This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Infodemiology, is properly cited. The complete bibliographic information, a link to the original publication on https://infodemiology.jmir.org/, as well as this copyright and license information must be included.
Skin cancer is among the deadliest forms of cancer in the United States. The American Cancer Society reported that 3 million skin cancer cases could be avoided every year if individuals are more aware of the risk factors related to sun exposure and prevention. Social media platforms may serve as potential intervention modalities that can be used to raise public awareness of several diseases and health conditions, including skin cancer. Social media platforms are efficient, cost-effective tools for health-related content that can reach a broad number of individuals who are already using these spaces in their day-to-day personal lives. Instagram was launched in 2010, and it is now used by 1 billion users, of which 90% are under the age of 35 years. Despite previous research highlighting the potential of image-based platforms in skin cancer prevention and leveraging Instagram’s popularity among the priority population to raise awareness, there is still a lack of studies describing skin cancer–related content on Instagram.
This study aims to describe skin cancer–related content on Instagram, including the type of account; the characteristics of the content, such as the kind of media used; and the type of skin cancer discussed. This study also seeks to reveal content themes in terms of skin cancer risks, treatment, and prevention.
Through CrowdTangle, a Facebook-owned tool, we retrieved content from publicly available accounts on Instagram for the 30 days preceding May 14, 2021. Out of 2932 posts, we randomly selected 1000 posts for review. Of the 1000 posts, 592 (59.2%) met the following inclusion criteria: (1) content was focused on
Of the 592 posts, profiles representing organizations (n=321, 54.2%) were slightly more common than individual accounts (n=256, 43.2%). The type of media included in the posts varied, with posts containing photos occurring more frequently (n=315, 53.2%) than posts containing infographics (n=233, 39.4%) or videos (n=85, 14.4%). Melanoma was the most mentioned type of skin cancer (n=252, 42.6%). Prevention methods (n=404, 68.2%) were discussed in Instagram posts more often than risk factors (n=271, 45.8%). Only 81 out of 592 (13.7%) posts provided a citation.
This study’s findings highlight the potential role of Instagram as a platform for improving awareness of skin cancer risks and the benefits of prevention practices. We believe that social media is the most promising venue for researchers and dermatologists to dedicate their efforts and presence that can widely reach the public to educate about skin cancer and empower prevention.
Skin cancer, in general, and melanoma, specifically, are among the deadliest forms of cancer in the United States [
Social media platforms may serve as potential intervention modalities that can be used to raise public awareness of several diseases and health conditions, including skin cancer. Social media platforms are efficient, cost-effective tools for health-related content that can reach a broad number of individuals who are already using these spaces in their day-to-day personal lives [
Instagram was launched in 2010, and it is now used by 1 billion users, of which 90% are under the age of 35 years [
Social media is becoming a potential intervention modality to raise skin cancer awareness, especially via image-based platforms [
Accordingly, the following research questions (RQs) were proposed:
RQ1: What are the source and content characteristics of Instagram posts related to skin cancer?
RQ2: To what extent are different types of skin cancer covered on Instagram?
RQ3: How do messages frame causes and solutions regarding skin cancer causes, treatments, and prevention?
RQ4: To what extent do Instagram posts on skin cancer address the susceptibility and severity of skin cancer; benefits and barriers associated with diagnosis, prevention, and treatment; call to action; and readers’ self-efficacy?
We used CrowdTangle [
The initial search for our main sample yielded 2982 posts. We reordered the full set of posts descending from the highest number of “total interactions,” defined by CrowdTangle as an indicator of engagement—total reactions, comments, and shares combined. We selected the top 1000 posts in terms of total interactions, and then used Research Randomizer [
The following inclusion criteria for post content were assessed by the coders: (1) directly related to human skin cancer, (2) in the English language only, and (3) originating from the United States. Content mentioning skin cancer in animals was excluded. Of the sample of 1000 posts, 408 (40.8%) did not meet the inclusion criteria, resulting in an analytic sample of 592 (59.2%) Instagram posts.
Although there has been increased interest in applying machine learning methods to the analysis of social media data, it has also been suggested that these techniques may present challenges when applied to qualitative coding in social science research [
Coding was conducted independently by 2 trained undergraduate students, in multiple stages, to assess the reliability of the codebook constructs. As a precursor to coding the final sample, the coders underwent a pilot stage, in which posts from the time period of February 24 to March 25, 2021, were downloaded. A total of 1173 posts were downloaded and sorted in descending order by “total interactions.” The first 60 posts were used for coder training purposes and to provide additional insight on the reasons for exclusion/inclusion. Once this initial training took place, 35 posts were reviewed and coded by the 2 raters, of which 16 posts met our inclusion criteria. The 2 raters and a moderator met and held a discussion for 1-2 hours every 2 weeks to identify problematic variables and reach perfect agreement [
The sample for the main study was downloaded from CrowdTangle, including posts from the 30 days preceding May 14, 2021, as described above. The 2 undergraduate raters independently coded 250 posts each. Each rater double-coded 10% (n=25) of posts from the other rater’s sample of 250 posts. Of the 50 posts double-coded by the raters, 45 met the inclusion criteria. Interrater reliability was assessed using Cohen κ statistics to determine the level of consistency in codes between each rater. Of the 66 codebook variables for which a Cohen κ could be calculated, 82% (n=54) of the variables resulted in a moderate or higher agreement. The median Cohen κ was 0.78, reflecting substantial agreement between the raters on identifying codebook constructs [
Institution Review Board approval was not required as CrowdTangle only imports data from public accounts on Instagram.
Of the 592 posts that met the inclusion criteria, the sources of content originated from 2 different types of profiles. Profiles representing organizations (n=321, 54.2%) were slightly more common than individual accounts (n=256, 43.2%). For posts that could not be clearly classified as being derived from either an individual or organization, an “other” category (n=15, 2.5%) was selected.
As displayed in
For posts that could not be clearly classified as being derived from either an individual or organization, an “other” category (15/592, 2.5%) was selected, and these posts were not included in
Source characteristics of skin cancer–related posts on Instagram (N=577).
Source characteristica | Post, n (%) | ||
|
|||
|
Influencer or public figure | 152 (59.4) | |
|
Physician | 107 (41.8) | |
|
Dermatologist | 83 (32.4) | |
|
Parent | 51 (19.9) | |
|
Business owner | 37 (14.4) | |
|
Journalist | 14 (5.5) | |
|
Esthetician | 11 (4.3) | |
|
Nurse or other health worker | 4 (1.6) | |
|
Health educator | 2 (0.8) | |
|
|||
|
Business | 222 (69.2) | |
|
Health information provider | 53 (16.5) | |
|
Nonprofit | 39 (12.1) | |
|
Health care organization | 34 (10.6) | |
|
News organization | 8 (2.5) | |
|
Government entity | 5 (1.6) | |
|
School | 1 (0.3) |
aSource characteristics for each broader type of profile (individual or organization) exceed 100% when added together because multiple categories may have been selected for a particular post (eg, a business and a health care organization).
More than half (318/592, 53.7%) of the posts analyzed mentioned skin cancer generally but did not specify the type (
Posts mentioning types of skin cancer on Instagram (N=592).
Type of skin cancer | Post, n (%) |
Skin cancer mentioned generally, but type was not specified | 318 (53.7) |
Melanoma | 252 (42.6) |
Basal cell carcinoma | 29 (4.9) |
Squamous cell carcinoma | 29 (4.9) |
Just under half of all posts (271/592, 45.8%) included information regarding some kind of skin cancer risk factor. The “sun” was coded as the top named risk factor for skin cancer (227/271, 83.8%), followed by artificial tanning (eg, indoor tanning; 48/271, 17.7%) and genetics (15/271, 5.5%).
Information regarding prevention methods (404/592, 68.2%) was included in Instagram posts more often than risk factors (271/592, 45.8%). Within prevention methods, wearing sunscreen (280/404, 69.3%) was the most commonly mentioned method, followed by getting checked by a physician (131/404, 32.4%) and wearing protective gear/clothes (101/404, 25%;
Posts discussing prevention methods of skin cancer (N=404).
Prevention method | Post, n (%) |
Wearing sunscreen | 280 (69.3) |
Getting checked by a physician | 131 (32.4) |
Wearing protective gear/clothes | 101 (25) |
Self-examination | 50 (12.4) |
Staying away from the sun | 41 (10.1) |
Mentioning warning signs | 37 (9.2) |
Not using tanning beds | 21 (5.2) |
Using self-tanning products | 19 (4.7) |
Types of content portrayed in Instagram posts.
Content typea | Post, n (%) |
Information urging readers to adopt a certain behavior (N=592) | 209 (35.3) |
Benefits of skin cancer prevention (n=402) | 120 (29.9) |
Prevalence of skin cancer (N=592) | 96 (16.2) |
Seriousness of skin cancer (N=592) | 75 (12.7) |
Diagnostic method (N=592) | 29 (4.9) |
Benefit of diagnostic method (n=29) | 4 (13.8) |
aDue to skip logic in the extraction survey, some posts were not rated for certain content types.
Top 10 hashtags appearing in the study sample posts (N=592).
Hashtag | Number of mentions, n (%) |
#skincancerawareness | 165 (27.9) |
#skincancer | 134 (22.6) |
#skincare | 98 (16.6) |
#melanoma | 89 (15) |
#sunscreen | 88 (14.9) |
#skincancerawarenessmonth | 87 (14.7) |
#melanomaawareness | 78 (13.2) |
#melanomamonday | 76 (12.8) |
#spf | 67 (11.3) |
#dermatology | 65 (11) |
Types of sources cited in posts (N=81).
Citation source | Post, n (%) |
Cancer organization | 34 (42) |
Health or web source (eg, WebMD) | 21 (26) |
Physician | 14 (17) |
Research community | 11 (14) |
Centers for Disease Control and Prevention or federal organizations | 5 (6) |
Celebrity | 3 (4) |
World Health Organization | 2 (2) |
This study aimed to describe the content landscape of skin cancer on Instagram, specifically focusing on the content source and type of information posted by users. The study also reveals the themes of the posts in terms of skin cancer causes, diagnosis, and prevention methods. To the best of our knowledge, this is the first study to identify the sources of skin cancer content on Instagram.
Overall, slightly more content originating from organizational accounts was posted than content from accounts owned by individuals. Business-owned accounts (eg, Skin Store, Baby Bum) tended to post about skin cancer more than those of a medical background, such as health organizations or health information providers. Among individual account owners, influencers (not from a medical background) posted skin cancer–related content more than individuals who possessed medical expertise (eg, dermatologists, other physicians). The term
Although sunscreen promotion was the most commonly identified prevention method represented in Instagram posts (that discussed skin cancer prevention methods), such posts also contained other prevention methods included in prevention guidelines [
Given that the majority of the posts originated from nonmedical accounts, it is concerning that only 13% of the posts in this sample cited their information from a credible source, such as a cancer organization. A behavioral intent study found that, of its participants, 91% said that online communities (such as Instagram) play a role in their health decisions [
In contrast to skin cancer content on Pinterest [
As with other research, this study is not devoid of limitations. First, we were only able to collect information from publicly available Instagram accounts. Second, our inclusion criteria may have limited the generalizability of our results. For example, we only included content in English. We also excluded profiles that mentioned they were originating from a non–US-based location. A final potential drawback of this work may be limiting the study to only a single time period within the year. For example, the sample of Instagram posts was extracted during the month of May, which was Skin Cancer Awareness Month. Posts in Winter months may look different than those in spring and summer months, and we suggest that future research consider such possible issues of seasonality.
Despite these limitations, this study has several strengths and implications. The study findings illuminated a paucity of medically credible sources related to skin cancer on Instagram, at least in our sample. For example, a relatively high percentage of the sources of skin cancer content in our sample were of a nonmedical nature. In addition, almost 90% of the skin cancer information in this sample was not backed by any credible citations, which is concerning. As linking health behavior to misinformation is difficult to observe, further research is needed to confirm that the spread of misinformation among users could lead to poor decision-making as it relates to skin cancer prevention, as well as public confusion [
Although Instagram is the most viewed social media platform regarding skin cancer [
As a possible intervention strategy, given that influencers were prevalent in our sample, medical experts could partner with celebrities and influencers to lead awareness campaigns on skin cancer. In considering the overall presence of physicians on social media, dermatologists are among the top [
An example of a nonmedical celebrity who has raised awareness about skin cancer is Australian actor Hugh Jackman, who was diagnosed 6 times with basal cell carcinoma on his nose, which required a surgical treatment. Jackman took advantage of his popularity and used his social media platforms to advise followers regarding the risks of exposure to the sun by openly sharing his experience and medical process. This led to increased public awareness, verified by a spike in online searches for “Basal Cell Carcinoma” at the time of his skin cancer–related post [
In summary, this study’s findings highlight the potential role of Instagram as a platform for improving awareness of skin cancer risks and the benefits of prevention practices. As skin cancer remains one of the most common cancers in the United States [
Codebook.
research question
ultraviolet
The authors wish to acknowledge Ryli Hockensmith for her assistance in coding a portion of the Instagram posts reported in this manuscript, John Ferrand for his guidance in determining the Cohen κ statistics, and the Indiana University Observatory on Social Media for providing access to CrowdTangle.
None declared.