The first speaker, Prof. Catherine Harmer from Oxford, had a thought-provoking talk on the early correlates of antidepressant treatment response. Since negative affective bias is a key factor in maintaining depression, it is not a surprising result that an improvement in recognition of happiness precedes a reduction of depression level in response to treatment. Another important result is that this early improvement of positive affect recognition acts as a crucial moderator of environmental effects: interpersonal support for the patient decreases depression only if this early improvement is present. Early changes in emotional processing in response to treatment also has brain correlates within amygdala and ACC (see picture). These early psychological and imaging correlates of antidepressant treatment response may have a crucial clinical translational value.
The second speaker, Prof. Michael Browning, also from Oxford, presented cross-Europe PReDict Study results of a randomised-controlled trial on the efficiency of a machine learning algorithm that guides GPs to change or not to change antidepressant medication after one week. 8-week treatment outcomes of 913 depressed patients demonstrated no effect of the algorithm on response rates, but highlighted that the algorithm was superior to treatment-as-usual in decreasing patients’ anxiety.
Finally, Prof. Richard Morriss from Nottingham, introduced us into the acceptability of PReDict tests, based on patients’ and clinicians’ evaluation. Patients found the tests as an important catalyst for self-reflection, but would need more feedback on results. Clinicians felt more confident in decision-making because of the tests. Overall, PReDict tests have promoted the use of antidepressants.