Web-based dietary assessment and advice helps inflammatory bowel disease patients to improve their diet quality

Time to evaluate diet quality and give dietary advice is limited in clinical inflammatory bowel disease (IBD) practice. The Eetscore is a web-based tool that assesses diet quality according to the Dutch dietary guidelines and provides personalised dietary advice. We aimed to assess diet quality of IBD patients using the Eetscore and to study changes in diet quality, health-related quality of life (HRQoL) and clinical disease activity over time. A prospective cohort study was performed in 195 adult IBD patients. Participants were invited to fill out questionnaires (Eetscore-FFQ, short Inflammatory Bowel Disease Questionnaire and Patient Harvey Bradshaw Index/Patient Simple Clinical Colitis Activity Index) at baseline and after 1 and 4 months. The Eetscore calculates diet quality based on sixteen food components (ten points per component, total score 0–160; the higher the better) and provides dietary advice per component based on the assessment. At baseline, mean diet quality was 98 (sd 19). Diet quality was positively associated with age, female sex and level of education. Component scores were highest for red meat, wholegrain products and sweetened beverages, and lowest for legumes, nuts and processed meat. Over time, diet quality increased to 107 (sd 21) at 4 months (P < 0·001). Each ten-point improvement in diet quality was associated with an increase in HRQoL (β = 0·4 (95 % CI (0·02, 0·7), P = 0·04). Clinical disease activity did not change. In conclusion, diet quality of IBD patients significantly improved following personalised dietary advice of the Eetscore. Improvement of diet quality was associated with a slight improvement in HRQoL. The Eetscore is a practical and useful tool to monitor and support a healthy diet in IBD patients.

Crohn's disease and ulcerative colitis are chronic inflammatory bowel disease (IBD) with a complex aetiology including genetic, microbial, immune and environmental factors (1) . Increasing evidence suggests that a person's diet is an important environmental factor contributing to the development of IBD and its disease course (2) . Dietary components may influence the pathogenesis of IBD through their effects on the microbiome, mucosal barrier and immune response (2) . Although many IBD patients believe diet affects the course of their disease and is at least as important as medication in their IBD treatment, evidence supporting beneficial effects of specific dietary patterns in adult IBD patients is limited (3,4) .
Recently, the International Organization for the Study of Inflammatory Bowel Diseases attempted to establish a consensus document stating which foods may be either beneficial, harmful or safe to consume (5) . Due to lack of evidence, recommendations were limited but in line with the general guidelines for a healthy diet, such as to consume more fruit, vegetables and n-3 fatty acids, and to consume less saturated, trans, and dairy fat and red and processed meat (5) . Therefore, general dietary guidelines may also be recommended to patients with IBD.
In the Netherlands, the composition of a healthy diet is described by the Health Council in the Dutch dietary guidelines (6,7) . These guidelines advise to adopt a more plant-based and less animal-based dietary pattern by eating sufficient fruit, vegetables and wholegrain products, moderating the intake of † Shared first author. ‡ Shared last author. meat, varying between fish, legumes, nuts and eggs and taking sufficient dairy products and fluids. In clinical practice, IBD patients are interested in the beneficial effects of diet on the course of their disease. However, time is limited to assess a patients' food intake and to support a healthy diet. It would be helpful if patients could assess their own diet quality and receive personal advice outside the consultation room.
A web-based tool that may be used to assess diet quality and to support a healthy diet is the Eetscore ('Eatscore') (8) . This is a validated tool to assess adherence to the Dutch dietary guidelines, that is, diet quality (9) . Based on the assessment, the Eetscore provides personalised dietary advice to eat healthier and more in line with the Dutch dietary guidelines. It can also be used to monitor diet quality over time. So far, the Eetscore has only been used in healthy populations and CVD patients, but not in IBD patients (10,11) .
We used the Eetscore tool to assess diet quality of IBD patients, to evaluate if diet quality improves after personalised dietary advice and to study factors associated with diet quality in IBD patients. In addition, we aimed to evaluate whether changes in diet quality resulted in changes in health-related quality of life (HRQoL) and clinical disease activity. We also evaluated the experiences of IBD patients with the Eetscore tool.

Study population and design
A prospective cohort study was performed in adult IBD patients. Participants were recruited between October 2020 and February 2021 from the outpatient clinics of Hospital Gelderse Vallei in Ede and Rijnstate Hospital in Arnhem, the Netherlands. Both are secondary care referral centres, with IBD-dedicated gastroenterologists and nurses, and are collaborative partners of the Nutrition & Healthcare Alliance, an expert centre in nutrition and healthcare. Inclusion criteria were diagnosis of Crohn's disease, ulcerative colitis or IBD-unclassified, 18 years of age or older, ability to read and understand the Dutch language and access to an e-mail address and a device to complete the online questionnaires. Exclusion criteria were current dietary counselling by a dietician or lifestyle coach, allergy for nuts, peanuts, fish or cows' milk protein (to protect them from potential harmful recommendations to consume these products), adherence to a vegan lifestyle (to avoid invalid dietary assessment as the Eetscore does not assess plant-based alternatives) and participation in another intervention study. This study was approved by the Medical Ethical Committee of Wageningen University (METC nr. 20/16) and conducted in accordance with the Declaration of Helsinki and registered at trialregister.nl (NL8784). All participants provided written informed consent.

Data collection
After the informed consent form was signed, participants were invited via e-mail to complete questionnaires assessing diet quality, HRQoL and clinical disease activity. Participants who completed the baseline assessment were invited to complete the questionnaires again after 1 and 4 months. These time points were chosen to assess short-term changes in diet quality as well as to evaluate effects at longer term. At baseline, participants were also asked about their level of education, smoking behaviour, dietary restrictions, and height and weight. At 4 months, participants were also invited to complete an evaluation about the Eetscore. Non-responders were reminded 5, 10 and 15 d after each invitation.
Relevant participant and disease characteristics were retrieved from medical records, including age, sex, diagnosis, disease duration, disease phenotype according to the Montreal classification, extra-intestinal manifestations, IBD medication used and IBD-related surgeries. The highest treatment step was documented, classified as 5-aminosalicylic acids, steroids, immunomodulators and biologics.
The Eetscore-FFQ assesses intake of the following sixteen components: vegetables, fruit, wholegrain products, legumes, nuts, dairy products, fish, tea, fats and oils, coffee, red meat, processed meat, sweetened beverages and fruit juices, alcohol, salt and unhealthy choices. Participants reported their intake frequency of each component during the past month ('never' to 'every day') together with portion size (standard and natural portions (e.g. piece of fruit) or commonly used household measures (e.g. spoons or cups). It takes approximately 10-15 min to complete the Eetscore-FFQ.
The DHD15-index is calculated based on a person's food intake as assessed by the Eetscore-FFQ. The criteria to calculate the DHD15-index have been described in detail elsewhere and are summarised in Table 1 (9) . Components are categorised into five types: adequacy, ratio, optimum, qualitative and moderation component. Scores range from 0 to 10 points with cut-offs for each component based on the dietary guidelines (Table 1). In the total score, all sixteen components are combined, resulting in a score range of 0-160. Higher scores indicate better adherence to Dutch dietary guidelines.
The Eetscore tool provides personalised dietary advice every time the Eetscore-FFQ is completed and aims to improve a person's adherence to the Dutch dietary guidelines. The dietary advice is automatically personalised based on the calculated diet quality score and is provided per component. It consists of general healthy diet information and practical advice to improve the intake of each component. Participants are presented their total diet quality score (online Supplementary Fig. S1) and diet quality score per component together with dietary advice (online Supplementary Fig. S2). In addition, the tool presents an overview of results to monitor changes in diet quality over time.
Health-related quality of life. HRQoL was assessed with the validated Short Inflammatory Bowel Disease Questionnaire. It consists of ten items each with a 7-point Likert scale resulting in a possible score range of 10-70. Higher scores indicate better HRQoL (12) .  (13,14) . Clinical remission was defined as a Patient Harvey Bradshaw Index score ≤4 and a Patient Simple Clinical Colitis Activity Index score ≤2 (15,16) .
Evaluation. Participants' experiences with the Eetscore tool were evaluated with a self-composed evaluation consisting of questions about the Eetscore-FFQ and the personalised dietary advice.

Statistical analysis
Participants who completed baseline and at least one follow-up assessment were included in the analysis. Normally distributed data are presented as mean values and standard deviations, skewed data as median with interquartile range (IQR) and categorical data as counts and percentages.
Linear mixed models were performed to assess changes in diet quality within subjects over time (fixed main factor) and to account for missing values. Baseline values were used as reference. If effect of time was significant, pairwise comparisons between the different time points were conducted with Bonferroni correction to adjust for multiple comparisons.
Multivariate linear regression analysis was performed to identify factors associated with diet quality at baseline. First, the following variables were evaluated in univariable analysis: age, sex, BMI, level of education, smoking behaviour, diagnosis, disease duration and clinical disease activity. Factors with a P-value < 0·2 in the univariable analysis, and BMI and clinical disease activity were included in the multivariable analysis. Linear mixed models were performed to identify factors associated with diet quality over time with an identity covariance structure and indicating time as repeated measure. Diet quality was the dependent variable and the same variables as in the regression analysis were added as fixed main effects to the model. Linear mixed models were performed to assess changes in HRQoL and clinical disease activity within subjects over time (fixed main factor) and to account for missing values. Pairwise comparisons were performed similar to the analysis of diet quality. If HRQoL or clinical disease activity significantly changed over time, further linear mixed model analyses were performed to assess if those changes resulted from changes in diet quality (main fixed covariate) with an identity covariance structure and indicating time as repeated measure. After univariate analysis, the following variables were evaluated as fixed main effects: age, sex, BMI, level of education, biologics as highest step-up in IBD medication and clinical disease activity. Variables with a P-value < 0·2 were included in the multivariable analysis.
Linear mixed model data are reported as fixed effect estimates with 95 % confidence intervals. A P-value of <0·05 was considered statistically significant. Statistical analysis was performed using IBM SPSS Statistics version 25.0.

Results
In this study, 212 participants were invited for baseline questionnaires (Fig. 1). Participants who completed the baseline questionnaires (n 204) were invited for follow-up assessments after 1 and 4 months. In total, 195 participants completed at least one follow-up assessment and were included in the analysis.
Of the 195 participants, 60 % was female, median age was 47 years (IQR 32-57) and BMI was 25 kg/m 2 (IQR 22-28) ( Table 2). The non-responders (n 17) were slightly different in terms of sex (41 % female) and age (37 years (27-47)) while BMI was comparable. Median disease duration of participants was 10 years (IQR 4-18) and half of the participants had Crohn's disease. At baseline, 77 % of participants used IBD medication and 72 % was in clinical remission. In their habitual diet, 28 % of participants applied certain dietary restrictions. Most common restrictions were exclusion or limitation of meat and lactose products. Other frequently mentioned restrictions were exclusion or limitation of gluten, carbohydrates, added sugar and ultra-processed food products.
Mean diet quality at baseline was 98 (SD 19) points out of 160 points. Component scores were highest for red meat, alcohol, salt, sweetened beverages and wholegrain products, suggesting that the intake of these food groups was (almost) in line with dietary recommendations (Fig. 2). Component scores were lowest for legumes, nuts, processed meat and unhealthy choices. Median intake per food component in g/d is described in online Supplementary Table S1. As reflected in diet quality scores, participants' median intake of vegetables, fruit, legumes, nuts, dairy products and fish was lower than recommended, while the median intake of processed meat was higher than recommended.
After using the Eetscore, diet quality increased to 103 (SD 21) points after 1 month and 107 (SD 21) after 4 months (Table 3).  This increase was significant between baseline and 1 month (P < 0·001), baseline and 4 months (P < 0·001) and 1 month and 4 months (P = 0·005). Biggest improvements were seen in legumes, unhealthy choices, nuts and sweetened beverages (Fig. 2). Diet quality at baseline was positively associated with age, female sex and level of education (Table 4). For example, each 10-year increase in age was associated with a three-point higher diet quality. The same factors were positively associated with diet quality over time. BMI and clinical disease activity were not associated with diet quality.
After the last assessment at 4 months, 180 participants completed an evaluation about the Eetscore tool (Fig. 3). Of these participants, 23 % reported the Eetscore-FFQ lacked regularly consumed food components such as eggs, and lactose-free and soya products. Also, 55 % did not eat or drink all food components of the Eetscore-FFQ given allergies, intolerances or dietary restrictions (28 %), personal taste (52 %) or gastrointestinal symptoms (40 %). Almost all participants (96 %, n 173) had reviewed their Eetscore results of which 92 % believed these results gave insight into how healthy their intake was per food component.

Discussion
In this study, we used the web-based Eetscore tool to assess diet quality of IBD patients and to provide personalised dietary advice supporting a healthy diet. At baseline, intake of red meat, wholegrain products and sweetened beverages was close to the Dutch dietary guidelines, while intake of dairy products,  legumes, nuts and processed meat was not. Diet quality significantly improved following personalised dietary advice. IBD patients who were older, female and higher educated had a better diet quality over time. Improvement of diet quality was associated with a small improvement in HRQoL. Clinical disease activity did not change. The Eetscore can provide IBD patients insight into how healthy their habitual diet is and what to do to better adhere to the Dutch dietary guidelines.
To date, the current Eetscore-FFQ has only been used once in a general, healthy population cohort of 751 Dutch participants (10) . Compared with this general cohort, diet quality was lower in our IBD cohort (mean 98 (SD 19) v. 111 (SD 18)) with IBD-U, inflammatory bowel disease-unclassified. * Total mean ± SD Eetscore.  lower scores on fruit, legumes, processed meat, sweetened beverages and unhealthy choices. This may be explained by differences in characteristics of the study populations. Many of their participants were highly educated (65 %), few smoked (6 %) and few were overweight or obese (27 % and 8 %) indicating a health-conscious population. In line with our study, older age, female sex and higher level of education were positively associated with diet quality (10) . Using the web-based Eetscore, diet quality of participants improved by ten points (i.e. 6 % out of a maximum of 160 points). This better adherence to the Dutch dietary guidelines has been associated with a lower allcause mortality and a reduction in the risk of stroke, depression and colorectal cancer (17) . Consistent with our findings, previous studies in IBD patients also reported a consumption of dairy products and nuts (18)(19)(20) below the recommendations, an overconsumption of processed meat and an adequate consumption of red meat (18) . In contrast to our results, previous studies in IBD patients indicate a sufficient intake of legumes (18,19) . IBD patients often avoid lactose products and legumes as they believe this prevents abdominal pain, bloating and disease flares, which may explain the low dairy products and legume intake (3,21) . In our study, 7 % of participants reported to consume little or no lactose products and while they may have compensated for this with alternatives such as soya milk, this cannot be entered in the Eetscore-FFQ. The reason for a nut consumption below the recommendations is unclear. In our study, participants did not report specific reasons for avoiding nuts and those allergic to nuts were excluded.
A recent study showed that long-term dietary patterns enriched with legumes, vegetables, fruits and nuts, more plant-based instead of animal-based products and avoidance of alcoholic beverages, high-fat processed meats and soft drinks have the potential to prevent intestinal inflammatory processes via the gut microbiome (22) . This potentially favourable dietary pattern for IBD patients is in line with the Dutch dietary guidelines and the Eetscore can provide insight in adherence to these guidelines. Moreover, IBD patients consider dietary guidance to be important, while only a minority feels to have received adequate information from their physician (23) . The Eetscore can fulfil this need for dietary guidance of IBD patients.
Several studies investigated dietary patterns and their association with HRQoL and clinical disease activity (18,24,25) . A Western dietary pattern characterised by the intake of refined grains, red and processed meat, condiments and sauces, and unhealthy choices was associated with flare occurrence (18) , while adherence to a Mediterranean dietary pattern characterised by a high consumption of vegetables, fruit, wholegrains and nuts, and a low consumption of red and processed meat and processed foods was associated with a higher HRQoL and lower disease activity (24,25) . Our observations support that better adherence to the Dutch dietary guidelines, which share characteristics with the Mediterranean diet, is associated with improved HRQoL. In contrast to previous studies, we did not find an association between diet quality and clinical disease activity. This might be explained by the relatively short follow-up time and low number of participants with clinically active disease.
Relatively few participants reported that use of the Eetscore tool helped to reduce their gastrointestinal symptoms. During this study, the majority of participants did not experience gastrointestinal symptoms. Besides, many participants commented to already know very well which foods work best for them to reduce or prevent gastrointestinal symptoms after several years of living with IBD. Also, the Eetscore tool is not designed to help reduce gastrointestinal symptoms. For example, IBD patients often avoid certain food components because of abdominal complaints and the Eetscore does not provide alternatives (e.g. for dairy products or legumes) (26) . Dietary advice of the Eetscore may be optimised for IBD patients by suggesting alternative products to achieve a better diet quality while limiting abdominal complaints.
The main strength of this study is its prospective nature with repeated dietary assessment and advice in a clinical setting using the Eetscore. In addition, the Eetscore is a quick and easy tool to assess diet quality compared with commonly used extensive FFQ. Besides, we included a large, representative cohort of IBD patients with varying ages and different disease phenotypes. Finally, the response rate was very high (92 %), thereby limiting the risk of non-response bias. However, this study was limited by the absence of a control group. Therefore, we cannot conclude whether diet quality improved as a result of the personalised dietary advice of the Eetscore or as a result of increased dietary awareness. In line with this, studies on lifestyle factors are prone to socially desirable answers, which may have led to an overestimation of diet quality. However, participation was anonymous and results were not shared with participants' gastroenterologist or IBD nurse. Also, completing the Eetscore several times may have induced a learning curve with participants being aware of the correct answers when completing the Eetscore for the second or third time. However, a previous study showed no change in diet quality when the Eetscore-FFQ was completed twice within 4 months in the absence of dietary advice (10) . Additionally, memory bias and estimation error may have occurred as these are inherent to using FFQ (27) . Lastly, the Eetscore-FFQ covers approximately 85 % of dietary intake, is based on regularly consumed food products and does therefore not assess plant-based alternatives of meat and dairy products (8,9) .
In conclusion, web-based dietary assessment and advice helps IBD patients to improve their diet quality. The Eetscore tool is easy to use, gives practical insight into dietary intake and supports a healthy diet in IBD patients. It may be further optimised by adding healthy alternatives for food products that are commonly avoided by IBD patients.