Effectiveness of guided self-help, guided internet-delivered cognitive behavioral therapy, and face-to-face cognitive behavioral therapy for depression and anxiety: protocols of four parallel randomized controlled non-inferiority trials of the Finnish First-Line Therapies –Initiative (FLT-Step) - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Psychiatry . 2026 Mar 11;26:329. doi: 10.1186/s12888-026-07962-w Search in PMC Search in PubMed View in NLM Catalog Add to search Effectiveness of guided self-help, guided internet-delivered cognitive behavioral therapy, and face-to-face cognitive behavioral therapy for depression and anxiety: protocols of four parallel randomized controlled non-inferiority trials of the Finnish First-Line Therapies –Initiative (FLT-Step) Eeva-Eerika Helminen Eeva-Eerika Helminen 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland 2 Department of Psychiatry, University of Helsinki, Helsinki, Finland Find articles by Eeva-Eerika Helminen 1, 2, ✉, # , Suoma E Saarni Suoma E Saarni 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland 6 Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland 7 Department of Psychiatry, Wellbeing Services County of Pirkanmaa, Tampere, Finland Find articles by Suoma E Saarni 1, 6, 7, # , Kasperi Mikkonen Kasperi Mikkonen 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland 3 Department of Psychology and Logopedics, University of Helsinki, Helsinki, Finland Find articles by Kasperi Mikkonen 1, 3 , M Katariina Mattila M Katariina Mattila 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland Find articles by M Katariina Mattila 1 , Tom H Rosenström Tom H Rosenström 3 Department of Psychology and Logopedics, University of Helsinki, Helsinki, Finland Find articles by Tom H Rosenström 3 , Max Karukivi Max Karukivi 4 Department of Adolescent Psychiatry, University of Turku, Turku, Finland 5 Department of Adolescent Psychiatry, Turku University Hospital, Turku, Finland Find articles by Max Karukivi 4, 5 , Erkki Isometsä Erkki Isometsä 2 Department of Psychiatry, University of Helsinki, Helsinki, Finland Find articles by Erkki Isometsä 2 , Jan-Henry Stenberg Jan-Henry Stenberg 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland Find articles by Jan-Henry Stenberg 1 , Jesper Ekelund Jesper Ekelund 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland Find articles by Jesper Ekelund 1 , Samuli I Saarni Samuli I Saarni 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland 6 Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland Find articles by Samuli I Saarni 1, 6 Author information Article notes Copyright and License information 1 Psychiatry, HUS Helsinki University Hospital, Helsinki, Finland 2 Department of Psychiatry, University of Helsinki, Helsinki, Finland 3 Department of Psychology and Logopedics, University of Helsinki, Helsinki, Finland 4 Department of Adolescent Psychiatry, University of Turku, Turku, Finland 5 Department of Adolescent Psychiatry, Turku University Hospital, Turku, Finland 6 Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland 7 Department of Psychiatry, Wellbeing Services County of Pirkanmaa, Tampere, Finland ✉ Corresponding author. # Contributed equally. Received 2025 Jul 3; Accepted 2026 Mar 5; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13088592 PMID: 41814249 Abstract Background Low-intensity cognitive behavioral therapy (CBT) based guided self-help (GSH) and guided internet-delivered CBT (GiCBT) have demonstrated equivalent effectiveness and superior cost-efficiency compared to traditional face-to-face CBT (fCBT) for treating depression and anxiety. This study addresses critical gaps in the current understanding of the effectiveness and cost-effectiveness of various CBT interventions for depression and anxiety within a stepped care model. Methods We describe a pragmatic multi-center randomized controlled trial (RCT) study with four parallel study protocols (the Finnish First-Line Therapies –Initiative study, FLT-step) for examining three widely used CBT interventions in public healthcare using a stepped care approach according to the FLT-Initiative. The study was preregistered in spring 2024, and participant recruitment began in September 2024. We compare the effectiveness and cost-effectiveness of three treatment approaches for depression (protocol 1) and anxiety (protocol 2) in a non-inferiority setting within the Finnish public healthcare: (A) stepped care (GSH followed by fCBT for non-responders), (B) fCBT, and (C) GiCBT. Non-inferiority margins reflect patient-detectable improvement: 1.7 points on the Patient Health Questionnaire (PHQ-9, protocol 1) and 1.5 points on the Generalized Anxiety Disorder 7-item scale (GAD-7, protocol 2). We plan to recruit 948 adults (≥ 16 years old) with depression (PHQ-9 ≥ 10 p) and 948 adults with anxiety (GAD-7 ≥ 10 p). A randomized substudy will examine the effect of waiting time (≤4 or ≥ 5 weeks) for the treatment outcomes of depression ( n = 115, protocol 3) or anxiety ( n = 115, protocol 4), comparing the stepped care model (A) and fCBT (B). In all four RCTs, the primary outcome measures are the within-individual change in depression (PHQ-9) or anxiety (GAD-7) symptoms at six months. Secondary outcomes include wellbeing, work and social ability, costs associated with illness, and quality of life. The follow-up will extend up to 20 years. Finnish national registry data will be used to supplement participant data and create population-matched controls to evaluate whether the interventions can prevent clinical episodes, reduce long-term societal costs, and decrease somatic morbidity. Discussion This extensive RCT will provide robust evidence on the comparative effectiveness and cost-effectiveness of low-intensity CBT treatments for depression and anxiety, and clarify the impact of waiting times on outcomes. Trial registrations (Registration date) ISRCTN14296278 (18 Sep 2024), ISRCTN63914711 (8 Oct 2024), ISRCTN10064801 (20 Sep 2024), ISRCTN14990924 (8 Oct 2024). Supplementary Information The online version contains supplementary material available at 10.1186/s12888-026-07962-w. Keywords: Depression, Anxiety, Randomized controlled trial, Cognitive behavioral therapy, Guided self-help, Internet-delivered CBT, Stepped care, Cost-effectiveness, Waiting-time Background Mental disorders are at the core of the public health care crisis. Given that depression and anxiety are the most common mental disorders in western countries [ 1 ], it is crucial to effectively organize the management of these disorders within primary care settings. An increasing amount of research supports the cost-effectiveness of psychotherapeutic interventions for depression and anxiety disorders [ 2 – 5 ]. Cognitive-behavioral therapy (CBT) has been found effective in the treatment of depression [ 6 , 7 ], subclinical depression [ 8 ], and anxiety disorders [ 9 ], but access to evidence-based treatments in primary care settings is often inadequate, also in Finland [ 10 , 11 ]. Furthermore, along with conventional face-to-face CBT (fCBT), low-intensity CBT-based guided self-help (GSH) [ 6 , 12 , 13 ], and guided internet-delivered CBT (GiCBT) [ 14 , 15 ] have been reported to be as effective as and more affordable than the traditional forms of psychotherapeutic treatment for depression and anxiety. The use of GSH, fCBT, and GiCBT as a part of a stepped care model could provide an applicable solution to make evidence-based treatment accessible in public healthcare. In stepped care, patients are treated at the lowest appropriate service level and stepped up only when clinically needed. In recent years, attention has been directed toward identifying the most effective strategies for establishing an optimally cost-effective sequence and equilibrium among various steps [ 16 – 19 ]. However, these optimization efforts tend to focus on immediate gains, often assuming that similar short-term average efficacies will result in analogous long-term outcomes. Thus, thoroughly evaluating the non-inferiority, cost-effectiveness, and long-term health economic impacts of three presumably equally effective treatments (GSH, fCBT, GiCBT), each with varying costs, can potentially provide valuable information for policymakers to optimize resource allocation and improve accessibility. In publicly funded healthcare systems, the goal is to optimize the cost-effectiveness on system level while ensuring access to necessary services to all. This calls for a modified stepped-care model that integrates a stratified approach, allowing patients to skip steps based on professional assessments of their individual needs. Modified stepped care has been implemented since 2008 in the British NHS Talking Therapies program, previously known as IAPT (Improving Access to Psychological Treatments) [ 20 ]. Subsequently, similar programs have been piloted and implemented in other countries including Norway [ 21 ], Ireland [ 22 ], France [ 23 ], Spain [ 24 ], and Australia [ 25 ] in various clinical settings including communities, schools as well as private and public healthcare services. In Finland, a stepped-care model has been launched through the First-line Therapies -Initiative (FLT), a comprehensive nationwide program aimed at providing early, evidence-based treatment for common mental health problems. This initiative also focuses on systematizing pre-treatment assessments within the national public health care system. Stepped-care models can differ in how strictly patients are required to proceed step-by-step through increasing treatment intensity. The Finnish First-Line Therapies (FLT) model is best characterized as stratified or modified stepped care, where a structured initial assessment guides patients directly to the most suitable level of care. This means that some patients may receive fCBT without first undergoing GSH, if clinically appropriate. At the same time, many patients with mild to moderate symptoms are initially offered GSH, with the option to step up to fCBT if symptoms persist. The present trial specifically evaluates this stepping process by comparing outcomes of GSH followed by fCBT when needed versus direct allocation to fCBT. In the British NHS Talking Therapies program, longer waiting times to treatment have been linked to lower recovery rates within service-providing units [ 26 ]. Consequently, addressing delays in these low-cost, easily accessible treatments is crucial for enhancing the cost-effectiveness of stepped-care systems at a societal level. This study protocol details the FLT-Step trial (Effectiveness of Psychosocial Interventions for Depression and Anxiety in the Stepped Care Model of the Finnish First-Line Therapies), which comprises four parallel multicenter randomized controlled trials (RCTs). The study was preregistered (ISRCTN14296278, ISRCTN63914711, ISRCTN10064801, ISRCTN14990924) [ 27 ], and participant recruitment began in autumn 2024. The protocol is described in adherence to the SPIRIT statement [ 28 ], TIDieR recommendations [ 29 ], and PRECIS-2 [ 30 ]. Methods/design This study consists of four parallel study protocols for four multicenter, randomized controlled trials (RCTs) aimed at treating depression or anxiety symptoms with CBT interventions in public healthcare routine settings. Primary hypothesis The primary hypotheses of protocols 1 (main study , depression) and 2 (main study , anxiety) : A stepped care model (sequential GSH followed by fCBT for non-responders) and GiCBT are both non-inferior to fCBT for treating depression and anxiety symptoms. Protocols 3 (substudy , depression) and 4 (substudy , anxiety) : Longer waiting time for the treatment is associated with poorer treatment response in depression (protocol 3) or anxiety (protocol 4) symptoms in both study interventions: (A) a stepped care model (sequential GSH followed by fCBT for non-responders) and (B) direct admission to fCBT. In addition to hypothesis testing, we strived to estimate the quantitative causal effect of waiting time on treatment efficacy, as only observational estimates exist thus far to our knowledge (see power calculations -section). The primary hypothesis in all four protocols will be tested at the primary outcome measurement point, which is six months after enrollment, for patients with a baseline score of ≥ 10 points on either the Patient Health Questionnaire [ 31 , 32 ] (PHQ-9, protocols 1 and 3) or Generalized Anxiety Disorder 7-item Scale [ 33 ] (GAD-7, protocols 2 and 4). The treatment response is assessed with the within-individual change in depression/anxiety symptoms measured by the PHQ-9/GAD-7, depending on the primary symptom being studied. Secondary hypotheses protocols 1, 2, 3 and 4 If non-inferiority is demonstrated, effectiveness of the stepped care model (sequential GSH followed by fCBT for non-responders) is superior compared to directly admitting patients to fCBT when treating (1) depression symptoms (baseline score of ≥ 10 p on PHQ-9) or (2) anxiety symptoms (baseline score of ≥ 10 p on GAD-7), assessed six months after enrollment. Stepped care is more cost-effective than directing all patients with depression symptoms (baseline score ≥ 10 p on PHQ-9) or anxiety symptoms (baseline score ≥ 10 p on GAD-7) directly to fCBT assessed six months after enrollment. GiCBT is more cost-effective than directing all patients with depression symptoms (baseline score ≥ 10 p on PHQ-9) or anxiety symptoms (baseline score ≥ 10 p on GAD-7) directly to fCBT (protocols 1 and 2 only) assessed six months after enrollment. Data collected by the Finnish Therapy Navigator (FTN) [ 34 ], a digital tool to help assess individual needs and symptom profile for psychotherapy, can be used to predict responses to treatment by using a multivariate model (an ability that would allow for treatment personalization) [ 35 ]. All treatment approaches studied are cost-saving in the long term compared to matched population controls, when direct and indirect health care, social care, employment, and societal costs are considered (see section “health economic evaluation”). Patients seeking treatment with subclinical depressive or anxiety symptoms (baseline score of 5–9 on the PHQ-9 or GAD-7) benefit from the studied treatment approaches, in terms of reduced risk of developing clinical episodes, reduced total long-term societal costs, and decreased somatic morbidity. Longer waiting times for the study treatments are associated with poorer overall long-term outcomes when direct and indirect health care, social care, employment, and societal costs are considered (protocols 3 and 4 only, see section “health economic evaluation”). Design Protocols 1 (depression) and 2 (anxiety) each include three treatment arms of equal sample size. The following treatment approaches are included in research arms: (A) GSH + fCBT (sequential GSH followed by fCBT for non-responders, planned n = 316), (B) fCBT (planned n = 316), and (C) GiCBT (planned n = 316). Non-responder is a participant who still scores above the clinical cut-off on the respective symptom measure i.e. PHQ-9 ≥ 10 p [ 32 ] or GAD-7 ≥ 10 p [ 36 ]. The evaluation of non-responder status will occur at the end of the GSH intervention. The treatment sites participating in this trial are in public primary mental health care services in Southern and Western Finland across nine wellbeing service counties (population approx. 3.4 million), covering 60% of the total population of Finland. Alongside the RCTs of protocols 1 and 2, we will collect a convenience sample of patients with subclinical symptoms of depression (PHQ-9 5–9 p) or anxiety (GAD-7 5–9 p) to receive treatment in all three treatment arms (A, B, or C). Protocols 3 (depression, planned n = 115) and 4 (anxiety, planned n = 115) concentrate on a substudy examining the impact of waiting time, featuring two equally sized treatment arms: (A) GSH followed by fCBT for non-responders, and (B) fCBT. Participants in both arms are randomly assigned (1:1) to two groups: (I) those commencing treatment in less than 4 weeks, and (II) those starting treatment after 5 weeks or more. These substudies are being carried out in the Pirkanmaa wellbeing service county, which has a population of approximately 540,000, located in Central Finland. The regional sites with sufficiently long waiting times for the studied interventions at the study initiation were included in the trial. For all study participants, including those involving patients seeking help for subclinical symptoms, the study data will be combined with data from Finnish national registries. Matched population controls will be identified to conduct a comprehensive registry study among these patients. The flow of patients through the protocols 1–4 is presented in Fig. 1 . Fig. 1. Open in a new tab The flow of patients through the study. Left panel: The main study protocol 1 (depression) and protocol 2 (anxiety). Right panel: The substudy protocol 3 (depression) and protocol 4 (anxiety) Recruitment and patients To achieve maximal external validity in the study, all patients evaluated by a professional in primary care as suitable for the studied treatments for depression or anxiety (e.g., GSH, GiCBT, and fCBT) are invited to participate. The FTN is used to assess treatment needs as part of routine practice by the trained health care professional responsible for clinical signposting. If the patient is interested in participating in this trial, the clinician will schedule an appointment with a research nurse. During this appointment, the research nurse will provide information about the study, review the inclusion and exclusion criteria, and give the patient a written study information sheet. The inclusion and exclusion criteria used in the study are described in Table 1 . Table 1. Inclusion and exclusion criteria in the FLT-Step trial (protocols 1–4) Inclusion Criteria Exclusion Criteria • ≥ 16 years of age • Suitable for studied treatments (GSH, GiCBT or fCBT intervention) for depression/anxiety (cf. exclusions) • Depression protocol: PHQ ≥ 10 p • Anxiety protocol: GAD-7 ≥ 10 p • General exclusion criteria for studied treatments (i.e. recommended more intensive or other treatment forms) • Serious suicidal thoughts, plans or any self-harming act or suicidal attempt within the past 2 months. • Ongoing other psychological treatment for depression and/or anxiety • Cognitive impairment • Inability to speak, read and write Finnish • Currently symptomatic psychotic illness or bipolar disorder • Drug or alcohol dependence In a sub-clinical convenience sample adjacent to protocols 1 and 2: • Depression protocol: PHQ-9 5–9 p • Anxiety protocol: GAD-7 5–9 p Open in a new tab GSH guided self-help; GiCBT guided internet-delivered cognitive-behavioral therapy; fCBT face-to-face cognitive-behavioral therapy; PHQ-9 Patient Health Questionnaire; GAD-7 Generalized Anxiety Disorder 7-item scale Patients deemed eligible for the study will be asked by the research nurse to provide informed consent to participate. If the patient does not wish to participate, permission to use the information gathered in the FTN during the prescreening process of the study is requested. This approach ensures a more comprehensive understanding of the representativeness of the sample of primary care patients, thereby enhancing the overall quality and applicability of the study findings. Further, the pragmatic vs. explanatory nature of the trial was evaluated using the PRECIS-2-tool [ 30 ]. Five members of the research team (EEH, SES, KM, MKM, THR) representing expertise in the CBT interventions, health service settings and statistical methods independently rated each of the nine PRECIS-2 domains. The agreement between raters across nine items was high (ICC 0.92 (95% CI 0.80–0.98). All discrepancies were discussed until consensus was reached. The PRECIS-2 consensus is illustrated in Fig. 2 . Fig. 2. Open in a new tab Level of pragmatism of the current study (PRECIS-2) Power calculations Protocols 1 and 2 Primary estimands in protocols 1 and 2 were simple differences in mean PHQ-9 change scores between treatment arms, with treatment-policy (intention-to-treat) primary treatment outcome assessment at 6 months, irrespective of therapy completion (Fig. 1 ). Three types of power calculations were essential to these protocols. First, we tested the hypothesis that difference between PHQ-9 score change in the stepped care model (GSH + fCBT) and fCBT groups is no greater than a margin of 1.7 points, which amounts to the estimated threshold at which patients with moderate-severity symptoms can detect the difference between “feeling the same” and “feeling better” after a treatment (37). Second, the analogous margin for GAD-7 was 1.5 points. This threshold is slightly more stringent (leads to larger sample-size requirements) than a previously used statistically motivated margin (38), but more clinically meaningful in our opinion. We computed statistical power to detect non-inferiority (39) assuming no efficacy difference exists between the treatment arms and assuming the relevant clinical population standard deviation of PHQ-9 is 6 points and that of GAD-7 is 3.6 points, as previously observed in Finnish context (40, 41). Third, we computed statistical power to detect superiority between treatment groups when their standardized mean difference is d = 0.3 (i.e., small but larger than the standardized subjective-difference detection threshold of PHQ-9 from above 1.7/6 ≈ 0.28). We used significance level α = 0.025 for one-sided non-inferiority test, which corresponds to 95% two-sided confidence interval being wholly on the better side of the margin. We used level α = 0.05 in the two-sided tests of superiority. Figure 3 shows the results from the first three statistical power calculations. We observed that a treatment group size 263 leads to at least 90% statistical power for all the three types of study questions. We inflated this estimate by 20% to allow for some participant follow-up attrition, resulting in a target of 316 patients per intervention group. Fig. 3. Open in a new tab Statistical power as a function of the treatment-arm size. Power for the PHQ-9-based non-inferiority test is shown with the thick solid line, the same for GAD-7 with the dashed line, and the power of the superiority test with the dotted line The detailed scripts for the study’s power calculations are available from the authors upon request. Additionally, similar matching and cross-validation will be conducted on the registry data linked to the study as described in Rosenström et al. [ 15 ]. Permission for the use of collected background information (FTN) will be requested from those who decline to participate. The impact of non-adherence will be investigated using multiple imputation [ 42 ]. Note that dropout from treatment does not imply non-adherence to treatment protocol as the 6-month assessment is independent of the treatment. We have much background data, and in multiple imputation, “[i]ncluding as many predictors as possible tends to make the MAR assumption more plausible. [but] .it is expedient to select a suitable subset of data that contains no more than 15 to 25 variables” [ 42 ]. To avoid subjectivity, the prespecified plan is to follow established standards and otherwise use the algorithmic quickpred- and flux-functions of mice [ 42 ]. Post-hoc sensitivity analyses, such as tipping-point analyses, may be used to investigate sensitivity of conclusions for possible unobservable non-random missingness. Regarding observation clustering, a standard solution is to divide the sample size with the design effect 1 + ( m − 1) ρ to arrive at effective sample size, where the intraclass correlation ρ typically “is small (often < 0.05)” and m is average cluster size [ 43 ]. However, the definition of a “cluster” is not straightforward here: some treatments (fCBT) have physical sites, whereas others do not obey geology (GiCBT). Therapist effects are relatively meaningless because so many are involved ( m becomes very small) and difficult to track (e.g., due to substitutes). Therefore, we computed non-clustered sample size estimates but required greater power than is typical (i.e., we aimed for power 0.9 rather than the typical 0.8). A sceptical reader could, e.g., divide our target sample size n by 1 + ( n /45 − 1)×0.05 for the estimated 45 sites of fCBT, meaning that one of the compared two groups would have an effective sample size 212 while the other would remain at 263 (cf. Figure 3 for power estimates falling between these limits, containing the effective power of the comparison under such clustering assumption). The power for the comparison should remain adequate for our primary estimand that will not explicate clustering, whereas sensitivity analyses may explore it. Protocols 3 and 4 Fourth, and finally, we estimated statistical power to detect the (previous) observational effect of waiting time of treatment on the subsequent treatment effect (assuming that the observational effect of Clark et al. [ 26 ] is also a causal effect; i.e., this effect is what our RCT tests). We digitized the Figure A of Clark et al. [ 26 ] (with minor loss in accuracy) using the “digitize” R package [ 44 ] and fitted a standard local regression smoother on the resulting data points (Fig. 4 A; result looks much like Clark et al., 2018, r was computed from the digitized data). We simulated treatment success per waiting time by taking the mean-prediction of this local regression plus a random draw from a normal distribution with standard deviation corresponding to the model residuals (i.e., 4.14). We can also investigate what happens in more noisy data, having double the residual standard deviation, as Clark et al. [ 26 ] modeled averages over clinical commissioning groups rather than individual patients’ scores (Fig. 4 B; r computed from 10 thousand simulated observations, of which first 195 are plotted). Fig. 4. Open in a new tab A ) Data (circles) digitized from Clark et al. (26; their Figure A ) using the R software package “digitize”, and a local regression fit (solid line) with default parameters of R (loess function). B ) Data simulated from the local regression model (of panel A ) with twice the residual standard deviation compared to original. Wait times were simulated from a truncated exponential distribution (see text for details) For research-logistic reasons, we were requested to use two waiting-time groups instead of a continuous-valued allocation, for simplicity. We created two waiting-time groups by simulating uniformly distributed waiting times on 6 to τ days and on τ + g to 100 days. By varying the parameters τ (threshold between short and long wait) and g (gap between the groups) in the simulations, we ensured that random allocation of 48 patients to a group who waits 1 to 4 weeks (but no longer than 4 weeks) and 48 patients to a group who waits over 5 weeks resulted in > 90% power even in the above-modeled high-noise case when using a standard t-test. With the 20% sample inflation to guard against attrition, we strived to collect 115 patients per studied diagnoses (115 with depression for protocol 3 and 115 with anxiety symptoms for protocol 4). This is the power calculation for our primary waiting-time hypothesis test. Besides verifying the existence of a causal effect of waiting time on treatment outcome in the simplest possible way, we also want a quantitative estimate on the causal effect of waiting days on symptom change (to support various later service-optimization efforts). For this, we will use instrumental variable analysis, with the random waiting-time group allocation being the instrument. We studied this setting by simulating the above-defined 96 patients (2 × 48) using the high-noise case of Fig. 4 B, to verify the procedure and sufficient statistical power. Across 10 000 repeated simulations, ordinary least squares regression of outcome on waiting time was − 0.12 (95% CI = − 0.15 to − 0.09). Two-stage least squares estimate using waiting-time group allocation as the instrumental variable was − 0.13 (CI = − 0.17 to − 0.10), indicating good agreement. We then investigated a scenario where a normally distributed confounding variable is added to waiting times and outcome after allocation, such that it cancels their correlation. This procedure fully diluted the ordinary least squares estimate, to − 0.00 (CI = − 0.07 to 0.06), but not the two-stage least squares estimate, which was still − 0.14 (CI = − 0.24 to − 0.05). Statistical power to detect the causal effect was still 83%. Thus, we can be relatively certain that our procedure provides a valid causal estimate, even in the presence of quite catastrophic failures of experimental control. We used R package “ivmodel”, version 1.9.1, in these analyses. Register trial sample size and statistical analysis It is not ethical and feasible to (a) monitor patients with heavy questionnaires over multiple years and (b) include a group of untreated patients. Therefore, we collect register-based data on the participating patients and population controls. (a) Because of the passive register sampling, we do not expect additional attrition for the long-term follow up of patients and, thus, above power calculations pertain here too. (b) It is possible to create balanced control groups representing untreated patients without risking over-fitting to data (cf. [ 15 , 45 ]). In this approach, more data, including data on familial risks (cf. [ 15 ]), allows for learning more comprehensive balancing models and the ‘true’ balancing model is always unknown. Therefore, we strive to sample as much from the general population as feasible with the pertinent future funding, at the minimum, as large a population sample as the clinical sample. Randomization Randomization will be performed using the random permuted block method with SAS software, Version 9.4 of the SAS System for Windows (SAS Institute Inc., Cary, NC, USA). Stratification (disorder, severity of disorder, and area of study site) will be used. A biostatistician (EL) outside the study team has prepared the randomisation lists, which have been entered into the REDCap software. Patient randomization will be conducted in REDCap by the research nurse at the press of a button, after they have been deemed eligible and have read the study information sheet and signed the informed consent. A biostatistician outside the study team (EL) will keep the randomization key confidential until the analyses for the primary follow-up time point (six months after enrolment) have been completed. The treatment arms differ in both format and duration (e.g., guided self-help vs. face-to-face CBT) making blinding of participants and intervention providers not feasible. Given the lack of blinding, risk of bias will be mitigated by pre-specifying statistical analyses and using validated, standardized outcome measures. Protocols 1 , 2 and the subclinical sample : In the initial phase of the study, participants in Southern Finland and Western Finland collaborative areas will be randomized (1:1:1) into the following three (A-C) treatment arms separately for depression (protocol 1) and anxiety (protocol 2): (A) a stepped care model (GSH, followed by fCBT for non-responders); (B) fCBT; (C) GiCBT. To minimize potential randomly occurring biases in the treatment-group compositions, we employ stratified randomization. The stratification will be based on two criteria: (1) symptom severity (protocols 1 and 2 ≥ 10 points on the PHQ-9/GAD-7; subclinical sample 5–9 points on the PHQ-9/GAD-7;) and (2) research location (wellbeing service county). Protocols 3 and 4 : In Pirkanmaa wellbeing service county, where we are investigating the effect of waiting time on treatment efficacy among patients with clinical symptoms of depression or anxiety (PHQ-9/GAD-7 ≥ 10 points), participants will be randomized into two treatment arms (1:1) separately for depression and anxiety: (I) those with a waiting time of less than 4 weeks and (II) those with a waiting time of more than 5 weeks. Before waiting-time randomization, the patients are randomized to two treatment arms (A) stepped care model: GSH, followed by fCBT non-responders or (B) fCBT. Assessments Finnish Therapy Navigator (FTN) Patients seeking help for mental health concerns in primary care services are asked to complete the FTN [ 34 ], a digital tool designed to help assess individual needs and preferences for psychosocial treatments. FTN, in brief, collects information on current symptoms, as well as previous treatment, current social and working ability and treatment preferences. The FTN also covers possible traumatic events, recent crises and significant life events. The FTN is freely available on the internet ( https://www.terapianavigaattori.fi/ ). The FTN comprises several well established and validated screening questions and accordingly applied self-reported symptom-specific measures: PHQ-9 [ 32 ], GAD-7 [ 36 ], Social Phobia Inventory (SPIN) [ 46 ], Panic Disorder Severity Scale-Self Report (PDSS-SR) [ 47 ], Alcohol Use Disorders Identification Test-Concise (AUDIT-C) [ 48 ], Alcohol Use Disorders Identification Test (AUDIT) [ 49 ] for those scoring ≥ 6/5 p (men/women) in AUDIT-C, Drug Use Disorders Identification Test (DUDIT) [ 50 ], Obsessive-Compulsive Inventory –Revised (OCI-R) [ 51 ], Burnout Assessment Tool (BAT-12) [ 52 ], and Trauma Screening Questionnaire (TSQ) [ 53 ]. In January 2025, the Emotional and Psychological Single-Item Outcome (EPO-1) [ 54 ] was also added to the FTN. All patients suitable for the studied treatment approaches for depression or anxiety are offered the possibility to participate in the study. The suitability is assessed by a trained health care professional utilizing the information gathered by FTN. The FTN creates a case summary, which the clinician complements with a manualized semi-structured interview. The FTN interview manual supports treatment selection in line with clinical practice guidelines, but treatment decision is always left for the assessing clinician to negotiate with the patient. Interventions Guided self-help (GSH) Separate web-based self-help programs for depression and anxiety, both based on evidence-based CBT models, are used in the study. The programs are focused on the core elements of CBT: identifying and restructuring dysfunctional cognitions and emotional responses, and their interaction on patients’ current behavior through psychoeducation and exercises. The online materials contain: (1) psychoeducation and illustrations about the symptoms, severity, treatment, and causes of mental health problems and the cognitive model of the described problem, (2) multimedia exercise instructions and materials which target the change in thought and behavior patterns, and (3) information of several lifestyles and habits which generate and maintain the problems or symptoms. In the self-help material for depression, the patient is tasked to monitor their mood and recognize their thoughts, emotions and behaviors and their inter-related connections, which maintain depressive symptoms. With increased attentive skills and knowledge of one’s own beliefs, the patient is tasked to challenge and modify these beliefs and to form plans to increase pleasant and value-based activities in their life. The self-help material for anxiety focuses on modifying the catastrophic thinking patterns and dysfunctional beliefs that worrying is serving a useful function. The behavioral techniques include relaxation training, scheduling specific ‘worry time’ as well as planning pleasurable activities, and controlled exposure to thoughts and situations that are being avoided. GSH is administered utilizing symptom-specific (depression/anxiety) self-help materials, accessible at Mentalhub.fi website ( https://www.mielenterveystalo.fi/en ), in conjunction with 1–5 (an average of 3) face-to-face, video conference, or telephone sessions with a trained healthcare professional (e.g. nurse, clinical psychologist, social worker) working in primary healthcare [ 55 ]. These professionals assist patients in a collaborative manner by establishing goals, reviewing the self-help materials, and practicing the included exercises. Session topics are guided by the client’s preferences and shared decision-making with the healthcare professional. Progress, outcomes, and possible side effects are routinely monitored, but intervention adherence or fidelity is not. The use of digital materials helps ensure a degree of treatment fidelity by standardizing the delivery and structure of core content. Guided Internet-delivered cognitive behavioral therapy (GiCBT) The HUS GiCBT programs are therapist supported and diagnosis-specific, each designed by an expert group including leading national level experts, selected separately for each disorder. During the development phase of GiCBT programs, patient feedback was systematically collected and incorporated into the final refinement of the intervention. The programs are divided into three phases: (1) introduction including psychoeducation, (2) actual treatment phase, and (3) summarizing phase including a plan for relapsing prevention and maintenance of the treatment results. Treatments consist of altogether 7–12 weekly sessions, which the patient can proceed at her own pace. The exact informational materials used in the programs are not publicly available, but they are similar with GSH and fCBT programs. The therapists are trained mental health professionals working at HUS specialized in certain treatment programs and disorders. The patient is supported by the therapist throughout the treatment by asynchronous written messaging, in which the therapist provides feedback and topics tailored to the patient’s needs. GiCBT for depression : The program for Depressive Disorder consists of seven treatment sessions and one follow-up session. Sessions include different topics each and build a proceeding treatment program. The topics and exercises are for example goal setting, behavioral activation, cognitive restructuring, advice on balanced life and relapse prevention. Throughout the program the outcomes and possible side effects are assessed with validated symptom measures [ 15 ]. GiCBT for generalized anxiety : The program for Generalized Anxiety Disorder (GAD) consists of 12 consecutive sessions and a follow-up session 3 months after treatment completion. The program is theoretically based on several models of GAD and anxiety, including aspects of the cognitive avoidance model, model of intolerance of uncertainty, metacognitive therapy, and acceptance and commitment therapy, as well as social aspects, such as assertiveness training. The sessions include text and videos, as well as educational illustrations, example stories, therapeutic exercises, and homework [ 40 ]. Throughout the program the outcomes and possible side effects are assessed with validated symptom measures. Face-to-face cognitive behavioral therapy (fCBT) The recommended length of the fCBT intervention is5–10 (the average in routine care is 7 sessions) weekly face-to-face sessions with a therapist trained to deliver symptom-specific structured fCBT. The intervention protocols are specific to depression and anxiety symptoms and are grounded in evidence-based CBT practices. The content of the sessions is guided by session-specific materials (Supplement 2 ) but is also tailored based on patient’s preferences and shared decision-making with the therapist. They encompass the same core elements as the GSH and GiCBT models, including goal setting, psychoeducation, exercises, and homework assignments [ 56 ]. The treatment programs utilize the same informational materials used in the GSH programs. Therapists providing fCBT are mental health professionals employed in public sector primary care units, including nurses (over 80%), clinical psychologists, and social workers working in primary healthcare. All therapists have received specialized training in administering CBT-based interventions for anxiety and depression. The one-year (15 ECTS) training program includes asynchronous learning activities on a digital learning platform, as well as supervised interventions: 30 h of supervision and a minimum of 70 h of clinical casework using at least two specific treatment protocols [ 56 ]. Supervision takes place in small groups (3–5 pers) biweekly or once a month. To ensure competence and fidelity supervision is provided exclusively by professionals with formal training in CBT methods and FLT CBT training — licensed CBT psychotherapists, psychologists, or other mental health care professionals with documented CBT training and experience in delivering the present intervention. The therapists participating in this study have either completed or are currently undergoing their supervised therapy training according to this curriculum, and this information about the intervention provider’s training status will be collected during the study. Adherence to the intervention manual is not systematically audited due to the project’s scope and naturalistic nature. Professionals conducting all three treatment approaches receive clinical supervision to reinforce adherence to treatment protocols. Also, the fidelity is supported by the extensive use of written materials (e.g. describing exemplar structure of each session, suggested exercises, documentation). Intervention will be discontinued if the participant requests it, if exclusion criteria are met, or if other reasons arise that require discontinuing the intervention, such as a somatic illness or relocation to another area. The trial does not interfere with standard care practices, such as pharmacotherapy. The information regarding other treatments, including pharmacotherapies is collected. The discontinuation of studied interventions and reasons for discontinuation are monitored, and all study participants will be analysed within the groups to which they were randomized, following the intention-to-treat principle. In order to minimize interference with the treatment protocol, we will not implement active retention procedures for the study interventions. However, follow-up questionnaires will still be sent to all participants who do not explicitly withdraw their participation. The full content, duration, and intensity of the intervention protocols is available in Supplement 2 . Evaluation Pretreatment information is gathered in the clinical assessment together with FTN or clinical assessment during the recruitment process described above. Baseline information (as well as all data except registries used in the study) is gathered at the beginning of the treatment via the secure REDCap [ 57 ] software. Baseline self-reported measures include the PHQ-9, GAD-7, AUDIT-C, and EPO-1 measure applicable to all participants. Additional measures are administered only to those who screen positive for the corresponding symptoms, which include the PDSS-SR, SPIN, TSQ, OCI-R, BBGS, BAT-12, AUDIT, and DUDIT. Participant schedule of enrolment, interventions, and assessments throughout the study are described in Fig. 5 . Fig. 5. Open in a new tab Timeline of participant enrolment, interventions, and assessments throughout the study Primary outcome measure The primary outcome measure for protocols 1 and 3 is the within-individual change in depression symptoms, as assessed by the PHQ-9, and for protocols 2 and 4, the within-individual change in anxiety symptoms, as assessed by the GAD-7, from baseline to six months after enrolment. The PHQ-9 (protocols 1 & 3) or GAD-7 (protocols 2 & 4) is administered weekly for the first 16 weeks of the intervention and at follow-up time points (e.g., 4, 6, 8, and 12 months, as well as 2, 5, 10, 15, and 20 years) to facilitate intention-to-treat (ITT) analysis and to model symptom changes over time. Cohen’s d effect sizes will also be reported for the main outcome measures. PHQ-9 and GAD-7 were selected as primary outcome measures due to their widespread use in international stepped-care systems (e.g., UK Talking Therapies [ 20 ], PMHC in Norway [ 21 ], PsicAP in Spain [ 24 ]), enabling direct comparability across service models. Both instruments have demonstrated good sensitivity to change in routine-care contexts, including Finnish CBT service settings [ 58 ]. We also acknowledge recent discussions regarding measurement invariance in PHQ-9 [ 59 ] and address this in the Discussion. Secondary outcome measures The secondary outcome measure is the proportion of patients in remission [ 32 , 36 ], defined as scoring below the clinical cut-off (< 10 p in PHQ-9 or GAD-7 respectively), and those experiencing a clinically significant change in depression symptoms, indicated by a change of ≥ 5 [ 60 ] points on the PHQ-9 or ≥ 4 points on the GAD-7 [ 61 ]. For protocols 1 and 3, the assessment is conducted using the PHQ-9 and for protocols 2 and 4, the assessment is conducted using the GAD-7. Data collection at various time points during the trial is illustrated in Table 2 . Table 2. Overview of measures and assessment times Trial period Enrollment Postrandomization Long-term FUs Pre-screen ing Base line Treatment initiation Dur ing treatment Post-treatment 4 mos 6 & 8 mos 1 , 2 , 5 , 10 , 15 and 20 yrs Background information History of depression Somatic comorbidities x Exercise, sleep, smoking, BMI x x x x Marital status, socioeconomic information x x x Previous psychotherapeutic interventions x Use of psychotropic medication x x x x Employment status x x x x Baseline diagnosis-specific measures (only for screen positives in FTN) PDSS-SR x SPIN x TSQ x OCI-R x BBGS x BAT-12 x AUDIT-C and AUDIT (if AUDIT-C ≥6/5 p, men/women) x DUDIT x Primary outcome measure (6 months) NOTE: 16 times weekly after treatment initiation PHQ-9/GAD-7 according to the target symptom x x x x x x x Secondary outcome measures / covariate information PHQ-9/GAD-7 other than the target symptom x x x x x x x AUDIT-C and AUDIT (if AUDIT-C ≥6/5 p, men/women, starting from 6 mos) x x x x PSSS-R x x x x WSAS x x x x EQ-5D-5 L x x x x Euro-HIS (Quality of Life) x x x x EPO-1 x x x x x Number of sessions attended x Patient experience x Subjective work ability x x x x x Healthcare visits (previous 12 months) x x Income in the previous year x x Open in a new tab FU, follow-up; FTN, Finnish Therapy Navigator; GSH, Guided self-help; fCBT, face-to-face cognitive-behavioral therapy; PHQ-9, Patient Health Questionnaire; GAD-7, Generalized Anxiety Disorder 7-item scale; iCBT, Internet-delivered cognitive-behavioral therapy; PDSS-SR, Panic Disorder Severity Scale-Self Report; SPIN, Social Phobia Inventory; TSQ, Trauma Screening Questionnaire; OCI-R, Obsessive-Compulsive Inventory –Revised; BBGS, Brief Biosocial Gambling Screen; BAT-12, Burnout Assessment Tool; AUDIT-C, Alcohol Use Disorders Identification Test-Concise; AUDIT, Alcohol Use Disorders Identification Test; DUDIT, Drug Use Disorders Identification Test; PSSS-R, Perceived Social Support Scale-Revised; WSAS, Work and Social Adjustment Scale; EQ-5D-5 L, Health Related Quality of Life; Euro-HIS, Quality of Life; EPO-1, Emotional and Psychological Outcome Register data collection Registered data is collected for all patients, matched population controls and family members in order to (a) conduct detailed analyses of consequences and societal costs of investigated interventions and (b) modelling of alternative courses of treatments on population level. Register data is collected from the following registries: Digital and Population Data Services Agency (DVV), Finnish Centre for Pensions (ETK), The Social Insurance Institution of Finland (KELA), Finnish Institute for Health and Welfare (THL) and Statistics Finland’s Register. The main groups of variables are presented in Table 3 . Table 3. Main groups of register data from patients, matched population and family controls Main groups of register data to be collected Health care service use (e.g. date, procedure, diagnosis, provider’s professional group, psychotherapeutic treatments such as rehabilitative psychotherapy, online therapy, other) Diagnoses (any ICD diagnosis) Medications (prescriptions and purchases) Rehabilitation (rehabilitation services and benefits) Physiological measurements and lifestyle factors (depending on availability: e.g., height, weight, smoking, laboratory tests) Cause and date of death Employment (e.g. education, work classification, income, sick leaves, labor market status, employment) Pension benefits (incl. disability and old-age pensions) Other social benefits (e.g. economic benefits like housing support and reimbursed travel expenses) Income information (incl. work and pension income) Open in a new tab Health-economic evaluation The study includes a wide range of data collection that allows for comprehensive evaluation of both costs and consequences of treatments. This study adopts a societal perspective to estimate the economic impact of investigated treatments and waiting time for treatment of depression and anxiety. The analysis includes direct healthcare costs, indirect costs from lost productivity, and broader societal costs such as social services and illness-related changes in income. Health-utility analyses are conducted using EQ-5D-5 L based health utility and register-based mortality data during follow-up. Incremental cost-effectiveness ratios (ICERs) will compare the costs and effectiveness of different interventions. Additionally, a budget impact analysis will estimate the potential savings of implementing the optimal treatments at a national level. Cost-related data is collected with questionnaires from patients at baseline and follow-up visits by asking about their use of different health services, medications, employment status, sickness absence days and household income by using standard questionnaires employed in current national health surveys [ 62 ]. Healthcare visits will be converted to money using standard Finnish healthcare unit cost Table [ 63 ] or more recent official estimates. All costs are reported in euros (€). Costs from previous years will be adjusted for inflation based on the Finnish Consumer Price Index (CPI). Future costs and benefits will be discounted at a rate based on national recommendations (currently 3%) with sensitivity analyses for different rates. For long-term follow-up and wide societal perspective of costs and consequences, register data will be used. Detailed modelling of total societal costs and their constituent factors from register data will be planned, tested and reported before the health economic analyses are conducted. This can be done in detail when registered data is obtained, as it is not fully transparent what kind of register data will be available for long-term follow-up. In brief, the study estimates societal costs as widely as possible using Finnish registry data. We categorize costs into direct healthcare costs, productivity losses, and other societal costs. Some costs are directly available from registries, whereas others will be converted using unit costs recommended for national health economics evaluations. Direct health care costs are available in detail from registries. The costs included will include at least hospitalizations, outpatient visits to different healthcare professionals, medications, and procedures. Cost-conversions will be done based on national unit costs for different interventions. Indirect costs from productivity loss arise from sickness absences, disability pensions, unemployment and loss of income due to other factors. The economic impact of sickness benefits and disability pensions is assessed using the duration of absence and average monthly earnings, but direct payments from registers are used where applicable. Individual income data is used to model changes in income. Other societal costs available from registries including rehabilitation, social service and patient travel costs. Travel costs are estimated by multiplying the number of reimbursed trips by standard trip costs or using direct payment data where applicable. Rehabilitation and social service costs are derived from unit costs of long-term care and rehabilitation programs. Housing benefits and social assistance payments are accounted for using direct register data. Monitoring Data collection and storage The data will be collected by digital surveys and research nurses working for the wellbeing counties and/or for Helsinki University Hospital (HUS). The data will be collected using REDCap software [ 57 ] for HUS and stored on a secure server of HUS. Built-in data quality rules are applied to minimize data entry errors. Entries made by research will be monitored regularly by study coordinator both manually and using REDCap’s quality control tool. All identified discrepancies will be documented in a separate error log. Research nurses will register a new patient in REDCap only after the patient’s identity has been verified in accordance with the protocol of the respective wellbeing county. Study participants will provide the research nurse with a personal code from the FTN, which allows access to their results for a limited time. The research nurse will manually transfer the information from the Therapy Navigator to REDCap as part of the study’s baseline data collection. Patients will complete the questionnaires directly in REDCap via a link sent to the email address they have provided. Automated reminders and weekly symptom questionnaires will also be sent through REDCap starting from the date of the first intervention session and continuing for 16 weeks. The research group of FLT-Step will supervise and support the research nurses from various wellbeing counties to ensure the proper conduct of the trial. To maintain compliance with the study protocol—such as inclusion and exclusion criteria, informing participants, and obtaining consent—the research nurses meet biweekly with members of the clinical and research team to address any questions or concerns that may arise. Additionally, continuous consultation is available. In the event of a protocol modification, the trial registries will be updated, and the research nurses as well as the participating wellbeing counties will be informed by the research group. Since this trial takes place in a naturalistic setting within public healthcare, we do not interfere with the treatment practices, which are aligned with national treatment guidelines. Patient recruitment for the study began in October 2024 and is planned to continue at least until June 2026. The main results of the study will be published in scientific journals. The research team is responsible for preparing the articles that present the study findings. Authorship will be determined in accordance with the criteria set by the International Committee of Medical Journal Editors (ICMJE). Safety All participants will receive comprehensive information about the study protocol to ensure they can provide truly informed consent. Guardians of participants aged 16–17 will receive a written notification of their ward’s participation. All patients in the study will receive at least one evidence-based treatment (GSH, GiCBT or fCBT). Patients who do not respond to the initial GSH will be offered fCBT. The level of care provided to study patients will be at least equivalent to what they would receive in current clinical practice. Participants are only randomized to a waitlist arm in clinical contexts (Pirkanmaa) where waiting is expected regardless of the research project. Since HUS GiCBT treatment does not entail any waiting times, it is not included in the protocols 3–4. We do not anticipate increased risks for patients participating in this study, based on existing research literature on low-threshold CBT-based interventions (see background). To minimize the risk of adverse effects, we have excluded individuals with severe acute suicidality or a recent suicide attempt from the study. If a participant reports suicidal thoughts at any point on the PHQ-9 questionnaire, REDCap will automatically provide instructions on where to seek help and additional support. These procedures align with the operating principles of the FTN. During the treatment, providers will clinically and through questionnaires monitor patients’ wellbeing and suicidality. All patients participate voluntarily in accordance with the Helsinki Declaration, and any adverse events will be closely monitored and reported. In the event of adverse events or symptom deterioration, therapists are trained to refer patients for appropriate evaluation and treatment. If serious suicidal thoughts, plans, or any self-harming acts are reported at follow-up points, patients will be instructed to contact healthcare emergency phone services for further evaluation and instructions. Participation, non-participation, or opting out of the study after initial consent will have no impact on the patient’s ongoing treatment or access to other treatment modalities. The possible risks associated with these forms of treatment are minimal and fully comparable to current standard treatments. Compensation for any potential patient injuries can be sought from the Patient Insurance Centre in accordance with standard Finnish healthcare procedures. The study has received approval from the ethics committee of HUS Helsinki University Hospital (HUS/6234/2023). The committee includes lay representatives participating in the evaluation process to ensure consideration of public and patient perspectives. The necessary research permits have been obtained from HUS Helsinki University Hospital and all participating wellbeing service counties. Discussion The FLT-Step study aims to address critical gaps in the current understanding of the effectiveness and cost-effectiveness of various CBT interventions for depression and anxiety within a stepped care model. The primary hypothesis posits that a stepped care model (sequential GSH followed by fCBT for non-responders) and GiCBT are non-inferior to fCBT for treating depression and anxiety symptoms. This hypothesis is grounded in existing literature that supports the efficacy of CBT-based interventions of different delivery formats [ 5 – 8 , 13 , 14 , 64 , 65 ]. The study’s design (protocols 1 and 2), which includes three treatment arms, allows for a comprehensive comparison of these interventions within a naturalistic stepped-care approach. One secondary hypothesis suggests that the effectiveness and cost-effectiveness of the stepped care model (sequential GSH followed by fCBT for non-responders) may be superior to directly admitting patients to fCBT when treating depression or anxiety symptoms [ 19 ]. This aspect of the study is crucial for public healthcare systems that aim to optimize resource allocation while maintaining high-quality care. The study also hypothesizes that longer waiting times for the study interventions will be associated with poorer treatment responses in depression or anxiety symptoms (protocols 3 and 4). This, in turn, offers valuable information to policymakers about the effects of service system organization and delays in accessing care [ 26 ]. Furthermore, this hypothesis highlights the importance of timely access to mental health services and the need for efficient triage systems and resource allocation to optimize treatment timing. Across all 4 RCTs (protocols 1–4), we plan to evaluate the predictive validity of the FTN in assessing individual needs and symptom profiles for therapeutic interventions. The data collected by FTN could potentially be used to predict treatment responses, facilitate personalized care and improve outcomes. This digital tool’s integration into routine practice could enhance the precision of treatment allocation in addition to streamlining the assessment process. Long-term follow-up data collection is planned to evaluate the cost-saving potential of the studied stepped care model compared to matched population controls. By considering direct and indirect healthcare, social care, employment, and societal costs the study aims to provide a comprehensive assessment of the economic viability of implementing stepped care models in public health systems. Finally, the study will assess whether patients with subclinical depressive or anxiety symptoms benefit from psychotherapeutic interventions in terms of reduced risk of developing clinical episodes, decreased somatic morbidity, and reduced total long-term societal costs. This aspect of the study underscores the importance of knowledge in planning early intervention and preventive measures in the management of mental health conditions. Strenghts and limitations The large scale and strong external validity rising from the pragmatic multi-site design covering a substantial proportion of Finland’s population and public primary care, are the strengths of this study. The sample size is based on careful power calculations that explicitly account for expected attrition, particularly relevant for iCBT interventions, where dropout rates as high as 55% have been reported [ 58 ]. Attrition will also be addressed statistically using intention to treat (ITT) analysis. Further, dropout from treatment does not imply non-adherence to the treatments, as the follow-up questionnaires are sent to all participants. In addition, participants who withdraw and individuals who decline participation at recruitment are asked to permit the use of data collected up to that point, minimizing information loss. Pragmatic design inevitably requires certain compromises compared to more controlled or explanatory RCT’s. As mentioned earlier, the duration of the studied interventions – and, for GSH and fCBT, the dose (number of sessions) – may vary, which may introduce diverse confounding factors, such as changes in life circumstances or concurrent treatments during the follow-up. To mitigate these risks, both survey and national registry e.g. on service use, medication, social benefits and sickness absence will be collected to adjust for relevant confounders. Prioritising ecological validity means we do not conduct formal fidelity auditing for GSH or fCBT, and both allow flexible dosing within recommended minimum and maximum session numbers. By contrast, GiCBT is a standardised therapy with a fixed number of sessions. We will record the number and timing of sessions in all arms. From an estimand perspective consistent with pragmatic, treatment-policy approach, variability of dose and absence of formal fidelity auditing reflect real-world implementation. However, in a non-inferiority framework, together with the lack of formal fidelity monitoring in the reference arm (fCBT)—and similarly in GSH—may attenuate the reference-arm effect and thereby increase the probability of concluding non-inferiority when comparing GSH and GiCBT to fCBT. This trade-off between internal control and ecological validity should be considered when interpreting the results. In conclusion, the FLT-Step trial (protocols 1–4) outlines a rigorous approach to evaluating the effectiveness and cost-benefit of integrating GSH, GiCBT, and fCBT into a stepped care model within the public healthcare system. The study’s findings are expected to provide valuable evidence that could inform the broader implementation of stepped care models, enhancing accessibility, cost-effectiveness, and long-term societal benefits. Electronic Supplementary Material Below is the link to the electronic supplementary material. 12888_2026_7962_MOESM1_ESM.pdf (186.4KB, pdf) Supplement Material 1: SPIRIT 2025 checklist 12888_2026_7962_MOESM2_ESM.pdf (136.4KB, pdf) Supplement Material 2: Intervention contents Supplement Material 3: TIDieR checklist (116.4KB, pdf) 12888_2026_7962_MOESM4_ESM.pdf (100.5KB, pdf) Supplement Material 4: CONSORT-NI item-by-item report Acknowledgements We would like to thank biostatistician Eliisa Löyttyniemi (EL) for her implementation of the randomization process for the study and her valuable advice on data collection. We also extend our gratitude to Bishwesvar Singh for his practical work on the construction of the data collection platform. Abbreviations CBT Cognitive-behavioral therapy GSH Guided self-help GiCBT Guided internet-delivered cognitive-behavioral therapy fCBT Face-to-face cognitive-behavioral therapy PHQ-9 Patient Health Questionnaire GAD-7 Generalized Anxiety Disorder 7-item scale FLT First-Line Therapies FTN Finnish Therapy Navigator FU Follow-up SPIN Social Phobia Inventory PDSS-SR Panic Disorder Severity Scale-Self Report AUDIT-C Alcohol Use Disorders Identification Test-Concise AUDIT Alcohol Use Disorders Identification Test DUDIT Drug Use Disorders Identification Test BBGS Brief Biosocial Gambling Screen OCI-R Obsessive-Compulsive Inventory –Revised BAT-12 Burnout Assessment Tool TSQ Trauma Screening Questionnaire EPO-1 Emotional and Psychological Outcome PSSS-R Perceived Social Support Scale-Revised WSAS Work and Social Adjustment Scale Author contributions SES is the principal investigator and, together with SIS, led the planning and development of the full study with support from all other authors. SES and EEH have coordinated the preparation of the study protocol with support from all other authors. EEH, SES and KM wrote the first draft of this paper. THR specifically revised the statistical analyses sections of this paper. SIS provided health economic expertise and wrote the related sections of the manuscript and is the program director for the Finnish First Line Therapies -Initiative. Funding acquisition was primarily carried out by SES, SIS and KM. The design of data collection platforms was mainly managed by MKM and EEH. MKM is responsible for organizing and monitoring practicalities in patient recruitment. MK is responsible for managing the participation of minors in this study and addressing specific issues related to this patient group, as well as overseeing the recruitment of all patients from the Western Finland welfare area. SES, SIS, KM, EEH, and JHS contributed to the development of GSH and fCBT interventions, while J-HS contributed to the development of GiCBT protocols. JE oversees the HUS Psychiatry Department and the implementation of this study under its auspices. All authors conceived the study, revised the manuscript for relevant scientific content, and approved the final version. Funding Open Access funding provided by University of Helsinki (including Helsinki University Central Hospital). The has received funding from the European Union – NextGenerationEU, Research Council of Finland (372876, 372982, 372878), Department of Psychiatry HUS Helsinki University Hospital (EEH), and The Medical Society of Finland/Finska Läkaresällskapet rf (SES). Roles and responsibilities—sponsor and funder. Funders or organizational sponsors had no role in the design of this study and will not have any role in the data management, analyses and interpretation of the data, nor in the decision to submit the reports for publication. The study is coordination is led by PI, supported by the research team. No additional formal steering committee is established as this is a pragmatic, researcher-initiated trial conducted within routine public health care services and poses minimal risk. Data management is overseen by the research team, with institutional support from the respective wellbeing services counties and HUS. Outcomes are collected using validated self-report instruments or registry data, minimizing the need for external adjudication. Data availability Public access to trial protocols is available through the trial registrations. Access to the study data is restricted to members of the research team only. According to Finnish legislations and institutional research permissions access to pseudonymised research data is granted only to members of the research team and do not allow sharing of data with third parties directly.All data will be pseudonymized prior to statistical analyses and reported at an aggregate level to ensure that individual patients cannot be identified. The statistical code will be reported alongside the publication of study results, in accordance with standard principles of good scientific writing. Members of research group will have access to the final trial dataset. No contractual agreements exist that would limit investigator access. Declarations Ethics approval and consent to participate The study has been approved by the Helsinki University Hospital Regional Committee on Medical Research Ethics (HUS/6234/2023). Written informed consent will be obtained from all participants, fully informing them of the aims and procedures of the study. Participants will also be asked for permission for the further use of their data, which will be anonymized to ensure confidentiality and privacy. Consent for publication Not applicable. Data monitoring A data monitoring committee (DMC) will not be established, as this is a low-risk, researcher-initiated trial involving non-invasive psychosocial interventions, with no planned interim analyses. Competing interests EEH, SES, KM, JHS, SIS, JE have participated in the development of Finnish First-Line Therapies model, guided self-help and fCBT training protocols but receive no financial gain from them. JHS has participated in the development of GiCBT but receives no financial gain from it. SIS, KM, JHS, JE and SES have participated in the development of Finnish Therapy Navigator but receive no financial gain from it. Other authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Eeva-Eerika Helminen and Suoma E. Saarni contributed equally to this work. References 1. 2019 Mental Disorders GBD Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry. 2022;9(2):137–50. [ DOI ] [ PMC free article ] [ PubMed ] 2. Smith ORF, Clark DM, Hensing G, Layard R, Knapstad M. Cost–benefit of IAPT Norway and effects on work-related outcomes and health care utilization: results from a randomized controlled trial using registry-based data. Psychol Med. 2025;55:e86. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Castelnuovo G, Pietrabissa G, Cattivelli R, Manzoni GM, Molinari E. Not Only Clinical Efficacy in Psychological Treatments: Clinical Psychology Must Promote Cost-Benefit, Cost-Effectiveness, and Cost-Utility Analysis. Front Psychol. 2016;7:563. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Layard R, Clark DM. Why More Psychological Therapy Would Cost Nothing. Front Psychol. 2015;6:1713. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Chisholm D, Sweeny K, Sheehan P, Rasmussen B, Smit F, Cuijpers P, et al. Scaling-up treatment of depression and anxiety: a global return on investment analysis. Lancet Psychiatry. 2016;3(5):415–24. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Cuijpers P, Noma H, Karyotaki E, Cipriani A, Furukawa TA. Effectiveness and acceptability of cognitive behavior therapy delivery formats in adults with depression: a network meta-analysis. JAMA Psychiatry. 2019 July 1;76(7):700. [ DOI ] [ PMC free article ] [ PubMed ] 7. Cuijpers P, Miguel C, Harrer M, Plessen CY, Ciharova M, Ebert D, et al. Cognitive behavior therapy vs. control conditions, other psychotherapies, pharmacotherapies and combined treatment for depression: a comprehensive meta-analysis including 409 trials with 52,702 patients. World Psychiatry. 2023;22(1):105–15. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Cuijpers P, Koole SL, van Dijke A, Roca M, Li J, Reynolds CF. Psychotherapy for subclinical depression: meta-analysis. Br J Psychiatry. 2014;205(4):268–74. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Szuhany KL, Simon NM. Anxiety Disorders: A Review. JAMA. 2022;328(24):2431. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Kasteenpohja T, Marttunen M, Aalto-Setälä T, Perälä J, Saarni SI, Suvisaari J. Treatment received and treatment adequacy of depressive disorders among young adults in Finland. BMC Psychiatry. 2015;15:47. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Santomauro DF, Vos T, Whiteford HA, Chisholm D, Saxena S, Ferrari AJ. Service coverage for major depressive disorder: estimated rates of minimally adequate treatment for 204 countries and territories in 2021. Lancet Psychiatry. 2024;11(12):1012–21. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Powell CLYM, Chiu CY, Sun X, So SH wai. A meta-analysis on the efficacy of low-intensity cognitive behavioural therapy for generalised anxiety disorder. BMC Psychiatry. 2024;24(1):10. [ DOI ] [ PMC free article ] [ PubMed ] 13. Cuijpers P, Donker T, van Straten A, Li J, Andersson G. Is guided self-help as effective as face-to-face psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of comparative outcome studies. Psychol Med. 2010;40(12):1943–57. [ DOI ] [ PubMed ] [ Google Scholar ] 14. Hedman-Lagerlöf E, Carlbring P, Svärdman F, Riper H, Cuijpers P, Andersson G. Therapist-supported Internet-based cognitive behaviour therapy yields similar effects as face-to-face therapy for psychiatric and somatic disorders: an updated systematic review and meta-analysis. World Psychiatry. 2023 June;22(2):305–14. [ DOI ] [ PMC free article ] [ PubMed ] 15. Rosenström TH, Saarni SE, Saarni SI, Tammilehto J, Stenberg JH. Efficacy and effectiveness of therapist-guided internet versus face-to-face cognitive behavioural therapy for depression via counterfactual inference using naturalistic registers and machine learning in Finland: a retrospective cohort study. Lancet Psychiatry. 2025;12(3):189–97. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Jeitani A, Fahey PP, Gascoigne M, Darnal A, Lim D. Effectiveness of stepped care for mental health disorders: An umbrella review of meta-analyses. Pers Med Psychiatry. 2024;47–48:100140. [ Google Scholar ] 17. Delgadillo J, Ali S, Fleck K, Agnew C, Southgate A, Parkhouse L, et al. Stratified Care vs Stepped Care for Depression: A Cluster Randomized Clinical Trial. JAMA Psychiatry. 2022;79(2):101–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Rivero-Santana A, Perestelo-Perez L, Alvarez-Perez Y, Ramos-Garcia V, Duarte-Diaz A, Linertova R, et al. Stepped care for the treatment of depression: a systematic review and meta-analysis. J Affect Disord. 2021;294:391–409. [ DOI ] [ PubMed ] [ Google Scholar ] 19. Salomonsson S, Santoft F, Lindsäter E, Ejeby K, Ljótsson B, Öst LG, et al. Stepped care in primary care - guided self-help and face-to-face cognitive behavioural therapy for common mental disorders: a randomized controlled trial. Psychol Med. 2018 July;48(10):1644–54. [ DOI ] [ PubMed ] 20. Clark DM. Implementing NICE guidelines for the psychological treatment of depression and anxiety disorders: The IAPT experience. Int Rev Psychiatry. 2011;23(4):318–27. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Knapstad M, Nordgreen T, Smith ORF. Prompt mental health care, the Norwegian version of IAPT: clinical outcomes and predictors of change in a multicenter cohort study. BMC Psychiatry. 2018;18(1):260. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. McHugh P, Martin N, Hennessy M, Collins P, Byrne M. An evaluation of Access to Psychological Services Ireland: year one outcomes. Ir J Psychol Med. 2016;33(4):225–33. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Gandré C, Rosenberg S, Coldefy M, Or Z. Experimenting locally with a stepped-care approach for the treatment of mild to moderate mental disorders in France: Challenges and opportunities. Health Policy. 2019;123(11):1021–7. [ DOI ] [ PubMed ] [ Google Scholar ] 24. Cano-Vindel A, Ruiz-Rodríguez P, Moriana J, Medrano L, González-Blanch C, Aguirre E, et al. Improving Access to Psychological Therapies in Spain: From IAPT to PsicAP. Psicothema. 2022;1(34):18–24. [ DOI ] [ PubMed ] [ Google Scholar ] 25. Australian Government Department of Health and Ageing. Primary Health Networks (PHN) primary mental health care guidance – stepped care [Internet]. Australian Government Department of Health and Aged Care; 2021 [cited 2025 June 11]. Available from: https://www.health.gov.au/resources/publications/primary-health-networks-phn-primary-mental-health-care-guidance-stepped-care?language=en 26. Clark DM, Canvin L, Green J, Layard R, Pilling S, Janecka M. Transparency about the outcomes of mental health services (IAPT approach): an analysis of public data. Lancet. 2018;391(10121):679–86. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Helminen EE, Mikkonen K, Saarni SI, Mattila K, Rosenström T, Isometsä E, et al. Effectiveness of Guided Self-Help Versus Internet-Delivered or Face-to-Face Cognitive Behavioral Therapy for Depression and Anxiety: a Randomized Controlled Non-Inferiority Trial of the Finnish First-Line Therapies –Initiative (FLT-Step). Eur Psychiatry. 2025;68(S1):S320–320. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Chan AW, Boutron I, Hopewell S, Moher D, Schulz KF, Collins GS, et al. SPIRIT 2025 statement: updated guideline for protocols of randomised trials. Lancet. 2025;405(10491):e19–27. [ Google Scholar ] 29. Cruchinho P, López-Franco MD, Capelas ML, Almeida S, Bennett PM, Miranda da Silva M, et al. Translation, Cross-Cultural Adaptation, and Validation of Measurement Instruments: A Practical Guideline for Novice Researchers. J Multidiscip Healthc. 2024;17:2701–28. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Loudon K, Treweek S, Sullivan F, Donnan P, Thorpe KE, Zwarenstein M. The PRECIS-2 tool: designing trials that are fit for purpose. BMJ. 2015;350:h2147. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Kroenke K. Enhancing the clinical utility of depression screening. Can Med Assoc J. 2012;184(3):281–2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001 Sept;16(9):606–13. [ DOI ] [ PMC free article ] [ PubMed ] 33. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092–7. [ DOI ] [ PubMed ] [ Google Scholar ] 34. Saarni SI, Nurminen S, Mikkonen K, Service H, Karolaakso T, Stenberg JH, et al. The Finnish therapy navigator – digital support system for introducing stepped care in Finland. Psychiatr Fenn. 2022;2022(53):120–37. [ Google Scholar ] 35. Luedtke A, Sadikova E, Kessler RC. Sample Size Requirements for Multivariate Models to Predict Between-Patient Differences in Best Treatments of Major Depressive Disorder. Clin Psychol Sci. 2019;7(3):445–61. [ Google Scholar ] 36. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092–7. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Kounali D, Button KS, Lewis G, Gilbody S, Kessler D, Araya R, et al. How much change is enough? Evidence from a longitudinal study on depression in UK primary care. Psychol Med. 2022 July;52(10):1875–82. [ DOI ] [ PMC free article ] [ PubMed ] 38. Richards DA, Ekers D, McMillan D, Taylor RS, Byford S, Warren FC, et al. Cost and Outcome of Behavioural Activation versus Cognitive Behavioural Therapy for Depression (COBRA): a randomised, controlled, non-inferiority trial. Lancet Lond Engl. 2016;388(10047):871–80. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Julious SA. Sample sizes for clinical trials with Normal data. Stat Med. 2004 June;30(12):1921–86. [ DOI ] [ PubMed ] 40. Ritola V, Lipsanen JO, Pihlaja S, Gummerus EM, Stenberg JH, Saarni S, et al. Internet-Delivered Cognitive Behavioral Therapy for Generalized Anxiety Disorder in Nationwide Routine Care: Effectiveness Study. J Med Internet Res. 2022;24(3):e29384. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Saarni SE, Rosenström T, Stenberg JH, Plattonen A, Holi M, Ekelund J, et al. Finnish psychotherapy quality register: rationale, development, and baseline results. Nord J Psychiatry. 2023 July 3;77(5):455–66. [ DOI ] [ PubMed ] 42. van Buuren S. Flexible Imputation of Missing Data [Internet]. 2nd edition. Vol. 2018. 2018 [cited 2025 June 10]. Available from: https://stefvanbuuren.name/fimd/ 43. Campbell MK, Piaggio G, Elbourne DR, Altman DG. Consort 2010 statement: extension to cluster randomised trials. BMJ 2012 Sept 4;345:e5661. [ DOI ] [ PubMed ] 44. Poisot T. The digitize package: extracting numerical data from scatterplots. R J. 2011;3(1):25–6. [ Google Scholar ] 45. van der Laan MJ, Rose S. Targeted learning: causal inference for observational and experimental data. New York: Springer; 2011. (Springer series in statistics). [ Google Scholar ] 46. Connor KM, Davidson JR, Churchill LE, Sherwood A, Foa E, Weisler RH. Psychometric properties of the Social Phobia Inventory (SPIN). New self-rating scale. Br J Psychiatry. 2000;176:379–86. [ DOI ] [ PubMed ] [ Google Scholar ] 47. Furukawa TA, Katherine Shear M, Barlow DH, Gorman JM, Woods SW, Money R, et al. Evidence-based guidelines for interpretation of the Panic Disorder Severity Scale. Depress Anxiety. 2009;26(10):922–9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Kriston L, Hölzel L, Weiser AK, Berner MM, Härter M. Meta-analysis: are 3 questions enough to detect unhealthy alcohol use? Ann Intern Med. 2008;149(12):879–88. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Allen JP, Litten RZ, Fertig JB, Babor T. A review of research on the Alcohol Use Disorders Identification Test (AUDIT). Alcohol Clin Exp Res. 1997 June;21(4):613–9. [ PubMed ] 50. Berman AH, Berman H, Palmstierna T, Schlyter F. [cited 2024 Nov 11]. Drug Use Disorders Identification Test (DUDIT). 2024. Available from: https://euda.europa.eu/drugs-library/drug-use-disorders-identification-test-dudit_en https://www.file://C:/Users/HUS672~1/AppData/Local/Temp/MicrosoftEdgeDownloads/37be8d7a-a787-4455-b2f8-e3232829d4ed/FI_DUDIT%25;20suomalainen%2520versio.pdf [ DOI ] [ PubMed ] 51. Foa EB, Huppert JD, Leiberg S, Langner R, Kichic R, Hajcak G, et al. The Obsessive-Compulsive Inventory: development and validation of a short version. Psychol Assess. 2002;14(4):485–96. [ PubMed ] [ Google Scholar ] 52. Schaufeli W, De Witte H, Desart S. Burnout Assessment Tool (BAT) – Test Manual [Internet]. 2020. Available from: https://burnoutassessmenttool.be/wp-content/uploads/2020/08/Test-Manual-BAT-English-version-2.0-1.pdf [ DOI ] [ PMC free article ] [ PubMed ] 53. Brewin CR, Rose S, Andrews B, Green J, Tata P, McEvedy C, et al. Brief screening instrument for post-traumatic stress disorder. Br J Psychiatry J Ment Sci. 2002;181:158–62. [ DOI ] [ PubMed ] [ Google Scholar ] 54. Gonçalves MM, Lutz W, Schwartz B, Oliveira JT, Saarni SE, Tishby O, et al. Developing a European Psychotherapy Consortium (EPoC): towards adopting a single-item self-report outcome measure across European countries. Clin Psychol Eur. 2024 Sept 30;6(3):e13827. [ DOI ] [ PMC free article ] [ PubMed ] 55. Mikkonen K, Bombino A, Villa A, Nurminen S, Roiha RM, Roslund P, et al. Guided self-help in the treatment of common mental health disorders - the development of the finnish guided self-help (F-GSH) model. Foundation for Psychiatric Research in Finland; 2024. 56. Mikkonen K, Saarni SE, Nurminen S, Stenberg JH, Ekelund J, Villa A, et al. Kognitiivinen lyhytterapia tuo apua mielenterveyskriisiin. Suom Lääkäril. 2024;79:e39914. Available from: https://www.laakarilehti.fi/e39914 57. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inf. 2009;42(2):377–81. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Measurement Invariant but Non-Normal Treatment Responses in Guided Internet Psychotherapies for Depressive and Generalized Anxiety Disorders. - Tom H. Rosenström, Ville Ritola, Suoma Saarni, Grigori Joffe, Jan-Henry Stenberg, 2023 [Internet]. [cited 2025 Nov 4]. Available from: https://journals.sagepub.com/doi/10.1177/10731911211062500 . [ DOI ] [ PMC free article ] [ PubMed ] 59. Hlynsson JI, Skúlason S, Andersson G, Carlbring P. Why are we still using the PHQ-9? A historical review and psychometric evaluation of measurement invariance. Psychiatr Q. 2025 Sept 3. [ DOI ] [ PubMed ] 60. McMillan D, Gilbody S, Richards D. Defining successful treatment outcome in depression using the PHQ-9: A comparison of methods. J Affect Disord. 2010;127(1–3):122–9. [ DOI ] [ PubMed ] [ Google Scholar ] 61. Toussaint A, Hüsing P, Gumz A, Wingenfeld K, Härter M, Schramm E, et al. Sensitivity to change and minimal clinically important difference of the 7-item Generalized Anxiety Disorder Questionnaire (GAD-7). J Affect Disord. 2020;265:395–401. [ DOI ] [ PubMed ] [ Google Scholar ] 62. Finnish Institute for Health and Welfare (THL). Finland [Internet]. 2025 [cited 2025 June 5]. Healthy Finland Survey - THL. Available from: https://thl.fi/en/research-and-development/research-and-projects/healthy-finland-survey 63. Mäklin S, Kokko P. The unit costs of health and social care in Finland in 2017 [Internet]. THL; 2021 [cited 2025 June 5]. Available from: https://www.julkari.fi/handle/10024/142882 64. Szuhany KL, Simon NM. Anxiety Disorders: A Review. JAMA. 2022;328(24):2431. [ DOI ] [ PubMed ] [ Google Scholar ] 65. Powell CLYM, Chiu CY, Sun X, So SH wai. A meta-analysis on the efficacy of low-intensity cognitive behavioural therapy for generalised anxiety disorder. BMC Psychiatry. 2024;24(1):10. [ DOI ] [ PMC free article ] [ PubMed ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials 12888_2026_7962_MOESM1_ESM.pdf (186.4KB, pdf) Supplement Material 1: SPIRIT 2025 checklist 12888_2026_7962_MOESM2_ESM.pdf (136.4KB, pdf) Supplement Material 2: Intervention contents Supplement Material 3: TIDieR checklist (116.4KB, pdf) 12888_2026_7962_MOESM4_ESM.pdf (100.5KB, pdf) Supplement Material 4: CONSORT-NI item-by-item report Data Availability Statement Public access to trial protocols is available through the trial registrations. Access to the study data is restricted to members of the research team only. According to Finnish legislations and institutional research permissions access to pseudonymised research data is granted only to members of the research team and do not allow sharing of data with third parties directly.All data will be pseudonymized prior to statistical analyses and reported at an aggregate level to ensure that individual patients cannot be identified. The statistical code will be reported alongside the publication of study results, in accordance with standard principles of good scientific writing. Members of research group will have access to the final trial dataset. No contractual agreements exist that would limit investigator access. Articles from BMC Psychiatry are provided here courtesy of BMC ACTIONS View on publisher site PDF (2.5 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top