Generating practice-based evidence for common mental health problems using routine outcome monitoring: a protocol for a longitudinal naturalistic study in Ecuadorian mental health services Authors: Clara Paz, Itaty Benalcázar-Betancourt, Jorge Valdiviezo-Oña, Chris Evans Published: 2026 ABSTRACT AbstractIntroductionPractice-oriented research uses scientific methods to generate practice-based evidence on psychotherapy processes and outcomes. The Consorcio Latinoamericano de Investigación en Psicoterapia (CLIP) unifies data collection across multiple sites to enhance the value of routine outcome monitoring. This project aims to generate practice-based evidence on the implementation, progress and predictors of change in psychological interventions within Ecuador’s routine mental health settings, using CLIP’s unified data forms.Methods and analysesThis is a protocol for an observational, naturalistic, longitudinal study. From February 2025 to February 2028, we will describe implementation processes and examine intervention progress and outcomes across 11 Ecuadorian mental health services. Participants will include adolescents (11–17 years) and adults (≥18 years) seeking psychological intervention along with their therapists and trainees. Clients’ change will be monitored through psychological distress measures. Socio-demographic and intervention satisfaction data will be collected alongside information on therapists and services. Analyses will include initial contact characterisation, reliable change, outcome predictors and the examination of how time, therapist and service characteristics each contribute to clients’ change.Ethics and disseminationThe protocol for this study was approved by the Human Research Ethics Committee of the Universidad de Las Américas, Ecuador (ID: 2024-OBS-033). The results will be disseminated within the services for service improvement, in peer-reviewed scientific articles and at conferences and workshops.Trial registration numberNCT07257848. STRENGTHS AND LIMITATIONS OF THIS STUDY This project will enrol adults and adolescents from 11 mental health services in Ecuador, enabling both intragroup and intergroup comparisons of intervention trajectories and outcomes. This project will enrol adults and adolescents from 11 mental health services in Ecuador, enabling both intragroup and intergroup comparisons of intervention trajectories and outcomes. The assessment procedures are based on the core requirements of the Consorcio Latinoamericano de Investigación en Psicoterapia (CLIP). The assessment procedures are based on the core requirements of the Consorcio Latinoamericano de Investigación en Psicoterapia (CLIP). Data loss will be minimised through consistent supervision of the data collection process. Data loss will be minimised through consistent supervision of the data collection process. While the design does not support causal inferences, the findings can inform decision-making for service improvement. While the design does not support causal inferences, the findings can inform decision-making for service improvement. This study protocol adheres to practice-oriented research standards to ensure transparency and to offer a model for other services interested in this approach. This study protocol adheres to practice-oriented research standards to ensure transparency and to offer a model for other services interested in this approach. Introduction Psychological interventions with established efficacy in controlled trials often show variable effectiveness when implemented in routine mental health settings1 2 or their findings are simply not inspected in routine practice. This efficacy–effectiveness gap is particularly pronounced in low- and middle-income countries (LMICs), where resource constraints, diverse client populations and contextual factors can significantly influence intervention delivery and outcomes.3 Generating evidence that is both scientifically rigorous and directly applicable to real-world contexts is therefore a priority for global mental health. The practice-based evidence (PBE) paradigm addresses this gap by prioritising the systematic study of routine clinical practice.46 Unlike traditional efficacy trials, which control variables tightly to maximise internal validity, PBE studies embrace clinical complexity to enhance ecological validity. Key characteristics of PBE include: (1) data collection integrated into routine clinical workflows, (2) assessment processes that reflect everyday practice and (3) investigation of interventions and client pathways of change without research-imposed constraints, such as strict protocol adherence or randomisation.7 This approach supports a collaborative tripartite model involving clinicians, researchers and administrators to answer contextually relevant questions about what works, for whom and under what conditions.7 8 The generation of PBE represents a powerful paradigm shift, moving beyond the primary goal of knowledge generation toward the direct and continuous improvement of mental health services.7 PBE relies on systematic and longitudinal data collection, which enables the critical evaluation of clinical and administrative decisions. This contributes to the creation of a self-correcting, continuous feedback loop: evidence informs practice and practice, in turn, generates new evidence for further refinement. Routine outcome monitoring (ROM) provides the methodological infrastructure for generating PBE. It involves the repeated, unified measurement of client change and other key variables throughout intervention.9 10 The nested structure of ROM data enables analysis at multiple levels, generating distinct benefits for different stakeholders.7 At the client level, repeated assessments of psychological distress allow for the creation of individualised intervention trajectories, facilitating timely clinical adaptations.1113 At the therapist level, the nesting clients within clinicians makes it possible to explore ‘therapist effects’,14 15 supporting professional development and continuous quality improvement. At the service level ROM enables analyses of organisational and systemic factors associated with opting out of intervention and changes in self-report data across interventions.16 The PBE generated by ROM can be also used within formal, protocol driven research: termed practice-oriented research (POR),17 18 which seeks to bridge the gap between researchers and clinicians. POR benefits from a collaborative research ethos characterised by transparency, such as the open publication of study protocols, and a methodological approach that balances a priori analytical planning with the flexibility to address emergent, practice-informed questions. This ensures the research process remains responsive to the clinical context from which the evidence originates. While such studies are not primarily focused on hypothesis-testing, we specify a set of analyses to help avoid accusations of post hoc ‘cherry picking’ of findings. As well as these analyses any POR also remains open to emergent questions arising from ongoing dialogues with therapists and service administrators, all directed by the ultimate goal of understanding data to improve mental health services but will always distinguish the prespecified from the emergent analyses. The scientific value of ROM can increase significantly when data collection is unified across multiple sites. The Consorcio Latinoamericano de Investigación en Psicoterapia (CLIP; https://www.psychotherapyresearch.org/page/CLIP-About_engl) was established to build such a collaborative network across Latin America.19 The CLIP’s primary aim is to unify the collection of clinical data to enhance the quality and effectiveness of mental health services in the region through three core functions: (1) developing a shared, practice-based database for joint analysis; (2) fostering collaboration in implementation and training; and (3) establishing a common framework for data collection on therapies, clients, therapists and outcomes. While ROM implementation has advanced in high-income countries,9 a significant evidence gap remains in LMICs20 such as Ecuador. Building on a prior study conducted in a single Ecuadorian service,21 this protocol describes a multicentre initiative to implement a ROM system across a diverse network of mental health services in Ecuador, in alignment with the CLIP framework. This study aims to generate strong PBE to bridge the local efficacy–effectiveness gap and contribute a contextualised perspective on ROM implementation and its outcomes from the Andean region to the broader Latin American context. This initiative is grounded in lessons learnt from our previous project,21 particularly the need to minimise client and therapist burden while maintaining scientific rigour. The current protocol strategically retains common metrics from the Clinical Outcomes in Routine Evaluation (CORE) system22 to ensure data comparability with prior and future projects. To maintain collaborative alignment with the European Psychotherapy Consortium (EPoC),19 23 a European based consortium partner to the CLIP, the Emotional and Psychological Outcome (single-item measure) (EPO-1) measure was adopted to match their assessment schedule. The transparent presentation of this protocol is, therefore, a direct response to a recognised need for greater openness in PBE projects. It serves as a detailed roadmap for colleagues and services interested in undertaking similar POR, demystifying the procedures and operational challenges.2427 Study objectives General objective To generate PBE on the implementation, progress and predictors of change for psychological interventions in routine settings in Ecuador, with implications for similar contexts in Latin America. Specific objectives Implementation process Objective 1: To describe the implementation of a routine outcome monitoring system across participating psychological services in Ecuador, including the main challenges and opportunities of its use in psychotherapeutic practice within a Latin American context Client progress and intervention characteristics Objective 2: To identify the characteristics of clients attending psychological therapy, as well as the characteristics of the services and therapists providing it. Objective 3: To explore the change trajectories and predictors of change for clients attending the services. Objective 4: To establish one national, Ecuadorian dataset and findings to be used to support comparisons between the services, the therapies they offer, the clients and the clients’ score change trajectories. These will support future comparison with CLIP-affiliated services in other countries in Latin America. Methods and analysis Study design This study adopts a non-experimental, longitudinal and primarily quantitative design with an exploratory and descriptive purpose, conducted in naturalistic clinical settings in Ecuador. The mental health services included are Centro de Psicología Aplicada of the Universidad de Las Américas (UDLA), Equimente, Historie, Jorge López Consultorios, Kumpana, Mentis Psicología, Psico Estratégico, Psico Neuro Manta, Renova, Sentir Psicología Integral and Solo Psique. Any other interested mental health services which may ask to join the work within the country will be included following the same procedures, with their inclusion date recorded and reported in dissemination materials. Participants Clients All adults (18 years and older) and adolescents (11–17 years old) clients who are admitted to psychotherapy at a participating service centre will be invited to participate. Recruitment began on 21 February 2025 and will continue until 29 February 2028. Data collection for each enrolled client will continue until the conclusion of their intervention. Elegibility criteria are: seeking psychotherapy, having sufficient ability to communicate in Spanish and being 11 years of age or older. These broad criteria reflect the routine practice setting and the demographics of the country, where Spanish is spoken by approximately 93% of the population28 and literacy rates are high (94%).29 As a practice-oriented study, this protocol adopts the inclusion and exclusion criteria for therapies used in each participating centre. The study expects to gather data from at least 300 adults and 40 adolescents. Psychotherapists, trainees and supervisors Clinicians, including licensed professionals and psychology trainees providing psychotherapy under supervision from different therapeutic orientations, will be invited to participate, with at least 25 expected across all sites. In addition, clinical supervisors and service directors will be included to provide organisational and contextual data. At least one supervisor and one administrative authority per participating centre will complete the relevant service information forms and participate in follow-up procedures. Instruments Implementation process Objective 1: to describe the implementation of a routine outcome monitoring system across participating psychological services in Ecuador, including the main challenges and opportunities of its use in psychotherapeutic practice within a Latin American context To address this objective qualitative data will be documented throughout the project using two tools: (1) structured field notes recorded by the research team, and (2) minutes from annual coordination meetings held with therapists, supervisors and service managers. These tools will serve to identify, document and summarise the principal barriers and facilitators to ROM implementation across participating services. Client progress and intervention characteristics The following measures are used to collect information and track client progress over the course of the intervention, informing objectives 2, 3 and 4. The measures are presented in order of their relevance to each objective. Specifically, they address client, therapist and service characteristics (objective 2); change trajectories and predictors of change (objective 3); and the creation of a national dataset for future comparisons with CLIP-affiliated services (objective 4). Some measures contribute to more than one objective. All instruments are administered in Spanish. Objective 2: to identify the characteristics of clients attending psychological therapy, as well as the characteristics of the services and therapists providing it The following instruments will be used to describe client, therapist and service characteristics. Additionally, the CORE measures and the EPO-1 baseline measures (described in objective 3) will also serve to describe client psychological distress at intake. Client Information Form This form is completed by clients when they begin intervention. It gathers basic socio-demographic information relevant to their clinical care. All items included were previously agreed on by the CLIP consortium for use within the participating services. See online supplemental material for the form. CORE Therapy Assessment Form The CORE Therapy Assessment Form (CORE-A-TAF) is a pragmatic form completed by therapists at the beginning of the intervention. It collects referral information, socio-demographic characteristics of the client and data regarding the nature, severity and duration of the presenting problems.5 30 In this study, the Spanish version adapted for use within the CLIP will be administered. See online supplemental material for the form. CORE End of Therapy The CORE End of Therapy (CORE-A-EoT) is a pragmatic form completed by therapists immediately after the end of intervention. It collects information about the completed therapeutic process, including the number of sessions, type of therapy, session duration and frequency, whether the ending was planned or unplanned and the perceived benefits of therapy.5 30 The Spanish version adapted for the CLIP will be used in the present study. See online supplemental material for the form. Personal Style of the Therapist Questionnaire The Personal Style of the Therapist Questionnaire (PST-Q) is a self-report instrument designed to assess characteristic ways therapists engage in their professional practice across five stylistic dimensions: instructional, expressive, involvement, attentional and operative.31 It has 20 items rated on a 7-point Likert scale from 1 ‘Completely disagree’ to 7 ‘Completely agree’. The Spanish psychometric exploration32 showed adequate psychometric properties and confirmed the multidimensional structure of the measure, supporting its usefulness for characterising therapists’ interpersonal and technical preferences in clinical settings. The Spanish version of the PST-Q can be accessed in Prado-Abril et al.32 Therapist Information Form This form is completed by each therapist when joining the study and then annually thereafter. It gathers basic demographic information along with details about their professional background and clinical orientation. The Spanish version used in this study has already been reviewed and approved for implementation within the CLIP context. See online supplemental material for the form. Service Information Form The Service Information Form is completed by supervisors or administrative staff from each participating centre. It gathers key organisational details, including the type of population served, the approximate number of users attended annually and other relevant characteristics of the service. It is administered at the beginning of participation in the study and subsequently once per year. See online supplemental material for the form. Objective 3: to explore the change trajectories and predictors of change for clients attending the services The following instruments will be used to track change throughout the intervention. However, the CORE-A-TAF, CORE-A-EoT and Client Information Form (described above) will provide the predictor variables used to understand change. Clinical Outcomes in Routine Evaluation-Outcome Measure The CORE-Outcome Measure (CORE-OM)22 is a 34-item self-report instrument intended for use in varied clinical contexts, built on a pan-theoretical framework for assessing psychological distress in adults; it has four domains: well-being (4 items), risk (6 items), problems (12 items) and functioning (12 items). Items are scored on a Likert scale from 0 (never) to 4 (always or almost always); higher scores indicate greater psychological distress. The Spanish version of this measure was translated by Feixas et al33 and has demonstrated good psychometric properties in both Spain34 and Ecuador.35 In the Ecuadorian context, internal consistency was excellent, with Cronbach’s alpha for the total score reported at 0.94 (0.93, 0.95).36 The Spanish version used in this study is available at: https://www.coresystemtrust.org.uk/wp-content/uploads/2024/03/CORE-OM_Spanish_FM.pdf. Clinical Outcomes in Routine Evaluation-10 The CORE-1037, a 10-item version, is a brief, easy-to-administer and generic psychological distress assessment measure for adults, whose items were derived from the CORE-OM. It is considered an acceptable and feasible instrument that has shown solid psychometric properties in its original version in the UK.37 Its psychometric performance has been evaluated in Ecuador indicating that it is a suitable measure for use in this context, showing Cronbach’s alpha of 0.81 (0.78, 0.84) as well as convergent and discriminant validity.36 The Spanish version of the scale can be accessed at: https://www.coresystemtrust.org.uk/wp-content/uploads/2024/03/CORE-10_Spanish_FM.pdf. Young Person’s-Clinical Outcomes in Routine Evaluation The Young Person’s-CORE (YP-CORE) is a brief self-report instrument designed to assess psychological distress in adolescents aged 11–16 years, capturing both overall functioning and distress across a broad range of difficulties.38 It has 10 items, 1 related to well-being, 1 to risk, 4 to problems and 4 to general functioning; its items are answered on a 5-point scale (0–4). The Spanish translation by Feixas et al39 demonstrated adequate psychometric properties. In Ecuador, the instrument has shown adequate psychometric properties, with Cronbach’s alpha reported at 0.85 (0.82, 0.87),40 supporting its suitability for use in this context. The version applied in the present study corresponds to the adaptation approved for Latin America, available at: https://www.coresystemtrust.org.uk/wp-content/uploads/2025/03/YP-CORE_SpanishLA_FM.pdf. Emotional and Psychological Outcome The Emotional and Psychological Outcome-1 (EPO-1) is a single-item outcome measure in which clients are asked to rate their current emotional and psychological state by responding to the question: ‘At this moment, how are you feeling emotionally and psychologically?’.23 Responses are scored on a 5-point scale ranging from 0 (‘Very bad; I can barely cope’) to 4 (‘Very good; I have no major complaints’). This item was adopted by the EPoC as part of its effort to unify outcome monitoring across participating clinics in Europe.23 It has been translated into multiple languages and used in the context of routine mental health services, offering a pragmatic solution to the challenges of implementing multi-item instruments in busy practice-based settings. The EPO-1 therefore enables therapists and researchers to obtain a rapid, session-by-session indicator of change in emotional and psychological functioning, facilitating the integration of outcome monitoring into routine psychotherapy. The Spanish version used in this study is available at: https://www.psychotherapyresearch.org/page/epo1. Consumer Reports Effectiveness Scale The Consumer Reports Effectiveness Scale (4-item version) (CRES-4) is a self-report brief four-item measure assessing satisfaction with psychological interventions, including overall satisfaction (one item), perceived problem resolution (one item) and emotional change throughout therapy (one item on the emotional state looking back before attending the service and one question on the emotional state when completing the questionnaire at the end of the intervention).41 Items are rated on a Likert scale and transformed into scores ranging from 0 to 100 for each component, with a total score between 0 and 300. Higher values indicate greater satisfaction and perceived improvement. The Spanish version of the scale, translated by Feixas et al41 was used in this study, and the instrument can be accessed in that publication. Objective 4: to establish one national, Ecuadorian dataset and findings to be used to support comparisons between the services, the therapies they offer, the clients and the clients’ score change trajectories. These will support future comparison with CLIP-affiliated services in other countries in Latin America To address this objective, data from instruments that are part of the basic assessment agreed on by the CLIP will be aggregated into a single national dataset. These include the Client Information Form, CORE-A-TAF, CORE-A-EoT, Therapist Information Form, Service Information Form, PST-Q, CORE-OM, CORE-10, YP-CORE and EPO-1. The dataset will be composed of pooled data from all participating Ecuadorian services. This national dataset will then be available for future comparison with data from other CLIP-affiliated services across Latin America. Procedure Implementation process All participating centres have already granted approval to implement the project. Annual coordination meetings with service authorities and treating clinicians will be held to collaboratively identify clinical-research questions of interest to each centre, which will later be explored through quantitative analyses. During these meetings, we will also actively engage stakeholders to describe their experiences, perceived barriers and identified facilitators regarding the implementation of routine outcome monitoring. Throughout the implementation period, the research team will document structured field notes and meeting minutes to capture ongoing observations and discussions. These qualitative insights will be systematically documented and will directly inform the comprehensive qualitative synthesis planned on project completion. Client progress and intervention characteristics Recruitment began on 21 February 2025 and it will conclude on 29 February 2028. No experimental manipulation of independent variables, random assignment of participants to specific interventions or standardisation of therapeutic interventions will occur, as the research is practice-based and naturalistic in design. Data are being collected through MarBar, a free online software available at https://www.marbarsystem.com/ for Spanish-speaking therapists and centres. This system allows clinicians to view individualised results for each session, along with graphical depictions of change over time. Therapists can also access aggregated data showing the evolution of all their clients. Additionally, services can use the system to review aggregated data describing the main outcomes and trajectories of change for all clients, both overall and disaggregated by therapist. Clients complete their information and brief measures using a secure online link generated by clinicians through the MarBar System, which can be accessed from their personal devices. Figure 1 provides an example of the sequence and type of questionnaires administered at different points throughout the intervention for both adult and adolescent clients. Assessment schedule for adult and adolescent clients throughout the intervention. CORE, Clinical Outcomes in Routine Evaluation; CORE-A-EoT, CORE End of Therapy form; CORE-A-TAF, CORE Therapy Assessment Form; CORE-OM, CORE-Outcome Measure; CORE-10, CORE (10-item version); CRES-4, Consumer Reports Effectiveness Scale (4-item version); EPO-1, Emotional and Psychological Outcome (single-item measure); YP-CORE, Young Person’s-CORE. At intake, adult participants receive information about the study, and they are asked whether they wish to participate. For adolescent participants, guardians and adolescents are informed about the study. If the guardian consents to the adolescent’s participation, the adolescent is asked to provide assent. Before the first session, all participants are asked to complete the Client Information Form. Following this, participants complete the designated CORE measure according to their age: adults complete the CORE-OM, and adolescents complete the YP-CORE. After completing the CORE measure, both adolescents and adults are asked to fill out the EPO-1. At the same time, therapists complete the CORE-A-TAF based on the clinical information available for each participant. From session two onward to reduce burden, adults are asked to complete a briefer version of the CORE-OM—the CORE-10—while adolescents continue to complete the YP-CORE. All participants also complete the EPO-1 at each session. During the final session, adults complete the CORE-OM, adolescents complete the YP-CORE and all participants complete the EPO-1 together with the CRES-4. At this point, therapists complete the CORE-A-EoT using the client’s clinical information. At enrolment, the person in charge of each service completes the service approval consent. Supervisors then complete the Service Information Form, which is updated annually. Therapists are provided with information about the study and asked to provide their consent to participate. Once enrolled, they complete the Therapist Information Form and the PST-Q. These are also completed annually. Figure 2 provides the sequence of the therapists’ and services’ assessment schedule. Assessment schedule for therapists and services throughout their participation in the study. PST-Q, Personal Style of the Therapist Questionnaire. Data analysis Implementation process Specific objective 1: to describe the implementation of a routine outcome monitoring system across participating psychological services in Ecuador, including the main challenges and opportunities of its use in psychotherapeutic practice within a Latin American context To address this objective, we will conduct a qualitative synthesis and comparative analysis of data collected on project completion. The information gathered through structured field notes and meeting minutes will be summarised to identify the principal barriers and facilitators to ROM implementation. This synthesis will then form the basis of an analysis comparing the findings with documented experiences from prior ROM implementations in three contexts: (1) earlier within Ecuador, (2) elsewhere in Latin America and (3) in other global regions. This comparison will aim to identify key points of convergence and divergence, highlighting common themes and unique contextual factors that characterise the implementation of ROM in Latin American practice settings. Client progress and intervention characteristics The analyses for objectives 2, 3 and 4 are tailored to the descriptive and exploratory nature of the quantitative data, with the goal of producing PBE that can inform service improvement and guide future investigations. For each objective, we summarise the use of findings for both practice and research. There are four nested levels of analysis: time, clients, therapists and services. The data at each level are: (1) date and session count with scores on change measures; (2) socio-demographic and psychological data from the Client Information Form and CORE-A-TAF and CRES-4 scores; (3) psychotherapist data from the PST-Q and Therapist Information Form; and (4) service description data from the Service Information Form. The core analytical plan is outlined below by objective, with further details for objectives 2 and 3 in table 1. Pre-planned analyses are specified here. Additional exploratory analyses may be conducted, and any post hoc findings will be clearly identified as such in all future reports. All quantitative analyses will be carried out in R.42 Table 1Detailed analysis plan for objectives 2 and 3: specific analyses, data sources and descriptors/estimatorsSpecific objectivesType of analysisData sourceDescriptor/estimator2. To identify the characteristics of clients attending psychological therapy, as well as the characteristics of the services and therapists providing it.DescriptiveScores from all quantitative measures (CORE-OM, YP-CORE, EPO-1); CORE-A-TAF and CORE-A-EoT socio-demographic and clinical variables, therapists’ and services’ characteristicsFrequencies, median, mean, bootstrapped CIs, range, SDAcceptabilityAll quantitative measuresRate of completed items per measureInternal consistencyAll quantitative measuresCronbach’s alphaMcDonald’s omegaCorrelations and associationsAll quantitative measures, gender, agePearson’s rSpearman rhoMean differencesHedges’s gDescriptives and predictors of satisfaction with interventionCRES-4, data from the client information form and the CORE-A-TAF and CORE-A-EoT; therapists; servicesFrequencies, median, means with bootstrapped CIs, range, SD, differences, multiple linear regressions, mean group differencesDescriptivePST-Q; therapists; servicesFrequencies, median, means with bootstrapped CIs, range, SD, differences, mean group differences3. To explore the change trajectories and predictors of change for clients attending the services.Individual improvement and deteriorationCORE-OM, CORE-10, YP-CORE and EPO-1 scoresReliable change indexTrajectories of changeScores from measures used at least three times (CORE-10, YP-CORE, EPO-1)Cat’s cradle plots and raincloud plotsMultilevel analyses (contingent on the adequacy of the study data)CORE-10, YP-CORE and EPO-1 scores; therapists and servicesIntrapersonal, intragroup and intergroup differencesRandom and fixed effects modelsIntraclass correlation coefficientOutcome predictorsInitial distressData from the client information form and the CORE-A-TAFFixed effects MLM (see above)CORE-10, Clinical Outcomes in Routine Evaluation (10-item version); CORE-A-EoT, CORE End of Therapy Form; CORE-A-TAF, CORE Therapy Assessment Form; CORE-OM, CORE-Outcome Measure; CRES-4, Consumer Reports Effectiveness Scale (4-item version); EPO-1, Emotional and Psychological Outcome (single-item measure); MLM, multilevel modelling; PST-Q, Personal Style of the Therapist Questionnaire; YP-CORE, Young Person's-Clinical Outcomes in Routine Evaluation. Detailed analysis plan for objectives 2 and 3: specific analyses, data sources and descriptors/estimators CORE-10, Clinical Outcomes in Routine Evaluation (10-item version); CORE-A-EoT, CORE End of Therapy Form; CORE-A-TAF, CORE Therapy Assessment Form; CORE-OM, CORE-Outcome Measure; CRES-4, Consumer Reports Effectiveness Scale (4-item version); EPO-1, Emotional and Psychological Outcome (single-item measure); MLM, multilevel modelling; PST-Q, Personal Style of the Therapist Questionnaire; YP-CORE, Young Person's-Clinical Outcomes in Routine Evaluation. Data management and preparation Data entry via the MarBar system will pre-empt transcription errors. To address missing item data, we will use prorating: a total or subscale score will be calculated if ≥90% of items are complete, using the average score of completed items.43 44 Records below this threshold will be excluded. To limit information loss from therapy discontinuation, primary measures will be collected session-by-session.43 44 Analysis by specific objectives Specific objective 2: to identify the characteristics of clients attending psychological therapy, as well as the characteristics of the services and therapists providing it These will be simple descriptive summaries at the four levels of the data using tabulated aggregate summary statistics and plots to facilitate visualisation and interpretation. Client-level analysis will summarise socio-demographic and clinical characteristics using intake data from client information forms, CORE-A/TAF data and baseline scores from the CORE-OM/YP-CORE and EPO-1 measures. We will examine measure acceptability (completion rates), explore correlations between questionnaires and describe associations between client characteristics and baseline distress scores. Post-intervention satisfaction will be summarised using the median, mean (with 95% CI), range and SD. Mean differences in satisfaction (CRES-4) across therapists and services will be estimated with 95% CIs, and its predictors will be evaluated using multilevel models. Therapist-level analysis will describe socio-demographic characteristics and professional styles using data from the PST-Q. Therapist styles will be explored through summaries stratified by service and individual practitioner. Annual style scores from enrolment through year 3 will be summarised within and across services, reporting distributions, means with 95% CIs, SDs and medians. Trajectories of change will be visualised using plots of yearly service means and therapist-level raincloud plots, with within-therapist change characterised relative to the first score. Service-level analysis will provide a descriptive profile of each participating organisation using data from the annual Service Information Form, presented in simple tabulations and plots. To facilitate cross-service visual comparison for all analyses, forest and caterpillar plots will be used. To guard against the overinterpretation of differences, especially where data points per service are few, 95% CIs will be reported around summary statistics. Bootstrap CIs will be applied where cell sizes are sufficient (n>40), with parametric CIs used otherwise. These descriptive analyses will enable participating services to understand and compare their client populations, therapist profiles and organisational characteristics. The primary research purpose is to reflect the diversity of routine mental healthcare in Ecuador. The results will be shared with stakeholders during annual coordination meetings to inform service planning, identify areas for quality improvement and guide the development of targeted interventions. Specific objective 3: to explore the change trajectories and predictors of change for clients attending the services For addressing this objective, we will focus on measures completed across multiple sessions (CORE-10, YP-CORE, EPO-1). CORE measures will be reported separately for adult and youth age groups; EPO-1 data will be analysed both split by and aggregated across age. Reports will prioritise visualisations, such as plots, to convey the diversity of raw data and all aggregate estimates will include 95% CIs. Individual pre–post change on the CORE-10 and YP-CORE will be evaluated per service using the Reliable Change Index by Jacobson and Truax.45 This will be conducted for each measure and service. Individuals will be classified as reliably improved, reliably deteriorated or showing no reliable change. Service-level and pooled proportions with 95% CIs will be reported, with exploratory comparisons across services. For trajectory analyses, multilevel modelling (MLM) will be employed to account for the nested data structure across four levels: repeated measurements (occasion), client, therapist and service. This approach will estimate client change and partition variance at therapist and service levels, enabling within-therapist and between-therapist and within-service and between-service comparisons. The implementation will be adaptive: where cell sizes permit, services will be included as a level in the model. For services with enough therapists (approximately n≥8), therapist-level variance will also be examined. In services with too few therapists for a full four-level model, therapists meeting minimum data criteria (eg, eight clients each with at least five session measures) may be aggregated into a pseudo-service for analysis, with appropriate caveats noted regarding the statistical implications of this aggregation. Complementing the MLM analyses, trajectories will be shown using two plots: cat’s cradle plots for individual and aggregated data, and raincloud plots for group distributions, with assessment occasion on the horizontal axis and scores on the vertical axis. Group trajectories will display mean scores with 95% CIs for each occasion. Predictor and subgroup analyses will incorporate socio-demographic and clinical variables, with trajectories stratified by service and therapist. This study will produce nested data across the four analytical levels: time, clients, therapists and services. Where cell sizes allow, associations within and between levels will be modelled with multilevel mixed-effects models to account for differences at the service, therapist, client and repeated measurements levels. Multilevel models permit estimation of client change and partitioning of variance at therapist and service levels, enabling within-therapist and between-therapist and within-service and between-service comparisons. Given the planned nested analyses, when conducting three-level analyses, achieving 0.80 power with five measurements per participant requires at least 22 therapists, each with a minimum of 8 clients, for a total of at least 176 clients.14 At the service level, conventional rules of thumb indicate that estimating variance components and higher-level random effects is imprecise with fewer than 30–50 clusters; so, with 11 participating services, service-level fixed effects can be estimated but random effects and cross-service variance will be interpreted cautiously.46 47 The modelling approach will address the variable duration of intervention by employing a conditional, time point-specific framework. For example, to analyse client status at a specific milestone, such as the fifth session, the model will condition the analysis on the subset of clients who reached at least that point in therapy. This conditional logic will apply sequentially: the analysis for session six will include only clients who persisted to a sixth session, the analysis for session seven will include only those who reached a seventh, and so forth. This framework ensures that referential comparisons are based on appropriate peer groups defined by actual intervention duration, rather than on an aggregated sample that includes clients with shorter therapies. The findings will provide clinicians and supervisors with actionable feedback on client progress, supporting timely clinical adjustments and service improvement discussions. Service-level comparisons will help identify organisational factors linked to better outcomes, guiding quality improvement initiatives across centres. For research, the findings will advance understanding of factors associated with change at the client, therapist and service levels, with multilevel models revealing the relative contributions of each to intervention outcomes. Specific objective 4: to establish one national, Ecuadorian, dataset and findings to be used support comparisons between the services, the therapies they offer, the clients and the clients’ score change trajectories. These will support future comparison with CLIP-affiliated services in other countries in Latin America To address this objective, we will adopt a benchmarking approach. Following the established definition of benchmarking as ‘the establishment of reference points to aid the interpretation of data’,48 we will generate comparative reference points to be compared against aggregated data from services participating in the CLIP network in other Latin American countries. Where participating CLIP countries agree, non-Ecuador data may also be presented by pseudonymised country to illustrate the diversity across the region. Comparisons will cover key indicators including initial distress severity, distress outcomes, outcome measure completion rates, number of sessions, proportions of planned and unplanned endings, service and therapist characteristics and rates of reliable change. National benchmarking data will allow Ecuadorian services to compare their outcomes and processes with those of other Latin American services, identifying strengths and areas for improvement. These comparisons will inform decision-making by administrators and policymakers, highlighting contextual factors that influence service delivery and outcomes. For research, the findings will enable cross-country comparisons that reveal how sociocultural, organisational and health system factors shape psychotherapy implementation and effectiveness across Latin America. Client and public involvement This research was developed in collaboration with various stakeholders. Therapists, supervisors and decision-makers from the participating services contributed to the study design, drawing on their experience from a previous similar study conducted within one of the included services.21 Furthermore, the protocols were discussed with members of the CLIP consortium, whose feedback, grounded in their experience implementing ROM in their respective services, helped refine the current procedures. Regarding client involvement, clients’ feedback on the experience of completing the questionnaires was collected by therapists and the research team and this feedback is being used to inform the ongoing conduct of the study. In the later stages of the project, clients, therapists, supervisors and decision-makers will be actively involved in interpreting the results. Their insights will inform the best options to disseminate the findings within the consortium and to other services, ensuring that the outputs are relevant and accessible to both clinical and public audiences. Discussion The presentation of this protocol serves two purposes. First, it exemplifies the unified assessment schedule adopted by services within the CLIP.19 By making our methods public, we aim to facilitate replication, encourage collaboration and potentially attract new services to join the consortium. Second, the multicentric nature of this study is particularly significant, as it represents one of the few initiatives of its kind in Latin America. In the short term, we expect this project to generate robust evidence about the effects of psychological interventions across diverse service settings in Ecuador. This evidence will directly inform decision-making through structured meetings with stakeholders. In addition, we aim to cultivate a stronger research culture within routine mental health services, demonstrating that clinical practice and scientific inquiry can be synergistic. Ultimately, the most significant long-term contribution of this work is to support the broader endeavour of the CLIP consortium: to build large, representative datasets that reflect the state of psychological interventions in Latin America, thereby guiding future policy, training and clinical innovation. Despite these strengths, weaknesses inherent to the PBE paradigm must be acknowledged. First, due to the observational and naturalistic design, causal claims about the effectiveness of specific therapeutic interventions cannot be made. Any associations identified between client, therapist or service characteristics and outcomes will not be interpreted as causal. Second, as in most experimental or non-experimental studies of therapies, the absence of long-term post-intervention follow-up prevents understanding whether therapeutic gains are sustained after discharge. Current resource constraints preclude including such work in the current protocol. We hope that findings from this initial project will provide the foundation for future research with additional resources to examine the longer-term sustainability of therapeutic change. Third, although this multicentre study includes 11 diverse mental health services across Ecuador, the findings can only be generalised cautiously. Fourth, sample size and statistical precision considerations also warrant caution about generalisation. This is particularly relevant for estimating service-level effects, given the anticipated variability in the number of clients per service.46 47 However, the use of an estimation paradigm and reporting CIs builds caution about these considerations into the reporting. Finally, missing data and attrition are inherent challenges in naturalistic research. Clients may discontinue therapy unexpectedly, forget to complete measures or decline participation at certain time points. While we have implemented strategies to minimise missingness, such as session-by-session data collection, prorating for incomplete item-level data and using brief measures to reduce burden, attrition will occur and may introduce bias. Clients who drop out of therapy may differ systematically from those who complete intervention, potentially leading to overestimation of positive outcomes if only completers are analysed. These issues will be clearly noted in reports. Despite these limitations, this study represents an important step toward building a sustainable routine outcome monitoring infrastructure in Ecuador and contributing PBE from a Latin American context to the broader CLIP network. By being transparent about these methodological constraints, we aim to position our findings appropriately and provide a foundation for future research that can address the questions this study is not designed to answer. Ethics and dissemination The study protocol received approval from the Human Research Ethics Committee of the UDLA (ID: 2024-OBS-033). Given the project’s dual relevance to both research and clinical practice, a comprehensive dissemination strategy has been developed. Key findings will be shared directly with participating services through formal reports for therapists and administrators, as well as dedicated workshops to facilitate collaborative decision-making for service improvement. To broaden the impact of the project, results will also be presented at academic conferences, published in peer-reviewed journals and communicated to the public via social media through accessible infographics and plain-language summaries. Supplementary material