We created online video content on breast cancer (Fig. 1). Content development was discussed by multiple breast medical oncologists and surgeons. First, the key messages and target audience were discussed among the BC Tube editorial board members. Then, the main writer and the second-in-command staff were selected. The BC Tube editorial board reviewed the draft. If the BC Tube editorial board members agreed that the content was well-produced, the content was reviewed by an external peer review group of independent breast medical oncologists and surgeons to ensure scientific validity. The narration text was also reviewed. After external peer review, the revised content was reviewed by all BC Tube editorial board members again. The content was converted to a video format with narration and animation. The narration text was inserted at the bottom of the video. Videos were no more than 10 min to promote sustained concentration while viewing the YouTube viewing screen. The video was reviewed by the BC Tube Support Group to ensure readability and ensure that no offensive expressions were included. The BC Tube Support Group is a general citizen group consisting of 110 volunteers who were selected through open recruitment, including women and men who have not been diagnosed with breast cancer, breast cancer patients, breast cancer survivors, and medical staff other than breast medical oncologists/surgeons.

Strategies for creating online video content by BC Tube

Finally, the videos were uploaded to the YouTube channel “Breast Cancer Encyclopedia [BC Tube Editors]” (https://www.youtube.com/@-BCTube), and the content was spread through social media, including the BC Tube Website, Twitter, Facebook, and Instagram. A link to a questionnaire form was placed in the summary section of the YouTube video to obtain opinions and feedback from viewers.

The strategy to select the videos included in the analysis is outlined in the CONSORT diagram (Fig. 2). Ninety-nine duplicated videos were excluded, and 107 videos without information about hereditary breast cancer were excluded, including content with general information about breast cancer treatment, introduction to statistical analysis using genetic breast cancer data, and content introducing cars with HBOC in the name. Nine videos in English without any Japanese information were excluded. Thus, 85 videos were eligible for this study.

Consort diagram on study-selection process

A majority of the videos (43 videos, 50.6%) were provided by public-interest organizations/companies, followed by hospitals/governments (18 videos, 21.2%), individual physicians (7 videos, 8.2%), breast cancer survivors (6 videos, 7.1%), and others (6 videos, 7.1%) (Fig. 3A, Table 1). Five videos about genetics and breast cancer produced by the BC Tube were included in the top 100 for “hereditary breast cancer (in hiragana)”; 4 videos were included in the top 100 for “hereditary breast cancer (in kanji),” and 2 videos were included in the top 100 for “HBOC.”

Characteristics of online video content by creator

The average length of the BC Tube (7.1 ± 2.3 min), individual physician (8.9 ± 3.0 min), and breast cancer survivor (9.0 ± 4.7 min) videos tended to be shorter compared with the length of the hospital/government (26.9 ± 43.3), public-interest organization/company (23.3 ± 27.0 min), and “other” (24.0 ± 11.7 min) videos. However, differences between groups were not significant (Fig. 3B, Table 1). The time since the upload of videos on YouTube was significantly longer for videos posted by public-interest organizations/companies compared with videos posted by hospitals/governments (p = 0.017) and breast cancer survivors (p < 0.01) (Fig. 3C, Table 1). The time since upload of videos from the “others” group was longer than the times for other groups (Fig. 3C, Table 1). These results suggest that public-interest organizations and companies focused earlier on providing information on hereditary breast cancer via YouTube than other sources. The same trend was evident in the “others” group, which included three recorded videos from public-interest organizations. Most videos created by hospitals and governments and breast cancer survivors were sent out around 2020, when health insurance coverage for BRCA1/2 genetic testing and HBOC treatment began in Japan.

We used YouTube Analytics to check the viewing status of BC Tube videos. The view counts per day decreased after the videos were posted. Therefore, we compared the total view counts instead of the number of views per unit of time. Public-interest organizations/companies exhibited the highest number of views (26,686.8 ± 156,763.4), followed by BC Tube (5614.8 ± 1690.2) (Fig. 3D, Table 1). Hospital/ governments (1115.7 ± 1307.7), breast cancer survivors (1343.7 ± 1572.3), and the “others” group (1213.0 ± 1530.2) showed similar view counts whereas individual physicians (540.5 ± 395.6) showed the lowest view counts although the differences were not significant (Fig. 3D, Table 1). Three videos from hospitals/governments and 10 videos from public-interest organizations/companies disabled the view counters on YouTube. Public-interest organizations/companies (65.0 ± 280.1) had the highest number of “likes,” followed by BC Tube (56.2 ± 6.9) and breast cancer survivors (34.2 ± 44.8) (Fig. 3E, Table 1).

All BC Tube videos turned off the comments to avoid personal consultations but employed a survey form in the summary section to receive feedback. Five BC Tube videos, 8 videos from hospitals/governments, and 15 videos from public-interest organizations/companies disabled comments. Excluding these videos, breast cancer survivors had the highest number of comments (5.7 ± 8.5) followed by public-interest organizations/companies (3.8 ± 17.0); however, the differences were not significant (Fig. 3F, Table 1). The viewers may find assessing the impartiality of content creators who display advertisements on YouTube difficult. To address this issue, the BC Tube does not show advertisements. Aside from the BC Tube, no significant differences in the number of videos with advertisements among the four groups were detected (Fig. 3G, Table 1).

To elucidate the quality of hereditary breast cancer information content, we applied two validated assessment instruments: PEMAT and DISCERN quality criteria. All 5 videos made by the BC Tube had perfect scores for PEMAT understandability scores; the PEMAT scores for BC Tube videos were significantly higher than the scores of the other groups (Fig. 4A). The videos from breast cancer survivors had significantly lower PEMAT understandability scores compared with the other groups (Fig. 4A). Videos from BC Tube, hospitals/governments, individual physician, public-interest organizations/companies and “others” had significantly higher PEMAT actionability scores compared with breast cancer survivors (p < 0.01 in all) (Fig. 4B). Videos made by the BC Tube had higher PEMAT actionability scores than public-interest organizations/companies (p < 0.05) (Fig. 4B).

PEMAT understandability score

Question 16 of the DISCERN score is an overall rating of the content quality. The BC Tube videos had higher overall quality ratings compared with all other groups (Fig. 4C). The breast cancer survivor videos had significantly lower DISCERN score compared with the other groups (Fig. 4C). Collectively, these results indicate that videos from BC Tube, hospitals, and governments were high quality, leading viewers to take action. Videos sent by individual contributors, such as breast cancer survivors, were generally of lower quality based on the qualitative scale (Tables 2 and 3).