Digital versus face-to-face gut-directed psychotherapy for irritable bowel syndrome: a protocol of systematic review and network meta-analysis

Abstract

Introduction

Irritable bowel syndrome (IBS) is a common functional gastrointestinal disorder worldwide. Although not life-threatening, its chronic and recurrent nature greatly impacts patients’ quality of life. There is strong evidence that gut-directed psychotherapies (GDPs) help improve IBS symptoms. With technological advances, digital GDP is increasingly used as an alternative to traditional face-to-face GDP. This study will compare the clinical effectiveness of digital versus face-to-face GDP for IBS through network meta-analysis.

Methods and analysis

We will search English databases (PubMed, Cochrane Library, EMBASE and Web of Science) and Chinese databases (China National Knowledge Infrastructure, Wanfang, VIP and Chinese Biomedical Database) for randomised controlled trials (RCTs) of digital or face-to-face GDP for IBS. The search will cover the period from database inception to May 2025. We will perform multivariate network meta-analyses within a frequentist framework, using the mvmeta command in STATA V.16 software, and traditional pairwise meta-analysis using the DerSimonian-Laird random-effects model. The Cochrane Risk of Bias (RoB) tool (V.2) will be used to assess the RoB of each RCT, and the Confidence in Network Meta-Analysis (CINeMA) tool will be used to evaluate the certainty of the evidence.

Ethics and dissemination

Ethical approval is not required for this systematic review, as it involves the collection and synthesis of data from previously published primary studies.

OSF registration

Open Science Framework (OSF) registration: DOI 10.17605/OSF.IO/87463.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study will employ network meta-analysis to simultaneously compare the efficacy of digital and face-to-face gut-directed psychotherapies, incorporating both direct and indirect evidence.

• The search strategy encompasses both English and Chinese databases to ensure comprehensive coverage of relevant randomised controlled trials.

• The Confidence in Network Meta-Analysis framework will be used to rigorously assess the certainty of evidence for each network estimate.

• Our results will be limited by the quantity and quality of eligible studies included.

• The paucity of direct head-to-head trials comparing digital versus face-to-face interventions may limit the precision of direct comparative estimates.

Introduction

Irritable bowel syndrome (IBS) is a functional disorder of the gut-brain interaction,¹ characterised by abdominal pain and altered bowel habits. According to the Rome Foundation Global Study, the global prevalence of IBS is 4.1%. In China, prevalence varies according to diagnostic criteria, with a large national study reporting a pooled prevalence of 8.34% using Rome III criteria and 1.40% using the stricter Rome IV criteria.² The chronicity of symptoms significantly affects quality of life, work and social functioning.³ IBS is also associated with substantial economic costs, including direct medical expenses and indirect costs due to reduced productivity.^{2 4 5} In the USA, IBS-related costs have exceeded US$1 billion and are rising with the increasing disease burden.⁶ Therefore, effective IBS treatment is essential.

Gut-directed psychotherapies (GDPs), integrating cognitive behavioural therapy (CBT) and gut-directed hypnotherapy, can effectively alleviate symptoms in patients with IBS. Studies indicate that up to 80% of patients experience significant symptom relief following GDP, which can also reduce visceral hypersensitivity and anxiety by modulating the brain-gut axis,⁷ highlighting the bidirectional communication between the central nervous system and the gastrointestinal tract. The UK’s National Institute for Health and Care Excellence guidelines on IBS diagnosis and management recommend psychological interventions, such as hypnotherapy and CBT, as effective treatments for refractory IBS.⁸ However, face-to-face GDP has limitations that hinder its widespread implementation. These include high treatment costs, challenges posed by comorbidities in therapists and patients,9,11 a shortage of trained therapists, uneven geographical distribution and licensing regulations that restrict cross-border practice.^{12 13} Consequently, despite the demonstrated effectiveness of GDP, its practical application faces obstacles.9,11

Digital GDP is an emerging approach that can extend support beyond face-to-face settings,¹⁴ delivering core components of GDP, such as cognitive restructuring, emotion regulation training, gut health education, stress management techniques and mindfulness-based interventions via digital technology.¹⁵ The most apparent difference between digital and face-to-face GDP lies in the mode of interaction. Face-to-face GDP requires synchronous interaction between patients and therapists at a specific time and place, enabling therapists to assess patients’ emotional and cognitive states in person, adjust intervention strategies promptly and provide highly individualised guidance.¹⁶ In contrast, digital GDP relies on an online platform developed through collaboration between psychological experts and software engineers, and it has been applied in clinical practice.¹⁷ Digital GDP can be delivered through real-time online therapist guidance¹⁷ or fully automated programs.¹⁸ Both approaches are based on the principles of GDPs.¹⁹ The accessibility of digital GDP, facilitated by widespread internet use, can address the uneven distribution of treatment resources, such as the scarcity of specialist therapists.²⁰ Digital GDP requires less therapist time than face-to-face GDP, substantially reducing treatment costs.²¹ Digital GDP offers a viable alternative when face-to-face GDP is impractical due to geographical or time constraints. However, there is currently a lack of direct comparative studies assessing the efficacy of digital versus face-to-face GDP.

Several systematic reviews and meta-analyses have previously confirmed the overall efficacy of psychological therapies for IBS.22,24 While existing network meta-analyses, such as that by Goodoory et al,²³ have successfully ranked different brain-gut behavioural treatments, they generally group digital and face-to-face delivery modes or do not prioritise the direct comparison between these two formats. Furthermore, no previous synthesis has specifically compared the efficacy of digital and face-to-face delivery modes of GDP using a network meta-analysis framework.²⁴ This distinction is crucial for clinicians and policymakers making decisions about resource allocation and treatment accessibility. A direct comparison is needed to validate digital GDP as a scalable, cost-effective alternative that can expand service reach without compromising efficacy and to empower patients with evidence-based choices aligned with their lifestyles. Therefore, this study aims to fill this evidence gap by providing a direct comparative evaluation of digital and face-to-face GDPs for IBS based on available evidence, which will offer more nuanced insights for clinical practice.

Methods

This study was prospectively registered on OSF (DOI:10.17605/OSF.IO/87463) and will adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for network meta-analysis.²⁵

Data sources and searches

We will search English databases (PubMed, Cochrane Library, EMBASE and Web of Science) and Chinese databases (China National Knowledge Infrastructure, Wanfang, VIP and Chinese Biomedical Database) for all RCTs on digital or face-to-face GDP for IBS. The search strategy will employ Medical Subject Headings (MeSH), which is a hierarchically-organized controlled vocabulary maintained by the National Library of Medicine for indexing and retrieving biomedical information. These terms will be combined with entry terms to cover the period from database inception to May 2025. The study is planned to start in December 2025 and is expected to be completed by December 2026. We will also search ClinicalTrials.gov to identify additional relevant trials. Furthermore, we will examine the reference lists of relevant reviews and articles to identify potential studies. The data retrieval strategy from the PubMed database is shown in table 1.

Table 1. Search strategy in the PubMed database.

Number	Search terms
#1	Cognitive Behavioral Therapy (MeSH)
#2	Behavioral Therapies, Cognitive (title/abstract)
#3	Cognitive Behavioral Therapies (title/abstract)
#4	Cognitive Therapy (title/abstract)
#5	Cognition Therapies (title/abstract)
#6	Cognitive Behavior Therapy (title/abstract)
#7	Psychotherapy, Cognitive (title/abstract)
#8	Gut-directed hypnotherapy (title/abstract)
#9	Hypnotherapy (MeSH)
#10	Hypnotism (title/abstract)
#11	Self-Hypnosis (title/abstract)
#12	Autohypnosis (title/abstract)
#13	Gut-directed psychotherapy (title/abstract)
#14	Therapies, Cognitive Behaviour (title/abstract)
#15	#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14
#16	Irritable bowel syndrome (MeSH)
#17	Syndromes, Irritable Bowel (title/abstract)
#18	Irritable Colon (title/abstract)
#19	IBS (title/abstract)
#20	#16 OR #17 OR #18 OR #19
#21	Randomized Controlled Trial (title/abstract)
#22	Controlled Clinical Trial (title/abstract)
#23	Randomized (title/abstract)
#24	RCT (title/abstract)
#25	#21 OR #22 OR #23 OR #24
#26	#15 AND #20 AND #25

IBS, irritable bowel syndrome; MeSH, Medical Subject Headings; RCT, randomised controlled trial.

Eligibility criteria

Inclusion criteria

1. Study population: adults (aged ≥18 years) diagnosed with IBS according to expert opinion or Rome criteria.

2. Study design: only RCTs will be included. All patients must have received at least 4 weeks of treatment.

3. Intervention: any GDP used to treat IBS, delivered either face-to-face (individual or group) or digitally (eg, internet-based, smartphone applications or telephone-based). All patients must have received a full course of treatment as defined by the study protocol.

4. Comparisons: trials must compare psychological therapies with each other or with a control intervention. Eligible control interventions include waiting lists, usual care, education/support, active controls and sham therapy (eg, sham digital programmes, non-therapeutic apps or sham meditation) to ensure robust comparisons.

Exclusion criteria

1. Duplicate publications.

2. Studies combining multiple GDP interventions.

3. Reviews, commentaries, conference abstracts or case reports.

Outcomes

Main outcomes

The primary outcome is the efficacy of all psychological therapies and control interventions in IBS, in terms of effect on global IBS symptoms or abdominal pain after completion of therapy. We defined an improvement of ≥30% as a response.

Additional outcomes

We will also report adverse events occurring as a result of therapy (total number of adverse events, adverse events leading to study withdrawal and individual adverse events, if reported). Changes in gastrointestinal symptoms will also be taken into consideration.

Study selection

All retrieved studies will be imported into EndNote X9 (Clarivate Analytics, Philadelphia, Pennsylvania, USA), and duplicates will be removed. The screening process will involve three stages. First, two authors will independently screen titles and abstracts based on the eligibility criteria, excluding irrelevant studies. Second, the two authors will cross-check the screening results, resolving disagreements through discussion; any remaining discrepancies will be adjudicated by a third author. Third, studies meeting the eligibility criteria will undergo full-text review. Articles lacking sufficient data for meta-analysis will be excluded, as will abstracts and duplicate publications.

Data extraction

Two authors will extract the following information from each included study using a predesigned, standardised form: title, authors, publication year, country where the study was conducted, participant characteristics, intervention details and outcome measures (mean and SD for continuous data and number of events for dichotomous data). If both per-protocol and intention-to-treat data are reported, we will prioritise intention-to-treat data. If necessary data are unavailable in the published manuscripts, we will attempt to obtain them from the authors or the trial registry.

If the mean and SD are not provided directly in the study, they will be calculated using formulas from the Cochrane Handbook. The scale of all abdominal pain or IBS symptom scores will be converted to 0–100. The process of screening the garnered literature is shown in figure 1.

Figure 1. Flow diagram.

Quality assessment of evidence

The Cochrane Risk of Bias (RoB) tool (V.2) will be used to assess the RoB in each RCT. This tool comprises five domains: randomisation process, deviations from intended interventions, missing outcome data, measurement of the outcome and selection of the reported result.²⁶ Two authors will independently assess the RoB and cross-check their assessments, resolving disagreements through discussion. Comparison-adjusted funnel plots will be used to assess potential publication bias, and Egger’s test will be used for quantitative assessment of publication bias. The Confidence in Network Meta-Analysis (CINeMA) tool will be used to assess confidence in the network meta-analysis results. CINeMA is an online platform rooted in the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework—a standardized and transparent system used to appraise the certainty of evidence and strength of recommendations. It facilitates the rigorous evaluation of network meta-analysis results across six core domains: within-study bias, reporting bias, indirectness, imprecision, heterogeneity, and inconsistency.²⁷

Data synthesis and statistical analysis

The DerSimonian-Laird random-effects model will be used for traditional pairwise meta-analysis. Changes in pain and gastrointestinal symptoms (continuous outcomes) will be expressed as standardised mean differences, overall response (dichotomous variable) as relative risks and adverse events as ORs, with 95% CIs calculated. For studies with zero events in both groups, a continuity correction of 0.5 will be applied. Statistical heterogeneity between pooled data will be assessed using the I² statistic.

We will conduct multivariate network meta-analyses within a frequentist framework, using the mvmeta command in STATA V.16 software, and will generate network diagrams. Inconsistency will be tested using the node-splitting method. A p value of ≥0.05 will indicate good consistency, and the consistency model will be used for analysis. Heterogeneity in network models will be evaluated using τ² (with τ² values of <0.04 indicating low heterogeneity, 0.04–0.16 low-moderate, 0.16–0.36 moderate-high and >0.36 high heterogeneity).^{28 29} For each outcome measure, we will also use the surface under the cumulative ranking curve (SUCRA) to evaluate the magnitude and uncertainty of the interventions; larger SUCRA values indicate greater effect sizes.³⁰

Subgroup and sensitivity analyses

Based on previous IBS research,^{22 23 31} we anticipate that treatment duration, frequency and specific intervention details may be the primary sources of heterogeneity. Therefore, we will conduct subgroup analyses of GDP treatment duration and frequency, categorising them into low, moderate and high. We will also perform subgroup analyses based on the type of GDP. Additionally, we will conduct subgroup analyses based on different IBS diagnostic criteria to account for sample variations. To assess the robustness of our findings, we will also conduct a sensitivity analysis using a Bayesian Markov Chain Monte Carlo consistent model (implemented in GeMTC V.0.14.3 software).

Patient and public involvement

Patients and the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Ethics and dissemination

This study presents no ethical concerns, as the systematic review design does not require informed consent or direct participant involvement. The findings of the protocol and final review will be disseminated through publication in peer-reviewed journals.