Article Text
PDF
Abstract
Podcasts are an area of innovation in the neurointerventional space that has the potential to convey information in ways that traditional journal articles in peer-reviewed journals do not. BMJ maintains an archive of all of its podcasts on the Journal of NeuroInterventional Surgery (JNIS) website. We sought to analyze this early JNIS podcast experience and assess the impact of content elements and an increased presence in social media.
- Technology
Statistics from Altmetric.com
Introduction
The Merriam-Webster Tenth International Collegiate defines ‘podcast’ as: a program (of music or talk) made available in digital format for automatic download over the internet.1 A portmanteau is a linguistic blend of words that derives from a suitcase that opens into two equal sections.2 A podcast is a portmanteau of ‘pod’ and ‘broadcast’.3 The term ‘pod’ had been popularized by the ubiquitous presence of i-Pods in the mid-2000s. The term ‘podcasting’ was first used in a British newspaper in February 2004.4
Prior to the emergence of the i-Pod, podcasting as we now think of it did exist and was referred to as ‘audioblogging’.5 In the 1980s, radio computing services provided audioblogs to radio stations. Websites began to provide audio subscription services in the 1990s. Carl Malamud launched Internet Talk Radio in 1993 with audio files pulled one at a time. The digital download of music became widely popular with the emergence of Napster. Napster was a peer-to-peer audio file-sharing program that became popular soon after it was launched. Perceived loss of control and/or revenue led to Napster being immersed in legal quagmires from soon after it launched. In the early 2000s, the first portable MP3 players began to appear. Simultaneously, pioneers began to experiment with the concept of ‘enclosure’. In this context, enclosure refers to passing a web address to a media aggregator in order to share ‘fetchable’ content. Over the next few years, additional believers in the technology adopted this format. Seven months after Hammersley at The Guardian suggested the term ‘podcasting’ as described above, Dannie Gregoire used the term ‘podcasting’ to describe the automatic download and synchronization of audio content.6 The term ‘podcast’ rapidly took hold.
The JNIS experience
The Journal of NeuroInterventional Surgery (JNIS), the official journal of the Society of Neurointerventional Surgery (SNIS), was launched in 2009 in partnership with the publishing house BMJ. BMJ, the journal, has a robust online presence including access to high-quality podcast production facilities. This podcast service is enabled for its constituent publications.
The first JNIS podcast featured Dr Turner describing an Editor's Choice paper on a novel non-occlusive neck protection device that had been evaluated in a canine model.7 This pattern continued for the next seven podcasts with Editor-in-Chief Dr Tarr reviewing Editor's Choice articles with a single discussant—typically an important contributor to the manuscript. Since its inception and continuing to the present day, JNIS has featured a focus on socioeconomic and policy topics impacting neurointerventionalists. The ninth podcast featured several innovations. These included having two (rather than one) discussants being interviewed by an associate editor (rather than the Editor-in-Chief) and focused on a socioeconomic (non-traditional) topic. Also, the ninth podcast was based on two distinct articles rather than the one Editor's Choice paper.8 ,9 Future podcasts varied in the number of underlying articles that were discussed. At the top of this scale, the MR CLEAN podcast made reference to five underlying JNIS articles.10–14
In addition to expanding beyond the print horizons into podcasts, JNIS has embraced new media through an active presence on both Facebook and Twitter. More recently, the JNIS masthead expanded to include two social media assistant editors. It is now common for these editors to share production of JNIS podcasts through these vibrant media. The JNIS social media editors began expanding the online presence and advertising for articles and podcasts in early 2015 after the podcast discussing the endovascular treatment for large vessel stroke. One of the advantages of social media is the ability it affords to track website views. This makes it possible to continually assess the effect of social media advertisement on article and podcast downloads.
Materials and methods
The number of podcast downloads as of November 3, 2015 16:16 h was placed into Table A and the per cent of total downloads individually tabulated utilizing public domain information from http://jnis.bmj.com/site/podcasts/. The podcasts were then divided into two groups: Group 1 is represented by content related to ‘non-classic’ neurointerventional article topics and Group 2 is important clinical and research ‘classic’ articles/podcasts in neurointervention. Next, the groups were divided by pre- and post- introduction of JNIS social media editors.
For the analysis of non-classic versus classic articles as well as the impact of the addition of the social media editors, the number of downloads of each podcast is a continuous variable and was examined for each group of podcasts. The number of downloads for the total articles and for each group is not normally distributed. The Wilcoxon rank sum test was used to compare the median number of downloads between the two groups. The statistical software package STATA (V.13) was used for all analyses with a p value <0.05 considered to be statistically significant.
This analysis does not require approval of an Institutional Review Board and is not impacted by the Health Insurance Portability and Accountability Act.
Results
There were 30 total podcasts that were downloaded a total of 15 526 times. The least downloaded podcast was the chronological first (#30 in table 1), which accounted for 0.57% of total downloads. The most downloaded podcast accounted for slightly more than one-third of the total. The number of downloads per individual podcast and the percentage of the total are shown in table 1 and figure 1.
Title, number and percentage of downloads for each podcast
Graphic representation of frequency of individual podcast downloads.
There were seven non-classic neurointerventional podcasts in Group 1 and 23 in Group 2. The podcasts in Group 2 were considered traditional clinical or research neurointerventional ‘scientific’ content. Group 1 included podcasts 3, 4, 12, 14, 17, 20 and 22 and Group 2 included all the others (see table 1). The median number of downloads for Group 1 was 644 (IQR 346–745) and the median number of downloads for Group 2 was 304 (IQR 126–414). Group 1 podcasts were downloaded more than Group 2 podcasts (644 vs 304) and the difference was statistically significant (p=0.0087; see table 2 and figure 2).
Non-traditional versus traditional neurointerventional podcasts
Graphic representation of number of downloads per podcast.
Downloads have been qualitatively higher in the five podcasts since the addition of the social media editors. Group 1 includes the five podcasts that have occurred since that intervention. There are five articles in Group 1 and 25 articles in Group 2. The median number of downloads for Group 1 was 542 (IQR 423–974) compared with 311 (IQR 147–414) for Group 2. Group 1 articles were downloaded more than Group 2 articles (542 vs 311), but the difference is not statistically significant (p=0.0584, table 3).
Impact of addition of social media editors
Discussion
The expanding SNIS social media presence couples seamlessly with the growing JNIS podcast library. With reinforcement through announcements at the annual meeting, various suite talks, and other methods, the number of SNIS members active on Twitter and social media continues to increase. These followers also tend to be interested in SNIS Facebook posts and Tweets resulting in a positive feedback loop.
SNIS has approximately 550 senior members out of 800 total members. JNIS reaches a broader audience by its placement in over 2000 academic libraries throughout the world. By its nature, the multiplicative effect of a library subscription allows access to a far larger group of potential readers. The podcasts have also clearly reached an audience that is larger and probably different from traditional SNIS members. Put differently, there are eight podcasts that have been individually downloaded by over 500 listeners, which is noteworthy in the context of the number of senior members. Some of the most downloaded podcasts are in fact those on socioeconomic topics, which again might indicate that these podcasts are allowing JNIS authors to reach a broader audience.
A podcast featuring JNIS Editor-in-Chief Robert W Tarr and BMJ Editor-in-Chief Fiona Godlee has been our most downloaded effort by a significant margin with over 5000 unique downloads (figure 1). The topic of the paper in question was peer review and studied many of the elements on this issue of burgeoning interest.15 The manuscript highlighted the JNIS editorial process and the podcast featured a unique ‘give and take’ between the respective Editors-in-Chief. The robust interest in this podcast suggests the potential of far larger audiences beyond our traditional neurointerventional topics. Indeed, the only data point to reach statistical significance was that non-classic neurointerventional topics—most commonly socioeconomic—were downloaded more commonly than podcasts on classic topics of clinical and/or research related to neurointervention.
The interest generated by podcast content can be analyzed in many ways. One potential method is to consider topics as being classic neurointerventional (both clinical and research) and other (figure 1). The percentage total downloads of the seven non-traditional podcasts is over 50%, although this is skewed by the Tarr/Godlee podcast on peer review. A different analysis would evaluate the period before and after initiation of advertising for social media content by the new social media editors. This occurred in February 2015 and was considered for the last five podcasts in this analysis. The active dialog on Facebook and Twitter, excluding the two most recent podcasts, appears to have increased podcast visibility and listenership. While it is important to note that the recent and most heavily downloaded podcast may bias these results, that podcast also represents the most heavily advertised podcast by the social media editors. While it is early, these data do suggest that advertising on the journal's social media platforms has been helpful in further expanding the audience for its podcasts. We believe that this statistical trend will become significant as familiarity with the social media platform grows.
We acknowledge that this retrospective study is subject to several limitations due in part to its small size. There are a limited number of podcasts in total and, as a result, in each subgroup analyzed. Additionally, besides the variable we analyzed with each group, there are other factors such as alternative advertising or time since podcast creation that would also impact the actual number of downloads. We did not control for those other factors, in part due to the small sample size. Over time, as we gain greater experience and a larger sample size of JNIS podcasts, it should become possible to control for additional variables. Finally, the analysis suffers from possible selection bias as the inclusion criteria for each category were chosen retrospectively and not randomly. Despite the limitations, we believe this pilot study does provide insights into the potential size of the opportunity for JNIS podcasts as well as the value of focused ‘advertising’ through social media.
Future directions
Podcasts did not meaningfully exist when many of the editors were training in neurointervention. The growth of these new media may even be more explosive than advancements in our underlying field. The last 5 years have seen significant advances in JNIS, including the growth of the podcast series. While today podcasts are conducted by participants dialing into a number based in the UK and having their discussion digitally recorded, technologic advances continue. BMJ and thus JNIS already feature the capability of having video uploads associated with articles. It is not difficult to imagine JNIS podcasts transitioning to include simultaneous slideshows (webinars). Beyond that, video chats are likewise within the realm of possibility. One can imagine allowing audience members to participate in question and answer sessions through our social media outlets, as has been trialed at the SNIS annual meeting. Moreover, while podcasts have typically focused on JNIS-specific content thus far, there is editorial discretion. As we think about our journal's future, podcasts in whatever form they exist will serve to bring our content to a growing cadre of diverse participant listeners.
Footnotes
Contributors JAH and RWT wrote the preliminary draft. All authors contributed to the final manuscript.
Competing interests None declared.
Provenance and peer review Not commissioned; internally peer reviewed.