AbstractObjective:To evaluate the use of biofeedback intervention in the levels of depression. The main hypothesis tested if the use of biofeedback improves depression levels compared to the control group.Methods:A randomised clinical trial. The final sample was composed of 36 participants (18 in the experimental group, receiving 6 training, once a week, with biofeedback; and 18 in the control group, who received conventional treatment in the service).Outcome measures were assessed in two stages: pre-test and post-test. The research used the following instruments: demographic survey data, Mini International Neuropsychiatric Interview 5.0.0 and Beck Depression Inventory (BDI). The factors and variables were presented in terms of descriptive and inferential statistics. Fisher’s exact test (p < 0.05) was used to verify the existence of an association between the counting variables. The multinomial logistic regression model was adopted, and the Logit link function was used, as the software RStudio version 3.6.2.Results:The factors that remained in the final model were group, sex, partner, atypical antidepressant, benzodiazepines, mood stabiliser, antiepileptic and antihistamine, according to the levels of depression based on the BDI. The group that did not receive biofeedback intervention had 16 times more chances of increasing the depression levels compared to participants in the experimental group.Conclusion:The use of biofeedback reduces depression, thus, representing a complementary alternative for the treatment of moderate and severe depression, and dysthymia.