Open AccessHarnessing Clinical Psychiatric Data with an Electronic Assessment Tool (OPCRIT+): The Utility of Symptom Dimensions
Author(s) -
Philip J. Brittain,
Sarah Lobo,
James Rucker,
Myanthi Amarasinghe,
Anantha P. Anilkumar,
Martin Baggaley,
Pallavi Banerjee,
Jenny Bearn,
Matthew Broadbent,
Matthew Butler,
Colin Campbell,
Anthony J. Cleare,
Luiz Dratcu,
Sophia Frangou,
Fiona Gaughran,
Matthew Goldin,
Annika Henke,
Nikola Kern,
Abdallah Krayem,
Faiza Mufti,
Ronan McIvor,
Humphrey Needham-Bennett,
Stuart Newman,
Dele Olajide,
David O’Flynn,
Ranga Rao,
Ijaz Ur Rehman,
Gertrude Seneviratne,
Daniel Ståhl,
Sajid Suleman,
Janet Treasure,
John Tully,
David Veale,
Robert Stewart,
Peter McGuffin,
Simon Lovestone,
Matthew Hotopf,
Günter Schumann
Publication year - 2013
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0058790
Subject(s) - medical diagnosis , psychopathology , anxiety , mania , psychiatry , medicine , depression (economics) , clinical psychology , psychology , bipolar disorder , mood , pathology , economics , macroeconomics
Progress in personalised psychiatry is dependent on researchers having access to systematic and accurately acquired symptom data across clinical diagnoses. We have developed a structured psychiatric assessment tool, OPCRIT+, that is being introduced into the electronic medical records system of the South London and Maudsley NHS Foundation Trust which can help to achieve this. In this report we examine the utility of the symptom data being collected with the tool. Cross-sectional mental state data from a mixed-diagnostic cohort of 876 inpatients was subjected to a principal components analysis (PCA). Six components, explaining 46% of the variance in recorded symptoms, were extracted. The components represented dimensions of mania, depression, positive symptoms, anxiety, negative symptoms and disorganization. As indicated by component scores, different clinical diagnoses demonstrated distinct symptom profiles characterized by wide-ranging levels of severity. When comparing the predictive value of symptoms against diagnosis for a variety of clinical outcome measures (e.g. ‘Overactive, aggressive behaviour’), symptoms proved superior in five instances (R 2 range: 0.06–0.28) whereas diagnosis was best just once (R 2 ∶0.25). This report demonstrates that symptom data being routinely gathered in an NHS trust, when documented on the appropriate tool, have considerable potential for onward use in a variety of clinical and research applications via representation as dimensions of psychopathology.