Sugar‐Sweetened Beverages, Artificially Sweetened Beverages and Sugar Forms With Long‐Term Risk of Irritable Bowel Syndrome: A Large‐Scale Prospective Cohort Study
Shanshan Wu, Zhirong Yang, Si Liu, Qian Zhang, Shutian Zhang, Shengtao Zhu
TL;DR
Drinking more sugar-sweetened beverages is linked to a higher long-term risk of developing irritable bowel syndrome, while artificial sweeteners and natural juice show no significant risk.
Contribution
This study is the first large-scale prospective cohort to link sugar-sweetened beverage and added sugar intake with increased IBS risk.
Findings
Higher sugar-sweetened beverage intake is associated with a 3% increased IBS risk per 100g/day.
Added sugar consumption is linked to higher IBS risk compared to naturally occurring sugar.
Artificially sweetened beverages and natural juice show no significant association with IBS.
Abstract
We aimed to examine the prospective association of sugar‐sweetened beverages (SSB), artificially sweetened beverages (ASB), natural juice, and sugar forms with irritable bowel syndrome (IBS). Participants free of IBS, celiac disease, inflammatory bowel disease, and any cancer at recruitment were included (N = 178,711, 53.1% female). SSB, ASB, natural juice, and different sugar forms' consumption were measured via a 24‐h dietary recall questionnaire. The primary outcome was incident IBS. A Cox proportional hazard model adjusting for age, sex, BMI, Townsend deprivation index, education, ethnicity, smoking, alcohol drinking, physical activity, total energy intake, type 2 diabetes, depression, and anxiety was conducted to assess the relationship. Mean consumption of SSB, ASB, and natural juice was 90.0, 72.4, and 105.7 g/day, respectively. During a median of 11.3‐year follow‐up, 2690…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 1
FIGURE 2| Characteristic | Total ( | 0 ( | Quartile 1 ( | Quartile 2 ( | Quartile 3 ( | Quartile 4 ( |
|---|---|---|---|---|---|---|
| Age (years) | 55.81 ± 7.96 | 56.30 ± 7.83 | 56.10 ± 7.76 | 55.40 ± 8.04 | 54.63 ± 8.19 | 53.12 ± 8.20 |
| Sex | ||||||
| Male | 83,731 (46.9) | 52,326 (45.3) | 7533 (44.5) | 6980 (47.9) | 9761 (50.9) | 7131 (50.7) |
| Female | 94,980 (53.1) | 63,128 (54.7) | 9392 (55.5) | 7598 (52.1) | 9420 (49.1) | 5442 (49.3) |
| Nutrient and food intake | ||||||
| Total energy intake (KJ/day) | 8648 ± 2447 | 8425 ± 2422 | 8671 ± 2020 | 8792 ± 2296 | 9110 ± 2535 | 9786 ± 2779 |
| Total daily intake (g/day) | 3221 ± 799 | 3177 ± 786 | 3167 ± 675 | 3207 ± 750 | 3304 ± 821 | 3588 ± 972 |
| Protein (g/day) | 80.9 ± 24.5 | 80.2 ± 24.7 | 80.8 ± 19.7 | 81.1 ± 22.9 | 82.2 ± 25.5 | 85.0 ± 28.4 |
| Fat (g/day) | 73.1 ± 28.4 | 71.6 ± 28.5 | 74.3 ± 23.8 | 74.6 ± 26.8 | 76.2 ± 29.7 | 79.3 ± 31.8 |
| Carbohydrate (g/day) | 254.3 ± 78.5 | 243.3 ± 76.0 | 253.8 ± 63.4 | 262.0 ± 71.9 | 277.7 ± 79.7 | 312.0 ± 90.6 |
| Englyst fiber (g/day) | 17.8 ± 6.7 | 17.8 ± 6.8 | 17.8 ± 5.6 | 17.7 ± 6.3 | 17.7 ± 6.8 | 17.8 ± 7.3 |
| Alcohol intake (g/day) | 17.3 ± 22.2 | 18.1 ± 23.0 | 16.9 ± 19.0 | 15.9 ± 20.1 | 15.5 ± 21.1 | 14.4 ± 22.0 |
| Sugar‐sweetened beverages (g/day) | 90.0 ± 181.5 | 0.0 ± 0.0 | 73.4 ± 22.2 | 143.6 ± 19.2 | 265.0 ± 44.6 | 609.4 ± 263.7 |
| Artificially sweetened beverages (g/day) | 72.4 ± 198.3 | 68.6 ± 202.4 | 59.8 ± 144.4 | 76.4 ± 177.3 | 80.4 ± 190.2 | 107.7 ± 247.6 |
| Natural juice (g/day) | 105.7 ± 144.5 | 103.0 ± 145.3 | 112.0 ± 119.4 | 107.5 ± 131.0 | 107.8 ± 139.3 | 115.9 ± 185.2 |
| Total sugars intake (g/day) | 124.8 ± 48.7 | 115.9 ± 45.5 | 124.2 ± 38.5 | 130.4 ± 43.4 | 142.6 ± 47.7 | 172.9 ± 59.6 |
| Ethnicity | ||||||
| Non‐White | 8199 (4.6) | 4655 (4.0) | 610 (3.6) | 773 (5.3) | 1217 (6.3) | 944 (7.5) |
| White | 169,930 (95.1) | 110,439 (95.7) | 16,257 (96.1) | 13,759 (94.4) | 17,895 (93.3) | 11,580 (92.1) |
| Unknown | 582 (0.3) | 360 (0.3) | 58 (0.3) | 46 (0.3) | 69 (0.4) | 49 (0.4) |
| Education level | ||||||
| Non‐university | 100,832 (56.4) | 65,107 (56.4) | 8630 (51.0) | 8204 (56.3) | 11,206 (58.4) | 7685 (61.1) |
| University | 77,000 (43.1) | 49,767 (43.1) | 8234 (48.6) | 6313 (43.3) | 7863 (41.0) | 4823 (38.4) |
| Unknown | 879 (0.5) | 580 (0.5) | 61 (0.4) | 61 (0.4) | 112 (0.6) | 65 (0.5) |
| Townsend deprivation index | ||||||
| Mean (SD) | −1.55 (2.88) | −1.59 (2.86) | −1.65 (2.81) | −1.63 (2.83) | −1.43 (2.98) | −1.21 (3.07) |
| Q1 (≤ −3.71) | 44,592 (25.0) | 29,041 (25.2) | 4339 (25.6) | 3720 (25.5) | 4661 (24.3) | 2831 (22.5) |
| Q2 (−3.71 to −2.30) | 44,659 (25.0) | 29,099 (25.2) | 4262 (25.2) | 3651 (25.0) | 4687 (24.4) | 2960 (23.5) |
| Q3 (−2.30 to 0.11) | 44,618 (25.0) | 28,885 (25.0) | 4248 (25.1) | 3686 (25.3) | 4752 (24.8) | 3047 (24.2) |
| Q4 (> 0.11) | 44,614 (25.0) | 28,293 (24.5) | 4065 (24.0) | 3502 (24.0) | 5050 (26.3) | 3704 (29.5) |
| Unknown | 228 (0.1) | 136 (0.1) | 11 (0.1) | 19 (0.1) | 31 (0.2) | 31 (0.2) |
| Smoking status | ||||||
| Never | 101,498 (56.8) | 64,398 (55.8) | 9985 (59.0) | 8654 (59.4) | 11,233 (58.6) | 7228 (57.5) |
| Previous | 62,545 (35.0) | 41,528 (36.0) | 5824 (34.4) | 4837 (33.2) | 6336 (33.0) | 4020 (32.0) |
| Current | 14,191 (7.9) | 9234 (8.0) | 1081 (6.4) | 1037 (7.1) | 1555 (8.1) | 1284 (10.2) |
| Unknown | 477 (0.3) | 294 (0.3) | 35 (0.2) | 50 (0.3) | 57 (0.3) | 41 (0.3) |
| Alcohol drinking | ||||||
| Never | 5741 (3.2) | 3340 (2.9) | 497 (2.9) | 504 (3.5) | 804 (4.2) | 596 (4.7) |
| Previous | 5221 (2.9) | 3088 (2.7) | 467 (2.8) | 428 (2.9) | 659 (3.4) | 579 (4.6) |
| Current | 167,571 (93.8) | 108,915 (94.3) | 15,947 (94.2) | 13,628 (93.5) | 17,698 (92.3) | 11,383 (90.5) |
| Unknown | 178 (0.1) | 111 (0.1) | 14 (0.1) | 18 (0.1) | 20 (0.1) | 15 (0.1) |
| IPAQ | ||||||
| Low | 27,668 (15.5) | 17,667 (15.3) | 2632 (15.6) | 2285 (15.7) | 2989 (15.6) | 2095 (16.7) |
| Moderate | 64,193 (35.9) | 41,589 (36.0) | 6364 (37.6) | 5247 (36.0) | 6846 (35.7) | 4147 (33.0) |
| High | 60,110 (33.6) | 38,618 (33.4) | 5574 (32.9) | 4921 (33.8) | 6509 (33.9) | 4488 (35.7) |
| Unknown | 26,740 (15.0) | 17,580 (15.2) | 2355 (13.9) | 2125 (14.6) | 2837 (14.8) | 1843 (14.7) |
| BMI | ||||||
| < 18.5 kg/m2 | 915 (0.5) | 645 (0.6) | 95 (0.6) | 60 (0.4) | 74 (0.4) | 41 (0.3) |
| 18.5–24.9 kg/m2 | 63,104 (35.3) | 41,540 (36.0) | 6605 (39.0) | 5028 (34.5) | 6210 (32.4) | 3721 (29.6) |
| 25.0–29.9 kg/m2 | 75,429 (42.2) | 48,647 (42.1) | 6960 (41.1) | 6301 (43.2) | 8235 (42.9) | 5286 (42.0) |
| ≥ 30 kg/m2 | 38,765 (21.7) | 24,297 (21.0) | 3224 (19.0) | 3154 (21.6) | 4606 (24.0) | 3484 (27.7) |
| Unknown | 498 (0.3) | 325 (0.3) | 41 (0.2) | 35 (0.2) | 56 (0.3) | 41 (0.3) |
| Type 2 Diabetes | 3498 (2.0) | 2408 (2.1) | 260 (1.5) | 243 (1.7) | 323 (1.7) | 264 (2.1) |
| Depression | 12,297 (6.9) | 7810 (6.8) | 1102 (6.5) | 935 (6.4) | 1376 (7.2) | 1078 (8.6) |
| Anxiety | 5581 (3.1) | 3555 (3.1) | 507 (3.0) | 474 (3.3) | 597 (3.1) | 448 (3.6) |
| Sugar‐sweetened, artificially sweetened beverages/natural juice consumption, quartiles | 100 g/day increment |
| ||||||
|---|---|---|---|---|---|---|---|---|
| 0 | Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 |
| |||
| Sugar‐sweetened beverages | ||||||||
| No. of participants | 115,454 | 16,925 | 14,578 | 19,181 | 12,573 | 178,711 | ||
| No. of incident IBS | 1697 | 265 | 225 | 291 | 212 | 2690 | ||
| Follow‐up, person‐years | 1,282,797 | 190,553 | 162,134 | 213,406 | 139,458 | 1,988,348 | ||
| Adjusted model 1 | Reference | 1.05 (0.92–1.20) | 1.07 (0.93–1.22) | 1.07 (0.94–1.21) | 1.24 (1.07–1.43) | 0.003 | 1.03 (1.01–1.05) | 0.001 |
| Adjusted model 2 | Reference | 1.07 (0.93–1.22) | 1.07 (0.93–1.23) | 1.05 (0.93–1.19) | 1.19 (1.03–1.37) | 0.017 | 1.03 (1.01–1.05) | 0.009 |
| Adjusted model 3 | Reference | 1.08 (0.95–1.23) | 1.08 (0.94–1.24) | 1.06 (0.94–1.20) | 1.19 (1.03–1.37) | 0.017 | 1.03 (1.01–1.05) | 0.010 |
| Artificially sweetened beverages | ||||||||
| No. of participants | 141,688 | 8659 | 9401 | 9322 | 9641 | 178,711 | ||
| No. of incident IBS | 2059 | 133 | 153 | 167 | 178 | 2690 | ||
| Follow‐up, person‐years | 1,576,106 | 97,695 | 104,774 | 103,020 | 106,752 | 1,988,348 | ||
| Adjusted model 1 | Reference | 1.02 (0.86–1.22) | 1.10 (0.93–1.29) | 1.20 (1.02–1.40) | 1.24 (1.07–1.45) | 0.001 | 1.03 (1.01–1.05) | < 0.001 |
| Adjusted model 2 | Reference | 1.02 (0.85–1.21) | 1.08 (0.91–1.27) | 1.15 (0.98–1.35) | 1.16 (0.99–1.36) | 0.018 | 1.02 (1.00–1.04) | 0.015 |
| Adjusted model 3 | Reference | 1.01 (0.85–1.21) | 1.07 (0.91–1.26) | 1.14 (0.97–1.33) | 1.12 (0.95–1.31) | 0.062 | 1.02 (1.00–1.03) | 0.055 |
| Natural juice | ||||||||
| No. of participants | 86,255 | 18,204 | 32,018 | 29,198 | 13,036 | 178,711 | ||
| No. of incident IBS | 1384 | 285 | 452 | 383 | 186 | 2690 | ||
| Follow‐up, person‐years | 957,877 | 204,999 | 355,413 | 324,456 | 145,603 | 1,988,348 | ||
| Adjusted model 1 | Reference | 0.96 (0.84–1.09) | 0.90 (0.81–1.01) | 0.84 (0.75–0.95) | 0.94 (0.80–1.09) | 0.017 | 0.98 (0.95–1.01) | 0.160 |
| Adjusted model 2 | Reference | 1.00 (0.88–1.14) | 0.95 (0.85–1.06) | 0.89 (0.79–1.00) | 0.99 (0.85–1.16) | 0.193 | 0.99 (0.97–1.02) | 0.659 |
| Adjusted model 3 | Reference | 1.01 (0.89–1.15) | 0.97 (0.87–1.08) | 0.91 (0.81–1.02) | 1.01 (0.87–1.18) | 0.363 | 1.00 (0.97–1.03) | 0.951 |
- —National Key Research and Development Program of China 10.13039/501100012166
- —Beijing Nova Program 10.13039/501100005090
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGastrointestinal motility and disorders · Diet, Metabolism, and Disease · Helicobacter pylori-related gastroenterology studies
Introduction
1
Irritable bowel syndrome (IBS) is the most common gut‐brain interaction disorder, which is impaired communication between the gut and the brain via the nervous system in both directions (from gut to brain and brain to gut). It is reported that IBS affects 4.1% and 10.1% of adults worldwide according to Rome III and IV criteria (Sperber et al. 2021). It is characterized by recurrent abdominal pain, altered bowel habits, and bloating/distension owing to intestinal barrier damage and visceral hypersensitivity, accordingly resulting in substantial adverse impact on quality of life and significant economic burden (Lacy et al. 2021; Frändemark et al. 2018). Hence, it is of high priority to identify modifiable contributing factors to help develop targeted strategies on reducing IBS burden.
Interestingly, there is growing concern about the role of diet in the development and treatment of IBS. Several lines of evidence demonstrate that a diet high in fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAPs), certain short‐chain carbohydrates that the small intestine absorbs poorly, may trigger or aggravate IBS symptoms via adverse impacts on gut motility, gut microbiota, and intestinal permeability (Lacy et al. 2021; McKenzie et al. 2016). Sugar‐sweetened beverages (SSB), artificially sweetened beverages (ASB) and natural juice, as the largest sources of FODMAPs in the diet, are consumed at a high level worldwide, accompanied by urbanization. Generally, SSBs are the largest source of added sugar, including soft drinks, fruit drinks, sports drinks, and energy drinks (Malik and Hu 2022). By contrast, pure natural juice is the major source of naturally occurring sugar, instead of added sugar (Malik and Hu 2022). ASBs are marketed as healthier options with low or no calories and suggested as alternatives to SSBs, containing artificial sweeteners such as aspartame, saccharin, and sucralose, etc. Emerging data suggest a possible link between dietary sugar and IBS through plausible mechanisms, including increased mucosal inflammation, gut microflora disturbance, impaired gut motility, and increased secretion of incretin hormones and serotonin, which may further induce intestinal barrier damage and visceral hypersensitivity and eventually lead to IBS development (Malik and Hu 2022; Spencer et al. 2016; Singh and Nee 2021). However, to the best of knowledge, there is a remarkable paucity of epidemiological evidence on the relationship between SSBs, ASBs, and natural juice consumption and the risk of incident IBS. Additionally, different types of beverages, as well as sugar forms, and their link with IBS occurrence remain to be answered yet.
To address these knowledge gaps, we prospectively examined the association of sugar‐sweetened beverages, artificially‐sweetened beverages, and natural juice intake with the risk of incident IBS in a large‐scale population‐based cohort with long‐term follow‐up, in order to provide more evidence for restricting their consumption and improving the overall health of the general population.
Materials and Methods
2
Study Population
2.1
This study was based on an ongoing large‐scale prospective cohort, UK Biobank (UKB), with over 500,000 participants from 22 assessment centers across England, Wales, and Scotland being enrolled between 2006 and 2010. All participants completed baseline questionnaires with anthropometric assessments and medical conditions reports. Over 210,000 participants finished a web‐based 24 h diet recall questionnaire (Oxford WebQ) in five cycles between April 2009 and June 2012 (Sudlow et al. 2015). The UKB study was approved by the North West Multicenter Research Ethical Committee(21/NW/0157), and all participants or their proxy respondents provided written informed consent.
Participants who were free of IBS before the 24 h diet recall questionnaire with available SSB, ASB, and natural juice intake information were included. Those who had a previous cancer, inflammatory bowel disease (IBD) or celiac disease diagnosis prior to baseline were excluded. All diagnoses were identified through International Classification of Disease‐10th revision (ICD‐10) codes. In order to ensure data accuracy and reduce reporting bias, those with total energy intake > 5000 kcal or < 500 kcal (i.e., implausibly high or low energy intake) were excluded. Additionally, 10 participants who withdrew from the study was excluded. Therefore, a total of 178,711 participants (53.1% female, mean [SD] age of 55.8 [8.0] years) were included (Figure S1), which was relatively older than the whole general population in the UK.
Assessment of Baseline SSB, ASB, Natural Juice, and Different Forms of Sugar Intake
2.2
Dietary information was collected via Oxford WebQ from five cycles (cycle 1: April 2009–September 2010, cycle 2: February 2011–April 2011, cycle 3: June 2011–August 2011, cycle 4: October 2011–December 2011, cycle 5: April 2012–June 2012), in order to maximize the coverage of all seasons and days of the week. Participants were asked how many units (glasses/cans/250 mL/cartons) of beverages they drank yesterday. Beverages' intake (g/day) was generated by multiplying the number of portions consumed by the set quantity of each portion size (Piernas et al. 2021). SSB referred to fizzy sugary drinks, squash, and fruit smoothies. ASB included low‐calorie fizzy drinks, whereas natural juice referred to pure orange juice, grapefruit juice, and other 100% fruit juice. In order to ensure accuracy, an average value of SSB, ASB, and natural juice intake as well as total energy was calculated for participants who completed ≥ 2 dietary cycles (arithmetic mean value of all dietary cycles). To examine the association of SSB, ASB, and natural juice intake with incident IBS, quartiles of each beverage intake for those who consumed > 0 g/day were defined, with no intake of this beverage as the reference group. Meanwhile, the continuous variable (i.e., per 100 g increment of SSB, ASB, and natural juice intake) was also considered.
Additionally, total sugar and different forms of sugar intake (added sugar or naturally occurring sugar) were measured from all types of food intake based on the 24‐h diet recall Oxford WebQ questionnaire, including both three types of beverages (SSB, ASB, and natural juice) and all other 206 types of food. Details of each type of food and beverages were described in Perez‐Cornago et al. 2021, using the food composition table from UK Nutrient Databank (UKNDB) (Perez‐Cornago et al. 2021). Naturally occurring sugar was defined as intrinsic and milk sugar, whereas added sugar was defined as the difference between total sugar and naturally occurring sugar. Since SSB/ASB and natural juice were the major sources of added sugar and naturally occurring sugar, the estimated total sugar and different forms of sugar intake (g/day) can also indirectly reflect the consumption of overall beverages, SSB/ASB, and natural juice. Hence, we considered this as a sensitivity analysis to verify the robustness of our results, with total sugar and different forms of sugar intake as additional exposures. The risk of incident IBS associated with total sugar intake, added sugar intake, and naturally occurring sugar intake was assessed separately by adjusting multiple covariates in the following.
Ascertainment of Incident IBS
2.3
Primary endpoint was incident IBS, which was determined via ICD‐10 codes (K58). ICD‐10 code (International Classification of Diseases, Tenth Revision) is a standardized system used to code diseases and medical conditions (morbidity) data, which is owned and published by the World Health Organization (WHO). IBS diagnosis was ascertained based on linkage to primary care and/or hospital admission data obtained from Hospital Episode Statistics for England, Scottish Morbidity Record data for Scotland, and Patient Episode Database for Wales. The censoring date (i.e., the end follow‐up date of the study) was set on May 30, 2022.
Covariates
2.4
Covariates were considered based on previous epidemiological evidence, including age (continuous variable), sex (male or female), BMI (underweight, normal, overweight, obese), ethnicity (white or nonwhite), Townsend deprivation index (TDI, quartiles), education level (university or non‐university), smoking status (never, current, previous), alcohol drinking (never, current, or previous), type 2 diabetes (Yes or No), depression (Yes or No), anxiety (Yes or No), physical activity, diet pattern, and total energy intake (continuous) (Lacy et al. 2021; McKenzie et al. 2016; Malik and Hu 2022). All covariates except total energy intake were collected at baseline (2006–2010) through touch‐screen questionnaires with anthropometric assessments and medical conditions reports. We assumed those covariates were unchanged from baseline (2006–2010) to the Oxford WebQ dietary assessment (February 2011–June 2012). TDI was a measure of socioeconomic status, with a lower value indicating relative affluence. Baseline comorbidity was identified via ICD‐10 code (E11 for type 2 diabetes, F32 for depression, F40 and F41 for anxiety) through linkage to primary care and/or hospital admission data. Self‐reported physical activity was divided into three levels (high, moderate, low) based on IPAQ (International Physical Activity Questionnaire). Diet pattern was defined based on the following 7 dietary components ([1] fruits ≥ 3 servings/day; [2] vegetables ≥ 3 servings/day; [3] fish ≥ 2 servings/week; [4] processed meats ≤ 1 serving/week; [5] unprocessed red meats ≤ 1.5 servings/week; [6] whole grains ≥ 3 servings/day; [7] refined grains ≤ 1.5 servings/day) (Mozaffarian 2016). Participants who fulfilled each of the above 7 items would achieve 1 healthy diet score. Therefore, the total diet score would range from 0 to 7. A healthy diet pattern was defined as a diet score ≥ 4. Total energy intake was obtained from a web‐based 24‐h dietary recall questionnaire.
Statistical Analysis
2.5
The cumulative incidence of IBS was calculated by the Kaplan–Meier method. Due to the longitudinal time‐to‐event data obtained during a long‐term follow‐up period, the Cox proportional hazard model, the most popular model for the analysis of time‐to‐event data in the cohort design, was performed to examine the associated risk of incident IBS with SSB, ASB, and natural juice consumption, respectively. The proportional hazard assumption was ascertained and satisfied by testing linear regression of scaled Schoenfeld residuals on functions of time (p = 0.120, 0.220, 0.790 for SSB, ASB and natural juice intake, separately) (Hess 1995). The follow‐up period started from the date of the 24‐h dietary recall questionnaire to the date of the first IBS diagnosis or was censored at the end of the study (May 30, 2022), the date of death, or losttofollow‐up for participants who did not develop IBS. The median follow‐up was 11.3 years for the whole cohort. Considering the very small percentage of missing values (0.1%–0.5%) for most variables, missing indicators were used.
For quartiles and per 100 g/day increment of SSB, ASB, and natural juice consumption, three models were conducted: (1) model 1, adjusted for age and sex; (2) model 2, additionally adjusted for BMI, TDI, education level, ethnicity, smoking status, and alcohol drinking; (3) model 3, additionally adjusted for IPAQ, total energy intake, type 2 diabetes, depression, and anxiety. Furthermore, subgroup analysis was conducted to assess whether the association varied by age (< 60, ≥ 60 years), sex, BMI (≤ 25, > 25 kg/m^2^), smoking status (never, current/previous), alcohol drinking (never/previous, current) and diet pattern (unhealthy, healthy). Effect modification was detected by adding interaction terms for each stratified variable and SSB/ASB/natural juice consumption. Trend analysis was performed by using median values of each quartile for SSB/ASB/natural juice.
Furthermore, the association of total sugar, added sugar, as well as naturally occurring sugar intake and incident IBS risk was also examined via model 3, with the lowest quartile intake as the reference group. Moreover, IBS risk associated with a per 10 g/day increment of total sugar, added sugar, and naturally occurring sugar intake was investigated. Similarly, subgroup analyses by age (< 60, ≥ 60 years) and sex, as well as effect modification and trend analyses were further conducted.
Additional sensitivity analyses were conducted. Firstly, participants who had an IBS diagnosis within 1 or 2 years after the 24‐h dietary recall questionnaire were excluded respectively, in order to avoid reverse causation. Secondly, SSB and natural juice were mutually adjusted to further control for confounding (i.e., additionally adjusting natural juice when assessing the risk of IBS associated with SSB, and additionally adjusting SSB when assessing IBS risk associated with natural juice). Thirdly, to rule out the mediation effect of total energy intake, models were re‐run without adjusting total energy intake as a covariate. Fourthly, a competing risk model considering lost‐to‐follow‐up and death as competing events was conducted, since those participants might develop IBS thereafter. Fifthly, the diet pattern was additionally adjusted to further rule out the influence of the diet pattern. Sixthly, considering different energy intake for males and females, we further excluded participants with total energy intake > 4200 kcal or < 800 kcal for males and > 3500 kcal or < 500 kcal for females. Finally, quintiles of SSB, ASB, and natural juice were used in the sensitivity analyses in addition to quartiles of those beverages to further verify the robustness of the results.
A two‐tailed p value < 0.05 was considered to be statistically significant. All analyses were conducted using SAS software Version 9.4 and R version 4.0.2 (forestplot, tableone, ggplot2, and survival packages).
Patient and Public Involvement
2.6
No patients and the public were involved in the design, conduct, reporting, and dissemination plans of this research.
Results
3
Baseline Characteristics
3.1
Overall, 63,257 (35.4%), 36,483 (20.4%), and 92,456 (51.7%) participants consumed SSB, ASB, and natural juice, respectively. SSB drinkers were more likely to be male, never/previous drinkers, had a lower level of education and socioeconomic deprivation, a higher BMI, higher total sugars, total energy intake, and ASB intake (Table 1). Regarding ASB consumption, ASB drinkers were more likely to be female, had a higher level of socioeconomic deprivation and BMI, a higher proportion of prevalent diabetes and depression, a lower alcohol, total sugars, total energy intake, and natural juice intake (Table S1). Natural juice consumers were more likely to be male, never smokers, had a higher level of education and socioeconomic deprivation, a lower BMI, a lower proportion of prevalent diabetes and depression, a higher total sugars and total energy intake, and a lower ASB intake (Table S2). The mean consumption of SSB, ASB, and natural juice was 90.0, 72.4, and 105.7 g/day, respectively. A total of 2690 incident IBS cases developed during 1,988,347 person‐years of follow‐up (median: 11.3 years; interquartile range: 10.8–12.1 years).
SSB Intake and Risk of Incident IBS
3.2
The 11‐year cumulative incidence of IBS was 1.7% (95% CI: 1.4%–1.9%) in the highest quartile group of SSB intake versus 1.5% (95% CI: 1.4%–1.5%) in no SSB drinkers. Compared with no SSB intake, the highest SSB quartile was associated with a significantly increased risk of IBS (HR = 1.19, 95% CI: 1.03–1.37; p trend = 0.017) according to the fully adjusted model 3 (Table 2, Table S3). Moreover, every 100 g/day increment of SSB was associated with a 3% higher risk of IBS (HR = 1.03, 95% CI: 1.01–1.05).
In subgroup analysis, higher IBS risk associated with SSB consumption (both per 100 g/day increment and quartiles) was generally observed across those with age ≥ 60 years, female, BMI > 25 kg/m^2^, never smoking, current alcohol drinking, and unhealthy diet pattern subgroups (Figure 1A, Tables [Link], [Link]). No significant interaction across these subgroups was observed (all p interaction > 0.05).
Association of sugar‐sweetened beverages, artificially sweetened beverages, and fruit juice consumption with incident IBS by age and sex. (A) Sugar‐sweetened beverages; (B) artificially sweetened beverages; (C) natural juice. All HRs were calculated by adjusting the following covariates: age, sex, Townsend deprivation index, education level, ethnicity, smoking status, alcohol drinking, IPAQ (International Physical Activity Questionnaire), total energy intake, type 2 diabetes, depression, and anxiety. p for trend was calculated by using median value (82.5, 130, 250, and 500 g/day) of each sugar‐sweetened beverages quartile, median value (82.5, 165, 330 and 660 g/day) of each artificially sweetened beverages quartile, and median value (62.5, 125, 250 and 417 g/day) of each natural juice quartile. Test for trend of per 100 g/day increment was performed by considering intake a continuous variable. CI, confidence interval; HR, hazard ratio; IBS, irritable bowel syndrome.
ASB Intake and Risk of Incident IBS
3.3
The 11‐year cumulative incidence of IBS was 1.8% (95% CI: 1.5%–2.1%) in the highest quartile group of ASB intake versus 1.5% (95% CI: 1.4%–1.5%) in no ASB drinkers. No significant association with IBS occurrence was detected in those who consumed the highest quartile of ASB compared with no ASB intake (HR_Q4 VS no intake_ = 1.12, 95% CI: 0.95–1.31, p trend = 0.062) after multivariable adjustment (Table 2, Table S8).
Similarly, consistent findings were observed in subgroup analyses by age, sex, BMI > 25 kg/m^2^, smoking, alcohol drinking, and diet pattern (Figure 1B, Tables [Link], [Link]). No significant modification effect was detected across those subgroups (all p interaction > 0.05) except for BMI subgroups (p interaction = 0.002), with higher IBS risk in those with BMI ≤ 25 kg/m^2^ (HR_Q4 VS no intake_ = 1.72, 95% CI: 1.30–2.27).
Natural Juice Intake and Risk of Incident IBS
3.4
Regarding natural juice intake, the 11‐year cumulative incidence of IBS was 1.4% (95% CI: 1.2%–1.6%) in the highest quartile group versus 1.6% (95% CI: 1.5%–1.7%) in no drinkers. Compared with no natural juice intake, no significant association with IBS occurrence was detected in those who consumed the highest quartile (HR_Q4 VS no intake_ = 1.01, 95% CI: 0.87–1.18, p trend = 0.363) (Table 2, Table S9).
Consistent results were demonstrated in subgroup analyses by age, sex, BMI, smoking, alcohol drinking, and diet pattern (Figure 1C, Tables [Link], [Link]). No significant interaction was detected across those subgroups (all p interaction > 0.05).
Total Sugar, Added Sugar, and Naturally Occurring Sugar Intake With Risk of Incident IBS
3.5
Participants with the highest quartile of total sugar intake compared with the lowest had a higher risk of incident IBS by 22% (HR = 1.22, 95% CI: 1.08–1.39, p trend = 0.001, Figure 2A). Similarly, a 2% increased risk of IBS was observed accompanied by a per 10 g/day increment of total sugar intake (HR = 1.02, 95% CI: 1.01–1.03, p trend < 0.001). Considering the intake of sugar forms, the increased risk of IBS was detected in added sugar (HR_Q4 VS Q1_ = 1.20, 95% CI: 1.05–1.36, p trend = 0.001, Figure 2B), instead of naturally occurring sugar (HR_Q4 VS Q1_ = 0.99, 95% CI: 0.88–1.11, p trend = 0.869, Figure 2C). Consistently, a 2% higher risk of IBS was observed with a per 10 g/day increment of added sugar intake (HR = 1.02, 95% CI: 1.01–1.04, p trend < 0.001, Figure 2B), whereas a null association was detected with a per 10 g/day increment of naturally occurring sugar intake (Figure 2C).
Association between total sugar, added sugar, and naturally occurring sugar intake and incident IBS. (A) total sugar; (B) added sugar; (C) naturally occurring sugar. All HRs were adjusted by age, sex, BMI, Townsend deprivation index, education level, ethnicity, smoking status and alcohol drinking, IPAQ (International Physical Activity Questionnaire), total energy intake, type 2 diabetes, depression, and anxiety; p for trend was calculated by using median value of each total sugar quartile (74.2, 106, 133, and 177 g/day), each added sugar quartile (28.2, 49.4, 69.6, and 104 g/day), and each naturally occurring sugar quartile (30.8, 48.9, 65.0, and 90.1 g/day), and considering intake a continuous variable for per 10 g/day increment of total sugar, added sugar, and naturally occurring sugar intake. CI, confidence interval; HR, hazard ratio; IBS, irritable bowel syndrome.
Meanwhile, the evidently elevated risk of incident IBS with increasing levels of total sugar and added sugar consumption persisted across all age and sex subgroups with significant dose–response relationships (all p trend < 0.05), while a null association with naturally occurring sugars was observed among those subgroups (all p values > 0.05, Figure 2A–C).
Sensitivity Analysis
3.6
Results of sensitivity analysis for SSB, ASB, and natural juice intake, either in quartiles or per 100 g/day increment, were all consistent with principal findings when excluding incident IBS cases within 1 or 2 years after the 24‐h dietary recall questionnaire, mutually adjusting SSB and natural juice, without adjusting total energy intake, performing competing risk models, additionally adjusting dietary pattern, excluding participants with different high and low cutoff points for energy intake, or using quintiles of beverage consumption (Table S10).
Discussion
4
In this large‐scale prospective cohort study, we found higher SSB intake was associated with a greater risk of IBS. By contrast, no association was detected between consumption of ASB as well as natural juice and the risk of IBS. More specifically, consumption of added sugar, rather than naturally occurring sugar, was associated with an increased risk of IBS. Every 10 g/day increment in added sugar consumption was associated with an increased incidence of IBS by 2%.
According to the national survey data from the USA, the average intake of SSBs remained high since 2003, contributing 145 kcal per day and corresponding to 6.5%–10% of daily calories (Marriott et al. 2019; Malek et al. 2018). As an important component of the diet, SSBs have been found to have a variety of adverse effects on human health (Malik and Hu 2022; Huang et al. 2023). However, the association of SSB consumption and IBS risk has long been controversial. Recent studies conducted in Norway and India revealed greater odds of IBS in individuals with higher consumption of carbonated beverages or aerated soft drinks (Ligaarden et al. 2012; Ghoshal and Singh 2017). By contrast, a null association with IBS or functional bowel disorders was observed in the French Nutrinet‐Santé study and other surveys in China (Xu et al. 2021; Guo et al. 2015; Torres et al. 2018). Nonetheless, all these studies were cross‐sectional with prevalent IBS cases rather than a cohort design with incident IBS, thereby leading to potential bias of reverse causation. Moreover, most studies had neither a clear definition of carbonated/soft drinks (i.e., combining SSB and ASB together) nor a thorough adjustment of important confounders, resulting in masking important differences and residual confounding. All of these may explain the discrepancies between previous studies and ours. Taken together, our study addressed these knowledge gaps and provided unique epidemiological evidence on the harmful effect of SSB as well as added sugar intake for IBS development.
Several potential biological mechanisms might explain the increased IBS risk associated with higher SSB consumption. Firstly, SSB, mainly sweetened by high‐fructose corn syrup (HFCS, containing 42%–55% fructose and 45%–58% glucose) in the USA, sucrose (50% fructose and 50% glucose) in Europe, or fruit juice concentrates (mainly fructose), is the major source of FODMAP in the diet. Excessive dietary intake of these monosaccharides and disaccharides can overwhelm small intestinal absorptive capacity, leading to malabsorption. Herein, unabsorbed sugars rapidly ferment in the bacteria‐rich colon and cause gas production, bloating, and diarrhea, which may contribute to IBS pathogenesis (Lacy et al. 2021; McKenzie et al. 2016; Singh and Nee 2021). Secondly, high intake of fructose in SSBs has been demonstrated to enhance the process of hepatic de novo lipogenesis, which may subsequently promote ectopic fat deposition in the liver (Stanhope and Havel 2008). A recent umbrella review showed a 39% greater risk of non‐alcoholic fatty liver disease (NAFLD) associated with high SSB intake (Huang et al. 2023). Thus, the excess risk of IBS associated with SSB may be mediated by an increased risk of NAFLD based on prior epidemiological evidence (Wu et al. 2022). Thirdly, SSB may promote the IBS pathogenic process through modulating gut microbiota (i.e., decreased diversity and alternations of amount), increasing mucosal inflammation, and disturbing the intestinal barrier (Malik and Hu 2022; Khan et al. 2020; O'Connor et al. 2018). Additionally, as IBS has been considered a disorder of gut‐brain interaction, SSB may increase IBS risk by modulating the gut‐brain axis via several neuronal pathways, including changes in the dopaminergic system and vagal neurons (Tan et al. 2020). However, the exact underlying mechanisms remain to be fully elucidated.
In line with other studies, our study did not observe any harmful or beneficial effect of natural juice on IBS risk, although natural juice also mainly contained fructose (Ligaarden et al. 2012; Xu et al. 2021; Guo et al. 2015; Torres et al. 2018). The large Nutrinet‐Santé study reported a null association between 100% vegetable/fruit juice consumption and prevalent IBS (Torres et al. 2018). This finding may be explained by the beneficial effects of vitamins and other bioactive substances in fruit juices, which might offset the harmful consequences of rapidly absorbed monosaccharides and disaccharides (Malik and Hu 2022). However, owing to data availability and a variety of complex bioactive substances in natural juice, we could not explicitly test the direct associations between these correlated factors and incident IBS risk and verify the hypothesized mechanism. Another possible explanation may be the different roles of different sugar forms in IBS pathogenesis, since the main component of SSB and natural juice is separately added sugar and naturally occurring sugars (Gonzalez 2024; Gillespie et al. 2023). Interestingly, our findings further confirmed the higher IBS risk associated with added sugars but not naturally occurring sugars, suggesting a potential role of sugar forms rather than simply the amount. Nevertheless, the exact mechanisms involved in such roles of intrinsic and extrinsic sugars remain to be investigated in detail.
Although ASB is considered a healthier alternative owing to its few calories and no sugar, little is known regarding its effect on the risk of IBS development. Interestingly, we did not detect any increased risk of IBS occurrence associated with ASB consumption except for individuals with BMI ≤ 25 kg/m^2^. However, evidence from cellular and animal studies demonstrated that artificial sweeteners may decrease gut mobility and delay intestinal transit by stimulating gut hormones release (i.e., glucagon‐like peptide‐1, gastric inhibitory polypeptide, Peptide YY, and cholecystokinin) and serotonin (Spencer et al. 2016; Ogawa et al. 2011). Meanwhile, accumulating evidence from animal studies indicated the effect of artificial sweeteners on alterations of gut microflora, with a significant decrease of beneficial intestinal bacteria (i.e., Lactobacillus and Bifidobacterium) (Spencer et al. 2016; Suez et al. 2014). Hence, all these may adversely affect intestinal permeability and visceral sensation, which, in turn, may potentially increase the risk of incident IBS and be a potential explanation for the higher IBS risk in those with BMI ≤ 25 kg/m^2^. Nonetheless, to date, a remarkable paucity of literature addresses this issue and whether ASB could be a safe substitute remains to be further validated. Besides, different artificial sweeteners are approved to be used in different countries. Currently, the Food and Drug Administration (FDA) allows 8 kinds of sweeteners to be used in the US, including Acesulfame potassium, Advantame, Aspartame, Neotame, Saccharin, Sucralose, Luo Han Guo, and Purified stevia leaf extracts. By contrast, the UK has more sweetener options than the US. In addition to the first 6 sweeteners in the US, other 15 kinds of sweeteners are allowed to be used according to the European Food Safety Agency (EFSA). Ongoing re‐evaluation of different kinds of sweeteners should also be necessary.
To the best of our knowledge, our findings for the first time highlighted the long‐term risk of IBS associated with SSB, ASB, and natural juice consumption, based on the well‐designed prospective cohort design. Moreover, we thoroughly examined the relationship with different sugar forms by adjusting for numerous potential confounders, confirming a greater IBS risk associated with SSB and added sugar intake. Additionally, various sensitivity analyses, by accounting for reverse causation and information bias, were conducted, verifying the robustness of results.
However, several potential limitations should be considered. Firstly, SSB, ASB, and natural juice consumption were self‐reported through a validated 24 h diet recall questionnaire, which was susceptible to measurement error and recall bias. However, we calculated the average value of dietary intake from multiple cycles after excluding participants with implausible energy intake, reducing the risk of measurement error. Secondly, SSB, ASB, and natural juice consumption were only measured once at baseline, instead of through repeated measurements. Hence, the IBS risk associated with changes in those beverages consumption could not be assessed. Thirdly, owing to unavailable IBS subtype data (i.e., IBS‐Diarrhea, IBS‐Constipation, IBS‐Mixed), we only investigated the associated risk of IBS as a whole, rather than distinguishing among various IBS subtypes, which may limit the applicability of the findings to specific patient populations. Future research should explore these subtypes individually to provide a more nuanced understanding of IBS. Fourthly, incident IBS cases ascertained via only ICD‐10 code may be underestimated in this study, since some patients may not seek medical consultation. Thus, this may lead to misclassification bias. Subsequent studies should incorporate more rigorous diagnostic criteria (i.e., Rome IV criteria) to enhance the robustness of the findings. Fifthly, other lifestyle factors or dietary habits may vary during the long‐term follow‐up, which may confound the association. Sixthly, although we have carefully controlled for various potential confounders, residual confounding may still exist owing to the observational design. Seventhly, despite the relatively large‐scale population, limited statistical power owing to insufficient subgroup sample size could not be avoided, leading to false‐negative results. Eighthly, owing to the unavailable data, we could not compare the risk of incident IBS among those who drink different types of 100% fruit juice (i.e., between those who drink mostly 100% orange juice and those who drink 100% grape juice) and different types of artificial sweeteners. Ninthly, the exact biological pathways or channels for the increased IBS risk of SSBs rather than ASBs and natural juice is still unclear and remain to be investigated in detail. Finally, our study was conducted among predominantly White Europeans of relatively older age, which may limit the generalizability of our findings to other ethnic populations and different age periods.
Conclusions
5
In summary, this large population‐based prospective cohort demonstrated that higher intake of sugar‐sweetened beverages was associated with a greater risk of incident IBS, with a significant dose–response relationship. Artificially sweetened beverages and natural juice intake appeared to have no association with the incidence of IBS. Meanwhile, higher consumption of added sugar, instead of naturally occurring sugar, was associated with a higher risk of IBS. Although causality cannot be inferred due to the observational design, these findings highlight the importance of limiting SSB consumption in diets to reduce the modifiable burden of IBS. Given the increasing trend in both SSB consumption and disease burden, future health‐related policies should incorporate these findings to alleviate the adverse impact of SSB on IBS development. Further studies are required to validate these findings in diverse ethnic populations and clarify underlying biological mechanisms.
Author Contributions
S Wu and S Zhu designed the study. S Wu drafted the manuscript. S Wu analyzed the data. Z Yang and Q Zhang revised the manuscript. S Wu, Z Yang, S Liu, Q Zhang, S Zhu, and S Zhang interpreted the results, incorporated comments from the co‐authors, and finalized the manuscript. All authors approved the final version of the paper.
Ethics Statement
The UKB study was approved by the North West Multicenter Research Ethical Committee(21/NW/0157), and all participants or their proxy respondents provided written informed consent.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1.
Table S2.
Table S3.
Table S4.
Table S5.
Table S6.
Table S7.
Table S8.
Table S9.
Table S10.
Table S11.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Frändemark, Å. , H. Törnblom , S. Jakobsson , and M. Simrén . 2018. “Work Productivity and Activity Impairment in Irritable Bowel Syndrome (IBS): A Multifaceted Problem.” American Journal of Gastroenterology 113: 1540–1549. 10.1038/s 41395-018-0262-x.30254230 · doi ↗ · pubmed ↗
- 2Ghoshal, U. C. , and R. Singh . 2017. “Frequency and Risk Factors of Functional Gastro‐Intestinal Disorders in a Rural Indian Population.” Journal of Gastroenterology and Hepatology 32: 378–387. 10.1111/jgh.13465.27262283 · doi ↗ · pubmed ↗
- 3Gillespie, K. M. , E. Kemps , M. J. White , and S. E. Bartlett . 2023. “The Impact of Free Sugar on Human Health‐A Narrative Review.” Nutrients 15, no. 4: 889. 10.3390/nu 15040889.36839247 PMC 9966020 · doi ↗ · pubmed ↗
- 4Gonzalez, J. T. 2024. “Are all Sugars Equal? Role of the Food Source in Physiological Responses to Sugars With an Emphasis on Fruit and Fruit Juice.” European Journal of Nutrition 63: 1435–1451. 10.1007/s 00394-024-03365-3.38492022 PMC 11329689 · doi ↗ · pubmed ↗
- 5Guo, Y. B. , K. M. Zhuang , L. Kuang , Q. Zhan , X. F. Wang , and S. D. Liu . 2015. “Association Between Diet and Lifestyle Habits and Irritable Bowel Syndrome: A Case‐Control Study.” Gut Liver 9: 649–656. 10.5009/gnl 13437.25266811 PMC 4562783 · doi ↗ · pubmed ↗
- 6Hess, K. R. 1995. “Graphical Methods for Assessing Violations of the Proportional Hazards Assumption in Cox Regression.” Statistics in Medicine 14: 1707–1723. 10.1002/sim.4780141510.7481205 · doi ↗ · pubmed ↗
- 7Huang, Y. , Z. Chen , B. Chen , et al. 2023. “Dietary Sugar Consumption and Health: Umbrella Review.” BMJ 381: e 071609. 10.1136/bmj-2022-071609.37019448 PMC 10074550 · doi ↗ · pubmed ↗
- 8Khan, S. , S. Waliullah , V. Godfrey , et al. 2020. “Dietary Simple Sugars Alter Microbial Ecology in the Gut and Promote Colitis in Mice.” Science Translational Medicine 12: eaay 6218. 10.1126/scitranslmed.aay 6218.33115951 · doi ↗ · pubmed ↗