Rob Saunders1, John Cape2, Pasco Fearon2, Stephen Pilling2. 1. Research Department of Clinical, Educational and Health Psychology, University College London (UCL), 1-19 Torrington Place, London WC1E 7HB, United Kingdom. Electronic address: r.saunders@ucl.ac.uk. 2. Research Department of Clinical, Educational and Health Psychology, University College London (UCL), 1-19 Torrington Place, London WC1E 7HB, United Kingdom.
Abstract
BACKGROUND: The outcomes of psychological therapies for anxiety and depression vary across individuals and symptom domains. Being able to predict treatment response from readily available patient data at presentation has potentially important benefits in aiding decisions about the most suitable interventions for a patient. This paper presents a method of identifying subgroups of patients using latent profile analysis, and comparing response to psychological treatments between these profiles. METHODS: All outpatients taken into treatment at two psychological treatment services in London, UK and who provided basic demographic information and standardized symptom measures were included in the analysis (n=16636). RESULTS: Latent Profile Analysis was performed on intake data to identify statistically different groups of patients, which were then examined in longitudinal analyses to determine their capacity to predict treatment outcomes. Comparison between profiles showed considerable variation in recovery (74-15%), deterioration rates (5-20%), and levels of attrition (17-40%). Further variation in outcomes was found within the profiles when different intensities of psychological intervention were delivered. LIMITATIONS: Latent profiles were identified using data from two services, so generalisability to other services should be considered. Routinely collected patient data was included, additional patient information may further enhance utility of the profiles. CONCLUSIONS: These results suggest that intake data can be used to reliably classify patients into profiles that are predictive of outcome to different intensities of psychological treatment in routine care. Algorithms based on these kinds of data could be used to optimize decision-making and aid the appropriate matching of patients to treatment.
BACKGROUND: The outcomes of psychological therapies for anxiety and depression vary across individuals and symptom domains. Being able to predict treatment response from readily available patient data at presentation has potentially important benefits in aiding decisions about the most suitable interventions for a patient. This paper presents a method of identifying subgroups of patients using latent profile analysis, and comparing response to psychological treatments between these profiles. METHODS: All outpatients taken into treatment at two psychological treatment services in London, UK and who provided basic demographic information and standardized symptom measures were included in the analysis (n=16636). RESULTS: Latent Profile Analysis was performed on intake data to identify statistically different groups of patients, which were then examined in longitudinal analyses to determine their capacity to predict treatment outcomes. Comparison between profiles showed considerable variation in recovery (74-15%), deterioration rates (5-20%), and levels of attrition (17-40%). Further variation in outcomes was found within the profiles when different intensities of psychological intervention were delivered. LIMITATIONS: Latent profiles were identified using data from two services, so generalisability to other services should be considered. Routinely collected patient data was included, additional patient information may further enhance utility of the profiles. CONCLUSIONS: These results suggest that intake data can be used to reliably classify patients into profiles that are predictive of outcome to different intensities of psychological treatment in routine care. Algorithms based on these kinds of data could be used to optimize decision-making and aid the appropriate matching of patients to treatment.
Authors: Akiva Kleinerman; Ariel Rosenfeld; David Benrimoh; Robert Fratila; Caitrin Armstrong; Joseph Mehltretter; Eliyahu Shneider; Amit Yaniv-Rosenfeld; Jordan Karp; Charles F Reynolds; Gustavo Turecki; Adam Kapelner Journal: PLoS One Date: 2021-11-12 Impact factor: 3.240
Authors: Joshua E J Buckman; Rob Saunders; Zachary D Cohen; Katherine Clarke; Gareth Ambler; Robert J DeRubeis; Simon Gilbody; Steven D Hollon; Tony Kendrick; Edward Watkins; Ian R White; Glyn Lewis; Stephen Pilling Journal: Wellcome Open Res Date: 2020-04-01
Authors: Joshua E J Buckman; Rob Saunders; Zachary D Cohen; Phoebe Barnett; Katherine Clarke; Gareth Ambler; Robert J DeRubeis; Simon Gilbody; Steven D Hollon; Tony Kendrick; Edward Watkins; Nicola Wiles; David Kessler; David Richards; Deborah Sharp; Sally Brabyn; Elizabeth Littlewood; Chris Salisbury; Ian R White; Glyn Lewis; Stephen Pilling Journal: Psychol Med Date: 2021-04-14 Impact factor: 7.723
Authors: Rob Saunders; Zachary D Cohen; Gareth Ambler; Robert J DeRubeis; Nicola Wiles; David Kessler; Simon Gilbody; Steve D Hollon; Tony Kendrick; Ed Watkins; David Richards; Sally Brabyn; Elizabeth Littlewood; Debbie Sharp; Glyn Lewis; Steve Pilling; Joshua E J Buckman Journal: J Pers Med Date: 2021-12-04