Approximate entropy of symptoms of mood: an effective technique to quantify regularity of mood

Objectives: Several psychiatric conditions are associated with lability of affect. In this study, we investigated regularity of mood using APEN (Approximate Entropy) on daily subjective ratings using a Visual Analog Scale with 11 items pertaining to mood. Methods: APEN was applied to the data in a double-blind placebo controlled investigation on the effects of fluoxetine (n=19), pemoline (n=18) and placebo (n=20) in normal controls. These subjects rated their subjective feelings daily at the end of each day. We analysed 56 data point time series (each value was obtained on each day) after the three experimental conditions. Results: While the mean or the SD of all the 56 points was not significantly different among the three conditions, APEN was highly and significantly lower for pemoline compared with fluoxetine and placebo. There was no significant correlation between average APEN and mean or SD (standard deviation). The one symptom that explained most of this difference among the groups after drug administration was the feeling of 'happiness'. Conclusions: This result indicates that there was a more consistent subjective sense of happiness during the 8-week period of pemoline administration compared with the other two drugs. Though this study was not designed to address the efficacy of mood stabilizing drugs, such daily subjective ratings may be useful in future studies that evaluate the stability of mood. APEN has been used in several different fields of research with small data sets and this study extends its possible use to evaluate changes in mood in certain populations such as patients with bipolar disorders.