When it comes to cancer prevention messaging, clear information from trusted organizations has greater reach on social media than personal accounts of patients, new University of California, Davis, research suggests.
Researchers looked at thousands of Twitter messages to identify the effects of the type of sender (individuals or organizations) and content type (basic information and facts or personal stories). They found that people shared informational tweets about cervical cancer prevention significantly more than personal-experience tweets. Furthermore, people shared information from organization senders, regardless of the content, rather than individual senders.
The findings were published in Preventive Medicine.
The paper’s lead author, Jingwen Zhang, assistant professor of communication at UC Davis, said the research shows that hospitals, public health organizations and other reputable entities may be able to use social media effectively for preventive care.
“Public health organizations may find social media an effective tool to raise awareness and deliver information,” she said. “If they make their messages simple and clear, people are more likely to share it.”
Patient uptake of cervical cancer prevention is low
Although early detection and treatment of cervical cancer and its precursors decrease cervical cancer mortality, people’s participation in prevention measures is low. Only about 83 percent of women receive preventive screenings, and a mere 43 percent of girls ages 13 to 17 receive the recommended number of vaccines for the human papillomavirus, or HPV, which causes cervical cancer, according to the research.
Women get the most cervical cancer prevention information through patient-doctor communication, but many women don’t have primary care physicians or a regular source of health care, researchers said. Public health campaigns traditionally use posters, websites and advertising, but with limited results. More than 4,000 deaths a year result from cervical cancer. About 13,000 new cervical cancer diagnoses are made each year, according to the researchers.
Social media, used by 88 percent of young adults and 78 percent of adults, might help spread prevention messaging, Zhang said.
To complete the research project, researchers first observed an archived Twitter dataset of almost 100,000 tweets mentioning such key words and phrases as “HPV vaccination,” “pap test” and “Gardasil,” which is the trade name for a common HPV vaccine. From that data, they obtained the most shared 3,000 tweets. Among those, 462 promoted cancer prevention and showed the sender type (individual or organization) and content type (anecdote or fact).
Researchers then created a controlled social media environment using an anonymous online discussion platform for U.S. women to discuss risks and prevention for five days in 2017. They gave the groups example tweets that consisted of both personal experience, such as women sharing on social media that they’d just had their first pap smear, as well as factual information, such as: “Most cervical cancers could be prevented by screening & HPV vaccination. Learn more… .”
The results showed that while a good anecdotal story can be shared many times, and in one case was the top tweet, most of the multiply tweeted messages contained factual information.
“Across these tweets, our consistent finding was that tweets were significantly more likely to be shared when they came from organization senders and contained factual information,” Zhang said.
“These findings suggest that practitioners can effectively design social media-based messages for cervical cancer prevention that significantly increase the reach of the messages to social media users,” Zhang concluded. “The findings reinforce the importance of public trust in organizations rather than individuals to share cancer prevention messages.
“The key strategy is to boost the credibility of the accounts and to develop messages that directly convey new factual information and resources.”
The research was supported by the National Cancer Institute (grant R01CA178875).