Ultraprocessed Foods and the Risk of Inflammatory Bowel Disease: Is it Time to Modify Diet?

Narula N, Wong ECL, Dehghan M, et al. Association of ultra-processed food intake with risk of inflammatory bowel disease: prospective cohort study. BMJ 2021;374:n1554. Intake of specific dietary macronutrients and micronutrients has been implicated in modulating the risk of inflammatory bowel disease (IBD) (Nat Rev Gastroenterol Hepatol 2018;15:525–535). Recent epidemiologic and preclinical studies have additionally examined the impact of non-nutritional dietary components on IBD. Ultraprocessed foods (UPF), as defined by the NOVA classification, are foods of mainly industrial origin, often containing little or no whole foods but many ingredients, including food additives such as artificial colors and flavors, stabilizers, emulsifiers, and preservatives (Public Health Nutr 2019;22:936–941). The objective of the study by Narula et al was to examine the association between intake of UPF and the risk of developing IBD. This was an observational study based on the Prospective Urban Rural Epidemiology (PURE) cohort, a multinational cohort of adults spanning 21 low-, middle-, and high-income countries across 5 continents. Dietary data, including intake of UPF, were collected using country- specific validated food frequency questionnaires. The outcome, incident IBD (Crohn's disease [CD] or ulcerative colitis [UC]), was self-reported and collected from follow-up questionnaires. Multivariable Cox proportional hazard models were used to estimate the risk of IBD. Covariates included age, sex, geographic region, education, alcohol intake, smoking status, physical activity, energy intake, body mass index, waist-to-hip ratio, and urban vs rural location. During a median follow-up of 9.7 years, IBD developed in 467 of 116,037 participants (90 CD and 377 UC). Higher intake of UPF was associated with an increased risk of IBD. Compared with 1 serving of UPF per day, 5 or more servings per day was associated with a hazard ratio of IBD of 1.82 (95% confidence interval, 1.22–2.72). Results were similar when diet quality, measured by the Alternate Health Eating Index, (J Nutr 2012;142:1009–1018) was additionally adjusted for. When analyzed separately, higher intake of processed meat, soft drinks, refined sweetened foods, and salty foods and snacks were each associated with higher risk of IBD. The study found no association between urinary sodium levels and risk of IBD. Also, intake of white meat, unprocessed red meat, dairy, starch, or fruit, vegetables, and legumes was not associated with IBD. In contrast, fried foods were associated with higher risk of IBD. Generally, results were similar for CD and UC, with no evidence of statistical heterogeneity. In summary, this multinational, prospective cohort study reported that higher intake of UPF is associated with higher risk of IBD, while individual food categories (meats, dairy, starch, and fruit and vegetables) are not. The findings presented by Narula et al are novel and intriguing. The fact that the risk of IBD was increased in individuals consuming higher amounts of UPF and fried foods, but not affected by specific food categories (meats, dairy, starch, and fruit and vegetables), underscores that the processing of foods is likely a major contributor to any dietary associations with IBD. However, the study did not examine the impact of specific non-nutritional components in UPF. Mice studies and a study using a simulator of the human intestinal microbiota have previously demonstrated that dietary emulsifiers alter the gut microbiome and promote intestinal inflammation (Gut 2017;66:1414–1427, Cancer Res 2017;77:27–40, Nature 2015;519:92–6). Likewise, the filler and thickener maltodextrin, titanium dioxide nanoparticles, and food colorants have been shown to promote intestinal inflammation in mice models (Cell Metab 2021;33:1358–1371.e5, Cell Mol Gastroenterol Hepatol 2019;7:457–473, Gut 2017;66:1216–1224). Although preclinical studies support a biological role of specific non-nutritional components in UPF in gut inflammation, the findings in the study by Narula et al should be interpreted with caution given the observational nature of the study. A diet containing high amounts of UPF may play a causal role in the development of IBD, but the observed association could also reflect confounding (Figure 1). The authors adjusted for several potential confounders, but residual confounding and confounding by unmeasured covariates, such as environmental factors associated with industrialized society, could influence the results. Interestingly, intake of salty foods and snacks was associated with increased risk of IBD, but urine sodium levels were not. The authors suggest that this may be explained by components other than sodium in UPF. However, the conflicting findings could also reflect that dietary information from questionnaires are more prone to confounding and bias compared with measurements of biomarkers such as urine sodium. Study of exposomic and metabolomic signatures of UPF can help bypass many limitations of food questionnaires (Annu Rev Pharmacol Toxicol 2019;59:107–127). This study found that higher intake of UPF was associated with increased risk of both CD and UC. The incidence of UC was substantially higher than CD, likely reflecting a higher incidence of UC, compared with that of CD in developing countries (Lancet 2018;390:2769–2778), and resulting in broader confidence intervals for hazard ratios of CD, but nominally, the association with CD was stronger than for UC. These findings are consistent with a recently published study based on data from the Nurses' Health Study that reported an association of UPF with CD but not UC (Clin Gastroenterol Hepatol 2021; doi:10.1016/j.cgh.2021.08.031). The same group also observed that individuals consuming a diet with higher inflammatory potential are at increased risk of CD but not UC (Gastroenterology 2020;159:873–883.e1). The study by Narula et al included individuals aged between 35 and 70 years; therefore, the impact of UPF on IBD observed in this study may only be applicable to middle-aged and older-aged individuals. Whether consumption of UPF between infancy and young adulthood similarly affects IBD risk remains to be explored. Considering that the early-life period is a critical window for immune maturation (EClinicalMedicine 2021;36:100884), dietary habits during childhood extend into adulthood, and IBD affects both children and young adults, this would be relevant. Also, diet was assessed once, at study entry, precluding examination of dietary changes longitudinally. The duration of UPF consumption and lag between exposure and IBD onset are also important unanswered questions. These data raise the question: Should we advise modifying diet to decrease the consumption of UPF? We think we should. Even though these data are observational, they are consistent with those from other studies and carry a biological plausibility. Certainly, IBD is on the rise in developing countries in parallel with urbanization and availability of UPF (Lancet 2018;390:2769–2778). In addition to IBD, UPF have been implicated in other diseases, including cardiometabolic diseases and cancer (Nutrients 2020;12:1955, BMJ 2019;365:l1451, BMJ 2018;360:k322). UPF have a negative environmental impact, both due to deforestation for ingredients, such as palm and soy oils, and due to the extensive processing, packaging, and distribution, and downstream effects (Lancet Planet Health 2020;4:e437–e438). Environmental health is critical to human health, and exposures like pollutants, chemicals, and climate change can impair health substantially (Cell 2021;184:1455–1468). Climate change is a global crisis. Another important consideration is that for patients and their family, evidence-based knowledge about the impact of diet on IBD risk and disease course will be key toward diet choices and this will help empower them. However, minimizing UPF intake will come with challenges. In the United States and United Kingdom, more than half of calories consumed originates from UPF (Nutrients 2018;10:587, BMJ Open 2018;8:e020574). UPF are often cheap, easily available and do not spoil, which makes them attractive to many people. Furthermore, access to and costs of fresh foods can be prohibitive for the less privileged (Public Health Nutr 2018;21:1639–1648). A study from the United States reported higher intake of UPF among younger, less educated, and lower-income groups (BMJ Open 2018;8:e020574). Developing strategies to improve access to quality fresh foods will be pivotal in shifting food cultures. We commend the authors for this intriguing study including dietary data from low-, middle-, and high-income countries across the globe and shedding light on an important question. Their findings of an increased risk of IBD in individuals consuming higher amounts of UPF is an important step toward improved understanding of IBD pathogenesis. Future studies of specific non-nutritional components of the diet will be informative toward understanding the specific triggers that may lead to development of IBD.