The control group is corrupted, the control-treatment pairs for comparison were selected by the researchers, and the provided “psychiatric treatment” includes

> In addition to weekly individual sessions the inpatient program at both groups contained two 75 min group sessions each week. In addition, VITA had shorter group meetings each morning (15 minutes). Patients in both treatments participated in two physical exercise sessions per week, weekly psycho-educational lectures and art-therapy groups, and both groups finish each week with end of the week status groups. On average, patients in both treatments received seven sessions of therapeutic activity each week.

“Both treatments” here is not control and treatment but both intensive treatments.

You can get large effects in almost everything by completely changing a person’s experiential environment from their prior environment, which is what they did here. In addition to everything listed, their sleeping conditions, diet, daily routine, and social environment were probably dramatically changed, although the researchers didn’t record that so we can’t know.

It’s impossible to tell if the actual psychiatric interventions were effective. This fact, though, is particularly damning as to the efficacy of the actual treatments provided:

>The analysis also showed minimal differences between the two intensive inpatient treatments, suggesting that the differences in effect may not be due to the theoretical rationale within the inpatient treatment but rather the treatment context.

There were no significant differences between two very different treatment modalities. This essentially admits that “something else” and not the psychiatric treatments were responsible for the uptick. My guess is regular schedule, connecting with people (community formation) and mattering to someone, none of which do you get from psychiatric treatments.