It is possible to use a mathematical method, based on the interconnection of symptoms, to predict the probability of recovery from major depression. This is stated in a study by the ISS published today in the journal Nature Mental Health. The research has developed a method to measure the so-called plasticity, which is the ability to modify brain activity and behavior, crucial for transitioning from psychopathology to mental well-being.
“To achieve this,” explains Igor Branchi of the Center of Reference for Behavioral Sciences and Mental Health at the Higher Institute of Health, who coordinated the study, “we employed a mathematical technique known as network analysis.
The goal was to demonstrate how plasticity can be mathematically measured by assessing the strength of connectivity in the symptom network, i.e., the frequency with which depression symptoms change together. The higher the synchrony of variations in different symptoms, the higher the coherence (connectivity) of the system and the lower its plasticity: this work shows how more connected configurations are more difficult to modify compared to configurations where connections between symptoms are weaker.”
To verify the method, researchers examined data from one of the most relevant studies on depression treatment strategies, known as STAR*D and provided by the National Institute of Mental Health in the United States, analyzing the improvement trajectory of over 4000 depressed individuals.
“The analysis,” Branchi continues, “confirmed that our mathematical approach can measure patients’ ability to change their depressive state. In particular, we demonstrated that the strength of symptom connectivity, measured at the beginning of the study, was weaker in patients who would later show greater plasticity, presenting significant improvement (responders), compared to those who would show less sensitive improvement (non-responders).”
“Furthermore,” Branchi informs, “we highlighted a highly significant correlation between symptom connectivity and both improvement in depressive state and the predisposition to change mood based on the perceived quality of life.”
The authors emphasize that this method allows estimating the probability of change but does not guarantee predicting with certainty the future health status of the individual, which depends on a multitude of factors.