidate in our department. She is from Lithuania and is working on genetic and environmental risk factors of allergic rhinitis. The paper in PLOS Medicine was in fact part of her PhD dissertation.
How did this research come about, and why did you do it?
We were interested in assessing whether one of the most common exposure factors (tobacco smoking) was related to one of the most common diseases (allergic rhinitis) and also to other allergic disease (dermatitis and food allergy). Since many studies had been carried out, it was natural to start this research by a meta-analysis. We were interested especially in knowing whether the effect of smoking on the risk of allergic rhinitis was different among children from the risk observed in adults.
What was the common knowledge on smoking and allergies before starting the study?
Mixed bag. Some studies discovered that smoking was a risk factor while others did not confirm this result. The best way to disentangle these discrepancies was to perform a meta-analysis.
This brings up the immediate question of how to go about accurately measuring exposure and outcomes. Dr Bahi Takkouche suggests the idea is to focus on carrying out studies that are bias-free.
Publication bias occurs when editors reject some studies because these studies do not show an effect of the risk factor. Apparently, publication bias was not an important issue in our meta-analysis. It’s no use spending huge amounts of money to carry out cross-sectional studies. It is better to have one good study than 10 bad ones.
How do you envision any clinical translations of the evidence your paper sets out?
Tobacco is already known as the cause of many diseases. People should know that it is also a risk factor of allergic diseases, especially among children. I believe that efforts should be made to reduce the risk of exposure to tobacco among children worldwide, especially in the non industrialized countries, where tobacco consumption is frequent among children and adolescents.
Background on “the Atopic March”
The human immune system is extremely complex, with many different types of cells that exist to perform different functions with the one goal of protecting the body from invaders. With monocytes, mast cells, neutrophils, macrophages, helper cells, T-cells, white blood cells, lymphocytes, B-cells, guards, soldiers, weapons factories, communicators, and intelligence — the force of the defending army is quite empowering. Every once in a while that overwhelming force is unleashed on something as benign as pollen or dust, and ends up with an allergic reaction.There is a growing consensus that there is a certain sequence and inevitability to the cascade of allergies we all seem to get from an early age. It starts with allergic dermatitis (such as eczema) during infancy, progresses to allergic rhinitis (for example hay fever) and food allergies, and finally exhibits with asthma. Like dominoes in a line, all these diseases are part of a continuous state, and the same cascade — known as the “atopic march.”
Not only are certain allergies and allergic reactions very common, they seem to be on the rise — an inexplicable and seemingly unstoppable march of allergic diseases. Allergic rhinitis, dermatitis, asthma, and food allergies are all household afflictions worldwide. And medical practitioners encounter them on a daily basis. Rhinitis affects up to 20% of the population in Europe and the United States — with a significant proportion of these being children. Allergic dermatitis alone can affect up to 20% in many places. Food allergies can be anywhere between 3% and 35% in prevalence.
Charles Ebikeme is a science journalist with a PhD in Parasitology, who serves as a Science Officer with the International Social Science Council of UNESCO. He writes frequently on global health, health policy, neglected tropical diseases and infectious diseases for The Huffington Post, The Guardian, Scientific American, and Think Africa Press. He is based in Paris. You can find him on Twitter @CEbikeme