Gluten Intake and All-Cause and Cause-Specific Mortality: Prospective Findings from the UK Biobank

Abstract

Background Gluten has been linked to adverse effects on metabolic and vascular health. Objectives The present study determines the association between dietary gluten intake and all-cause (primary objective), as well as cause-specific, mortality in people without celiac disease. Methods Gluten intake was estimated in 159,265 participants of the UK Biobank which is a large multicenter, prospective cohort study initiated in 2006. Cox proportional hazard regression models were used and HRs were determined for all-cause and cause-specific mortality. All models were adjusted for confounders and multiple testing. Results Median (IQR) age was 57 (49–62) y with 52.1% of participants being female. Gluten intake was 8.5 (5.1–12.4) g/d with significantly higher consumption in males [10.0 (6.3–14.1) g/d] than in females [7.2 (4.6–10.7) g/d] (P < 0.0001). During a median follow-up of 11.1 (10.6–11.9) y and 1.8 million person-years, 6259 deaths occurred. Gluten intake was not significantly associated with all-cause mortality after adjusting for confounders (HR: 1.00; 95% CI: 1.00, 1.01; P = 0.59). Dietary gluten was not significantly associated with cancer (HR: 1.00; 95% CI: 1.00, 1.01; raw P = 0.24) or noncancer (HR: 1.00; 95% CI: 0.99, 1.01; raw P = 0.56) mortality. However, gluten intake was positively associated with ischemic heart disease mortality (HR: 1.02; 95% CI: 1.01, 1.04; raw P = 0.003, Holm-adjusted P = 0.04). Conclusions Gluten intake is not significantly associated with all-cause and cancer mortality in adults without celiac disease. The findings support the hypothesis that limiting gluten intake is unlikely to provide significant overall survival benefits on a population level. The positive association between gluten intake and ischemic heart disease mortality requires further study.

Gluten and fructan intake and their associations with gastrointestinal symptoms in irritable bowel syndrome: A food diary study

“…In the present study, we aimed to determine the average daily intake of gluten in subjects with IBS, and to investigate the association between severity of GI symptoms and gluten intake. Our data show that the intake of gluten in IBS is similar to the reported average intake in Western countries [16,18,19]. IBS subjects that eat lower amounts of gluten were found to have more severe GI symptoms and less caloric intake.…”

“…in a geographical nearby location [19]. Furthermore, our observation that males (16.2 (11.5e18.8)) have a higher intake of gluten compared with females (10.3 (7.3e13.2)) is in line with the UK (11.7 (8.1e15.9) compared with 8.2 (5.2e11.6) g/ day respectively) [18] and Danish data (12.0 ± 4.6 compared with 9.0 ± 3.4 g/day respectively) [19], but we report a somewhat more pronounced difference. A possible factor that could explain this difference is the small sample size of males in our study, only 31 out of 147 subjects.…”

“…The comparison between our study in IBS subjects and data from the UK and Danish national dietary surveys that included random population-based samples, was surprisingly similar with regards to the total gluten intake. We demonstrate that IBS subjects in Sweden have similar gluten intake (11.0 (7.5e15.4)) compared with adults from the UK (9.7 (6.2e13.7)) [18], as well as with adult Danes (10.4 ± 4.4 g/day), i.e. in a geographical nearby location [19].…”

“…Rye, barley, and oats contain similar proteins, which are also referred to as 'gluten'. The average intake of gluten is estimated to be 5e20 g/day in Western countries [16,17], and recently, a national population-based study in the United Kingdom (UK) reported an average intake of 9.7 g/day [18]. More specifically from a Nordic point of view, a national Danish dietary populationbased study reported an average intake of 10.4 g/day in adults between 20 and 75 years of age [19].…”

Association of Antioxidants Use with All-Cause and Cause-Specific Mortality: A Prospective Study of the UK Biobank

“…Statistical analyses have been described recently [18,19]. In brief, the statistical software R version 3.6.3 [20] together with add-on packages was used to analyze the data.…”