The Use of Social Networking Sites for Public Health Practice and Research: A Systematic Review

1Public health has achieved great advances in improving population health over the past century through education, communication, policy development, and risk management. Public health practitioners and researchers have been exploring new information technologies to more effectively and efficiently communicate with, engage, and educate the public. Today, social media offers a range of possibilities for establishing multidirectional communication and interaction, as well as quickly monitoring public sentiment and activity. These new tools have the potential to help public health meet many of its modern challenges and mandates regarding communicating with, educating, engaging, and monitoring a diverse public. We conducted a systematic review of the use of social networking sites (SNSs) in public health practice and research to better understand the use of these technologies for public health purposes.

2An expanding number of people use the Internet in their daily lives, including for accessing health information. Recently, the growth of interactive and dynamic Web applications has allowed the growth of SNSs, such as Facebook and Twitter. In 2011, 65% of Internet users in the United States reported using SNSs; 61% of Americans aged 18 to 30 years reported using an SNS every day and daily usage by Americans aged 50 to 64 years rose from 20% in 2010 to 32% the following year. Facebook is currently the most popular SNS in the world, topping 1 billion active users, with 580 million who engage with the site daily. Twitter, with 500 million users worldwide, has gained a reputation as a way to detect and predict events and sentiments by observing users’ posts (tweets) in real time. In addition to these very large SNSs, innumerable smaller SNSs exist serving a range of interests and needs, from LinkedIn for professional networking to PatientsLikeMe, where patients of similar diseases connect to share treatment resources and support. These sites allow users to engage with and shape their Internet content, from creating a network of connections with other users, to posting their own content, and to reacting and adding to content posted by other users. SNSs are a subset of social media. Social media is a broader concept that encompasses sites that allow users to generate and share content, such as blogs, wikis, and content communities, such as YouTube. SNSs are also characterized by user-generated content, but their defining characteristic is the ability to generate direct communication and 2-way interaction between users, thus generating networks of users.

3Although the increasing popularity of SNSs is clear, there is no single canonical definition for an SNS. A social networking tool or site is generally distinguished by the creation of individual public profiles and multidirectional communication and collaboration, allowing users to connect to one another within the site. Public health researchers and practitioners are interested in the use of SNSs because of the quick and inexpensive access to a broad or specific population that they afford and the possibilities for multidirectional communication that they offer.

4These qualities give SNSs the potential to expand and enhance core public health functions. For instance, research, surveillance, health education, and linking people with health resources are essential public health services. These tasks are traditionally resource-heavy, requiring in-person participant recruitment and sometimes resulting in lag times in detecting or notifying the public of health threats. Furthermore, the public health community often has difficulty reaching certain vulnerable populations who may have the greatest need for services. The large user population and immediate nature of SNSs have the potential to increase the reach and efficiency of these core public health services. Some traditionally hard-to-reach populations, such as adolescents, Hispanics, and low-income Americans, use SNSs at a rate higher than the general population providing a new opportunity for inexpensive public health communication to key demographics. The Centers for Disease Control and Prevention (CDC) provides tools for public health departments to use SNSs to extend the reach of their campaigns. Previous publications have examined the use of SNSs for sexual health promotion, dissemination of health information, and recruitment of difficult-to-reach populations, such as individuals who seek sex online. Thus, SNSs may provide new opportunities to effectively achieve the aims of public health.

5Given the increasing popularity of SNSs, and the range of possibilities that they offer for public health practice and research, we conducted a systematic review to assess the current uses of SNSs for public health practice and research. This review will serve to inform public health practitioners and informatics researchers of the state of knowledge in the field and identify gaps where more research is needed.

6We conducted a database search that included PubMed, EMBASE, and CINAHL Plus using a query consisting of an extensive list of names of specific SNSs, as well as the terms “social networking,” “social network site,” and “public health.” Given the constant evolution of SNSs—with new sites being added while others disappear—we utilized a list of 199 specific SNS names from the corresponding Wikipedia entry current up to the search date. We eliminated all SNS titles that did not generate results. This process yielded the following PubMed search query:

7(“social networking”[All Fields] OR “social network site”[All Fields] OR twitter[All Fields] OR facebook[All Fields] OR patientslikeme[All Fields] OR myspace[All Fields] OR renren[All Fields] OR kaixin[All Fields] OR whyville[All Fields]) AND (“public health”[All Fields] OR “public health”[MeSH terms]) NOT “behavior, animal”

8The query was modified to fit specific requirements of each of the databases searched. We included articles published up until March 31, 2012. We also conducted a manual search through the references of the articles retrieved through the electronic search.

9Two researchers reviewed every article to determine if inclusion criteria were met. Discrepancies were resolved through discussion. We included articles that met the following inclusion criteria:

10Published in peer-reviewed journals,

11Featured SNSs as the primary subject of study or a main component of the study methodology, and

12Focused on a topic of public health practice or research or a disease of public health importance. We used the CDC’s 10 Essential Public Health Services and the Information Access for the Public Health Workforce’s list of public health disease processes as criteria to determine public health relevance.

13We excluded narrative reviews and articles that did not constitute original research communications, such as letters to the editor, whitepapers, and comments.

14After selecting the articles for inclusion, authors extracted data using an online form specifically designed for this purpose. The extracted data included publication year, location (defined by the location on the study population), study design, sample size, purpose of the study, specific SNS involved (ie, Facebook, Twitter), how the SNS was used (ie, recruiting study participants, promoting public health messaging), target population, and public health topic. We also categorized each article according to whether the study described the use of SNSs for activities currently performed by public health practitioners (we labeled this practice) or whether it described novel uses related to public health (we labeled this research). Because some studies analyzed individual posts on an SNS as opposed to including individuals or SNS page profiles corresponding to a single person, we also captured the unit of analysis. Data was entered into a specially constructed spreadsheet generated for analysis.

15During this step, every article was randomly assigned to 2 researchers to assess intercoder reliability and discrepancies were resolved through discussion with other members of the team.

16We ran our search in July 2012. The literature review retrieved 429 individual articles that were screened for eligibility. A total of 20 duplicates were excluded. After reviewing title and abstract, 318 articles did not meet our inclusion criteria. We reviewed 91 articles in full. Another 18 articles were removed after reviewing the full text because they did not meet our inclusion criteria. Therefore, our final review included 73 unique articles. See Figure 1 for the detailed description of the search process. Table 1 includes a complete description of the included articles.

17Despite the emergence of Internet SNSs in the late 1990s, the topic did not become prominent in the public health literature until the late 2000s. Since then, the number of publications describing the use of SNSs for public health research and practice has been steadily increasing. In 2007, there was only 1 publication, increasing to 9 in 2009, 12 in 2010, 31 in 2011, and 19 in the first 3 months of 2012. Regarding the location of the study, 43 of the 73 articles (60%) were conducted in the United States. Using the World Bank classification of countries, 14 were conducted in other high-income countries and 1 was conducted in a middle-high income country (South Africa). Only 1 study was conducted exclusively in a low- or low-middle income country (Haiti), but 13 articles explicitly included SNSs and their users without any reference to a particular region or country.

18Facebook was the most commonly used SNS (27%). In all, 10 of 73 studies (14%) used MySpace, 6 (8%) used Twitter, and 12 (17%) used other SNS, including Bebo, Friendster, LinkedIn, and PatientsLikeMe. There were 25 (34%) that studied multiple SNSs or did not limit by any specific site (Figure 2). Figure 2 shows the changes over time, with MySpace being most frequently used in the early years, and the later appearance of Facebook, Twitter, and other SNSs.

19Most studies (63/73, 86%) were cross-sectional observational studies providing descriptions of SNS usage or SNS users. Examples of cross-sectional studies include a study by Jenssen et al analyzing adolescents’ exposure to tobacco advertisements on SNS and a study by Kontos et al in which the authors studied inequalities in SNS usage and its implications for public health communications. Only 5 studies used an experimental design to test a specific intervention, and all 4 were randomized controlled trials. For instance, Centola randomized participants to different user groups on an SNS to study the effect of homophily (similar social contacts) on health behavior change. We found 1 systematic review of non-randomized controlled trials published by Gold et al, in which the authors explored the use of SNSs for sexual health promotion.

20Articles describing the use of SNSs for public health practice (n=15) focused primarily on hard-to-reach populations. Five articles used SNSs to reach youth and adolescents, or individuals at risk for sexually transmitted diseases (STDs) and human immunodeficiency virus (HIV), to connect them with available services. In an article by Feldacker et al, the authors describe their positive experience using SNSs to contact students at high risk for HIV and offer them HIV testing. Another significant use of SNSs for public health practice was to promote healthy behaviors. For example, Pemu et al designed a diabetes management and education SNS and evaluated it with diabetes patients. Other activities routinely conducted by local health departments for which SNSs are being used included disease surveillance and communications during natural disasters. See Table 2 for the frequency of SNSs in public health research publications.

21With respect to target population, 32 of 73 studies (44%) focused on young users, such as teenagers, adolescents, or college students. The second largest group of studies targeted the general population without specific restrictions (15/73, 20%). In all, 6 of 73 studies (8%) focused on individuals with specific diseases. Articles focused on a wide array of public health issues; 21 studies (29%) focused on sexual and reproductive health, 17 (23%) on general health promotion strategies, 14 (19%) on noncommunicable diseases, 13 (18%) on alcohol, tobacco, and substance abuse, and 8 (11%) on mental health (see Table 3).

22With over 1 billion active users worldwide on Facebook alone, SNSs have become a place where individuals express themselves and interact with one another. Multidirectional communication tools—a hallmark of these sites—allow for rapid and collaborative dissemination of information. This scenario opens up great opportunities for assessing a population’s risk factors and monitoring their health. Thus, SNSs represent a new avenue of action for public health researchers and practitioners. This systematic review provides an overview of the literature concerning the current use of SNSs in public health.

23Our results suggest that the application of SNSs to public health research and practice is still maturing. Most articles in this review describe relatively passive approaches to SNSs, rather than harnessing the full potential of SNSs in terms of multidirectional communication and networking. For example, most studies that used SNSs for recruiting study participants simply posted a recruitment ad or survey on the SNS, much as they could on a regular website. In contrast, 2 studies utilized SNSs not only to recruit study participants, but also to maintain contact with participants; thus, improving long-term participation. These findings are consistent with the gaps identified in a recently published systematic review. Although different in scope—that review explored the use of social media for health communication—the authors also identified a lack of studies evaluating the impact of such novel technologies. However, examples of more complex uses of SNSs are starting to emerge. Norman and Yip utilized a wide range of SNS capabilities to first recruit young adults to create their own SNS content and then use this content to recruit and communicate with additional young adults around a host of health promotion issues.

24Because the popularity of SNSs is a recent phenomenon, we are confident that these formative studies are the required foundation that will provide key knowledge to inform more interactive and experimental studies in the future. The formative study design of these initial SNS studies is suggestive of a developing domain. Most studies utilized a cross-sectional observational design; only 4 were experimental studies and 1 was a systematic review. This preponderance of nonexperimental designs is consistent with what is seen in other nascent domains, such as in the use of information technologies to help treat noncommunicable conditions. It is likely that in coming years we will see an increase in the number of experimental studies using SNSs. For example, one of the studies included in this systematic review was recently followed by a randomized controlled trial that tested the effectiveness of Facebook messages to decrease risky sexual behaviors. In the field of public health research and practice, SNSs may be an attractive low-cost tool to explore and test public health interventions with a large number of diverse participants—interventions that historically have been difficult to conduct using more traditional methods.

25One critical—and maybe limiting—issue to be considered when designing randomized trials involving SNSs is their rapid change to potential obsolescence. We were able to show an important number of studies were conducted using MySpace initially, a situation that changed after 2010 in parallel with a decline in the number of MySpace users. These rapid changes in SNS utilization might not be entirely compatible with the time it takes to obtain funding for a randomized trial.

26In addition to the types of studies, our findings suggest that although SNSs offer an opportunity to reach a wide range of the population at lower costs when compared to traditional communication methods, most published studies have been conducted in high-income countries—predominantly in the United States. Only one study was conducted in South Africa, a higher middle-income country. The cause of this phenomenon is unclear. With the accelerated adoption of mobile broadband and smartphones in both developed and developing countries, we will probably begin to see a broader use of SNSs for public health research and practice in low- and middle-income countries in the future.

27SNSs provide unique research and practice benefits beyond their low cost, high-reach, multidirectional communication. SNSs appear to be especially well-suited to research and practice on “taboo” public health topics, with nearly half (44%) of the studies included in our review addressing sexual and reproductive health and/or alcohol, tobacco, or substance use. Multiple aspects of SNSs make them suitable for such topics: SNSs afford a high level of anonymity; people display stigmatized behavior more freely online (visiting SNSs for sex, posting about their depression, etc) and youth and young adults, a population largely affected by these issues, has the highest rate of SNS usage of any age group. Not only do SNSs allow easier exploration of traditionally hard-to-discuss topics, they can also facilitate identification of and outreach to certain traditionally hard-to-reach populations, such as men who have sex with men (MSM) with risky sexual behavior, homeless youth, and young people suffering from mental illness. Finally, SNSs allow for the exploration of public health issues associated with the increasing use of SNSs, such as promotions for drugs and tobacco on SNSs, online harassment and sexual solicitation of youth, and depression associated with SNS use.

28The main limitations of this systematic review arise from our search strategy, which included articles only published in scholarly journals, thus excluding communications published in the grey literature. In addition, some articles might have been missed because of limitations of our search terms. We tried to minimize this by including a broad list of SNS keywords in our list of terms.

29An increasing number of studies involving the use of SNSs within the domain of public health have been published and the frequency and complexity of such studies parallels the growing popularity of SNSs in general. Most of the published studies are descriptive and take a limited approach to using SNSs. When the multiple possibilities that emerge from low cost are considered (eg, high visibility, multidirectional communications), it is obvious that there is a gap in range of possible study areas in public health and SNSs. As the field matures and our knowledge concerning the most effective ways to use these technologies to support public health research and practice matures, we expect to see additional innovative uses of SNSs to engage diverse and broad populations with the goal of better understanding and improving health.