Copyright © The British Psychological Society Reproduction in any form (including the internet) is prohibited without prior permission from the Society Dimensions of variation on the CORE-OM K. Jake Lyne 1 *, Paul Barrett 2 , Chris Evans 3 and Michael Barkham 4 1 Department of Psychology, University of York and Selby and York Primary Care Trust, UK 2 Department of Management and Employment Relations, University of Auckland, New Zealand 3 Rampton Hospital, Nottinghamshire Healthcare NHS Trust, UK 4 Psychological Therapies Research Centre, University of Leeds, UK Background. The Clinical Outcomes in Routine Evaluation-Outcome Measure (CORE-OM) is a self-report measure comprising 28 items tapping three domains; subjective well-being, psychological problems and functioning. In addition to the potential theoretical value of the domains for operationalizing the phase model of psychotherapy, when consulted, managers and clinicians considered the distinction between problems and functioning important for assessing case-mix and clinical outcomes. A further domain comprising six items was included to indicate possible risk. Subsequent analysis has suggested an alternative structure for CORE-OM with factors for risk and positively and negatively worded items (Evans et al., 2002). Methods. This study compares models for the interpersonal factor structure in data from the CORE-OM in 2,140 patients receiving psychological therapy in the UK. Results. A multi-method, multi-trait, nested factors solution accounted optimally for the CORE-OM item covariance, with a first-order general factor latent and residualized first-order factors of subjective well-being, psychological problems, functioning and risk and with positively and negatively worded methods factors. The general factor was labelled psychological distress. Scale quality for CORE-OM, using a scoring method in which non- risk items are treated as a single scale and risk items as a second scale is satisfactory. Implications. The CORE-OM has a complex factor structure and may be best scored as 2 scales for risk and psychological distress. The distinct measurement of psychological problems and functioning is problematic, partly because many patients receiving out-patient psychological therapies and counselling services function relatively well in comparison with patients receiving general psychiatric services. In addition, a clear distinction between self-report scales for these variables is overshadowed by their common variance with a general factor for psychological distress. An alternative strategy for operationalizing this distinction is proposed. * Correspondence should be addressed to Dr Jake Lyne, Department of Psychological Therapies, The Old Chapel, Bootham Park, York YO30 7BY, UK (e-mail: jake.lyne@sypct.nhs.uk). This paper, which is jointly authored by the current editor, was processed and accepted for publication by the previous editorial team. The British Psychological Society 185 British Journal of Clinical Psychology (2006), 45, 185–203 q 2006 The British Psychological Society www.bpsjournals.co.uk DOI:10.1348/014466505X39106