Skin cancer affects an estimated 5 million people annually in the United States, with a treatment cost of over $8.1 billion [1]. Ultraviolet (UV) radiation is a known carcinogen that causes direct and indirect DNA damage; intentional or unintentional UV exposure is responsible for the majority of skin cancer incidence [2]. Sunscreen is a primary sun protection strategy that is used by approximately one third of US adults [3]; however, sunscreen application rarely reflects the standards necessary to meet the advertised protection levels [4,5]. Furthermore, sunscreen is intended for use in combination with other sun protection strategies [6]. In 2012, the US Food and Drug Administration (FDA) issued new regulations for sunscreen labeling [7]; however, research has demonstrated that sunscreen application rarely reflects the thickness, evenness, and timing (of both initial application and reapplication) necessary to meet the advertised protection level of the sunscreen [8]. In 2014, the Surgeon General’s Call to Action to Prevent Skin Cancer included a renewed call to promote a combination sun protection strategy (eg, wearing sunscreen, a hat, and protective clothing) [5].

The Centers for Disease Control and Prevention (CDC) and the American Academy of Dermatology support the daily use of sunscreen with a sun protection factor (SPF) of at least 15 and 30, respectively, for UV protection, and dermatologists report recommending the use of sunscreen to their patients [9]. Although health care providers are trusted sources of health information, individuals increasingly supplement these sources with information garnered from the internet and social media [10]. In 2012, 72% of US adults had sought health information on the internet within the past 12 months [10]. The growing influence of social media in health decision making is not equally distributed across demographic and age groups; young age is a consistent predictor of social media use [11]; however, social media usage among adults is steadily increasing [12,13]. YouTube, the most prominent user-generated video-sharing website, allows any person to upload content on any topic [10]; based on traffic, it was the top-ranked website on the internet as of 2019 [14]. Furthermore, YouTube is increasingly being used as a platform to disseminate health information [15]. The American Cancer Society, National Institutes of Health, and National Cancer Institute have adopted this outlet to communicate health messages to broad audiences.

Despite these efforts, misleading or incorrect health information is common on YouTube [15]. For example, over one half of videos on the topic of immunization feature negative views or contradict the reference standard, which may lead viewers to make unsafe health decisions [16]. With respect to skin cancer, Basch et al [17] found YouTube content promoting untested home remedies for skin cancer, such as black salve; this finding highlights the challenge of health communication on social media. Moreover, a study of sun safety topics on YouTube [18] found that 17% of videos retrieved using the search term “sunscreen” included false or misleading information; however, less than a quarter of the videos in their sample were English language videos. Sunscreen use in practice is less thorough than recommended, and the question remains whether this is due in part to incomplete or misleading information available on social media platforms such as YouTube; this is especially likely to be the case if the content is not derived from governmental or professional medical organizations who create sun safety recommendations. Here, we analyzed a large sample of English language YouTube videos about sunscreen to address the following research questions: What primary sources, content types, positive/negative product attributions (valence), and adherence to standard recommendations [7] are included in YouTube videos about sunscreen? Do these videos contain the terms “sunblock” and “waterproof,” which the FDA prohibits on sunscreen packaging? What complementary sun protection strategies do these videos include?

These data may inform public health video messaging to improve sun safety practices for effective sunburn prevention.

We conducted direct YouTube searches between November 13 and November 15, 2017 (search terms: sunscreen cancer, sunscreen health, sunscreen information, sunscreen ingredients, and sunscreen natural), deleting cookies after each search to avoid influencing subsequent results [19]. The YouTube region was set to United States. We identified the 20 most frequently appearing videos for each search term and removed duplicates to obtain the analytic sample of videos (see Figure 1).

We developed an early draft codebook in consultation with sun safety and communication experts to reflect common metrics in YouTube-based health information studies and sunscreen recommendation reference standards. This draft was then expanded and refined to include the recommendation of nonsunscreen sun safety methods, use of the outdated terms “sunblock” and “waterproof,” and possible themes related to sunscreen safety. The codebook was pilot-tested using a test set of 5 videos by a team that included 3 experts in sun safety and 1 postbaccalaureate fellow. These videos were chosen using an extreme cases approach and reflected a diverse range of content and positions toward sunscreen. The codebook development team met to discuss and resolve coding discrepancies, generate additional constructs and codes, and refine coding response options.

After the initial codebook refinement, the coding team (MG, SS, JW, and AJ) double-coded all videos for the presence of 6 standard recommendations for effective sunscreen use based on FDA and CDC recommendations: SPF level, broad spectrum classification, application timing (20 minutes before exposure), liberal/generous application, and reapplication (every 2 hours and separately after toweling). We coded the overall valence toward sunscreen as negative or positive. We coded the use of the terms “sunblock” and “waterproof”; since 2011, the FDA has prohibited these terms on sunscreen labels, as they may be misleading [7]. We further coded the video source based on the posting source of the video and the posting channel’s thumbnail and description. Lastly, we coded recommendation of the following complementary sun protection strategies (operationalized as instruction to adopt): seeking shade, wearing a hat, wearing protective clothing, wearing sunglasses, and staying indoors during peak UV hours (Cohen kappa range 0.80-0.96). Viewer attention to social media content is short [20]; therefore, we stopped coding each video after 5 minutes. Any final disagreement was resolved through discussion. We used Stata/SE 14 [21] to calculate descriptive statistics (means, medians and standard deviations, ranges and percentages) overall and stratified by video source.

The videos had view counts of 1100 to 671,142 (mean 58,642, median 17,774, SD 109,651), were between 40 seconds and 41 minutes long (mean 5:54 minutes, median 4:19 minutes), and were uploaded between 2009 and 2017 (average post duration 1127 days, median 1101, range 64-3135, SD 784). Average daily views ranged from 1 to 1448 (mean 96.8, median 23, SD 234). Primary video sources were health care providers or their associations (24/111, 21.6%), end users (51/111, 45.9%), and corporate entities (15/111, 13.5%) (see Table 1 for the descriptive characteristics of the videos). No government health agencies were among the sources of videos retrieved. Valence toward sunscreen use was generally positive (82/111, 73.9%); however, positive content included promotion of mineral sunscreens that warned about health risks of chemical sunscreens. Videos that discouraged sunscreen use (29/111, 26.2%) included misleading content, such as overstatement of the benefits of vitamin D, and outright falsehoods, such as the claim that sunscreen causes cancer.

Of the 6 recommendations for effective sunscreen use, the most frequently mentioned were SPF (35/111, 31.5%) and broad-spectrum classification (30/111, 27.0%); meanwhile, application timing, generous application, reapplication every 2 hours, and reapplication after toweling were each mentioned in fewer than 20% of videos (see Table 2). Most videos included 0, 1, or 2 recommendations. SPF was mentioned in 5/9 (55%) of news and television coverage videos. Corporate entities, such as sunscreen or skin care companies, most frequently recommended generous application (4/14 videos, 29%); however, 6/14 (43%) of these videos paired this message with a poor demonstration of sunscreen application.

Few videos advocated the use of complementary sun protection strategies in conjunction with sunscreen (19/111, 17.1%). Of videos with positive valence toward sunscreen, 15/82 (18.3%) suggested complementary sun protection strategies, compared to 3/29 (10.3%) of videos with negative valence (see Table 2).

Few videos cited scientific sources for the presented information (30/111, 27.0%). Sources of scientific information were the Environmental Working Group (EWG), an environmental advocacy group (13/111, 11.7%), the FDA (7/111, 6.3%), and peer-reviewed articles (5/111 videos, 4.5%). Lastly, an unexpected finding was that 19/111 (17.1%) of videos demonstrated do-it-yourself recipes for how to make sunscreen at home.