Methods: Sixty-five participants were diagnosed with depression via structured clinical interview and subsequently completed questionnaires at three time points, including measures of cognition, defenses, and mental health symptoms. Reliabilities of all measures were good to excellent. Participants were recruited from the community, and 46% were receiving treatment for depression (medication, psychotherapy, or both). Of the 65 participants, 71% were women (n=46), 77% were white (n=50), the average age was 41.7 years (SD=14.4), and half of the participants (52.3%) were from a lower income socioeconomic status (n=34), The mean level of psychiatric symptoms exceeded published outpatient norms, and 66% had recurrent major depression.
Results: There were small statistically significant decreases across all time points in mean levels of depression, general psychiatric symptoms, automatic thoughts, and immature defenses, and increases in mature defenses. Early changes in automatic thoughts and immature defenses were associated with symptom change from time-one to time-three. The directionality of early automatic thought change predicting symptom change was partially supported, but immature defense change occurs simultaneously with, or after, symptom change. Changes in automatic thoughts were significantly correlated with changes in immature defenses.There were no differences between those receiving treatment and those not receiving treatment in any of these relationships.
Conclusions and Implications: Given the convergent evidence of cognitive change as a mediator of depression reduction, all depression therapies should consider how they address depressive cognition. To build a more complete understanding of how to ameliorate depression, future studies should continue to include constructs from multiple theories. Multitheoretical research may lead to the development of new, more effective treatments. This is the first study that tested CBT and PT mechanisms in untreated depression, suggesting change in cognition and defenses are generic processes not limited to theory-specific therapies. Limitations of the study include primary self-report measures, and partially cross-sectional analysis. Parallel dynamics were observed with this low SES sample, as previous studies with middle-class samples, suggesting similar dynamics of depression across these socioeconomic differences. Further research is needed on the application of evidence-based treatments across sociocultural differences.