Nonlinear Dynamics, Psychology, and Life Sciences, Vol. 14, Iss. 1, January, 2010, pp. 15-25
@2010 Society for Chaos Theory in Psychology & Life Sciences


Evidence of Reduced Complexity in Self-report Data from Patients with Medically Unexplained Symptoms

Christopher Burton, University of Edinburgh, UK
Rachel A. Heath, Syktek, Valentine NSW, Australia
David Weller, University of Edinburgh, UK
Michael Sharpe, University of Edinburgh, UK

Abstract: Physical symptoms which cannot be adequately explained by organic disease are a common problem in all fields of medicine. Reduced complexity, shown using nonlinear dynamic analysis, has been found to be associated with a wide range of illnesses. These methods have been applied to short time series of mood but not to self-rated physical symptoms. We tested the hypothesis that self-reported medically unexplained physical symptoms display reduced complexity by measuring the approximate entropy of self-reported emotions and physical symptoms collected twice daily over 12 weeks and comparing the results with series-specific surrogate data. We found that approximate entropy (ApEn) was lower for actual data series than for surrogate data. There was no significant difference in entropy between different types of symptoms and no significant correlation between entropy and the diurnal variation of the data series. Future studies should concentrate on specific symptoms and conditions, and evaluate the effect of treatment on the entropy of symptom patterns.

Keywords: affective disorders, approximate entropy, models, nonlinear dynamics, somatoform disorders