Introduction

The last two decades have seen an influx in Internet-based and digital mental health content. Such media offer unique opportunities to increase the availability and reach of mental health-related information and self-help content, often free or low-cost. For example, SMS-based mental health interventions have been used to target a range of outcomes including reducing symptoms of anxiety and depression, coping with suicidal ideation, and decreasing alcohol and cigarette use [1]. With the rise of the smartphone era, thousands of mental health-related smartphone applications have been released, with evidence supporting their effects on improving psychological symptoms and quality of life [2], as well as generally being perceived by users as acceptable and helpful [3]. Furthermore, as the COVID-19 global pandemic triggered a series of lockdowns from 2020 onwards, people have increasingly relied on digital mental health interventions in place of in-person supports, such as telehealth services, Internet-based support groups, and psychoeducation webinars [4].

Podcasts are another form of digital media with potential to positively impact upon mental health. Unlike traditional radio, a podcast is ‘a piece of episodical, downloadable or streamable, primarily spoken audio content, distributed via the internet, playable anywhere, at any time, produced by anyone who so wishes’ [5, p. 11]. Podcasts are typically distributed through an RSS (‘Really Simple Syndication’) Feed, which distributes the file to a range of streaming platforms [6]. As a result, podcast episodes can be streamed or downloaded to a computer or mobile device using distribution apps such as Apple Podcasts and Spotify. Podcasts are now extremely common, with an estimated four million podcasts series as of 2023 [7]. Uptake is also high, with around half of American and Australian adults listening to a podcast within the last month [7,8], and listeners reporting podcasts to be accessible, engaging, and diverse [9,10].

Podcasts are increasingly being used in the mental health field; like other mental health interventions [11], including media-based interventions, they may target the general public or whole population (i.e., as universal interventions). Or, at a smaller scale, they may target individuals who currently have, or are at risk of developing a mental health condition (i.e., selective or indicated interventions) [11]. According to podcast database ListenNotes, there are over 2000 podcast series publicly available with ‘mental health’ in the title [12]. Topics include psychoeducation (such as coping or self-help tips), marriage, couples and family relationship advice, and education and professional development aimed at mental health professionals [13,14]. Podcasts may feature hosts and guests with lived experience of mental health issues, mental health professionals, advocates, life coaches, celebrities and influencers, educators, journalists, and more. They may be used by mental health professionals as part of their therapeutic services, providing a resource to clients and patients to support their recovery [13]. Many prominent mental health organisations have released their own podcasts, for example, Not Alone by Beyond Blue (Australia), The Mind Podcast by Mind (UK), and In the Open by Mental Health America (USA).

Recent research has begun to map how health-related podcasts are being used and evaluated, including a recent scoping review by Robins et al. that identified three interventional studies for depression and anxiety with null findings [15]. But despite the plethora of mental health-related podcasts available to the public, these have not been systematically explored in terms of evaluation approaches, or for a wider range of mental health outcomes. There is a need to understand the format and structure of podcast-based interventions, to understand how mental health professionals, advocates, and researchers have utilised this medium to communicate with audiences and evaluate their impact. There is also a need to understand such podcasts’ impact upon listeners to date.

A systematic scoping review was chosen as the approach for the current study as the area had not previously been systematically reviewed, and was likely to be highly heterogeneous in terms of types of podcasts and mental health outcomes [16,17]. Furthermore, the authors wished to understand broad characteristics, concepts, and methodology, rather than answering a specific question about efficacy. The aim of this review was to determine how podcasts are being used as a platform for mental health-related interventions (‘podcast-based interventions’). This aim was achieved.

Methods

Results

Selection of sources of evidence

Please see Fig 1 for the PRISMA flowchart. Across all searches, there were 2958 records identified, with 2468 screened after removal of duplicates.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA flowchart.

https://doi.org/10.1371/journal.pmen.0000272.g001

The initial database and trial register search produced 1824 records, with 1717 records from databases, and 107 from trial registers. After removal of duplicates, 1510 titles and abstracts were screened, with 13 meeting inclusion criteria after full text review. Updated searches (searching between 1 November 2021 to 4 August 2024) identified a further 697 records, with 604 from databases and 93 from trial registers. After removal of duplicates, 566 unique titles and abstracts were screened, with seven new records meeting inclusion criteria after full text review.

An additional citation list search identified 437 citations, with 397 titles and abstracts screened after duplicates removal. One additional record met inclusion criteria after full text review.

Additional searches linked six further records from the unique studies (e.g., peer-reviewed publications published after trial registrations). After accounting for multiple reports and merging relevant records, in total, there were 20 unique studies included in the scoping review, with 27 associated records.

Characteristics of sources of evidence

Publication characteristics are presented in Table 1. Publication occurred between 2011 and 2024, with over half (11/20, 55%) published in the last five years. Study location was most commonly the USA (10/20, 50%) and Australia (5/20, 25%). Across the 27 records, there were 13 peer reviewed-journal articles, eight dissertations, five clinical trial registrations, and one conference abstract. Eleven studies (55%) were randomised controlled trials, with the remaining nine being pre-post studies including experimental studies or non-randomised controlled trials.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of included studies.

https://doi.org/10.1371/journal.pmen.0000272.t001

Podcast intervention features

Definitions.

Only five studies provided a clear definition of a podcast [23–27]; all referred to an audio recording or file, and all but one [27] referenced distribution through the Internet. Three definitions also highlighted podcasts as being on-demand [23–25]. Fernandez [23] further defined podcasts as also being distributed using RSS or Atom syndication formats. In their definition, Kissell et al. [26] specified that podcasts “frequently utilize storytelling or conversations between individuals to explore a topic” (p. 2), while Pownell’s [25] definition also mentioned that podcasts are “short” and “inexpensive to produce” (p. 46).

Episode descriptions.

Key intervention characteristics are presented in Table 2. Across studies, participants were asked to listen to between 1 episode [24,25,28–31] and up to 28 episodes (30). The shortest reported episode length was 3 minutes [29], while the longest episodes reported were 50-60 minutes [32]. Typically, podcast episode length was between 10-20 minutes long.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Podcast intervention characteristics.

https://doi.org/10.1371/journal.pmen.0000272.t002

Content of podcast intervention episodes varied significantly, with the majority seemingly being scripted or instructive; these were usually meditation or mindfulness-based podcasts. Only three [25,28,33] were explicitly described as being conversational in nature, or interview-based; two were documentary-style including clips from people with lived experience of mental health issues interspersed with narration or other audio elements [30,34]. One described a mixture of audio clips including interviews from people with lived experience and health professionals [26]. Though not always specified, episodes were described as typically involving 1-2 speakers (such as a single host guiding the listener through a mindfulness exercise). An exception was Kissel et al. [26], which described episodes as having a range of clips from different guests with various backgrounds.

As seen in Table 2, eight podcasts (40%) involved predominantly meditation or mindfulness exercises. These included guided body scans, guided imagery, relaxation and breathing exercises, attention training, and self-compassion meditations. Eight (40%) podcasts were classified as predominantly psychoeducational or therapeutic in nature. These podcasts included body image discussions based in cognitive behavioural therapy principles [25], coping skills and self-care [27,35]; psychoeducation about stress management [35,36]; brief interpersonal psychotherapy-based content around adjusting to a relationship separation [27]; parenting strategies for preventing and managing behavioural or emotional concerns in children [31,33]; and discussions about specific mental health conditions: psychosis [24] and opioid use disorder [26]. Two were focussed on storytelling with people from lived experience of mental health issues, with one person’s story presented per episode [30,34]. Other educational podcasts included audio-recorded university lectures, a.k.a. ‘coursecasting’ [23] while another involved short audio clips relating to advance care planning in the event of death [29]; both were eligible for the present review as they reported mental health-related outcomes as operationalised in S1 Protocol.

Comparison conditions.

As seen in Table 1, eight of the 20 studies (40%) involved only one group, while twelve studies (60%) involved at least one comparison condition. Nine (45%) had two groups, two (10%) had three groups, and one (5%) had four groups.

Of those with a control or comparison condition, half (6/12, 50%) involved some sort of comparison audio, five (5/12, 42%) involved a no-treatment or waitlist condition, and one was compared to an in-person meditation condition (the podcast was classified as the control condition [32]). Three studies (3/12, 25%) included at least one thematically different podcast as a control, externally produced and matched for episode length – including a science and philosophy podcast episode from Radio Lab [37], a history/pop culture podcast episode from the podcast Stuff You Should Know [28], and an unspecified nature podcast [30]. Others included at least one similarly themed podcast, but missing an active or key element of the intervention of interest, including different meditation exercises [38] or ‘sham’ mindfulness exercises that did not involve an attentional anchor [39], and episodes from This American Life and ABC Radio National about mindfulness and society [39]. Another involved a general interest psychology podcast created by the same institution but without an explicit focus on stigma or discrimination as seen in the intervention condition [34].

Meanwhile, two studies tested variations on the intervention itself. First, Wilks [30] compared three groups: one group were randomised to listen to any episode from a 9-part series, the second group were allowed to choose their own episode, and the third listened to a nature podcast. Seiter [29] compared different variations of similar content, testing variations on profanity, humour, and gender of the narrator.

Discussion

Across 20 studies, this systematic scoping review highlighted the breadth of podcast-related interventions targeting mental health-related outcomes in listeners, with most being published in the last five years. The review identified a variety of podcast types which have been studied experimentally, including meditation and mindfulness, psychoeducational content, other educational content, and lived experience non-fiction storytelling. These podcasts generally had positive impacts on a range of outcomes including increasing state mindfulness, self-compassion, stress, parenting-related and body image-related variables, and reducing stigma towards people living with mental illness. However, the relatively small number of podcast-based interventions identified in the search was somewhat surprising, considering podcasts have been produced for over 20 years at the time of writing [6]. The thousands of mental health-themed podcasts available to the public – including by well-known mental health organisations – far outnumber podcasts with published impact data. It is unlikely that this sample accurately represents mental health podcasts widely, and perhaps reflects a lack of interest in empirically measuring the impact of podcasts to date. However, as podcasts usually involve a passive, one-way listening experience – especially those designed to be universal interventions, delivered through mass media – not all mental health-themed podcasts are conceptualised as interventions with measurable mental health-related outcomes. Hence, in these instances, rigorous evaluation would not necessarily be plausible or warranted.

The podcasts identified in this review were heterogeneous, including diverse content aiming to affect a range of mostly self-reported mental health-related outcomes. Interestingly, nearly all podcast content was described or implied to be scripted – particularly those which involved meditation or mindfulness-based activities. This is despite podcasts being amenable to a range of formats, including semi-scripted or unscripted interviews and conversations, panel discussions, documentary-style storytelling, and even fictional narratives. In general, there was minimal description provided about the podcast content, limiting replicability. Few studies provided detailed descriptions about episode formats, production processes, or the usage of music, sound effects, or other elements that may contribute to a podcast’s tone or the listeners’ emotional experience or make content engaging [43]. There is a need to further explore which formats, guests, and creative elements might make podcast-based interventions engaging and effective for their intended listeners and outcomes. This might first require formative research with target audience members (e.g., co-design [44]). Such studies can help identify the ‘active’ elements of a podcast that can lead to the most change, prior to testing the podcast’s efficacy. Future mental health podcast creators and evaluators may benefit from drawing from communication and marketing literature [e.g., 45] to maximise podcast reach and impact in real-world settings, including when targeting broad populations or more selective groups [11].

This study also raises a question: what exactly can be labelled as a podcast? Most studies only involved exposure to 1-3 episodes, and the shortest podcast consisted of only a single episode of 3-5 minutes [29], which participants listened to once. This, alongside only five studies providing a clear definition of a ‘podcast’, illustrates significant conceptual overlap between podcasts and general audio-based stimuli used in interventions, and conceptual issues seen more broadly [5]. Single exposures to short clips are limited: they cannot build on messaging over time, or reinforce messages; nor can they be expected to foster a sense of connection between listeners, hosts and guests. Indeed, as podcasts typically use an RSS feed, this implies podcasts inherently have a serial nature, with multiple episodes [5,6]. Contrasting this, most studies included podcast episodes that were under 20 minutes in length [26]; this is interesting, considering the average podcast length range is between 30 and 42 minutes [46]. This likely reflects researchers’ attempts to minimise listening burden and attrition, though it may also be appropriate for their content (for example, meditation podcasts may require a shorter length). However, this does mean the present review cannot provide information about longer-form episodes and impact of listening to series over time. Ultimately, the ‘dosage’ required to make a meaningful impact on listeners is unknown.

Notably, nearly half of the studies involved meditation and mindfulness-based content, if they were described or contextualised as podcasts. This reflects a broader trend, as thousands of meditation podcasts are available, and they are popular among listeners [14]. Guided mindfulness naturally lends itself to the podcast format: these podcasts allow listeners to access a range of guided exercises, on demand, for free or at low cost. Of note, there is already a large body of research involving digitally guided meditation, including audio recordings, meditation apps, and virtual reality [47,48] that are not necessarily conceptualised as podcasts and therefore were not eligible for the current review. The positive impacts of meditation and mindfulness on mental health outcomes are well-established (e.g., [48–50], and the podcast-based studies included in this review generally also had positive findings.

Psychoeducational content was also popular, with 40% of the podcasts in the study being categorised as such. Interestingly, almost no studies used clinical samples, such as participants who have been diagnosed with a mental health condition and were seeking psychological support. This is despite podcasts increasingly being used as an adjunct to direct counselling or psychological support [13] or as part of multi-component mental health interventions [51–53]. The use of standalone or stepped care podcast-based self-help interventions is worthy of study, considering the potential to reach a wide audience at low cost. For example, Kazan [27] adapted brief interpersonal psychotherapy to the podcast format, whilst the podcast evaluated by Dunston [35] utilised cognitive behavioural therapy and mindfulness techniques; however, both studies had high attrition and neither had statistically significant outcomes. Whilst evidence-based psychological interventions can be adapted to this format, more research is needed to explore how these can be adapted effectively. As seen with smartphone apps, it is unlikely that listeners will use such podcasts to replace traditional psychological supports, as listeners may prefer individualised, tailored support rather than generic or transdiagnostic therapeutic advice [3].

Only four studies to date have explored podcasts for stigma reduction – exploring the impact of podcasts on stigmatising attitudes towards people living with psychosis [24], opioid use disorders [26] or people with lived experience of mental health issues in general [30,34]. These podcasts used both educational content and lived experience non-fiction storytelling, involving audio directly from people with lived experience. This highlights how podcasts can involve educational and contact-based stigma reduction messaging [54], which are known to be evidence-based stigma reduction strategies [55]. Podcasts can also be used for improving mental health literacy for stigma reduction purposes [34,44]. Indeed, all studies in the present review had promising findings, reporting improvements in stigmatising attitudes among listeners. However, only two of the four studies involved RCTs [30,34] and one of the four studies did not use any validated outcome measures [26]. However, as highlighted by Carrotte et al. [34,44], podcasts are a type of ‘opt-in’ media; more work is needed to understand how podcasts can maximise listener engagement, and attract listeners outside of closed research studies who hold stigmatising attitudes and may not be interested in the subject matter.

Attrition was high across the studies, with a median rate of 42%, and up to 84%. Attrition is a common issue in digital interventions [56] and raises questions about how to incentivise participants to complete such studies. Reimbursement (seen in 60% of studies) is one option, as is reducing the length or complexity of the intervention. In real world settings, many listeners will have low interest in engaging with a podcast unless it is entertaining, engaging, and relevant to their interests [10]. Consequently, listeners may not be interested in podcasts which explicitly aim to improve their wellbeing unless they are purposively help-seeking [43]. This review did not systematically extract feedback data, but as an example of uptake issues, in one study only 39% of participants reported that they would have been willing to practise the podcast-based meditation for another two weeks [41]. Indeed, some studies acknowledged target populations being time-poor [e.g., healthcare providers, 32]. Another issue is how to balance compliance with more pragmatic considerations. Most people listen to podcasts while multi-tasking, and the ability to multi-task while listening is a key feature of podcasts and a function that appeals to many listeners [10]. A laboratory setting (such as that used by Smyth et al. [28]) might increase compliance and attention, but not represent the real-life listening experience. Most studies involved participants listening online via the Internet, and usage of attention checks may help indicate how much attention participants paid to each episode. Embedding episodes in survey software can also help researchers collect data, such as time spent on a webpage, to help understand compliance.

Limitations

The authors acknowledge the limitations of this review. As a systematic scoping review, this study was purposefully broad, and unable to make claims about the systematic evidence for and against podcast-based interventions with mental health-related outcomes, or the quality of the evidence. Rather, it assessed the current state of both published research and grey literature, the value of this area of study, and explored methodological considerations for future research. The scope was also limited to English-language articles. Data relating to safety, risk of harm, or adverse events were not extracted; this may be of interest to future researchers. Furthermore, the authors had to carefully consider how ‘podcast’ was defined in the scoping review protocol (see S1 Protocol), considering the risk of conceptual overlap [5]. Hence, it was unable to assess evidence for podcasts with video content (i.e., webcasts, vodcasts) or those involved in multi-component or hybrid interventions (e.g., a podcast provided as a learning tool in addition to an in-person intervention). Though these are likely to have benefits, it would have been difficult to compare these with standalone, audio-only podcasts.

Conclusions

This systematic scoping review highlights how podcasts can be an effective medium for improving listeners’ wellbeing and attitudes across many types of mental health outcomes. However, podcast-based interventions were highly diverse with significant attrition. Future podcast-based intervention studies need to carefully consider the dosage, content, and pragmatic factors in order to evaluate their efficacy and ensure they are safe and appropriate for both selective and universal intervention efforts. This requires careful consideration of health communication literature to inform a theory of change relevant to the medium as well as the message [34,43]. Evaluation studies could also consider using objective or behavioural outcomes, and increasing the usage of control and comparison conditions, to improve study quality.

Future systematic reviews and meta-analyses should assess the quality of the emerging evidence base, and explore the features of podcasts which enhance the listener’s experience and engagement with the content, such as length, number of episodes, and types of content. They may also consider collating listener feedback and other acceptability data.

Supporting information