A result with a five-day horizon
The decisive time point in the Tehran trial is the end of the final session. Symptoms were reassessed immediately after electrical stimulation for treatment-resistant schizophrenia. Mean total PANSS scores fell from 84.95 to 59.35 in the active arm and changed from 78.21 to 80.26 with sham stimulation. The separation is large. A reader asking how the patient was doing a month later finds no answer here: there was no follow-up assessment.[1]
That time boundary defines the meaning of the measured benefit. Two sessions a day for five days were added to continuing antipsychotic treatment. The comparison therefore examines an immediate adjunctive effect within the same care setting. The design did not investigate durable recovery, functioning after discharge or a replacement for medication. Clinical expectations become less secure when they extend beyond the period actually measured.[1]
From 45 people to 39
The researchers used computer-generated random allocation and kept outcome assessors away from the allocation sequence. Sham current was briefly ramped up and down; active current continued through the session. This is a stronger design than simply comparing scores before and after treatment, because it supplies a comparison for the care setting and expectations surrounding the intervention. Improvement with time alone is therefore a less persuasive explanation.[1]
However, the analyzed group does not include all 45 people randomized. 3 people in each arm did not receive the allocated intervention, leaving 20 active and 19 sham participants in the analysis. Equal numbers leaving each arm do not establish equal outcomes among those who left. Whether a person actually receives an intervention matters in practice as well. A strong result among completers deserves more caution when translated into expected benefit across all eligible patients.[1]
From symptom scores to daily life
The primary outcome was change in total symptom score; quality of life was secondary. Quality-of-life differences did not remain significant after correction for multiple comparisons. That leaves open how far the symptom reduction carried into daily life. A small sample may have missed a quality-of-life benefit; alternatively, early symptom change may not yet have produced functional gains. The experiment cannot choose between those explanations.[1]
The next trial would gain more than statistical precision from enrolling additional people. Following everyone randomized, continuing the same measurements beyond the last session, and assessing symptoms alongside quality of life could broaden the clinical meaning. This study’s strength is a controlled demonstration of an immediate difference. The patient’s central question is whether that difference lasts. Replication becomes more useful when it measures duration as carefully as magnitude.[1]