Common dietary approaches used to reduce symptoms of IBD

Selected studies include a variety of dietary approaches. The results will be grouped based on dietary intervention similarities (Table 2). For example, seven of the research teams used dietary approaches that focused on eliminating foods from the diet, such as the low FODMAP diet, a highly restricted organic diet, and the immunoglobulin-G (IgG) diet, which requires testing an individual’s serum for a response of IgG antibodies after exposure to certain food items.^{(Reference Bodini, Zanella and Crespi14–Reference Bartel, Weiss and Turetschek20)} Other approaches, such as the Anti-Inflammatory Diet (AID), Autoimmune Protocol Diet (AIP), Dietary Modification Framework (DMF), and Dietary Modification Program (DMP) focused on eliminating specific foods associated with worsening symptoms while increasing intake of foods that may reduce GI inflammation, symptoms, and dysbiosis.^{(Reference Chandrasekaran, Groven and Lewis21–Reference Keshteli, Valcheva and Nickurak24)} Although vegetarian or vegan dietary patterns were not identified specifically, the last group of dietary approaches used focused on specific foods or food groups that may reduce inflammation such as the Mediterranean diet (MD) and high fibre diets that focus on including foods or food groups that promote intake of anti-inflammatory foods, such as fruits and vegetables, and improved gut health.^{(Reference Lewis, Sandler and Brotherton25–Reference Chicco, Magrì and Cingolani28)} Refer to Table 2 for descriptions of these diets.

Table 2.

Description of dietary approaches used across included studies

FODMAP, fermentable oligosaccharides, disaccharides, monosaccharides, and polyols.

IgG, Immunoglobulin G.

Measures of quality of life use in studies

Each study met inclusion criteria by exploring QOL as either a primary or secondary outcome measure. This was observed using validated QOL measures, most of which were specific to IBD (Table 3). For example, the Inflammatory Bowel Disease Questionnaire (IBDQ) is a validated, 32-item questionnaire that assesses the disease-related QOL for individuals with IBD^{(Reference Guyatt, Mitchell and Irvine29)} and was used most frequently within these studies.^{(Reference Bodini, Zanella and Crespi14,Reference Cox, Lindsay and Fromentin18,Reference Jian, Anqi and Gang19,Reference Bartel, Weiss and Turetschek20,Reference Kyaw, Moshkovska and Mayberry22,Reference Brotherton, Taylor and Bourguignon26,Reference Chicco, Magrì and Cingolani28)} The short IBDQ (sIBDQ).^{(Reference Irvine, Zhou and Thompson30)} is a shortened version of the IBDQ (containing 10 questions) and was utilized in six of the included studies.^{(Reference Gunasekeera, Mendall and Chan15,Reference Pedersen, Ankersen and Felding17,Reference Chandrasekaran, Groven and Lewis21,Reference Keshteli, Valcheva and Nickurak24,Reference Lewis, Sandler and Brotherton25,Reference Fritsch, Garces and Quintero27)} Additionally, the Short Form 36 Health Survey (SF-36)^{(Reference Brazier, Harper and Jones31)} is a validated questionnaire that assesses health-related QOL, though not specific to individuals with IBD, and was used in three studies.^{(Reference Melgaard, Sørensen and Riis16,Reference Tamizifar and Arab23,Reference Fritsch, Garces and Quintero27)}

Table 3.

Measures of QOL utilized across included studies

IBDQ, Inflammatory Bowel Disease Questionnaire.

QOL, quality of life.

Study design, setting, and duration

Of the 15 studies selected, 13 were randomized controlled trials (RCT).^{(Reference Bodini, Zanella and Crespi14–Reference Bartel, Weiss and Turetschek20,Reference Kyaw, Moshkovska and Mayberry22–Reference Fritsch, Garces and Quintero27)} Six studies were unblinded,^{(Reference Pedersen, Ankersen and Felding17,Reference Jian, Anqi and Gang19,Reference Kyaw, Moshkovska and Mayberry22,Reference Keshteli, Valcheva and Nickurak24,Reference Lewis, Sandler and Brotherton25,Reference Fritsch, Garces and Quintero27)} whereas three were double-blinded,^{(Reference Gunasekeera, Mendall and Chan15,Reference Melgaard, Sørensen and Riis16,Reference Bartel, Weiss and Turetschek20)} three were single-blinded,^{(Reference Cox, Lindsay and Fromentin18,Reference Tamizifar and Arab23,Reference Brotherton, Taylor and Bourguignon26)} and two were cross-over studies.^{(Reference Melgaard, Sørensen and Riis16,Reference Fritsch, Garces and Quintero27)} In addition, two studies were uncontrolled trials.^{(Reference Chandrasekaran, Groven and Lewis21,Reference Chicco, Magrì and Cingolani28)} These studies were conducted in multiple locations, globally (Tables 4–6). Duration of interventions ranged from four weeks to six months. Both crossover studies were 8 weeks with a 2-week washout period.^{(Reference Melgaard, Sørensen and Riis16,Reference Fritsch, Garces and Quintero27)} Intervention delivery varied across different studies. Six of the studies provided some or all of the food items within the diet intervention.^{(Reference Melgaard, Sørensen and Riis16,Reference Bartel, Weiss and Turetschek20,Reference Tamizifar and Arab23,Reference Brotherton, Taylor and Bourguignon26–Reference Chicco, Magrì and Cingolani28)} One cross-over study by Fritsch et al. provided participants with all meals based on menus created by a Registered Dietitian (RD).^{(Reference Fritsch, Garces and Quintero27)} Twelve of the included studies provided in-person or personalized diet education from an RD or nutritionist on how to follow the designated diet.^{(Reference Bodini, Zanella and Crespi14,Reference Melgaard, Sørensen and Riis16–Reference Lewis, Sandler and Brotherton25,Reference Chicco, Magrì and Cingolani28)} Additionally, two studies had access to online support while following the diets.^{(Reference Chandrasekaran, Groven and Lewis21,Reference Lewis, Sandler and Brotherton25)} Alternatively, three studies provided diet education via handouts or other written material.^{(Reference Gunasekeera, Mendall and Chan15,Reference Brotherton, Taylor and Bourguignon26)}

Table 4.

Characteristics and results of studies using elimination diets to improve quality of life among participants with IBD

CD, Crohn’s disease.

UC, Ulcerative colitis.

IBS, irritable bowel syndrome.

RCT, randomized controlled trial.

Short-form 36 (SF-36), An increase in the questionnaire score indicates an increase in Quality of Life (QOL).

Inflammatory Bowel Disease Questionnaire (IBDQ), The bowel II domain score of the IBDQ, measures GI symptoms impact on QOL.

Short Inflammatory Bowel Disease Questionnaire (SIBDQ), An increase in the questionnaire score indicates an increase in QOL.

Table 5.

Characteristics and results of studies using an elimination diet that also increased specific foods among participants with IBD

CD, Crohn’s disease.

UC, ulcerative colitis.

RCT, randomized controlled trial.

DMP, Dietary Modification Program.

DMF, Dietary Modification Framework.

Short-form 36 (SF-36), An increase in the questionnaire score indicates an increase in Quality of Life (QOL).

Inflammatory Bowel Disease Questionnaire (IBDQ), The bowel II domain score of the IBDQ, measures GI symptoms impact on QOL.

Short Inflammatory Bowel Disease Questionnaire (SIBDQ), An increase in the questionnaire score indicates an increase in QOL.

Table 6.

Characteristics and results of studies using dietary interventions that focus on increasing specific foods among participants with IBD

CD, Crohn’s disease.

UC, ulcerative colitis.

RCT, randomized controlled trial.

Short-form 36 (SF-36), An increase in the questionnaire score indicates an increase in Quality of Life (QOL).

Inflammatory Bowel Disease Questionnaire (IBDQ), The bowel II domain score of the IBDQ, measures GI symptoms impact on QOL.

Short Inflammatory Bowel Disease Questionnaire (SIBDQ), An increase in the questionnaire score indicates an increase in QOL.

Participant characteristics

Both sexes were included in all studies; however, more than half of the population studied were females. All studies included adult participants, 18 years and older, with a diagnosis of IBD. The IBD diagnosis was separated into subtypes: CD or UC. Where some studies included both subtypes,^{(Reference Bodini, Zanella and Crespi14,Reference Pedersen, Ankersen and Felding17,Reference Cox, Lindsay and Fromentin18,Reference Chandrasekaran, Groven and Lewis21,Reference Chicco, Magrì and Cingolani28)} most studies did not. Crohn’s disease was studied independently within four studies, including the highly restricted organic diet, IgG4 diet, Mediterranean Diet, and high-fibre diet.^{(Reference Gunasekeera, Mendall and Chan15,Reference Bartel, Weiss and Turetschek20,Reference Lewis, Sandler and Brotherton25,Reference Brotherton, Taylor and Bourguignon26)} Ulcerative colitis was studied independently in six studies including the low FODMAP diet, IgG diet, Dietary Modification Program, Dietary Modified Framework, and Anti-inflammatory diet interventions.^{(Reference Melgaard, Sørensen and Riis16,Reference Jian, Anqi and Gang19,Reference Kyaw, Moshkovska and Mayberry22,Reference Tamizifar and Arab23,Reference Keshteli, Valcheva and Nickurak24,Reference Fritsch, Garces and Quintero27)}

Among the disease types, disease activity was assessed as an inclusion/exclusion criteria for each study. Most studies only included those in remission or with mild to moderate symptoms as defined by a disease severity tool (Table 1). Depending on IBD type, researchers used appropriate screening tools to select participants in either remission or with mild to moderate to severe symptoms. Tools used to screen for CD included the Harvey Bradshaw Index,^{(4,Reference Bodini, Zanella and Crespi14,Reference Gunasekeera, Mendall and Chan15,Reference Pedersen, Ankersen and Felding17,Reference Cox, Lindsay and Fromentin18,Reference Chandrasekaran, Groven and Lewis21)} partial Harvey Bradshaw Index,^{(Reference Brotherton, Taylor and Bourguignon26)} and the Crohn’s Disease Activity Index.^{(Reference Best, Becktel, Singleton and Kern6,Reference Gunasekeera, Mendall and Chan15,Reference Bartel, Weiss and Turetschek20,Reference Chicco, Magrì and Cingolani28)} The Mayo Score,^{(Reference Bodini, Zanella and Crespi14,Reference Melgaard, Sørensen and Riis16,Reference Jian, Anqi and Gang19,Reference Chandrasekaran, Groven and Lewis21,Reference Fritsch, Garces and Quintero27)} Partial Mayo score,^{(Reference Keshteli, Valcheva and Nickurak24)} and Simple Clinical Colitis Index^{(Reference Pedersen, Ankersen and Felding17)} were used to screen and classify participants with UC.

Medications and supplements

Treatments for IBD may include surgeries and medications all of which can be very costly and have serious or unwanted side effects.^{(2,Reference Holt, Strauss and Moore9)} Medications are the most commonly used therapy when treating symptoms of IBD. The typical types include biologics, immunomodulators, aminosalicylate (5-ASA), corticosteroids, anti-TNF, and azathioprine.⁽²⁾ Each study had different, specific inclusion and exclusion criteria for medication type, dosage, and time frame of use. Researchers in 14 of the 15 studies allowed medication use as a concomitant treatment yet required that participants remain on a stable dose throughout the study duration.^{(Reference Bodini, Zanella and Crespi14–Reference Cox, Lindsay and Fromentin18,Reference Bartel, Weiss and Turetschek20–Reference Chicco, Magrì and Cingolani28)} Jian et al. did not report use of medication amongst participants, but they did report that changes in medical treatment would be considered for each participant.^{(Reference Jian, Anqi and Gang19)} Although supplements such as vitamins, minerals, and probiotics may affect QOL in this population, only four studies discussed the use of supplements as inclusion and exclusion criteria. Cox et al.^{(Reference Cox, Lindsay and Fromentin18)} and Tamizifar et al.^{(Reference Tamizifar and Arab23)} excluded those who used or changed the dose of probiotics or prebiotics 8 weeks prior to the start of the study. Fritsch et al.^{(Reference Fritsch, Garces and Quintero27)} excluded individuals who had used probiotics 4 weeks prior to the start of the study, whereas Bartel et al.^{(Reference Bartel, Weiss and Turetschek20)} excluded anyone who had used probiotics. The remaining authors did not mention the use of supplements as an inclusion or exclusion criterion.

Changes in quality of life after following an elimination diet

Four studies used the low FODMAP diet for the intervention (Table 4).^{(Reference Bodini, Zanella and Crespi14,Reference Melgaard, Sørensen and Riis16–Reference Cox, Lindsay and Fromentin18)} . Two of the four studies where the low FODMAP diet was compared to a sham or usual diet found significantly better and positive changes in QOL when compared to the control.^{(Reference Pedersen, Ankersen and Felding17,Reference Cox, Lindsay and Fromentin18)} In a four-week randomized, parallel, single-blinded placebo control study, Cox et al. provided diet advice from an RD to 52 participants with either UC or CD.^{(Reference Cox, Lindsay and Fromentin18)} At follow-up, the low FODMAP group experienced significantly greater IBDQ scores (81.9, SEM 1.2) compared to the sham diet group (78.3, SEM 1.2, P = 0.04). Pederson et al. also had nutritionists provide a one-hour counselling session on FODMAPs to the experimental group while the control group was asked to remain on the normal diet. After six weeks, the group following the low FODMAP intervention had a significantly greater increase in SIBDQ scores reported as median and interquartile ranges (IQR) compared to the normal diet group, (60, IQR 51–65 and 50, IQR 39–60, P < 0.01).^{(Reference Pedersen, Ankersen and Felding17)} Additionally, scores were compared within groups from baseline to the end of the study, where the low FODMAP group had a significant increase in median scores (47, IQR 42–55 and 60, IQR 51 to 65, P < 0.01), but the normal diet group did not (50, IQR 40–57 and 50, IQR 39–60, P = 0.54), respectively. Similar to Pederson, Bodini et al. utilized a six-week low FODMAP diet intervention compared to a Standard Diet (SD) to evaluate changes in QOL.^{(Reference Bodini, Zanella and Crespi14)} There was no significant difference between the scores reported as median and IQR from baseline to 6 weeks in the SD group (181, IQR 153–197 and 166, IQR 153–200, P = 0.88, respectively). However, there was a slight but significant increase in median survey scores in the low FODMAP diet group from baseline to 6 weeks (166, IQR 139–182; 177, IQR 155–188, P = 0.05, respectively). A positive outcome cannot be concluded as there was no significant difference found between groups at the end of 6 weeks (P = 0.88).

Lastly, in a three group, double blind RCT crossover study to test feasibility, Melgaard et al. examined changes in QOL in 19 adults with UC after completing three phases, including watchful waiting, low FODMAP with FODMAP provocation, and placebo provocation.^{(Reference Melgaard, Sørensen and Riis16)} Provocation arms were used to provide participants with food containing FODMAPs that appeared and tasted similar to foods with low-FODMAPs. These provocation foods were made in the lab and provided to participants. They found no significant differences in changes in QOL scores over 4 weeks between the low FODMAP and placebo provocation groups, respectively (3.2, IQR 1.1–7.5; 2.2, IQR −3.5–4.3, P = 0.24). Additionally, the watchful waiting group had a decrease in QOL scores after 4 weeks although no significance level was provided (−5.7, IQR −10.3–0.3). There was no significant difference found between the groups and therefore a positive outcome cannot be concluded (P = 0.24).

Another elimination diet to consider is the IgG diet. Using a randomized, open label, stratified prospective study, Jian et al. recruited individuals previously admitted to a local hospital.^{(Reference Jian, Anqi and Gang19)} Participants were assigned to follow the IgG diet intervention or the sham diet. Researchers examined changes in IBDQ scores in 97 participants with UC over a 6-month IgG diet intervention compared to a sham diet. There was no significant difference in the mean IBDQ scores between the groups at baseline (P > 0.05). Overall, the intervention group had significantly higher IBDQ scores compared to the sham diet group in all four dimensions; bowel symptoms, systemic symptoms, emotional function dimensions (P < 0.05), and the social function dimension (P < 0.01) after the 6-months. Similarly, Gunasekeera et al. studied an IgG4 diet, a subclass of the IgG diet, except only in those with CD.^{(Reference Gunasekeera, Mendall and Chan15)} Individuals in the intervention group, known as the True diet, were to exclude the four foods with the highest titre response, indicating a greater immune response to those foods. The control group was instructed to exclude the four foods with the lowest titre response. Significant improvements were found in the intervention group when compared to the control after adjusting for baseline in both intention to treat and per protocol analyses, respectively (−3.05; IQR −6.11 to −0.01, P = 0.05 and −4.728; IQR −8.10 to −1.36, P = 0.007). Refer to Table 4 for details on the study results.

While the previously mentioned diets are commonly used elimination diets, Bartel et al. used a highly restricted organic diet (Table 2) in a double-blind randomized controlled trial.^{(Reference Bartel, Weiss and Turetschek20)} Eighteen participants with mild-to-moderate CD were randomized to the active group (organic diet) or control group for six weeks with a follow-up at 24-weeks. The control group was instructed to eat a low-fat, high-carbohydrate diet while avoiding red meat and fibrous fruits and vegetables. Both groups received nutrition counselling by an RD at weeks 1, 3, and 6 of the intervention and weeks 12 and 24 of the follow-up period. The active group did not experience a significant change in IBDQ scores (reported as mean ± SD) from baseline to week 6, respectively (182 ± 23 and 196 ± 20; P = 0.204). Similarly, the control group did not experience a significant change in IBDQ scores from baseline to week 6 (180 ± 15 and 192 ± 20; P = 0.071). The changes in IBDQ scores were not compared between the intervention and control groups.

Changes in quality of life after following an elimination diet with increased specific foods

Research teams for four studies used an elimination diet paired with the addition of specific foods or nutrients (Table 5).^{(Reference Chandrasekaran, Groven and Lewis21–Reference Tamizifar and Arab23)} In two of the four studies, research teams created similar dietary programmes for UC; however, only one found significant improvements in QOL following the intervention.^{(Reference Kyaw, Moshkovska and Mayberry22,Reference Tamizifar and Arab23)} Tamizifar et al. created a 3-month Dietary Modification Program intervention (Table 2) for adults with UC, provided by RD instruction and an educational booklet.^{(Reference Tamizifar and Arab23)} The control group received usual dietary advice (UDA). After adjusting for differences in baseline data, when compared to the control group, the Dietary Modification Program group had significantly greater improvements in three of the eight scales within the SF-36, including role-physical, bodily pain, and general health. Similarly, Kyaw et al. utilized the Dietary Modified Framework (Table 2) intervention in adults with UC.^{(Reference Kyaw, Moshkovska and Mayberry22)} The diet advice was provided through an educational booklet over four to six weeks. Participants in the control group received healthy eating handouts. QOL was measured using the IBDQ, which was noted to increase slightly in the Dietary Modified Framework group with a mean difference of 7.173, yet this was not a significant change (P = 0.13). The control group was noted to have a decrease QOL scores with a mean difference of 3.454; however, this change was also not considered significant (P = 0.21). Between group changes were not noted.

In an open-label, randomized controlled, parallel study, Ketshteli et al. conducted a 6-month dietary intervention utilizing the Anti-Inflammatory Diet versus a control diet in adults with UC.^{(Reference Keshteli, Valcheva and Nickurak24)} Twenty-six participants randomized to the Anti-Inflammatory Diet group received dietary counselling, including menus and recipes, once a month from an RD. Twenty-seven participants randomized to the control group were instructed to follow the Canada Food Guide (CFG) 2007.^{(Reference Canada34)} There was no significant difference in SIBDQ scores reported as mean ± SD between the Anti-Inflammatory Diet group and the CFG group at baseline (5.5 ± 4.9–6.4 and 5.0 ± 5.0–6.0; P = 0.99). Furthermore, there was no significant difference in SIBDQ scores in the Anti-Inflammatory Diet group from baseline (5.5 ± 0.9) to the last visit (5.6 ± 0.8, P = 0.50) or in the CFG group from baseline (5.5 ± 0.7) to the last visit (5.5 ± 0.9, P = 0.80). There were no results reporting changes between the groups.

The Autoimmune Protocol Diet is an established diet used to reduce symptoms in autoimmune diseases. In an uncontrolled, open label study, Chandrasekaran et al. examined changes in QOL using the SIBDQ in 15 CD or UC participants after following the Autoimmune Protocol Diet over 12-weeks in an uncontrolled trial.^{(Reference Chandrasekaran, Groven and Lewis21)} SIBDQ mean scores increased steadily at all time periods of the intervention as follows: baseline (46.5; SD 12.5) to week 3 (54.0; SD 7.7, P = 0.02); week 6 (53.3 SD 10.9, P = 0.02); and week 9 (62.0; SD 3.3, P = 0.03). However, from all participants that completed follow-up measures (n = 4), there was not a significant increase in SIBDQ scores from baseline to week 11 (46.5 to 61.5, P = 0.07). Differences in SIBDQ scores between CD and UC were also explored and there was no significant difference found between the IBD types (P = 0.77). Finally, there was no control group for comparison in this study. Refer to Table 5 for details on the study results.

Changes in quality of life while following dietary patterns with a focus on including specific foods

Four studies used dietary pattern interventions with a focus on including specific foods, such as the Mediterranean Diet or a high-fibre diet, to examine changes in QOL.^{(Reference Lewis, Sandler and Brotherton25–Reference Chicco, Magrì and Cingolani28)} Control diets for three of the four studies varied, and one study was uncontrolled. All four studies had significant improvements in QOL following the intervention diets; however, not all results were compared between groups. Fritsch et al. used a randomized, parallel-group cross-over design in 18 adults with UC to examine changes in SIBQ scores following a low-fat, high-fibre diet (Table 6) and an improved standard American diet (iSAD) as the control.^{(Reference Fritsch, Garces and Quintero27)} All meals were provided to the participants. SIBDQ scores increased significantly in both the intervention (P = 0.001) and control groups (P = 0.02) after 4 weeks on each diet. There were significant improvements in many of the scales within the SF-36 within both groups, including limitations due to physical health scores, role limitations due to emotional problems scores, social functioning scores, and bodily pain scores. Overall, the general health scores significantly improved from a baseline average of 48.5 in both the intervention and the control (59.1, P = 0.03 and 57.6, P = 0.04).

Similar to Fritsch, Brotherton et al. examined changes in QOL using a high fibre diet.^{(Reference Brotherton, Taylor and Bourguignon26)} However, this was a randomized controlled, single-blinded study that used an attention control diet in 7 adults with CD over 4 weeks. The high fibre diet was not defined by the researchers; however, the intervention group received education on high-fibre, low-refined carbohydrate intake, and were encouraged to consume ½ cup of provided, whole wheat bran cereal and at least 48 oz of fluid per day, whereas the control group was instructed to avoid whole grains. They found that IBDQ scores significantly increased over time for the intervention group when compared to the control group, (P = 0.028). Clinical significance, defined by authors as a > 32-point increase, was also found in the high fibre diet group with a mean increase of 44.25-points in IBDQ scores whereas the control group had a mean increase of 19-points. This study collected additional data on diet adherence but did not report the findings. The small sample size of 7 should be considered when interpreting the results of this study.

Two studies explored changes in QOL after following the Mediterranean diet (MD).^{(Reference Lewis, Sandler and Brotherton25,Reference Chicco, Magrì and Cingolani28)} However, Lewis et al.^{(Reference Lewis, Sandler and Brotherton25)} used the Specific Carbohydrate diet (SCD), a diet commonly used within IBD, as the control diet in a parallel RCT with 194 participants with CD, whereas Chicco et al.^{(Reference Chicco, Magrì and Cingolani28)} did not use a control diet and included 142 participants with either UC or CD. Lewis et al. found improvements in SIBDQ scores in the SCD group which was associated with an increase in standardized mean difference of SIBDQ scores from baseline to 6 weeks, (8.85 to 10.4, P < 0.0001); however, there was no report on the results from 6–12 weeks.^{(Reference Lewis, Sandler and Brotherton25)} Additionally, the SCD did not have a significantly greater increase in median SIBDQ scores at week 6, (49.0, 39.0–58.0) compared to MD median scores at week 6, (49.5, 39.0–56.0, P = 0.55). Similarly, the SCD did not have a significantly greater increase in median SIBDQ scores at week 12, (50.0, 38.0–58.0) compared to MD median scores, (45.5, 36.0-57.0, P = 0.040). Chicco et al. found significantly improved IBDQ scores after the intervention (P < 0.0001) yet did not have a control group for comparison between groups. ^{Reference Chicco, Magrì and Cingolani28} Refer to Table 6 for additional details on study results.