Testing a Depression Treatment

Major depressive disorder is characterized by a depressed mood or loss of interest or pleasure that lasts for at least two weeks and is accompanied by other symptoms such as changes in sleep or appetite, fatigue, feelings of worthlessness and difficulty concentrating. One biological hypothesis proposes that depression is linked to lower activity of monoamine neurotransmitters, such as serotonin. Selective serotonin reuptake inhibitors (SSRIs) block the transporter that removes serotonin from the synaptic cleft, which raises the concentration of serotonin in the synapse. Cognitive-behavioral therapy (CBT) is a form of psychotherapy that teaches people to identify and change unhelpful patterns of thinking and behavior.

Part 1. Researchers conduct a randomized controlled trial with 400 adults who meet criteria for major depressive disorder. Participants are randomly assigned to four groups of 100: a placebo pill, an SSRI, CBT with a placebo pill, or an SSRI combined with CBT. Neither the participants who take pills nor the clinicians who rate their symptoms know whether a pill is a placebo or an SSRI. Symptoms are rated on a standard scale on which higher scores mean more severe depression. Table 1 shows the mean scores at the start and after 8 weeks, as well as the percentage of participants in each group who dropped out before the end.

Table 1. Mean depression scores and dropout.

Treatment response is defined in this trial as a fall in the depression score of at least half from the starting value. The trial runs at five clinics, and participants are assigned to groups by a computer program, so that the staff cannot predict which group a participant will join. Participants in the SSRI groups are asked to report side effects at every visit, and these reports are kept separate from the symptom ratings made by the blinded raters.

Part 2. The researchers follow the participants who responded to treatment in the SSRI group and in the CBT group for another year. In the SSRI group, treatment is stopped after 8 weeks, and in the CBT group the sessions end after 8 weeks. Table 2 shows the percentage of responders in each group who relapse into a depressive episode within the year.

Table 2. Relapse within one year after stopping treatment.

The researchers report the results for the participants who completed the 8 weeks of treatment, and in a separate analysis they include all of the participants who were originally assigned to each group, using their last recorded score for those who dropped out. The main reasons given for dropping out of the SSRI groups were side effects such as nausea and sleep problems, and in the placebo group the main reason was the feeling that the treatment was not working. The sponsors of the trial did not take part in rating the participants’ symptoms.