Feasibility and effectiveness of the Bergen 4-Day Treatment for obsessive-compulsive disorder in Australia: A pilot comparison with 3-week inpatient obsessive-compulsive disorder treatment Authors: Nathan L Dowling, Scott Blair-West, Keong Yap, Priscilla Yardley, Lachlan Cribb, Sophie Clement, Kerry Webster, Sebastian Wise, Thröstur Björgvinsson, Bjarne Hansen, Chee H Ng Published: 2026 ABSTRACT AbstractObjectives:Obsessive-compulsive disorder is a debilitating and chronic condition that, when untreated or unresponsive to treatment, imposes a significant health and economic burden on individuals and families. This prospective non-randomised inpatient study compared the acceptability and clinical outcomes of the Bergen 4-Day Treatment programme with those of a standard 3-week specialised treatment programme for obsessive-compulsive disorder in Australia.Method:Twenty-five participants diagnosed with obsessive-compulsive disorder were non-randomly assigned to Bergen 4-Day Treatment (n = 12) or a 3-week standard (n = 13) inpatient programme. Independent assessments were completed at pre-treatment, 10 days post treatment and at 3-month follow-up. The Yale-Brown Obsessive-Compulsive Scale was rated to assess obsessive-compulsive disorder severity, while secondary measures of depression, anxiety, obsessive beliefs and wellbeing were self-rated by participants.Results:Baseline characteristics of both groups were comparable, with obsessive-compulsive disorder symptom severity within the moderate to severe range. After treatment, obsessive-compulsive disorder symptoms as well as secondary depression and anxiety symptoms were reduced in both treatment groups. Participants receiving Bergen 4-Day Treatment had significantly lower Yale-Brown Obsessive-Compulsive Scale scores at 10 days (M = 13.46) and 3 months (M = 11.84), compared to standard treatment (M = 19.04 and M = 19.15, respectively). Response (91.9%) and remission (45.8%) rates for the Bergen 4-Day Treatment group were significantly higher at both post-treatment timepoints, compared to the standard treatment group. No dropouts occurred in the Bergen 4-Day Treatment group, and participant satisfaction was high.Conclusion:The findings of this pilot open-label study suggest that Bergen 4-Day Treatment shows promise as an acceptable, efficient and effective treatment for obsessive-compulsive disorder, warranting further investigation as a scalable alternative for improving access to specialised obsessive-compulsive disorder treatment in Australia.Keywords: Inpatient treatment, obsessive-compulsive disorder, exposure and response prevention, Bergen 4-Day Treatment, cognitive-behaviour therapy Obsessive-compulsive disorder (OCD) is a disabling and chronic mental disorder that imposes a substantial burden on individuals, their families and health systems. It is characterised by unwanted, recurrent and intrusive thoughts and images (obsessions), in which individuals feel driven to perform excessive repetitive behavioural and mental acts (compulsions) to neutralise distress or feared outcomes (American Psychiatric Association, 2022). The lifetime prevalence of Australians with OCD is 2–3% and generally follows a chronic and non-remitting course, which can significantly impact wellbeing, quality of life and functioning, especially when left untreated (Perris et al., 2023). Although the efficacy of existing OCD interventions, including cognitive-behavioural therapy (CBT), exposure and response prevention (ERP; Öst et al., 2022) and pharmacological treatments (Skapinakis et al., 2016), is well established, most individuals with OCD (40–60%) experience significant enduring symptoms, with a high risk of relapse (Eisen et al., 2013). Evidence-based OCD treatments in Australia remain underutilised and unavailable locally (Cooper et al., 2023) because of the high treatment costs, low awareness about available interventions and a few state-wide or national services for OCD treatments in Australia (Dyason et al., 2022; Gentle et al., 2014). In many cases, access to specialised care, optimal treatment approaches or intensive programmes for OCD remains lacking (Rowe and Kangas, 2020; Zisler et al., 2024). Patients with OCD treated within inpatient psychiatric settings often present with more severe, difficult-to-treat symptoms and complex comorbidities. A meta-analysis (Zisler et al., 2024) of inpatient OCD treatment found that inpatients demonstrated large improvements in OCD symptoms, but the length of stay did not justify the treatment effect size. A long length of stay with highly trained staff is often costly and resource-intensive, which may limit accessibility and affordability, even with health insurance and government support. Currently, to the best of our knowledge, there is only one service in Australia that provides standardised inpatient OCD treatment in Australia. Patients in this programme receive 3 weeks of specialised ERP treatment within a group setting (Dowling et al., 2016). Identifying briefer and targeted OCD treatment formats that provide high clinical outcomes is urgently needed to address the unmet need and reduce the burden of OCD and its associated disability and functional impairment. Standard psychological treatments for OCD involve either weekly 1- to 2-hour sessions over several months, or for more severe presentations, residential treatment for 3 or more weeks (O’Neill and Feusner, 2015). By contrast, the Bergen 4-Day Treatment (B4DT) model, developed in Norway, has emerged as a promising and effective option compared to conventional OCD treatments, where outpatients receive specialised concentrated treatment. Aligning with Öst and Ollendick’s (2017) definition of concentrated interventions, the total number of treatment hours is similar to that of standard outpatient treatment for OCD, but B4DT is delivered as a concentrated treatment over a much shorter period, over 4 consecutive days. Typically, 3–6 patients receive individualised and tailored exposure practice within a group setting, with a patient-therapist ratio of 1:1. The treatment is best described as ‘individual treatment delivered in a group setting’, with completion rates near 100% (Kvale et al., 2018) compared to 75–80% completion rates for standard OCD treatments (Öst et al., 2015). Clinical outcome data suggest that B4DT may have higher response and remission rates than existing treatments, with approximately 90% of the patients demonstrating full clinical response (>35% reduction in OCD), and around 70% achieving long-term remission (Hansen et al., 2018, 2019). It has also been shown to be effective for patients with difficult-to-treat OCD (Kvale et al., 2020; Launes et al., 2020). Larger treatment effect sizes have been demonstrated for B4DT (Cohen’s d = 4.60) as compared to more conventional ERP (d = 2.06; Öst et al., 2015), residential treatment (g = 1.87; Veale et al., 2016) and Australian inpatient OCD treatment (d = 1.42; Dowling et al., 2016) programmes. The B4DT format has also demonstrated significant improvements in comorbid depression, anxiety and social and occupational functioning (Hansen et al., 2019), with high patient-reported satisfaction and acceptability (Launes et al., 2020). With over 11,000 patients effectively treated in Norway in the past decade, B4DT developers have expanded their focus and reach to disseminate this treatment format internationally. Existing data demonstrate strong clinical outcomes of the B4DT format with comparable effect sizes observed in Sweden (Hansen et al., 2018), Germany (Jelinek et al., 2024), Iceland (Davíðsdóttir et al., 2019, 2025), Finland (Silver et al., 2023), Singapore (Gupta et al., 2024) and the USA (Björgvinsson et al., 2025). To date, B4DT has not been trialled in Australia, where there are long delays to professional care, limited expertise and poor access to specialised treatment programmes. Mogan et al. (2025) recently developed and evaluated an intensive 4-day outpatient OCD programme in Australia. The programme used ERP but was not based on the B4DT programme. Post-treatment and 12-month follow-up outcomes were comparable to other CBT programmes for OCD but were not as favourable as B4DT outcomes. The study was also limited by the lack of a control group. As such, feasibility studies of scalable and time-efficient treatment models, such as B4DT, are urgently needed in the Australian context. Furthermore, most B4DT programmes have been conducted in an outpatient setting. Only one study from Germany has examined B4DT for OCD within an acute inpatient psychiatric setting (Jelinek et al., 2024). They reported positive outcomes when compared to a historical inpatient dataset and large within-group effect sizes (pre to post d = 2.04, pre to follow-up d = 1.69). Further research into B4DT conducted in an inpatient setting is therefore also needed. In this pilot study, we assessed the treatment outcomes and acceptability of B4DT in a limited sample of inpatients with OCD, compared with the standard treatment over 3 weeks, within an Australian psychiatric setting. We hypothesise that both formats will demonstrate significant improvements in OCD symptoms and other domains of mental health (including depression, anxiety and wellbeing) from pre- to post-treatment, with improvements maintained at the 3-month follow-up. We also hypothesise that despite a treatment course over a shorter time frame, patients receiving B4DT would demonstrate significantly greater improvements in OCD and other mental health outcomes at post-treatment and at the 3-month follow-up, relative to the standard treatment. Method Participants Participants aged 18 and above diagnosed with primary OCD were referred to the inpatient OCD treatment programmes in Melbourne, Australia, by their treating doctor for specialised care. Patients were eligible if they had suitable private health insurance, or WorkCover insurance, coverage by a public health partnership, and all patients had an inadequate response or relapsed following outpatient OCD treatments, with current moderate OCD symptom severity or complexity and functional impairment. Study eligibility criteria were applied separately from clinical admission criteria. Participants were eligible if they are 18 years of age or older, have met Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) criteria for OCD, were clinically suitable for inpatient OCD treatment and were on stable psychotropic medication for at least 2 weeks. Exclusion criteria included moderate or high suicide risk or self-harm, unstable psychotic or mood episode, significant cognitive impairment or medical instability. To allow for the recruitment of participants who typically present at the programme for treatment, we did not set specific criteria for the range of baseline Y-BOCS scores, prior or concurrent treatments and prior length of stay before intervention. Table 1 provides the clinical characteristics of the sample of this study. Table 1.Baseline sociodemographic and clinical characteristics.CharacteristicStandardn = 13B4DTn = 12Statistic (t or Fisher’s exact test)AgeM = 34.15 (SD = 10.79)M = 40.92 (SD = 13.39)t = −1.40, p = .18Sex Female8 (61.5%)7 (58.3%) Male5 (38.5%)5 (41.7%)p = .87Highest education University degree5 (38.5%)11 (91.7%) High school or diploma8 (61.5%)1 (8.3%)p = .01No. of prior inpatient OCD programmes 08 (61.5%)6 (50.0%) 1+5 (38.5%)6 (50.0%)p = .56Years with OCD symptomsM = 20.62 (SD = 12.48)M = 26.10 (SD = 12.65)t = −1.09, p = .29Years with OCD diagnosisM = 7.62 (SD = 7.52)M = 11.33 (SD = 8.11)t = −1.19, p = .25History of psychotherapy for OCD Yes11 (84.6%)12 (100%) No2 (15.4%)0p = .16History of exposure therapy Yes6 (46.2%)8 (66.7%) No7 (53.8%)4 (33.3%)p = .30Any antidepressant Yes12 (92.3%)10 (83.3%) No1 (7.7%)2 (16.7%)p = .51Any benzodiazepine Yes3 (23.1%)4 (33.3%) No10 (76.9%)8 (66.7%)p = .67Any antipsychotic medication Yes8 (61.5%)3 (25.0%) No5 (38.5%)9 (75.0%)p = .11Any other psychotropic medication Yes7 (53.8%)1 (8.3%) No6 (46.2%)11 (91.7%)p = .03DSM-5 MINI Comorbid MDE Yes4 (30.8%)4 (33.3%) No9 (69.2%)8 (66.7%)p = 1.00DSM-5 MINI Comorbid Anxiety Yes8 (61.5%)6 (50.0%) No5 (38.5%)6 (50.0%)p = .70DSM-5 MINI Any Other Comorbid Yes4 (30.8%)1 (8.3%) No9 (69.2%)11 (91.7%)p = .32Y-BOCS total scoreM = 30.54 (SD = 4.16)M = 30.25 (SD = 4.05)t = 0.27, p = .79DOCS total scoreM = 40.62 (SD = 14.19)M = 34.83 (SD = 11.47)t = 1.11, p = .28PHQ-9 total scoreM = 13.92 (SD = 6.53)M = 12.17 (SD = 6.90)t = 0.65, p = .52GAD-7 total scoreM = 13.69 (SD = 4.97)M = 11.75 (SD = 5.31)t = 0.94, p = .36OBQ-20 total scoreM = 95.15 (SD = 31.57)M = 89.50 (SD = 23.35)t = 0.51, p = .62WEMWBS total scoreM = 37.38 (SD = 7.46)M = 38.00 (SD = 8.41)t = −0.19, p = .85Italicised p values indicate statistically significant differences between groups at baseline (p < .05). MDE = Major Depressive Episode, MINI = Mini International Neuropsychiatric Interview; Y-BOCS = Yale-Brown Obsessive Compulsive Scale; B4DT = Bergen 4-day Treatment; PHQ-9 = Patient Health Questionnaire-9; GAD-7 = Generalised Anxiety Disorder Scale-7; OBQ-20 = 20-item Obsessive Beliefs Questionnaire; WEMWBS = Warwick-Edinburgh Mental Wellbeing Scale. Baseline sociodemographic and clinical characteristics. Italicised p values indicate statistically significant differences between groups at baseline (p < .05). MDE = Major Depressive Episode, MINI = Mini International Neuropsychiatric Interview; Y-BOCS = Yale-Brown Obsessive Compulsive Scale; B4DT = Bergen 4-day Treatment; PHQ-9 = Patient Health Questionnaire-9; GAD-7 = Generalised Anxiety Disorder Scale-7; OBQ-20 = 20-item Obsessive Beliefs Questionnaire; WEMWBS = Warwick-Edinburgh Mental Wellbeing Scale. Patients admitted between September and December 2024 were informed of the study. Patients already scheduled for admission to standard treatment were contacted before admission and invited to participate. For the B4DT cohort, eligible patients were referred and contacted by the treatment team and offered participation in the study as a new treatment format when cohorts were scheduled. Participants were provided with information about both treatment formats, and allocation was primarily determined by programme availability and logistical considerations rather than randomisation. All participants enrolled voluntarily, were free to choose or decline either treatment and provided informed consent. Participants subsequently received either B4DT (n = 12) or standard treatment (n = 13). To ensure the quality of independent assessments, no more than four participants were recruited per cohort over a 3-week treatment period. Cohorts of standard treatment consisted of a 9:2 patient-to-therapist ratio. Approval was sought and granted by The Melbourne Clinic Research Ethics Committee (TMC REC #360), and all participants provided written informed consent before participation. Measures The following clinician-administered and self-report instruments were used to confirm participants’ OCD diagnosis; assess pre-treatment, post-treatment and follow-up OCD severity, depression, anxiety and wellbeing and to evaluate client satisfaction. The Mini International Neuropsychiatric Interview for DSM-5 (MINI-5; Sheehan et al., 1998) is a structured diagnostic interview used to confirm the diagnosis of OCD as well as other major psychiatric disorders according to the DSM-5 criteria. The MINI-5 is a relatively brief and standardised assessment that is comparable to longer structured interviews. Primary outcome The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS; Goodman et al., 1989) is the primary outcome measure used to assess OCD symptom severity. The Y-BOCS is a 10-item clinician-administered instrument that assesses on a 5-point scale (0 = none to 4 = severe), with total scores ranging from 0 to 40, with higher scores indicating more severe symptoms. The Y-BOCS has demonstrated strong reliability and validity in both clinical and research settings. The internal consistency for the Y-BOCS was acceptable (α = .77). Secondary outcomes The Dimensional Obsessive-Compulsive Scale (DOCS; Abramowitz et al., 2010) is a 20-item self-report measure of OCD severity across four symptom dimensions: contamination, responsibility for harm, unacceptable thoughts and completeness/symmetry. Items are rated on a 5-point scale from ‘0’ (no symptoms) to ‘4’ (extreme symptoms). Each DOCS subscale ranges from 0 to 20, and the total score ranges from 0 to 80. The DOCS has demonstrated adequate test-retest reliability (r = .55–.66) and strong convergent validity (Abramowitz et al., 2010). The internal consistency for the DOCS in our sample was good (α = .87). The Clinical Global Impression (CGI; Busner and Targum, 2007) measures symptom severity, treatment response and efficacy in treatment studies. It scales for severity and improvement. The Severity Scale (CGI-S) is a 7-point scale from 1 (normal) to 7 (among the most extremely ill patients). The Improvement Scale (CGI-I) is also a 7-point scale from 1 (very much improved) to 7 (very much worse). The Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001) is a nine-item depression module from the full PHQ and was designed to score each of the nine Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-4) criteria of major depression on a 4-point Likert-type scale from ‘0’ (not at all) to ‘3’ (nearly every day) over the previous 2 weeks. The internal consistency for the PHQ-9 was good (α = .87). The Generalised Anxiety Disorder Scale-7 (GAD-7; Spitzer et al., 2006) consists of seven questions based in part on the DSM-IV criteria for GAD and reflects the frequency of symptoms during the preceding 2-week period. The GAD-7 requires approximately 1–2 minutes to administer, and for each symptom queried, it provides the following response options: ‘not at all’, ‘several days’, ‘over half the days’ and ‘nearly every day’; and these are scored, respectively, as 0, 1, 2 or 3. The internal consistency for the GAD-7 was good (α = .87). The Obsessive Beliefs Questionnaire–20 (OBQ-20; Moulding et al., 2011) is a 20-item self-report measure assessing OCD-related beliefs across three domains: threat/responsibility, perfectionism/certainty, and importance/control of thoughts. Each item is rated on a 7-point scale, with higher scores indicating stronger endorsement of obsessive beliefs. The internal consistency for the OBQ-20 in our sample was excellent (α = .94). The Warwick-Edinburgh Mental Well-Being Scale (WEMWBS; Tennant et al., 2007) is a 14-item self-report measure of positive feelings and emotional wellbeing. Each item is rated on a 5-point scale, with scores ranging from 14 to 70, and high scores indicating greater wellbeing. The internal consistency for the WEMWBS was good (α = .89). The Client Satisfaction Questionnaire-8 (CSQ-8; Larsen et al., 1979) is an eight-item self-report measure to assess client satisfaction with the B4DT format. Each item is rated on a 4-point scale, with scores ranging from 8 to 32, and higher scores indicating greater satisfaction and acceptability. The CSQ-8 has demonstrated strong internal consistency and has been widely used across clinical settings due to its brevity, ease of use and robust psychometric properties. The internal consistency for the CSQ-8 was excellent (α = .94). Procedure The multidisciplinary OCD treatment team delivered both B4DT and standard inpatient treatments and comprise a psychiatrist, clinical psychologists, an occupational therapist and psychiatric nurses, all of whom have substantial experience in delivering ERP for OCD. Independent diagnostic interviews and assessment measures were administered prior to admission by an independent clinician-researcher who was not part of the treatment team. Assessments of outcomes were conducted at 10 days post treatment, and at 3 months after discharge from the programme. The assessments were completed at comparable timepoints, and most assessments were completed virtually. Bergen 4-Day Treatment As randomisation was not implemented for this study, the patients receiving B4DT were recruited by the treatment team. These individuals were not treatment-naïve, with all having had prior OCD treatments but remaining symptomatic. High levels of motivation could be inferred given their reported enthusiasm when offered a novel treatment approach, and all patients proceeded with the B4DT treatment protocol accordingly. Before hospital admission for the B4DT, patients completed a pre-treatment interview and were asked to clear their schedules to commit to the programme fully. The treatment was delivered with a 1:1 therapist-to-patient ratio, supported by five B4DT experts from Norway and Iceland who trained the Melbourne team (as observers). During the initial B4DT cohort, the B4DT experts were the primary clinicians while the Melbourne clinicians acted as observers. Across the second and third cohorts, the Melbourne clinicians progressively assumed the primary clinician role, under on-site supervision of the B4DT experts who transitioned to being observers. Treatment fidelity and adherence were supported by adherence to B4DT treatment training and materials, joint formulation and planning with the B4DT experts and ongoing supervision in B4DT and LET technique principles and group debriefs. No formal session-by-session fidelity ratings were conducted. A total of 12 patients received B4DT, divided into three groups of four, delivered sequentially. The treatment followed the established protocol described in detail elsewhere (see Launes et al., 2019) and was delivered over four consecutive days. Day 1 (4 hours) was focused on psychoeducation and the development of individualised exposure tasks. Days 2 and 3 (8 hours each) included psychoeducation refresh and therapist-assisted, individually tailored exposure exercises conducted in settings relevant to each patient’s OCD symptoms. On day 3, late afternoon, patients’ family members and close supports were invited to attend a psychoeducation session on the B4DT approach and how to support the patient after discharge. Day 4 focused on consolidating the treatment rationale, developing a personalised exposure plan for the following 3 weeks and covering relapse-prevention strategies. All patients were encouraged to practice exposure tasks between sessions and outside of formal therapy hours, consistent with the B4DT framework. Unlike typical B4DT programmes and other intensive outpatient OCD treatment programmes (Jónsson et al., 2015), in this B4DT format, patients stayed in hospital for all 4 days and were discharged at the conclusion of day 4 or the subsequent morning. B4DT builds on established exposure-based interventions, which remain first-line treatments for OCD, while incorporating a distinctive theoretical emphasis. Central to this approach is the role of intention: patients are guided to recognise how their regulatory focus- what they are trying to achieve or prevent in response to intrusive thoughts and anxiety-maintains and sustains maladaptive patterns of control. The Leaning into Anxiety Technique (LET) operationalises this principle, supporting patients to engage with feared situations in alignment with adaptive goals rather than the OCD control project. Unlike traditional CBT, which emphasises habituation, expectancy violation or disconfirmation of beliefs, LET prioritises deliberate modification of regulatory intention as the primary mechanism of change. Exposures are individually tailored and meaningful, and patients are coached to fully enact these tasks while identifying and disengaging from unnecessary regulatory control efforts. Success is defined by intentional, goal-aligned engagement rather than solely by reductions in subjective distress. The theoretical framework underlying LET and regulatory focus principles was developed by Bjarne Hansen prior to B4DT. Hansen and Kvale subsequently used this framework as the basis for designing the B4DT. This treatment is delivered over 4 consecutive days in a combined group and individual format with a one-to-one patient-to-therapist ratio, ensuring concentrated individualised guidance while also providing group-based social support. The programme is structured to optimise engagement with the model: the first day focuses on detailed planning and preparation, days 2 and 3 emphasise hands-on exposure practice using LET, and day four consolidates learning, reviews progress and plans continued application. The structure provides close supervision and continuous feedback to ensure that patients’ work aligns with the theoretical principles. Standard inpatient OCD treatment The 3-week inpatient programme follows an exposure-based treatment format in which patients engage in daily group-based ERP sessions, supported by psychoeducation and collaborative treatment planning. Sessions are conducted 5 days per week, from 9:00 a.m. to 4:00 p.m., with scheduled breaks. The 3-week programme was based on more traditional ERP/CBT treatment approaches, such as habituation, behavioural and interoceptive experiments and imaginal exposure. Patients were supported to approach feared thoughts, feared outcomes and sensations, while deliberately delaying, resisting and modifying compulsions. Exposures were ranked from moderately to highly distressing and approached hierarchically to learn to perform without compulsions, experience habituation and tolerance, accept uncertainty, disconfirm feared outcomes and facilitate corrective learning. Each treatment day typically included 1–2 hours of psychoeducation, followed by approximately 2 hours of group-based ERP practice and daily group check-ins (approximating 4 hours of treatment each weekday). ERP tasks were collaboratively planned between clinicians and patients, with patients expected to conduct exposures independently as part of a self-directed learning process, including practising outside of formal treatment hours. Patients were typically discharged on the weekend at the end of the 3-week programme. Standard treatment groups usually include up to nine patients, with a patient-to-therapist ratio of approximately 9:2. Patients were encouraged to practice exposure tasks outside of formal therapy sessions. For a detailed description of the Australian 3-week treatment programme, see Dowling et al. (2016). When comparing between treatment formats, the ERP dose differed. Specifically, B4DT patients received concentrated therapist-assisted ERP across days 2 and 3, approximately 14 hours in total. Conversely, the standard inpatient treatment distributed ERP across the 3-week admission through group-based ERP and self-directed practice, approximately 20 hours. Thus, B4DT delivered more concentrated and individualised ERP treatment within a shorter period. Statistical analysis Pre-treatment characteristics of the two groups were summarised with means and standard deviations for continuous variables and counts and percentages for categorical variables. Mean scores and mean between-group differences for each outcome were estimated using linear mixed-effects models. These models included the baseline score of the relevant outcome, treatment, follow-up time (10 days, 3 months) and treatment by follow-up time product terms. For binary treatment response and remission outcomes (e.g. full remission), we fitted logistic regression models, with adjustment for baseline Y-BOCS total score. Using Mataix-Cols et al.’s (2016) criteria, we defined full response as Y-BOCS total score reduction ⩾35% and CGI-I of 1 or 2, partial response as Y-BOCS total score reduction >25% to <35% and CGI-I of 3 and remission as a final Y-BOCS total score < 12 and CGI-S of 1 or 2. The proportion reaching response and remission was estimated using g-computation, with confidence intervals and p values obtained using the delta method (Arel-Bundock et al., 2024). As a sensitivity analysis for residual confounding for the primary outcome (total Y-BOCS score), we made additional adjustments for baseline clinical variables that differed between groups and were thought to be associated with the outcome. For this sensitivity analysis, each covariate was added separately (one at a time) to the primary model. Data analysis was performed in R version 4.4.1. Results Participant characteristics Sociodemographic and clinical characteristics of the sample at baseline are displayed in Table 1. The groups were similar in gender and scores on all baseline outcome measures. There was also no significant difference in the use of antidepressants and benzodiazepines. While participants in B4DT tended to be slightly older and less likely to use antipsychotic medication, these differences were also not significant. Participants in the B4DT group were, however, significantly more likely to have completed a graduate degree and less likely to be receiving any other psychotropic medication. Most participants were born in Australia, with only two in the B4DT group born overseas (Nigeria and Nepal) and two in the standard treatment group who did not report a country of birth. All participants had received prior treatments for OCD but continued to experience at least moderate OCD symptom severity with enduring functional impairment despite current or past treatments. There were no significant differences between groups on treatment history. One participant in the standard treatment group discontinued treatment overall. Primary outcome At 10-day follow-up, the mean change from baseline in the total Y-BOCS score for the B4DT and standard programmes was −17.00 (Cohen’s d = 2.61) and −11.31 (d = 2.16) points, respectively. At the 3-month follow-up, these were −18.58 (d = 4.15) and −11.33 (d = 2.40) points, respectively. Figure 1 displays the box plot of the distribution of Y-BOCS total scores at assessment periods. Box plot showing distributions of Y-BOCS total scores in treatment at pre-treatment, 10-day follow-up and 3-month follow-up. The central line in each box represents the median, boxes indicate the interquartile range (Q1–Q3) and whiskers extend to nonoutlier values. B4DT = Bergen 4-Day Treatment; Y-BOCS = Yale-Brown Obsessive Compulsive Scale. Results for the primary analysis are presented in Table 2. At 10 days after programme completion, participants in the B4DT programme had a greater reduction in Y-BOCS scores than those in the standard programme (mean difference [MD] −5.57, 95% confidence interval [CI] −9.57, −1.58, p = 0.006) (Figure 1). At the 3-month follow-up, those in the B4DT programme continued to have lower total Y-BOCS scores (MD −7.31, 95% CI [−11.37, −3.24], p < 0.001). These estimated MDs were equivalent to Cohen’s d effect sizes of −0.68 and −0.89 for the 10-day and 3-month follow-up, respectively. Table 2.Follow-up outcome estimated mean scores by treatment.Day 10 follow-up3-Month follow-upEstimated meanEstimated mean differences (95%, CI) p Estimated meanEstimated mean difference (95% CI) p OutcomeStandard (n = 13)B4DT(n = 12)Standard* (n = 12)B4DT(n = 12)Y-BOCS total19.0413.46–5.57 (−9.57, −1.58)0.00619.1511.84–7.31 (−11.37, −3.24)<0.001DOCS total24.1920.46–3.73 (−12.39, 4.93)0.4024.1515.02–9.13 (−17.89, −0.36)0.041CGI severity3.952.64–1.32 (−1.92, −0.71)<0.0013.752.48–1.27 (−1.88, −0.66)<0.001PHQ-9 total12.177.48–4.69 (−8.58, −0.80)0.01814.6310.66–3.97 (−7.89, −0.06)0.047GAD-7 total11.237.08–4.15 (−7.33, −0.97)0.01112.737.95–4.78 (−8.00, −1.56)0.004OBQ-20 total78.4763.82–14.65 (−35.21, 5.91)0.1683.1960.91–22.28 (−43.03, −1.53)0.035WEMWBS total40.4346.035.60 (−0.45, 11.66)0.07038.3043.985.69 (−0.45, 11.82)0.069Estimated means and mean differences are calculated by standardisation. *= One patient was not included from the 3-month follow-up as they were uncontactable. Y-BOCS, Yale-Brown Obsessive Compulsive Scale; B4DT, Bergen 4-Day Treatment; PHQ-9, Patient Health Questionnaire-9; GAD-7, Generalised Anxiety Disorder Scale-7; OBQ-20, 20-item Obsessive Beliefs Questionnaire; WEMWBS, Warwick-Edinburgh Mental Wellbeing Scale. Follow-up outcome estimated mean scores by treatment. Estimated means and mean differences are calculated by standardisation. *= One patient was not included from the 3-month follow-up as they were uncontactable. Y-BOCS, Yale-Brown Obsessive Compulsive Scale; B4DT, Bergen 4-Day Treatment; PHQ-9, Patient Health Questionnaire-9; GAD-7, Generalised Anxiety Disorder Scale-7; OBQ-20, 20-item Obsessive Beliefs Questionnaire; WEMWBS, Warwick-Edinburgh Mental Wellbeing Scale. The estimated proportion achieving full response, partial remission and full remission, based on criteria consistent with international expert consensus (Mataix-Cols et al., 2016), is shown in Table 3. The estimated proportion with full response (Y-BOCS reduction ⩾35% and CGI-I of 1 or 2) for the B4DT programme and standard programmes was 91.9% and 49.0%, respectively (proportion difference 42.9%; p = 0.001). At the 3-month follow-up, these were 93.0% and 47.4%, respectively (proportion difference 45.6%; p < 0.001). The remission rate criteria were met for 45.8% at 10-day and 53.6% at 3-month follow-up, compared to 6.6% and 16.1%, respectively, for standard treatment. Table 3.Estimated treatment response, partial and remission criteria by treatment.Day-10 follow-up3-month follow-upOutcomeStandard(n = 13)B4DT(n = 12) p Standard* (n = 12)B4DT(n = 12) p DefinitionsFull response49.0%91.9%0.00147.4%93.0%<0.001Y-BOCS total score reduction ≥35% and CGI-I of 1 or 2Partial response15.4%0%0.0860%0%-Y-BOCS total score reduction ≥25% to <35% and CGI-I of 3Remission6.6%45.8%0.00316.1%53.6%0.013Y-BOCS total score ≤ 12 and CGI-S of 1 or 2Estimated proportions and proportion differences are calculated by standardisation. Y-BOCS, Yale-Brown Obsessive Compulsive Scale; B4DT, Bergen 4-Day treatment; CGI-S, Clinical Global Impression Severity. * = One patient was not included in the 3-month follow-up assessment as they were uncontactable. Estimated treatment response, partial and remission criteria by treatment. Estimated proportions and proportion differences are calculated by standardisation. Y-BOCS, Yale-Brown Obsessive Compulsive Scale; B4DT, Bergen 4-Day treatment; CGI-S, Clinical Global Impression Severity. * = One patient was not included in the 3-month follow-up assessment as they were uncontactable. In sensitivity analyses assessing the no unmeasured confounding assumption, we found similar MDs for the total Y-BOCS score when additionally adjusting for graduate education, the number of prior inpatient OCD programmes, use of an antipsychotic and use of other psychotropic medications (Table 4). For example, when adjusting for antipsychotic medication use and other psychotropic medication use, estimated MDs were −3.72 (p = 0.053) and −6.15 (p = 0.007) at the 10-day follow-up, respectively. At the 3-month follow-up, all results were similar to the primary analysis (all p < 0.005). Table 4.Estimates mean differences for Y-BOCS outcome, with adjustment for potential confounding factors.Estimated mean difference (95% CI)Estimated mean difference (95% CI)Adjustment variable* p p Graduate education–5.64 (−10.28, −1.00)0.017–7.38 (−12.04, −2.72)0.002No. prior inpatient OCD programmes–4.87 (−8.99, −0.74)0.021–6.80 (−10.95, −2.64)0.001Antipsychotic medication–3.72 (−7.50, 0.05)0.053–5.81 (−9.58, −2.04)0.003Other psychotropic medication–6.15 (−10.61, −1.70)0.007–7.94 (−12.51, −3.36)0.001*All models were additionally adjusted for baseline Y-BOCS total score. Estimates mean differences for Y-BOCS outcome, with adjustment for potential confounding factors. All models were additionally adjusted for baseline Y-BOCS total score. Secondary outcomes Outcomes in the B4DT group were also substantially better than those for the standard programme for most of the secondary outcome scales (Table 2). At 10-day follow-up, participants in the B4DT group had lower scores for global impression (p ⩽ 0.001), depression (p = 0.018) and anxiety (p = 0.011), compared to standard treatment. At the 3-month follow-up, lower scores for B4DT were maintained for global impression (p < 0.001), depression (p = 0.047), anxiety (p = 0.004) and for the 20-item Obsessive Beliefs Questionnaire (OBQ-20) total (p = 0.035), compared to standard treatment. The total of WEMWBS failed to reach a significant difference between treatments at 10-day and 3-month follow-ups, p = 0.070 and 0.069, respectively. Participant satisfaction and acceptability All 12 patients who completed the B4DT completed the CSQ-8. The average participant satisfaction with B4DT was 31.08 (SD = 2.57), which was significantly higher (Welch’s t = 5.35, p < 0.001) than CSQ-8 norms reported in Nguyen et al. (1983; average CSQ-8 = 27.09, SD = 4.01, n = 3120). More detailed CSQ-8 results are presented in Table 5. Table 5.Participant responses of CSQ-8 for B4DT at post-treatment.M(n = 12)PoorFairGoodExcellentThe quality of the service received3.9200111Get the kind of service you wanted3.9200111The extent the service met my needs3.7501110Would recommend to a friend needing help4.0000111Satisfied with help received3.9200111Helped deal with problems effectively3.9200111Overall satisfaction with service3.8301011Come back if need3.8300210CSQ-8, Client Satisfaction Questionnaire-8; B4DT, Bergen 4-Day Treatment. Participant responses of CSQ-8 for B4DT at post-treatment. CSQ-8, Client Satisfaction Questionnaire-8; B4DT, Bergen 4-Day Treatment. Discussion This open-label pilot study is the first to evaluate the B4DT format in comparison with the standard 3-week OCD treatment programme within an inpatient psychiatric setting in Australia. It also represents the first implementation of B4DT that demonstrates favourable feasibility in an Australian context. While both treatments were effective, we found that B4DT demonstrated superior clinical outcomes across multiple domains, including OCD severity, global functioning and treatment response and remission rates. At both the 10-day and 3-month follow-up, participants in the B4DT exhibited significantly lower scores on OCD severity than those undergoing standard treatment despite having relatively similar baseline characteristics. Of note, the proportion of patients who experienced full clinical response at 10-day follow-up was 91.9% for the B4DT, compared to 49.0% in the standard treatment, with the nearly two-fold difference persisting at 3 months (93.0% for B4DT vs 47.4% for standard). Furthermore, the B4DT participants trended towards further OCD symptom improvement, whereas the standard treatment participants remained stable. Remission rates were also significantly superior for the B4DT at 10 days and 3 months, at 45.8% and 53.6%, compared to standard treatment at 6.6% and 16.1%, respectively. Proportion of response rates for the B4DT was also superior to our past findings with the standard treatment (Dowling et al., 2016) and comparable with international outcomes for B4DT (90% in Norway, Hansen et al., 2019; 68% in Iceland, Davíðsdóttir et al., 2025; 56% in Finland, Silver et al., 2023; and 91.7% in the United States, Björgvinsson et al., 2025). These findings strengthen the existing data that B4DT can lead to meaningful improvement within inpatient settings with active comparators (Jelinek et al., 2024). As for secondary outcome measures, improvements in comorbid depression, generalised anxiety and overall wellbeing were observed, in line with data showing that OCD treatment, particularly B4DT, yields benefits beyond the core OCD symptoms. Interestingly, the improvements were more pronounced at the 3-month follow-up for the B4DT. Overall self-reported wellbeing failed to reach statistical significance at each time point. This discrepancy, despite small sample sizes, warrants cautious interpretation. One possibility is that patients with more severe, chronic and complex symptoms with high comorbidities may be prone to reporting less subjective change due to higher functional impairment. Encouragingly, dropout rates across treatments were negligible for both treatments, with only one patient discharging from standard treatment after the first week. While no patient deteriorated in either treatment, the secondary measures revealed more modest improvements, compared to Y-BOCS, which have been observed in other studies (Launes et al., 2020) and clinical settings (Jelinek et al., 2024; Silver et al., 2023). Patients receiving the B4DT reported very high satisfaction levels, underscoring the feasibility and acceptability of this concentrated format within an inpatient setting. This aligns with findings from other clinical settings (Björgvinsson et al., 2025; Silver et al., 2023), including inpatient programmes (Jelinek et al., 2024), and contributes to the growing evidence that B4DT is adaptable and well-accepted across clinical and cultural settings, thus supporting its global dissemination. While B4DT is primarily delivered as an outpatient treatment internationally, it appears to be flexibly adapted across inpatient and outpatient settings, allowing treatment to be tailored to patient presentation, available supports and clinical context. For Australia, establishing both inpatient and outpatient services, alongside research and training, could allow a stepped-care model of care to maximise reach and impact. Limitations Several limitations should be noted. Importantly, the study had a small sample size in both treatment groups, which limits the reliability of our findings and introduces uncertainty around the true effect size. Participant characteristics unique to our sample, including high levels of psychiatric comorbidity and almost all having had previous psychological or psychiatric treatment for OCD, limit the generalisability of our study. Furthermore, the study employed a non-randomised design, raising the possibility of selection bias, since the B4DT participants were recruited by treating clinicians. Although there were a few significant baseline differences between groups, there may have been systematic baseline differences that were not detected, such as differences in motivation and readiness to change. Outcomes were assessed independently by a researcher who was unblinded to the treatment groups, as it was not feasible to implement blinded treatment allocation or assessments in this study. Therefore, expectancy effects remain possible given the novelty of the B4DT and participants’ motivation to receive a new intervention in Australia. The delivery of the B4DT by an expert Norwegian team may also have introduced demand characteristics, with participants perceiving overseas specialists as being more competent and innovative, potentially enhancing engagement and compliance. However, this may not have had a large impact given that recent research showed that pre-treatment motivation is not predictive of treatment outcomes in B4DT (Berg et al., 2025). Although the B4DT format was shorter, each patient received personalised ERP support by a B4DT expert and another clinician across relevant real-life contexts, whereas the standard treatment group had a higher patient-to-therapist ratio (9:2), with less access to individualised and personalised ERP. It is possible that the differential treatment response in the B4DT programme can be partly attributed solely to the lower patient-to-therapist ratio. While it is possible that treatment response would improve by decreasing the patient-to-therapist ratio in the standard 3-week treatment group, this would not be logistically feasible in a standard hospital setting. The lower patient-to-therapist ratio could be justified because the costs associated with a 4-day programme, including staff time and other non-staff hospital-related expenses (i.e. food, facilities, administration and beds), are substantially lower than those in a 3-week programme. Study strengths include the independent assessments, established OCD treatment formats, the use of validated measures and the direct comparison within a single established inpatient OCD treatment setting. The strong clinical response combined with high satisfaction highlights that the B4DT could be a viable option for some patients, which rivals longer treatment programmes. Our findings add to the existing literature that the B4DT might offer a scalable, time-limited and efficient alternative within inpatient OCD treatment settings with highly favourable clinical outcomes. This is especially relevant given that inpatient OCD programmes require specialised training and support, which can be resource-intensive and costly. Future research Future research should replicate our findings in larger, randomised trials and with patients from culturally diverse populations treated in various healthcare settings, including public mental health services. In addition, the mechanisms of change that drive significantly larger effects of the B4DT compared to standard ERP or CBT are still unclear. Unpacking the extent to which the treatment format (i.e. concentrated delivery of ERP), treatment approach (i.e. B4DT), non-specific factors (e.g. therapeutic alliance), or a combination of these factors contribute to treatment outcomes in OCD is important. Further research using a process-based idionomic approach may provide further answers (Hayes and Hofmann, 2022). Understanding change processes using patient-reported assessment of treatment experience and perceived mechanisms of change over the course of treatment may also provide further clarification on how the B4DT differs from traditional ERP and CBT approaches beyond differences in the treatment format. Recent research has shown that treatment adherence rather than pre-treatment motivation predicted the B4DT treatment outcomes (Berg et al., 2025). Future research examining predictors of treatment adherence in B4DT at an individual level using ecological momentary assessments may thus help improve treatment outcomes at an individual level. In addition, future research should investigate how the core principles of B4DT can be implemented in other treatment formats funded by health and government agencies, while maintaining high treatment response satisfaction and low dropout. Understanding how B4DT may be adapted to encompass patients with complex comorbidities, those with limited insight and low motivation, is relevant given the common clinical features within residential settings. Finally, future research should investigate the treatment perceptions with patients and clinicians, as well as the mechanisms of change for the B4DT, and how they differ from traditional CBT and ERP. Conclusion In summary, this pilot study supports the feasibility and acceptability of delivering B4DT in an inpatient setting, providing a potential alternative to longer standard 3-week inpatient treatment in Australia, with positive outcomes and high patient satisfaction. Despite the limitations, this study provides a foundation for broader implementation of how B4DT can be adapted within an Australian context to improve access to evidence-based care, reduce the burden of OCD on individuals and their communities and mitigate economic and societal impacts.