Robert Carreras-Torres, PhD, on Environmental Risk Factors in IBD

The rising incidence of inflammatory bowel disease (IBD) in developing economies has heightened interest in research into possible environmental contributors to IBD. Researchers in Spain took a new approach to studying these potential factors, using Mendelian randomization (MR) to analyze genetic markers as they relate to specific risk factors.

Gastroenterology Learning Network queried lead author Robert Carreras-Torres, PhD, a research fellow at Barcelona’s Bellvitge Biomedical Research Institute (IDIBELL), about some details of his research and how the results may benefit the practice of caring for patients with IBD.


GASTROENTEROLOGY LEARNING NETWORK: Last year you and your colleagues published a study in Scientific Reports1 in which you stated that the incidence and prevalence of IBD have stabilized in the most economically developed areas, but continue to rise in developing nations. What environmental factors did you focus on in your research to identify which may be driving risk of IBD?

Dr Carreras-Torres: In 2019, the journal Gastroenterology published an umbrella review of meta-analyses of observational studies describing the relation between environmental risk factors and IBD.2 In this type of observational setting, these relations are difficult to evaluate, because multiple other factors can be involved biasing the risk estimates. In order to complement that study, we decided to perform a Mendelian randomization (MR) study where genetic markers related to identified risk factors are analyzed rather than risk factors themselves. Because germ-line genetics are not affected by environmental exposures or disease status, this type of instrumental setting can provide further evidence of potential causal effect. Therefore, we provide for first time MR results for those environmental risk factors with available data from genome-wide associated studies (GWAS). Analyzed factors included measures of tobacco smoking, obesity and fat distribution, overall physical activity, and blood metabolite levels affected by a Western type of diet, such as vitamins and fatty acids.

GLN: Your article states that the 2-sample MR approach you took in your research allows for a more accurate estimate of the contribution of a putative risk factor to the risk of IBD, and also supports “the development of preventive disease strategies.” Can you explain more about this type of approach and how it enables better estimation and disease prevention?

Dr Carreras-Torres: In a 2-sample MR approach, genetic variants strongly associated with each exposure were identified as instruments in large-scale GWAS results (first sample). Subsequently, for the identified instruments, association data on IBD and subtypes were retrieved. GWAS data from large-scale IBD international consortia comprising more than 50,000 individuals (second sample). Using large-scale GWAS data allows to triangulate the effect of the exposure on the disease risk more accurately, since estimates of genetic association with the exposures and the outcome are as well more accurate; and with more power, since genetic variants associated with the exposures capture larger fractions of exposure variability. Indeed, the MR approach has recently enabled to refine the effect estimate for body-mass index on most of obesity-related cancers. 3 In summary, an MR approach can be a cost-efficient approach to corroborate, refine, or agnostically identify new potential risk factors, which, depending upon the factor, may be further prioritized for evaluating causal validity using the gold standard in epidemiological research: the randomized controlled trial or community intervention trial. The identification of causal risk factors for disease allows for the design of prevention strategies for improving public health.

GLN: There is a long-standing hypothesis—and some research that appears to confirm it—that smoking actually may confer some benefit in preventing or delaying ulcerative colitis, but not in Crohn disease. Did your findings confirm that? What is the mechanism underlying this protective effect?

Dr Carreras-Torres: This is an interesting issue, because the association between smoking and ulcerative colitis is a good example where the MR approach does not corroborate the observational case-control studies. Our genetic smoking instruments showed a null association with ulcerative colitis risk, and showed a small risk increase for Crohn disease. Thus, it is possible that the previously observed inverse association in traditional observation studies was confounded by a third factor.

GLN: Your study looked at obesity and smoking, among others factors, not only in isolation but also in relation to each other, leading you to hypothesize that the observed association between smoking and IBD subtypes could be confounded by obesity. Can you explain what you found and why it led you to this hypothesis?

Dr Carreras-Torres: We observed that obesity body mass index showed an inverse association with ulcerative colitis risk; and in a previous MR study, a positive association between body mass index and smoking status and intensity was observed.4 Therefore, any risk factor showing these patterns of associations with smoking status and ulcerative colitis could be considered a candidate to confound observational associations between smoking status and ulcerative colitis. To elucidate the causal relation of these 2 risk factors with ulcerative colitis, a population-based prospective cohort study in combination with a multivariate MR approach would be ideal. Our group is already working on this next objective.

GLN: Your research paper noted that although omega-3 fatty acids have been associated with a reduced risk of IBD, there’s been no real evidence of their therapeutic efficacy. However, you did find some evidence that omega-3 fatty acids are protective against Crohn disease. Would you tell us more about what you found?

Dr Carreras-Torres: The potential therapeutic effect of omega-3 fatty acids on IBD has been explored in observational studies and clinical trials. The majority of studies used diet as way of delivery of omega-3 fatty acids and showed a potential protective effect; however, some studies did not show any effect. This was partly attributed to study heterogeneity, including the estimation of total intake and effective administration due to a deficient intestinal absorption. Our study encourages the performance of clinical or intervention trials that pay close attention to the administration and absorption of omega-3 fatty acids to test the therapeutic efficacy of omega-3 fatty acids on IBD, especially on Crohn disease.

GLN: Exercise and its effects on IBD were also among the factors you examined. What did you conclude about whether exercise is protective against IBD, or if it helps mitigate the disease in patients?

Dr Carreras-Torres: Physical activity provides a wide range of beneficial effects for health. Our study was statistically powered to detect only large effects (OR ~ 3.0) of physical activity on IBD risk, but not small or moderate effects. Even though we observed an inverse association between physical activity and IBD, the association was not statistically significant.

GLN: Considering the key findings of your research, what do you think practicing gastroenterologists should advise their patients with IBD—or at risk of developing IBD due to genetics and lifestyle— in regard to diet, exercise, smoking, and other controllable factors?

Dr Carreras-Torres: The main result of our study is that we did not observe a protective effect of cigarette smoking on IBD risk. Therefore, quitting smoking should be promoted for all patients who currently smoke, due to its proven harmful effects on many aspects of health, and regardless of their IBD status or risk. Vitamin B and D supplementation seem not to play a role in IBD risk. Finally, even though our study did not definitively show a protective effect, regular physical activity and a diet rich in omega-3 fatty acids could be encouraged.




  1. Carreras-Torres R, Ibáñez-Sanz G, Obón-Santacana M. et al. Identifying environmental risk factors for inflammatory bowel diseases: a Mendelian randomization study. Sci Rep. 2020;10:1-11. 19273. doi:10.1038/s41598-020-76361-2
  2. Piovani D, Danese S, Peyrin-Biroulet L, Nikolopoulous GK, Lytras T, Bonovas S. Gastroenterology. 2019;157(3):P647-659.E4. doi:10.1053/j.gastro.2019.04.016
  3. Mariosa D, Carreras-Torres R, Martin RM, Johansson M, Brennan P. Commentary: what can Mendelian randomization tell us about causes of cancer? Int J Epidemiol. 2019; 48(3):816–821. doi:10.1093/ije/dyz151
  4. Carreras-Torres R, Johansson M, Haycock PC, et al. Role of obesity in smoking behaviour: Mendelian randomisation study in UK Biobank. BMJ. 2018;361:k1767  doi:10.1136/bmj.k1767