Introduction

Inflammatory bowel disease (IBD) is a chronic and debilitating inflammatory condition of the intestinal tract comprising of two related conditions: Crohn’s disease (CD) and ulcerative colitis (UC). CD can affect any area between the mouth and the anus, whereas UC is restricted to the colon. Multiple factors (i.e. genetic predisposition, environmental or microbial factors and dysregulated immune response) are known to contribute to the aetiology of IBD (Zhang & Li, 2014; Singh & Bernstein, 2022). Currently, there is no available cure for IBD. A proportion of IBD patients require lifelong medical treatment along with surgery at some point in their life time (Kilcoyne et al., 2016). Individuals with IBD report poorer quality of life (QoL; Bennebroek Evertsz et al., 2012). Moreover, symptoms of common mental disorders are prevalent and persistent in patients with IBD, with up to a third of patients reporting symptoms of anxiety and a quarter reporting symptoms of depression over the past 10–15 years (Barberio et al., 2021). More specifically, about 35% of adult IBD patients suffer from comorbid depression, during exacerbation of the disease, compared to 20% during illness remission. Additionally, 66% of the IBD patients had comorbid anxiety during active disease, compared with 28% during inactive IBD (Mikocka-Walus et al., 2016).

In line with these observations, a previous study on IBD patients who reported a poor mental QoL found the prevalence of psychiatric disorders to be 70% (Bennebroek Evertsz’ et al., 2017). Furthermore, another study found that patients with high levels of anxiety and depressive symptoms and poor mental QoL did not receive the required care. Only 18% of them had been treated by a psychologist and 21% had received psychotropic medication (Bennebroek Evertsz et al., 2012). This reinforces the observation that IBD patients with anxiety and depression are often insufficiently treated (Greywoode et al., 2023; Mikocka-Walus et al., 2016). This can be deemed especially relevant seeing that as in the case of other somatic diseases, poor QoL, comorbid depression and anxiety can have adverse effects in the expression and trajectory of the disease (Katon et al., 2007). According to a recent systematic review and meta-analyses, individuals with a history of depression may show an increased risk of IBD, even when the depression is diagnosed several years before the new-onset IBD (Piovani et al., 2023). Association between depression, anxiety and disease activity is well established, but the causal pathway is still unclear (Mikocka-Walus et al., 2016). Several studies found evidence for a bidirectional link between the brain–gut axis (Gracie, et al., 2018). This bidirectional link has been shown to affect the natural history of the disease along with psychological health (Fairbrass et al., 2022).

Cognitive behavioural therapy (CBT) has been shown to be effective in reducing anxiety and depressive symptoms in patients with chronic illnesses (Van Straten et al., 2010). Bennebroek Evertsz’ et al. (2017) showed in a randomized controlled trial that ‘IBD-specific CBT’ led to the reduction of symptoms of anxiety and depression, as well as an improved disease-specific and general QoL in IBD patients.

Recent meta-analyses and reviews point towards the short-term effectiveness of CBT for IBD on mental health (such a QoL, anxiety and depression) (Chen et al., 2021; Li et al., 2019; Reise-Filteau et al., 2021). In the current study, we evaluate whether ‘IBD-specific CBT’ from the benchmark study can be delivered and implemented with the same effectiveness in hospital settings, outside the research context of a RCT. Moreover, ‘IBD specific CBT’ is only available in a few specialized hospitals in the Netherlands and for many IBD patients it is not available. This study will therefore evaluate the treatment results of ‘IBD-specific CBT’ implemented in several clinical hospital settings in comparison with the aforementioned RCT (Bennebroek Evertsz’ et al., 2017), following the so-called “benchmark” strategy (Scheeres et al., 2008; Wiborg et al., 2014).

Implementation Studies

Implementation is an essential process to enable the dissemination of new therapies and treatment manuals within clinical routine, although this is often a challenge for healthcare institutions. It has been estimated that approximately 70% of healthcare organizations’ efforts to implement new treatments fail (Beer & Nohria, 2000), and thus it becomes crucial to understand the barriers and facilitators for implementation. These can occur at various levels: At an individual level, i.e. the professionals and patients (e.g. knowledge, skills or motivation) but also at broader social, organizational, economic and political levels (e.g. administration of care processes and policies). Furthermore, the design of the innovation itself (e.g. feasibility and attractiveness) (Grol & Wensing, 2004, 2005) can both facilitate and impede implementation. Many interventions found to be effective in research settings remain unused and do not end up being applied to the patients who need it (Damschroder et al., 2009). A carefully planned and monitored implementation process is required to reach this aim. Screening of high-risk groups of IBD patients allows us to recognize and treat those in need of mental care and treatment. Therefore, we preselected IBD patients with poor mental QoL and unmet care needs, like we did in the benchmark RCT (Bennebroek Evertsz’ et al., 2017). Not all IBD patients in the benchmark study with poor mental QoL indicated a need for psychological care for several reasons. The pre-treatment attrition rate (eligible people who decided to not participate in the study) was 67%. Furthermore, 17% of the patients dropped out during treatment (Bennebroek Evertsz’ et al., 2017). In the current implementation study, we compared the attrition rates (pre-treatment and during treatment) of the implementation study with those of the benchmark RCT (Bennebroek Evertsz’ et al., 2017).

It is important to consider that patient attrition is an expected outcome in trial studies and implementation processes and could be an important source of bias in evaluations if not properly accounted for (Dumville et al., 2006).

Current Study’s Objectives

Based on Bennebroek Evertsz’ et al. (2017), the current study examined the dissemination and implementation process of ‘IBD-specific CBT’ for QoL, anxiety and depression in a multi-centre setting. The study had the following objectives: (1) Evaluate the effectiveness of the implementation of ‘IBD-specific CBT’ on QoL, anxiety and depressive symptoms among patients with poor mental QoL; (2) Explore facilitators and barriers to the implementation process in different clinical settings by semi-structured interviews; (3) identify reasons of non-participation (pre-treatment) and attrition during treatment of the implementation study.

Methods

Study Design

This study employed a mixed methods approach to the implementation of ‘IBD-specific CBT’. It consisted of quantitative analysis, with a pre- and post-treatment assessment, and qualitative data collection and analysis happening simultaneously with data being integrated in a final stage. The methods used included a benchmark statistical analysis of treatment outcomes, semi-structured interviews with a sub-group of participating therapists and gastroenterologists, and an analysis of patient non-participation and attrition.

Patients and Recruitment

The departments of gastroenterology, hepatology and medical psychology of four participating hospitals (one academic (Amsterdam UMC) and three general hospitals (Noordwest Ziekenhuisgroep (Alkmaar), Isala (Zwolle) & Ziekenhuis Gelderse Vallei (Ede)) participated in the implementation study. IBD patients attending the outpatient clinic of the gastroenterology and hepatology department of any of the four hospitals were screened by gastroenterologists or IBD nurse specialists. Eligible patients had to meet the following inclusion criteria, as in the original benchmark RCT (Bennebroek Evertsz’ et al., 2017): (1) Diagnosis of Crohn’s disease (CD) or ulcerative colitis (UC), with or without inflammation; (2) age above 18 years; (3) score of ≤ 23 on the mental health sub-scale of MOS Short Form 36 (SF-36); (4) physically and mentally able to attend eight weekly sessions and (5) sufficient command of Dutch. Exclusion criteria were: (1) current psychotherapy and (2) other severe psychiatric diagnoses (e.g. substance use, bipolar disorder or psychoses). Patients who met the aforementioned criteria received additional information regarding the study design and were asked for informed consent. Patients were then approached by a specialized IBD nurse or a front desk assistant (part of the study team), who collected additional information on patients’ characteristics by administering a case report form (CRF). Thereafter, patients were scheduled with an assigned psychologist of the department of medical psychology from the participating hospitals.

Treatment Manual

The ‘IBD-specific CBT’ intervention used the same treatment manual as the benchmark study, designed and described in Bennebroek Evertsz’ et al. (2012). Treatment consisted of eight hourly sessions to be completed on a weekly basis, each lasting one hour. In the first session of ‘IBD-specific CBT’, therapists determined if patients met the criteria of anxiety and/or depression of the DSM V axis I disorders (See Table 1 overview treatment). A treatment manual based on the CBT model (Beck, 2005) was used to enhance treatment integrity. Every participant received three modules; writing assignments with regard to traumatic experiences in their illness history; cognitive interventions focussed on specific illness beliefs (for example ‘my illness has major consequences for life’ or ‘my illness causes difficulties for those who are close to me’) and on dysfunctional attitudes (for example ‘I have to be perfect to be happy’) along with a relapse prevention plan (for an extensive description see Bennebroek Evertsz et al., 2012). There were two additional optional modules depending on the specific psychiatric disorder: (1) a behavioural activation and an exposure-based intervention including response prevention (for depressive and anxiety disorders, respectively) and (2) imagery and rescripting (for posttraumatic stress disorders). In case of complaints of incontinence or uncontrollable diarrhoea attacks, we asked patients to compose a survival backpack (i.e. clean clothes, pants, spray, toilet paper, wipes, nappy, bin bag). The aim of this intervention is to diminish the patient's avoidance behaviour.

Table 1 Overview Treatment
Full size table

Skilled CBT therapists specialized in treating chronically ill patients with comorbid psychiatric disorders were trained in the application of the treatment manual‘IBD-specific CBT’ for two days with six-hour sessions each. Information about IBD was given and the ‘IBD-specific CBT’ treatment manual was shown alternatingly with video fragments of the treatment of IBD patients and role play, in which therapists practiced with specific interventions tailored for IBD (for example talking about faecal incontinence and introducing the ‘survival backpack’ and exposure to a diarrhoea attack in public using imagery and rescripting). After the training, the participating therapists received supervision on demand.

Measurements

Patients were asked to fill in questionnaires online before and after the CBT treatment. Follow-up measurements took place within a month after the completion of the last therapy session.

The primary outcome was the total score on the Inflammatory Bowel Disease Questionnaire (IBDQ), which comprises 32 items assessing the following four domains: bowel symptoms, systemic symptoms, and emotional and social functioning (Russel et al., 1997).

The secondary outcomes were depression and anxiety, which were assessed using the Hospital Anxiety and Depression Scale (HADS; Zigmond & Snaith, 1983), a measure specifically designed for patients with a somatic illness. A score ≥ 8 on either sub-scale indicates a possible anxiety and/or depressive disorder and a score ≥ 11 indicates a probable disorder in the same category. The Centre for Epidemiologic Studies Scale (CES-D) measuring depressive symptomatology in the general population was additionally used (Radloff, 1977).

Ethics Approval

This implementation study design has been approved by the Medical Ethics Committee of the Amsterdam UMC (location AMC: dossier number: XT4-148).

Statistical Analysis

Quantitative Analysis: Evaluation of Effectiveness of Therapy

All statistical analyses were conducted in IBM SPSS Statistics. A statistical benchmark analysis (see Scheeres et al., 2008; Wiborg et al., 2014) was performed comparing the outcomes of the current implementation study with those obtained in the context of a randomized clinical trial as reported by Bennebroek Evertsz’ et al. (2017). Employing the original data from the prior trial study, an analysis of covariance (ANCOVA) was performed using the difference in mean scores of the primary and secondary outcomes across two conditions, i.e. the experimental group of the trial study and the treatment group in the implementation study. Analyses were adjusted for age, sex and diagnosis (M Crohn, colitis ulcerosa). Analyses were done from an Intention-to-Treat (ITT) approach to avoid an over-estimation of the treatment effect. In addition, per protocol (PP) analyses were performed using the data of patients who attended at least five CBT sessions. Multiple imputation using chained equations (MICE) was employed to handle missing values, combining results from 5 imputed data sets into pooled estimates (White et al., 2011). To compare the relative magnitude of the effects, mean scores were standardized on the continuous outcomes to Cohen’s d using the pooled standard deviation of the baseline scores. Cohen’s d of 0.3, 0.5 and 0.8 indicating a small, moderate and large effect sizes (ES), respectively, were calculated. Effect size estimates were presented with a 95% confidence interval (CI). The level of significance (alpha) was set at a two-sided 0.05. We compared the percentage of patients who dropped out (pre-treatment and during treatment) between the benchmark RCT and the implementation study using Chi-squared tests.

All analyses were repeated omitting data of the therapist from the Academic Medical Center with the most experience in implementing the psychological intervention previous to the study (i.e. who designed the original ‘IBD-specific CBT’ treatment manual from the benchmark study, Bennebroek Evertsz’ et al. (2017).

Qualitative Analysis: Semi-Structured Interviews

Qualitative analysis of the implementation process is important to reveal strengths and limitations of the treatment manual and its implementation. Considering contributions of prior implementation studies (van der Vaart et al., 2019), a semi-structured interview for therapists was designed taking into account the following dimensions: the value of the therapy, factors of success, use of the treatment manual, patient heterogeneity, implementation challenges and recommendations for improvement (See Table 2). Furthermore, an interview for gastroenterologists was constructed, considering the screenings procedure, referral of patients, future improvements and the usefulness of ‘IBD-specific CBT’ for patients with poor OoL anxiety and depressive complaints (See Table 3). Therapists and gastroenterologists involved in the implementation of the ‘IBD-specific CBT’ treatment manual of the four participating hospitals were invited to be interviewed. Interviews were administered via email, telephone or face-to-face, according to the preference of the respondent. A thematic analysis was performed by the first author (Bennebroek Evertsz’) on the collected data in order to explore different dimensions of the implementation. The interviews of therapists and gastroenterologists were analysed separately.

Table 2 QL!C Implementation study – Semi-structured Interview for therapists
Full size table
Table 3 QL!C Implementation study – Semi-structured Interview for gastroenterologists
Full size table

Results

Sample Characteristics

Patients were recruited over a period of one and a half years between November 2018 and March 2020. We screened 280 patients for eligibility. Of these, 94 were eligible (meeting inclusion criteria a score of ≤ 23 on the mental health sub-scale of the SF-36 and willing to participate) and were allocated to the implementation study. At follow-up, 63 participants (67% compared to 81% for the benchmark RCT study) provided data for analysis for the IBDQ Total score (see Fig. 1).

Fig. 1
figure 1

Flow diagram for inflammatory bowel disease patients through the implementation study, including attrition. SF-36 = Medical Outcomes Study Short Form 36 Health Survey; CBT = cognitive behavioural therapy; IBDQ TOT = Inflammatory Bowel Disease Questionnaire total score; IBD = inflammatory bowel disease)

Full size image

Of the total group of patients in the implementation study, 26 patients were treated with ‘IBD-specific CBT’ (> = 5 sessions) in the University hospital and 37 patients were treated in the three general hospitals.

In the implementation study, a mean of 5.8 sessions (SD = 3.2) were completed, whereas 6.1 (SD = 3.0) sessions in the benchmark study (p = 0.513) were completed, with 66.3% versus 74.6% of all participants (59 of 89 vs 44 of 59) completed at least five sessions (p = 0.283).

The first participating hospital screened fifteen eligible patients (i.e. scored ≤ 23 on the mental health sub-scale of the SF-36), of whom nine started ‘IBD-specific CBT’ by participating therapists. In total five patients completed ‘IBD-specific CBT’ (followed more than five sessions). Four patients dropped out during treatment and ended therapy after one session; of them two patients had severe psychiatric disorders (i.e. PDD NOS) and two patients did not have enough complaints (no help requests). Six patients dropped out before the treatment started: one patient had no time/logistical problems, one patient was receiving psychotherapy elsewhere, three patients were not contactable and one patient was referred to a sleeping clinic because of problems with fatigue (other reason).

The therapists from this participating general hospital noticed that many of the patients referred by the specialized IBD nurse had other needs than ‘IBD-specific CBT’ (i.e. they had no need for treatment for anxiety and depression disorders, post-traumatic stress disorders, adjustment disorders and poor QoL). Initially, they proposed to extend the inclusion period by a half-year. Finally, the implementation at this hospital was stalled and has not been continued.

The second participating hospital screened thirty eligible patients (i.e. scored ≤ 23 on the mental health sub-scale of the SF-36), of whom twenty-eight patients started ‘IBD-specific CBT’ with participating therapists. In total, twenty-five patients completed ‘IBD-specific CBT’ (followed more than five sessions). Three patients dropped out during treatment (after less than five sessions); one dropped out after one session due to COVID and two others dropped out after one session due to other problems (other reasons). Two patients dropped out before starting treatment; one patient had no time/logistical problems and one patient did not have enough complaints.

The third participating hospital screened twenty eligible patients (i.e. scored ≤ 23 on the mental health sub-scale of the SF-36), of whom nine started ‘IBD-specific CBT’ by participating therapists. In total, seven patients completed ‘IBD-specific CBT’ (followed more than five sessions). Two patients dropped out during treatment (after less than five sessions); one had COVID after three sessions and one dropped out after two sessions because of other problems (other reasons). Eleven patients dropped out before starting treatment; two patients were receiving psychotherapy elsewhere, four patients did not have enough complaints, one patient was not contactable, two patients had COVID, one patient had inflammation problems (other reason) and one patient dropped out for unknown reason.

The last participating hospital screened ninety-six eligible patients (i.e. scored ≤ 23 on the mental health sub-scale of the SF-36), of whom forty-eight patients started ‘IBD-specific CBT’ by participating therapists. In total, twenty-six patients completed ‘IBD-specific CBT’ (followed more than five sessions). Twenty-two patients dropped out during treatment (after less than five sessions); of them three stopped because of no time/logistical problems, three did not have enough complaints, four had severe mental complaints, four had COVID, four dropped out because of other problems (other reasons) and four dropped out for unknown reasons. Forty-eight patients dropped out before starting treatment; seventeen patients had no time/logistical problems, seven patients were receiving psychotherapy elsewhere, six patients did not have enough complaints, six patients were not interested, one patient was not contactable, three patients dropped for other reasons and eight patients dropped out for unknown reasons.

Table 4 provides the sociodemographic and clinical characteristics of the implementation ITT group and the benchmark RCT experimental ITT group. There were significantly more women *(0.8% vs 66.1%) in the implementation study, patients were more often in a relationship (in a relationship 69.0% vs 50.8%) and treated in a university hospital (45.7% vs 67.8%) (see also Table 4).

Table 4 Comparison of study characteristics of the implementation ITT group and benchmark Randomized Controlled Trial experimental ITT group
Full size table

Treatment Effect

Both in the intention-to-treat and per protocol analyses, the only statistically significant difference between the benchmark RCT and the implementation study at follow-up was the mean score on the IBDQ sub-scale emotional following treatment (see Tables 5 and 6).

Table 5 Mean scores and mean differences between the experimental and implementation group at (3.5 months) follow-up according to intention-to-treat analysis ITT
Full size table
Table 6 Mean scores and mean differences between the experimental and implementation group at (3.5 months) follow-up according to Per Protocol analysis completers*
Full size table

All outcome measures improved significantly over time from baseline to 3.5 months of follow-up (see Tables 7 and 8).

Table 7 ITT analyses completers* implementation group; changes over time between baseline and at 3.5 months of follow-up
Full size table
Table 8 Per Protocol analyses completers* implementation group; changes over time between baseline and at 3.5 months of follow-up
Full size table

Following CBT, the IBDQ Total score enhanced significantly in the implementation ITT group with a moderate effect size (ES = 0.75), compared to the moderate effect size (ES = 0.64) of the benchmark RCT study (see Table 7). CBT had a significant improvement on the IBDQ sub-scales Systemic and Emotional, with moderate (ES = 0.57) and large effect sizes (ES = 0.83), respectively, compared with moderate (ES = 0.49) and large effect sizes (ES = 0.92) for the benchmark study. The results on the Bowel and Social sub-scales were significantly enhanced with small (ES = 0.40) to moderate effect sizes (ES = 0.56), compared to insignificant results of the benchmark RCT study with small effect sizes (ES = 0.33 and ES = 0.27, respectively).

Following CBT, the HADS Total score and the anxiety and depression sub-scales decreased significantly, all with a moderate effect size (ES = 0.76, ES = 0.67, ES = 0.70) compared to the moderate effect sizes of the benchmark RCT study (ES = 0.54, ES = 0.54, ES = 0.42). Similarly, CBT significantly reduced the CES-D scores, with a moderate effect size (ES = 0.76) compared to a fairly large ES of the benchmark RCT study (ES = 0.78) (see also Table 7). Following imputation, the pattern of results was similar (see also Table 7).

When we restricted the analyses to patients who attended at least 5 CBT sessions, the PP sample of the implementation group, results were similar to those in the ITT implementation group (see also Table 8). No adverse events related to CBT were identified in the implementation study.

We repeated all analyses without the therapist from the Academic Medical Center who had the most experience in implementing the ‘IBD-specific CBT’ treatment manual. The pattern of results (both significance and effect sizes) was similar. The effect sizes were slightly lower (data not shown).

The only different result is a significant effect in the mean differences between the experimental and the implementation group in the ITT analyses (not imputed) (resp. 12.02 vs 16.34), p = 0.008 and in the PP analyses (imputed) for the CES-D (resp. 12.84 vs 15.70), p = 0.028. In both cases, the mean on the CES-D in the implementation group was higher than in the RCT (data not shown).

We compared outcomes of participants who did versus did not receive CBT during the COVID-19 using ANCOVA. We did not find a reduced effect of the CBT for patients who were confronted with the pandemic during their treatment, compared to patients who were treated before the COVID-19 outbreak (data not shown).

Semi-structured Interview with Gastroenterologists and IBD Nurse Specialist

Three gastroenterologists and one specialized IBD nurse from three participating hospitals were interviewed (one working at the gastroenterology and hepatology department of the University hospital, and three working at the gastroenterology departments of the general hospitals) (see Table 9).

Table 9 Overview interview answers gastroenterologists 30-10-2023
Full size table

The gastroenterologists believed that CBT is necessary and helpful for patients with Inflammatory Bowel Diseases and a poor quality of life, depression and/or an anxiety disorder. They further remarked that it was beneficial for future quality of life and also improved compliance with medical treatment. Gastroenterologists argued that nurses are the professionals most suitable to screen IBD patients on poor QoL on a regular basis because they have got more time with patients to discuss psychological issues.

Two of the gastroenterologists did not experience any difficulties in the process of referring patients to the therapists of the department of medical psychology. One of them mentioned the lack of reimbursement of the department of medical psychology and lack of attention by the caregivers as an explanation for missed referrals of patients with mental disorders in the past.

Recommendations: paramedical care by a clinical nurse specialist, specializing in IBD care, should be the standard of care for IBD patients. A uniform workflow for a daily routine screening for anxiety and depression of IBD patients by a nurse specialist would be beneficial for the implementation of the CBT. Standard screening with the SF-36 could help overcome noticed time constraints, especially as the treating physician or nurse specialist would be alerted when the patient scores low (≤ 23) on the mental health sub-scale of the SF-36.

Semi-structured Interviews with Four Therapists

A total of twelve therapists were trained in the ‘IBD-specific CBT’ according to the treatment manual. All therapists received an invitation to be interviewed. Due to time constraints of most therapists, four therapists were interviewed from three out of four participating hospitals. The ‘IBD-specific CBT’ is considered successful and valuable for patients by the participating therapists. The therapy is not very strict and contains many different modules to fit the needs of individual patients. All modules were used (i.e. writing assignment, exposure part in imagery and rescripting). The cognitive therapy modules were found to be the most effective and were applied most. The strength of the treatment manual, in the experience of the therapists, is the availability of different modules which enables personalization. Also, the treatment manual is patient group specific (IBD), which can provide direction to therapists with respect to medical questions and complaints. However, according to one of the therapists, a limitation of the treatment manual is that there is a lack of psychoeducation about fatigue in the treatment manual.

The screening workflow (by gastroenterologists or IBD nurse specialists) also makes it more acceptable for patients to receive help from medical psychology, and it increases the contact between patients and the ‘MDL’ department. Therapists were not always able to follow the frequency of the sessions in the treatment manual as planned, (8 weekly sessions) because of full clinics, patients’ holiday plans or surgical operations. Therefore, some therapists introduced modifications like extending the duration of the treatment by changing the frequency of the sessions. Moreover, the number of sessions was not ideal for every patient, some patients needed more or less sessions than eight.

There are particular circumstances that make some IBD patients more or less challenging to treat according to the treatment manual than others such as remission of disease inflammation, comorbidity, severe psychiatric disorders (i.e. psychosis, addiction) or insufficient suffering. In some cases, problems unrelated to IBD were the reason patients scored high on depression or anxiety questionnaire (HADS).

There were a few other challenges faced by the therapists while implementing ‘IBD-specific CBT’. Therapists of different hospitals had limited contact with therapists of other participating hospitals. It was indicated by them that they would prefer more contact with other participating therapists (Table 10).

Table 10 Overview interview answers therapists 30-10-2023
Full size table

Patient Attrition Before Treatment and During Treatment

During the pre-treatment phase in the benchmark study, 67% dropped out for several reasons. Most of the patients refused help because they experienced barriers to obtaining help due to logistical and/or time problems. Some patients reported having no mental complaints and were not motivated to receive help or they received already psychological help. Some patients were physically not able to attend. In the implementation, study the percentage of non-participation (pre-treatment dropout) was 41.6% (of the 161 patients eligible for CBT, only 94 patients started CBT). Again, the most common reasons for non-participation were having no time/logistical problems (n = 19), having no mental complaints (n = 11) or receiving already psychological help (n = 10), were not interested (n = 6), not contactable (n = 5), other reasons (n = 7) or unknown (n = 9) (see also Fig. 1).

In the case of ‘IBD-specific CBT’, the benchmark RCT reported an attrition of 17%. The most common reasons for dropping out were no time/logistical problems and further health complications such as IBD inflammation. In the implementation study, the attrition rate was 33% (of the 94 patients eligible for ‘IBD-specific CBT’, 63 patients finished treatment). Reasons were having no time/logistical problems (n = 3), having no mental complaints (n = 5), having severe mental complaints (n = 6) or COVID-19 (n = 6), other reasons (n = 7), unknown (n = 4) (see also Fig. 1).

In summary, the pre-treatment attrition percentage was significantly lower in the implementation study compared to the benchmark RCT (41.6% vs 67%, p < 0.001).

The attrition percentage during treatment was significantly higher in the implementation study compared to the benchmark RCT (33% vs 17%, p = 0.046).

Discussion

The main aim of this study was to evaluate the effectiveness of ‘IBD-specific CBT’ on QoL, anxiety and depressive symptoms among patients with poor mental QoL when implemented in routine clinical care in hospitals.

The treatment effects of the implemented ‘IBD-specific CBT’ were comparable to those found in the benchmark RCT. Following ‘IBD-specific CBT’, specific IBD-related QoL improved significantly with a moderate effect size in the implementation group (ITT and PP). ‘IBD-specific CBT’ also attenuated anxiety and depressive complaints in the implementation group (ITT and PP), all with a moderate effect size. Effect sizes were similar to those of the benchmark RCT study.

In general, the treatment result of the implementation of ‘IBD-specific CBT’ was successful. In addition, there were relatively lower levels of eligible patients not starting CBT in the implementation group but a higher dropout of patients during treatment compared to the benchmark study.

The favourable effects of ‘IBD-specific CBT’ found in this study are in line with the findings of studies showing positive effects of CBT in patients with IBD (Chen et al., 2021; Reise-Filteau et al., 2021). However, it is unclear to what extent these favourable effects of CBT directly post treatment are sustained.

Previous research mentions the need for individualized interventions (McCombie et al., 2016). This is in line with the current study focussing on personal therapy goals of patients, challenging and changing basic living rules, and elective interventions (e.g. behavioural activation, exposure and response prevention for anxiety and depression (Gregory, 2021)).

Additionally, previous research proposes that the brain–gut axis consists of bidirectional links and thus not only would IBD have an effect on psychological state but also the other way around. This calls for equal focus on mental health alongside the physical well-being of patients of IBD in the form of collaborative care (Fairbrass et al., 2022; Lores et al., 2021; Peppas et al., 2021).

Obstacles and Limitations

Unfortunately, the limited treatment capacity of the therapists was a barrier in one of the hospitals. When implementing the CBT treatment manual, it is therefore important to set aside time in the schedule of the professionals in advance.

Moreover, some therapists had several questions during supervision about the selection and motivation of patients and the screening workflow. They indicated that there were many interventions per session in the treatment manual so they had a hard time conducting all interventions in time, whereas other therapists managed to conduct all interventions within the preplanned number of sessions.

The aforementioned problems could have been solved during pro-active consultations. More regular group supervision with all participating therapists would have allowed for better screening and selection of motivated patients, taking into account the exclusion criteria (e.g. severe psychiatric problems). In the University hospital, there was weekly supervision as part of the multidisciplinary meeting, if necessary. In the other general hospitals this could also have been better organized: a more pro-active consultation by the supervisor and not waiting for the therapist's wish and initiative could be beneficial.

The recommendation for regular supervision to increase adherence to the implemented treatment manual is also mentioned in the previous literature (Kramer & Burns, 2008). Kramer and Burns (2008) investigated factors influencing the success of the implementation of CBT for depressed adults in two healthcare centres, via post-study qualitative interviews with clinicians. Potentially crucial factors to increase adherence to the treatment manual are regular supervision and team meetings, which can aid in achieving a shared vision. This and organizational factors combined could facilitate a more effective implementation (Bruijniks et al., 2018).

During the implementation period, the World Health Organization (WHO) declared a global pandemic of COVID-19 on 11 March 2020 (Cucinotta & Vanelli, 2020). This global pandemic required containment measures to prevent contamination, like social distancing, quarantine and self-isolation. The effect of psychosocial consequences of this global pandemic can be particularly serious for people who are already affected by mental health problems, e.g. anxiety and depression (Fiorillo & Gorwood, 2020; Kumar & Nayar, 2020). As a result, the influx of patients slowed down. Fortunately, we moved on by giving online treatment instead of face-to-face. It was favourable that the digital ‘IBD-specific CBT’ treatment during the pandemic was as effective as the face-to-face ‘IBD-specific CBT’ treatment before the COVID outbreak.

A few more limitations of the screening workflow and the ‘IBD-specific CBT’ treatment manual came to light during this study.

To start with, the frequency of weekly therapy sessions was not feasible in reality, because of full clinics, patients’ holiday plans or surgical operations, inflammations, etc. Therefore, the duration of the trajectory differed between patients (a range from 8 weeks till 24 weeks). Flexible scheduling of the therapy sessions would be more convenient than a fixed duration of the trajectory.

Secondly, screening through the electronic patient files of the participating hospitals or standard online examination prior to consultation instead of screening during face-to-face consults by nurses or nurse specialists could help overcome noticed time constraints. Especially when the treating physician or nurse specialist receives a message in the software system when a patient scores low on the mental health sub-scale of the SF-36 (≤ 23).

Thirdly, in the benchmark RCT study, the Generic health status was assessed with the SF-36 (Ware, 1992); the 36 items can be aggregated into a Physical Component Summary (PCS) score and a Mental Component Summary (MCS) score. However, this questionnaire was not used as an outcome measurement in our implementation sample. As a consequence, we have no results regarding this outcome (SF-36, Ware, 1992) in the implementation study.

Lastly, a limitation of the qualitative part of the research is that it is explorative and descriptive on the basis of interviews of only a sub-group of therapists and gastroenterologists and one specialized IBD nurse.

Strengths and Facilitators

A strength of this study is the distribution of the implementation over one University hospital and three general hospitals, facilitating generalization of the results and comparison across healthcare settings.

This study contains quantitative research by the means of measuring the effectiveness of the implementation by the number of successful treatments, as well as qualitative research by post-study qualitative semi-structured interviews with gastroenterologists, an IBD nurse specialist and therapists. Both the research combined gives a good inclusive overview of the success of the implementation.

Post-study qualitative semi-structured interviews with gastroenterologists and therapists showed that they believe this ‘IBD-specific CBT’ is necessary and/or helpful for IBD patients with poor mental quality of life, depression and/or an anxiety disorder.

Patient Attrition

One of the aims of the current study was to investigate the reasons behind patient non-participation (pre-treatment dropout) or attrition during treatment in the implementation group. Compared to the benchmark study, the pre-treatment dropout percentage of IBD patients in the implementation study was significantly lower (respectively, 67% vs 42%). A possible explanation for this is that in the pre-selection of patients in the benchmark RCT, a trained psychologist carried out a telephone version of the validated SCID-I (First et al., 1999) to determine severe psychiatric disorders. In the implementation study, this SCID interview is omitted, because of time constraints. As a consequence, less patients in the implementation study compared to the benchmark study dropped out before treatment. Possibly, dropout was delayed during treatment in the implementation study. This is consistent with the finding that the attrition during treatment in the implementation study was significantly higher compared to the benchmark study (respectively, 33% vs 17%).

Another reason for the last difference was that six patients were affected with COVID in the implementation group and had to discontinue treatment. When leaving out these six patients, the percentage of attrition during treatment in the implementation group would become more or less equal (26.5%) to the average percentage of dropout during treatment mentioned in the previous literature.

A meta-analysis regarding the implementation of CBT conducted by Fernandez et al. (2015) on the dropout data of 115 primary empirical studies showed an average pre-treatment dropout rate of 15.9% and of 26.2% during treatment. Dropout was seen to be significantly associated to the following: the type of diagnosis, having depression the highest attrition rate; the format of delivery treatment; the treatment setting, seeing fewer dropouts in inpatient than in outpatient settings; and the number of sessions, being there a reduction in attrition rates of treatment starters as the number of sessions increased.

Practice Implications and Future Recommendations

Arranging more frequent and structured supervision of the participating therapists (and gastroenterologists and specialized IBD nurses) during the implementation of the ‘IBD-specific CBT’ would be beneficial for overcoming the earlier mentioned challenges. Moreover, regular group supervision with more opportunities for evaluation and consultation would be beneficial and presumably make the implementation more successful. Important issues include that the participants have insufficient suffering. On the other hand, the treatment manual has a limit on the severity of psychiatric problems (i.e. psychosis).

Although the current ‘IBD-specific CBT’ treatment manual is suitable for all IBD patients with poor mental QoL and in need of help, not every IBD patient (during a period of disease activity) will be physically able to attend face-to-face sessions. Especially for those IBD patients, an internet-based self-help manual might be of interest. Symptoms like bloody diarrhoea, abdominal pain, fatigue, flares and fistulas discourage some patients from visiting therapists. Therefore, a substantial proportion of IBD patients might prefer online self-help treatment during disease flare activity.

Finally, future research should focus on the long-term effects of CBT for IBD patients with anxiety and depression (Chen et al., 2021). This also applies to the effects of our ‘IBD-specific CBT’.

Conclusion

‘IBD-specific CBT’ can be successfully implemented in both general and University medical settings. Regular group supervision of psychologists performing the ‘IBD-specific CBT’ treatment is recommended.