Positive Affect Treatment for Depression, Anxiety, and Low Positive Affect: A Randomized Clinical Trial.

Importance: Targeting impaired reward processing that underlies anhedonia and diminished positive affect is essential for reducing key risk factors in depression and anxiety, including suicidality and relapse. However, mechanistic research in this area remains limited. Objectives: To assess whether a novel psychosocial intervention engages reward systems more than a mechanistically distinct comparison therapy, and whether alterations in reward and threat processing differentially mediate clinical outcomes. Design, Setting, and Participants: This was an assessor-blinded, parallel-group, multisite, 2-arm randomized clinical superiority trial. Study recruitment was from December 2021 to January 2024, with final assessment in July 2024. Participants were recruited from academic outpatient treatment centers in Los Angeles, California, and Dallas, Texas, and included treatment-seeking adults with severely low positive affect and moderate to severe depression or anxiety that was functionally impairing. Analyses were conducted with intent-to-treat principles. Intervention: Participants underwent 15 weekly individual therapy sessions of positive affect treatment (PAT) or negative affect treatment (NAT). Main Outcomes and Measures: Clinical status was self-reported positive affect (using the Positive and Negative Affect Schedule-Positive subscale), interviewer-rated anhedonia (embedded within the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition), and self-reported depression and anxiety (using the Depression, Anxiety, and Stress Scale). Target measures were 14 self-reported, behavioral, and physiological measures of reward anticipation-motivation, response to reward attainment, reward learning, and threat. Analyses included mixed-effects multilevel models. Results: In total, 98 participants (mean [SD] age, 32.8 [12.2] years; 65 [66.3%] female) were randomized to receive PAT (n = 51) or NAT (n = 47). Multivariate multilevel model analyses of the 3 clinical status variables as a multivariate outcome showed that clinical status improved more with PAT than NAT (b = -0.06 [95% CI, -0.11 to -0.01]; t3039 = 2.43; P = .02; d = 0.27) and that PAT had better (higher) scores on clinical status than NAT at the 1-month follow-up (b = -0.21 [95% CI, -0.41 to -0.02]; t3039 = 2.11; P = .04; d = 0.21). Improvements in reward anticipation-motivation (b = 0.02 [95% CI, 0.01-0.03]; t1307 = 4.36; P < .001; d = 0.40) and reward attainment (b = 0.04 [95% CI, 0.01-0.06]; t1405 = 3.16; P = .002; d = 0.18) targets were comparable for PAT and NAT. Of 7 reward and threat self-reported target measures, 6 mediated improvements in clinical status, but none of the behavioral or physiological measures did. There was limited evidence for moderated mediation. Conclusions and Relevance: In this randomized clinical trial of 98 adults with severely low positive affect, depression, and anxiety, findings suggested that modulation of reward and threat processes was a central mechanism of therapeutic improvement, with a reward-focused intervention producing superior clinical outcomes. Trial Registration: ClinicalTrials.gov Identifier: NCT05203861.