IMPORTANCE Compared with the treatment of physical conditions, the quality of care of mental health disorders remains poor and the rate of improvement in treatment is slow, a primary reason being the lack of objective and systematic methods for measuring the delivery of psychotherapy. OBJECTIVE To use a deep learning model applied to a large-scale clinical data set of cognitive behavioral therapy (CBT) session transcripts to generate a quantifiable measure of treatment delivered and to determine the association between the quantity of each aspect of therapy delivered and clinical outcomes. DESIGN, SETTING, AND PARTICIPANTS All data were obtained from patients receiving internet-enabled CBT for the treatment of a mental health disorder between June 2012 and March 2018 in England. Cognitive behavioral therapy was delivered in a secure online therapy room via instant synchronous messaging. The initial sample comprised a total of 17 572 patients (90 934 therapy session transcripts). Patients self-referred or were referred by a primary health care worker directly to the service. EXPOSURES All patients received National Institute for Heath and Care Excellence-approved disorder-specific CBT treatment protocols delivered by a qualified CBT therapist. MAIN OUTCOMES AND MEASURES Clinical outcomes were measured in terms of reliable improvement in patient symptoms and treatment engagement. Reliable improvement was calculated based on 2 severity measures: Patient Health Questionnaire (PHQ-9) and Generalized Anxiety Disorder 7-item scale (GAD-7), corresponding to depressive and anxiety symptoms respectively, completed by the patient at initial assessment and before every therapy session. RESULTS Treatment sessions from a total of 14 899 patients (10 882 women) aged between 18 and 94 years (median age, 34.8 years) were included in the final analysis. We trained a deep learning model to automatically categorize therapist utterances into 1 or more of 24 feature categories. The trained model was applied to our data set to obtain quantifiable measures of each feature of treatment delivered. A logistic regression revealed that increased quantities of a number of session features, including change methods (cognitive and behavioral techniques used in CBT), were associated with greater odds of reliable improvement in patient symptoms (odds ratio, 1.11; 95% CI, 1.06-1.17) and patient engagement (odds ratio, 1.20, 95% CI, 1.12-1.27). The quantity of nontherapy-related content was associated with reduced odds of symptom improvement (odds ratio, 0.89; 95% CI, 0.85-0.92) and patient engagement (odds ratio, 0.88, 95% CI, 0.84-0.92). CONCLUSIONS AND RELEVANCE This work demonstrates an association between clinical outcomes in psychotherapy and the content of therapist utterances. These findings support the principle that CBT change methods help produce improvements in patients' presenting symptoms. The application of deep learning to large clinical data sets can provide valuable insights into psychotherapy, informing the development of new tre...

BackgroundCommon mental health problems affect a quarter of the population. Online cognitive–behavioural therapy (CBT) is increasingly used, but the factors modulating response to this treatment modality remain unclear.AimsThis study aims to explore the demographic and clinical predictors of response to one-to-one CBT delivered via the internet.MethodReal-world clinical outcomes data were collected from 2211 NHS England patients completing a course of CBT delivered by a trained clinician via the internet. Logistic regression analyses were performed using patient and service variables to identify significant predictors of response to treatment.ResultsMultiple patient variables were significantly associated with positive response to treatment including older age, absence of long-term physical comorbidities and lower symptom severity at start of treatment. Service variables associated with positive response to treatment included shorter waiting times for initial assessment and longer treatment durations in terms of the number of sessions.ConclusionsKnowledge of which patient and service variables are associated with good clinical outcomes can be used to develop personalised treatment programmes, as part of a quality improvement cycle aiming to drive up standards in mental healthcare. This study exemplifies translational research put into practice and deployed at scale in the National Health Service, demonstrating the value of technology-enabled treatment delivery not only in facilitating access to care, but in enabling accelerated data capture for clinical research purposes.Declaration of interestA.C., S.B., V.T., K.I., S.F., A.R., A.H. and A.D.B. are employees or board members of the sponsor. S.R.C. consults for Cambridge Cognition and Shire. Keywords: Anxiety disorders; cognitive behavioural therapies; depressive disorders; individual psychotherapy

Background It is increasingly recognized that existing diagnostic approaches do not capture the underlying heterogeneity and complexity of psychiatric disorders such as depression. This study uses a data-driven approach to define fluid depressive states and explore how patients transition between these states in response to cognitive behavioural therapy (CBT). Methods Item-level Patient Health Questionnaire (PHQ-9) data were collected from 9891 patients with a diagnosis of depression, at each CBT treatment session. Latent Markov modelling was used on these data to define depressive states and explore transition probabilities between states. Clinical outcomes and patient demographics were compared between patients starting at different depressive states. Results A model with seven depressive states emerged as the best compromise between optimal fit and interpretability. States loading preferentially on cognitive/affective v. somatic symptoms of depression were identified. Analysis of transition probabilities revealed that patients in cognitive/affective states do not typically transition towards somatic states and vice-versa. Post-hoc analyses also showed that patients who start in a somatic depressive state are less likely to engage with or improve with therapy. These patients are also more likely to be female, suffer from a comorbid long-term physical condition and be taking psychotropic medication. Conclusions This study presents a novel approach for depression sub-typing, defining fluid depressive states and exploring transitions between states in response to CBT. Understanding how different symptom profiles respond to therapy will inform the development and delivery of stratified treatment protocols, improving clinical outcomes and cost-effectiveness of psychological therapies for patients with depression.

The UK’s Improving Access to Psychological Therapy Programme (IAPT) has improved transparency of primary mental health care in relation to the mandatory reporting of clinical outcomes. However, the data reveal a significant variance in outcomes. These findings have led to a growing body of research investigating to what extent therapist variables account for the difference in clinical outcomes. Previous studies have not had access to sufficient recordings or transcripts of therapy sessions in order to fully address this question. The purpose of this study was to use therapy transcripts derived from internet enabled cognitive behavioural therapy (CBT) treatment sessions in order to investigate whether and how therapist variables are associated with clinical outcome. A hierarchical log-linear analysis examined the relationship between therapist/patient variables and clinical outcome. Therapist fidelity to the CBT model and associated adherence to an evidence-based protocol were significantly related to clinical outcome. A graphical representation of the statistical model suggests that patient recovery is directly linked with fidelity and indirectly with adherence, after adjusting for patient attributes of age, gender and clinical presentation. Corroborating previous research, therapist competence and adherence to an evidence-based treatment protocol appear to be important in improving outcomes. These findings have implications for the continuing professional development of qualified therapists, potentially reinforcing the importance of reducing therapist drift. Key learning aims (1) To develop an understanding in relation to which therapist variables are associated with clinical outcome in IAPT. (2) To reflect on how fidelity to the CBT model and adherence to evidence-based treatment protocols may affect clinical outcomes. (3) To exemplify use of a statistical method for enhanced visual understanding of complex multi-factorial data.

This article reports on the experience of internet-enabled cognitive behavioural therapy (IECBT) for older people diagnosed with depression and anxiety. IECBT involves synchronous real-time communication between the therapist and patient via instant messaging and has been found to be effective in the treatment of patients over eighteen diagnosed with depression. While younger populations are an obvious focus for studies into the potential of internetbased therapies, older people's experience of such therapies can be overlooked due to assumptions about their relatively lower rates of internet use. However, rapid increases in this generation's levels of internet access make this an important avenue of enquiry. In addition, such therapies may offer a route to address the underdiagnosis and undertreatment of depression and anxiety among older people. Once older people are diagnosed, evidence suggests that they tend to prefer psychological therapies, and these can be effective in this age group. This article therefore builds on the positive findings about IECBT as a treatment option in general by, for the first time, analysing quantitative data relating to older people's use of this form of therapy. It analyses secondary data on patient characteristics, take-up of treatment, and treatment outcomes, finding that older men are over-represented among IECBT patients and that rates of self-referral are higher in this age group.