The tailored questionnaire incorporated culturally appropriate dietary references, language-specific phrasing and flexible response formats to enhance content validity and ensure accurate reflection of the characteristics and dietary behaviours of IBS sufferers in Malta. Building on these design considerations, the tool was structured into three domains: demography, symptomatology and characteristics. The demographic domain included nine items, covering anthropometric measures, smoking, drinking behaviours and psychological aspects.

The second domain adopted the validated Irritable Bowel Syndrome-Severity Scoring System (IBS-SSS), with permission from the Rome Foundation, to analyse symptoms [15]. The first part of the original tool was included in this questionnaire as a series of seven multiple-choice questions where respondents were required to rate the level of the indicated symptom experienced in the previous 10 days.

The third domain was designed to gather additional characteristics across a total of ten items, comprising yes/no questions, open-ended responses and multiple-choice formats. Item generation was guided by a review of current literature on IBS symptomatology, diagnostic pathways and treatment interventions, and further refined through consultation with a local multidisciplinary team of dietitians, nutritionists, statisticians and gastroenterologists with expertise in health research. This process ensured clinical relevance, methodological rigor and contextual appropriateness for the Maltese population. The resulting domain captured key constructs including diagnosis status, sources of information, food triggers and trialled treatment interventions. The Rome IV symptom-based diagnostic criteria [3], included with permission from the Rome Foundation, were selected to ensure inclusion of participants who met validated clinical thresholds despite lacking a formal diagnosis. This also allowed verification of self-reported physician-diagnosed IBS cases.

This yielded a questionnaire developed over three domains with a total of 26 items. Efforts focused on designing a questionnaire that was concise yet comprehensive, minimising respondent burden while ensuring sufficient data to address the main research questions [16]. Ethical approval for this study was granted by the Faculty of Research Ethics Committee and the University Research Ethics Committee (UREC FORM FHS-2024-00126).

To accommodate potential language barriers, the questionnaire was produced in both English and Maltese using a forward-back translation process to ensure linguistic and conceptual equivalence [17]. Content validity was assessed for both the English and Maltese versions of the questionnaire, the latter incorporating the translated and licensed IBS-SSS and Rome IV criteria. A structured validation form was completed by a purposive sample of 10 experts (7 registered dietitians, 2 statisticians, and 1 gastroenterologist), who evaluated the tools through a unified process to ensure consistency across versions. Relevance was rated on a 4-point scale and ratings were dichotomised for calculation of the item-level and scale-level Content Validity Index (CVI), following established procedures [18][19]. Although the modified kappa statistic (k*) was not computed, the expert panel size met the threshold where CVI and k* values are known to converge, reducing the likelihood of chance agreement.

Face validation was conducted following content validation to assess the clarity, relevance and acceptability of the questionnaire [20]. Experts provided written feedback at the end of each domain, focusing on language, item relevance and overall presentation. Validation occurred over one round where experts judged both the content and face validity of a total of 19 items across the first and third domains since the second section contained a validated scale [15].

To obtain respondent opinion, a purposive sample of the target population was invited via email to participate in face validation. A total of 6 respondents were identified based on eligibility criteria for the study and selected for convenience, as they were willing and available to contribute. Their qualitative feedback, informed by personal experience with IBS, was collected through open-ended questions addressing language clarity, ease of understanding and the relevance of the information provided.

Eligible participants were Maltese adults aged 18 to 65 years with IBS, either formally diagnosed by a physician or meeting the Rome IV diagnostic criteria. Individuals receiving dietetic treatment, as well as those not currently undergoing any treatment, were included. All participants were required to provide informed consent prior to participation.

Data collection was conducted through a dual recruitment strategy combining outpatient clinical settings and online outreach. Gastroenterologists and dietitians acted as intermediaries by inviting eligible patients attending their clinics to participate via the anonymous online questionnaire, while online platforms, including social media channels, similarly served as intermediaries to extend invitations to a wider audience.

The dataset was cleaned to exclude responses from non-Maltese participants and those not meeting the Rome IV diagnostic criteria. Body Mass Index (BMI) was calculated using self-reported weight and height values. Symptom severity data from the second domain, based on the IBS-SSS, were analysed according to the scoring protocol outlined15. Open-ended responses regarding perceived food triggers were categorised into eight food groups to facilitate analysis: carbohydrates; vegetables, fruit, beans and legumes; dairy; fats and oils; proteins; processed foods; miscellaneous; others or non-dietary triggers. All resultant data were then analysed using the Statistical Package for the Social Sciences Version 29.

Descriptive statistics were used to summarise sample characteristics. Inferential analyses (independent t-tests, one-way ANOVA, Mann-Whitney U tests, and Spearman’s rank-order correlation) were conducted to examine associations between IBS symptom severity and demographic and clinical variables. Effect sizes (Cohen’s d, eta-squared, and rank biserial correlation) were calculated to complement significance testing.

All items demonstrated excellent content validity, with I CVI values of 0.9 - 1 and an S CVI average of 0.98, exceeding established thresholds18.19. The universal agreement S CVI was 0.789, narrowly below the 0.80 benchmark but considered acceptable given the high average and the involvement of 10 experts.

Expert feedback resulted in the addition of three items on living situation, household meal preparation and trigger foods, and one response option was refined for clarity. Respondents confirmed the questionnaire’s accessibility and logical structure. After validation and pilot testing, the final instrument comprised 30 items across three domains: demography (12), symptomatology (7) and characteristics (11).

A total of 266 questionnaire responses were received, with no incomplete questionnaires. Following exclusion of non-Maltese respondents (n = 14) and those not meeting Rome IV criteria (n = 122), 130 valid responses were retained for analysis reducing the margin of error to 4.3%.

Demographic, lifestyle and anthropometric data from the study sample (N = 130) are summarised in Table 1. The sample was predominantly female (90.8%) and aged 36 - 45 years (60%). Educational attainment was high, with nearly half holding an undergraduate degree (47%) and a comparable proportion a postgraduate qualification (48%). Most participants were married (55%) and living with their partner and children (40%). Lifestyle indicators suggested a relatively healthy cohort: the majority prepared their own meals (52%), had never smoked (69%) and reported low alcohol consumption, typically once per month (49%). Despite these favourable health behaviours, perceived psychological comorbidities were common, with anxiety (68%) and stress (66%) most frequently reported, while 18% indicated no psychological health concerns. The mean BMI of participants was 27.5 ± 6.8 kg/m² (n = 129), indicating that most were classified as overweight.

Table 1. Demographic, lifestyle and anthropometric characteristics of the study sample (N = 130). Data are presented as frequency (percentage) for categorical variables and mean ± SD and mean (95% CI) for continuous variables. (a) Demographic and lifestyle characteristics; (b) Anthropometric characteristics.

(a) 1 = Participants could select more than one-response; (b) S. D. = Standard Deviation, CI = Confidence Interval.

Table 2 summarises outcomes from the symptomology and characteristics domains of the questionnaire. Participants reported a mean IBS SSS score of 258.1 ± 95.1, with moderate symptom severity being the most common classification (39%). The three participants classified as being in remission, according to the IBS‑SSS, remained eligible, as all met Rome IV diagnostic criteria for IBS and no minimum symptom severity threshold was required for eligibility. The remission category reflects low current symptom burden within individuals who still have IBS and is consistent with the fluctuating symptom patterns characteristic of the condition. Most respondents (72%) had received a physician diagnosis of IBS, typically at a mean age of 30.9 ± 12.6 years, with an average disease duration of 7.9 ± 7.6 years. Food intake was widely perceived as a symptom trigger (78.5%), with dairy products, plant-based foods (vegetables, fruits, beans, legumes) and carbohydrates most frequently implicated. Information sources were largely digital, with 80.8% relying on the internet or social media, while one third consulted general practitioners. In terms of management, pharmacological treatments were most trialled (85.4%), followed closely by dietary modifications (79.2%). Of the total sample (N = 130), disease duration was reported by 92 participants (7.9 ± 7.6 years), while age at diagnosis was available for 93 participants (31.1 ± 12.4 years). These findings contribute to the overall profile of the cohort, despite partial item response.

Table 2. Further characteristics of the study sample (N = 130). Data are presented as frequency (percentage) for categorical variables and mean ± SD and mean (95% CI, variance) for continuous variables. (a) IBS symptoms and diagnostic characteristics; (b) Symptom triggers and management; (c) Clinical characteristics.

(a) n = number of participants, IBS-SSS = Irritable Bowel Syndrome-Severity Scoring System; (b) n = number of participants, 1 = Participants could select more than one-response; (c) n = number of participants, S. D. = Standard Deviation, CI = Confidence Interval, IBS-SSS = Irritable Bowel Syndrome-Severity Scoring System.

Table 3 reports findings of associations between IBS-SSS scores and demographic and clinical factors. A sample size calculation was performed a priori to ensure adequate precision for estimating population characteristics, as mentioned in section 3.3. However, no separate power calculations were conducted for each inferential test presented in Table 3. As this was an exploratory study, effect sizes were reported to aid interpretation and findings should be considered within the context of potential limitations in statistical power for subgroup comparisons. Mean IBS-SSS scores were slightly higher among females (251.40 ± 96.30) compared to males (215.83 ± 77.98). Although the two-sided p-value (p = 0.218) was not statistically significant and the effect size was small (d = 0.2), these findings should be interpreted with caution due to the very small number of male participants (n = 12), which limits statistical power and reliability. No evidence of a significant association was found between IBS-SSS scores and age group (p = 0.417, η² = 0.01) whilst evidence of borderline significance was seen across diagnosis status (p = 0.052), although the effect size was small (r = 0.2).

Table 3. Associations between IBS-SSS scores and demographic and clinical factors.

n = number of participants, SD = Standard Deviation, d = Cohen’s d, η² = eta-squared, r = rank-biserial correlation coefficient, ρ = Spearman Correlation Coefficient.

Independent samples t-tests revealed no evidence of statistically significant differences in IBS-SSS scores between participants who reported psychological comorbidities and those who did not (stress p = 0.167, d = −0.3, anxiety p = 0.632, d = 0.1, depression p = 0.617, d = 0.1, none p = 0.671, d = 0.1). Corresponding effect sizes were small, with Cohen’s d values indicating minimal practical differences in symptom severity between groups. These results suggest that, within this sample, the presence of perceived psychological comorbidity was not significantly associated with IBS symptom severity.

Analysis also revealed no significant association between IBS-SSS and BMI (p = 0.518), with a very weak negative correlation (ρ = −0.057). This suggests that increases in BMI were associated with negligible decreases in symptom severity, indicating that BMI had minimal influence on IBS-SSS scores within the study sample. A summary of these results is presented in Table 3.