Podcasts are increasingly being used for mental health-related messaging. The objective of this scoping review is to understand how podcasts are being used as a platform for mental health-related interventions (‘podcast-based interventions’). Six databases were searched: CENTRAL, EMBASE, PsycINFO, Communication and Mass Media Complete, Web of Science, and ProQuest Dissertations & Theses Global. Journal articles, conference proceedings, and dissertations were eligible for inclusion. Eligible studies included an audio-only podcast-based condition with at least one quantitative mental health-related outcome, including symptoms, treatment or management of mental health issues, mental health literacy or knowledge, and mental illness stigma, prejudice or discrimination, and involved experimental and quasi-experimental designs. Databases were searched for English-language results up to August 4, 2024. Across all searches, there were 2958 records identified, with 2468 screened after removal of duplicates; 20 unique studies met inclusion criteria. Most common podcast intervention types were meditation or mindfulness exercises, comprising eight (40%) of the studies, and psychoeducational or therapeutic content, also comprising eight (40%) studies. Most podcasts were streamed or downloaded online, and interventions ranged 1-28 episodes. Studies typically involved university or general community convenience samples, and median attrition was 42% (IQR = 10-50%). Researchers most studied the impact of podcast-listening on anxiety, reported in 35% of studies, followed by depressive symptoms (30%), stress or psychological distress (25%), body image-related variables (20%), and stigma (20%). There is some evidence supporting the efficacy of podcast-based interventions on various mental health-related outcomes, including improvements in mindfulness, body image, and stigmatising attitudes. Results highlighted the breadth of experimental and quasi-experimental studies involving podcast-based interventions with mental health-related outcomes. These podcasts show promising mental health-related outcomes worthy of further study and refinement. Study generalisability was limited predominantly by self-report data, convenience samples, and high attrition rates.