Interventions Using Social Media for Cancer Prevention and Management

A Systematic Review

Claire Jungyoun Han, PhD, MSN, RN, Young Ji Lee, PhD, MS, RN, George Demiris, PhD, FACMI

Cancer Nursing

Background: Regarding cancer awareness, social media effectively promotes health and supports self-management. Given the diverse study designs, methodologies, and approaches of social media interventions in oncology, it is difficult to determine the effects of social media on cancer prevention and management.

Objective: We aim to systematically review intervention studies using social media for cancer care. Methods: A systematic search, using 7 electronic databases (PubMed, Web of Science, CINAHL, Cochrane Library, Scopus, EMBASE, and PsycINFO), was conducted to identify surveys and interventions using contemporary social media tools with a focus on cancer.

Results:Of the 18 selected studies, 7 were randomized controlled trials. Most studies were conducted for all types of cancer, and some were conducted for breast cancer in the United States, with mostly white female participants. Facebook was the most frequently used platform. Most studies targeted healthy participants providing cancer prevention education. With social media platforms as part of a larger intervention, or the main component of interventions, interventions were overall feasible and showed a significant improvement in cancer prevention and management.

Conclusions:Social media tools have the potential to be effective in delivering interventions for cancer prevention and management. However, there was a dearth of studies with rigorous study methodologies to test social media effects on various cancer-related clinical outcomes.

Implications for Practice: Social media use in cancer care will facilitate improved communication and support among patients, caregivers, and clinicians and, ultimately, improved patient care. Clinicians need to carefully harness social media to enhance patient care and clinical outcomes.

Cancer is a major public health problem worldwide and is the second leading cause of death in the United States.¹ Despite great progress in cancer prevention and management, nearly 15.5 million Americans are cancer survivors, approximately 1.6 million new cancer cases are expected to be diagnosed, and 60 million Americans are expected to die of cancer in 2017 in the United States.¹Clinicians and researchers are exploring the use of technology and other innovative methods to facilitate patient-centered interventions to improve cancer prevention and management. New strategies for delivering high-quality and high-value cancer prevention and management are suggested, and one of the proposed innovative technological approaches is social media.²

Social media are defined, according to Kaplan and Haenlein,³ as media using tools often seen as second generation or Web 2.0 software or Web site functionalities and, most importantly, as a platform for individuals to share perspectives, contents, insights, experiences, opinions, and other types of information. Various social media platforms include collaborative projects (eg, Wikipedia), blogs (eg, Blogger) or microblogs (eg, Twitter), content communities (eg, YouTube), and social network sites (eg, Facebook).³ Social media are different from traditional Web-based interventions. The Web-based interventions such as online forums and bulletin boards focus on discussion and conversation, and anonymity is common with users participating passively.⁴ In contrast, current contemporary (ie, broadly accessible) social media showcase more user-friendly designs and interfaces, easy to access, fast to communicate, and focus on individuals and active participations from the users.⁴ In addition, virtual worlds and social networking are associated with each other; yet, in the social media, the user need not create an avatar and simulation environment to interact with others, and most interaction takes place asynchronously with a time delay.⁵

There have been substantial growths in the uptake of social media by national healthcare organizations (eg, National Institutes of Health, Centers for Disease Control and Prevention [CDC]) and professionals and their use to network, share knowledge, promote, and engage in research.⁶ Social media enable connections and interactions between patients, caregivers, providers, and other stakeholders. Such a network can offer support, guidance, access to health information, education, and knowledge to people interested in learning more about preventing or managing cancer.⁷ Social media in oncology can also be an avenue to address and engage the public in relation to cancer-related research such as crowdsourcing.⁷ Because of its easiness of recruiting and engaging patients and low maintenance cost, social media interventions can be effective strategies to improve self-management and quality of life in patients with cancer and their caregivers.⁶

Despite the growing use of social media, the understanding of the use of social media in oncology is lacking and limited to guiding users (eg, patients, caregivers, and clinicians) on how best to use this tool for their interests.8 Recent reviews have addressed general information about health information technology in oncology.^2,8-10 However, there is a lack of intervention studies specifically exploring social media use for cancer prevention and management. Slev et al² reviewed a Web-based application used in cancer but did not specifically focus on social media. Patel et al⁸ reviewed social media use in chronic diseases including cancer. However, details of cancer-specific information such as types of cancer and cancer-related symptoms were not specifically addressed. McAlpine et al⁹ and Koskan et al¹⁰ reviewed online interventions for patients with cancer. However, none of the interventions used social media platforms,⁹ or the current literature on interventions using social media in cancer was not fully included.¹⁰

In addition, existing studies using descriptive or content analyses about social media contents are still too limited to provide reliable and useful information to improve cancer prevention and management.^2,8-10 Most of the social media lack documentation for the evidence of their online content and are missing standardized development protocols that can safeguard content quality, which in turn generates ambiguity and uncertainty among users.¹¹ Given the lack of a significant body of intervention studies using social media and diverse study methodologies (eg, designs, aims, outcome measures), it is difficult to determine the effects of social media or which social media interventions are efficacious in cancer prevention and management.⁷

The purpose of this systematic review is to comprehensively understand the characteristics of social media interventions in cancer prevention and management, to enhance the evidence base for fully harnessing the social media in oncology. Our objectives were (1) to provide an overview and synthesis of the extant literature on the interventions using social media tools for cancer prevention and management and (2) to identify gaps in current research and suggest further recommendations on the use of social media for improving cancer prevention and management. The research questions include the following: (1) ‘‘What are the characteristics specific to social media used as interventions in cancer prevention and management?’’ (2) ‘‘What are the components of the social media interventions?’’ and (3) ‘‘What are the outcomes assessed and the effects of social media use on interventions in cancer prevention and management?’’

Methods

Search Strategies

We conducted a systematic literature review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram¹² (Figure). We conducted systematic literature searches of 7 electronic databases: MEDLINE via PubMed, Web of Science, CINAHL, Cochrane Library (Review and CENTRAL), Scopus, EMBASE, and PsycINFO. We focused on contemporary social media for our review and searched for contemporary social media platforms based on the recent literature defining social media^3,7,8 and global social media Web traffic rankings.¹³ Key words using MeSH terms and manual searches were ‘‘cancer,’’ ‘‘prevention,’’ ‘‘management,’’ and ‘‘oncology,’’ in combination with ‘‘intervention,’’ ‘‘social media,’’ ‘‘Wikipedia or Wikis,’’ ‘‘Facebook,’’ ‘‘YouTube,’’ ‘‘blogs or microblogs,’’ ‘‘Twitter,’’ ‘‘Instagram,’’ ‘‘Pinterest,’’ ‘‘Google Plus,’’ ‘‘Rich Site Summary (RSS) feeds,’’ ‘‘Flickr,’’ ‘‘LinkedIn,’’ ‘‘Tumblr,’’ ‘‘Reddit,’’ and ‘‘MySpace.’’ Other eligible studies were also identified by searching the cited references from obtained published studies.

Inclusion and Exclusion Criteria

Studies were included if they met the following criteria: (a) the study was published in recent 20years from January 1997 (when the very first social media platform was launched)¹⁴ to February 2017, (b) the research report involved the outcomes regarding the effect of social media use in cancer interventions using contemporary social media, (c) preliminary surveys using contemporary social media for future cancer interventions, (d) original or experimental, and (e) the research targeted patients with cancer, caregivers, healthcare professionals, or other stakeholders. We excluded studies if they met the following criteria: (a) review articles or editorials, case studies, and not original research articles; (b) not published in English; (c) duplicated; (d) did not report a social media outcome; (e) studies for chronic diseases broadly, not only focusing on cancer; (f) studies with a descriptive design and description of contents or thematic analysis; and (g) intervention studies of Web-based, online forums; bulletin boards; and virtual worlds without incorporating characteristics of social media.

Study Selection and Screening

Two authors (C.J.H. and Y.J.L.) independently assessed and screened study eligibility and were involved in study selection. All titles, abstracts, and full-text articles were reviewed independently by C.J.H. and Y.J.L. against the inclusion and exclusion criteria. Discrepancies were resolved by discussion. A summary of the selection process is presented in the Figure. In case of disagreement among the 2 authors, a third author (G.D.) was available for arbitration.

Methodologic Quality Evaluation

Two authors (C.J.H. and Y.J.L.) independently evaluated the methodological quality of randomized and nonrandomized quantitative or mixed method studies using a Downs and Black¹⁵ checklist (Table 1). This checklist is used to appraise risk of bias in original or primary source research articles, and this tool contains 27 items across 5 sections.¹⁵ The internal consistency reliability scores were high overall on all 5 sections (Cronbach's α > .69), and criterion validity was high (r = 0.90).¹⁵ Of the 5 sections including reporting, external validity, internal validity (study bias, confounding), and power, the power section was excluded in our review because of item ambiguity.¹⁵ Reporting refers to the overall quality of the study (10 items; eg, Are the main findings of the study clearly described?). External validity refers to the ability to generalize the findings of the study (3 items; eg, Were the subjects asked to participate in the study representative of the entire population from which they were recruited?). Internal validity including study bias and confounding sections refers to how well a research is conducted, especially whether it avoids confounding. Study bias means the bias in the intervention and outcome measure(s) (7 items; eg, Was an attempt made to blind those measuring the main outcomes of the intervention?). Confounding section is used to determine cohort selection bias from sampling or group assignment (6 items; eg, Were the patients in different intervention groups or were the cases and controls recruited from the same population?). Each item is scored as 1 (“excellent quality”) or 0 (“poor quality”), except for 1 item in the reporting section, which is scored as 0, 1, or 2.¹⁵ Therefore, 26 items of 4 sections were evaluated in this review with the score ranging from 0 (minimum) to 27 (maximum) points. The overall score of the Down and Black checklist in each selected study was calculated and presented in Table 1. The κ statistic was used to measure the level of agreement between the 2 authors. In the case of continued disagreement among the 2 authors, a third author (G.D.) was available for arbitration.

Data Extraction and Data Synthesis

Data were extracted into standardized forms (Microsoft Excel; Microsoft, Redmond, Washington) by the first author (C.J.H.) and were verified for accuracy and completeness by the other 2 authors (Y.J.L. and G.D.). Discrepancies were resolved through discussion and consensus among the 3 authors. For all studies, the following variables were extracted: characteristics of intervention studies (eg, study methods, target participants, primary aims, social media platforms), feasibility and usability of the interventions, components of the interventions (eg, cancer prevention education, social support), outcome measures and effects of social media interventions (eg, psychological symptoms), and results and authors" conclusions. We synthesized extracted data and presented findings as narrative descriptions and simple descriptive statistics in tables.

Results

Selected Studies and Methodological Quality Assessment

Fifty-five studies met the criteria of the initial search, and the final number of studies included after a full review was 18.^16–33 All 18 studies were recently published between 2012 and 2016, and approximately half of the studies were published in 2016. The 18 included studies were primarily quantitative studies, and 3 quantitative studies^18,25,33 included thematic or content analyses of open-ended questionnaires. Overall, the methodological qualities of the selected studies were high (17.0 ± 4.0; range, 13-23 of 27 [total]). The lowest score was for a public cancer education intervention study via YouTube,²² and the highest score was for a randomized controlled trial (RCT) using personal blogs for breast cancer.³³ The score of confounding was low (2.7 ± 0.7) compared with the scores of reporting, external validity, and study bias. A mean κ statistic for interrater reliability across the 26 items was 0.97 ± 0.04 (very good agreement) in applying the evaluation criteria.

Characteristics of the Intervention Studies Using Social Media Tools

Study Designs

Most of the design approaches were quasi-experimental ones, as in 9 studies,^{16–20,22,24,26,30}followed by an experimental design for 6 RCTs^{21,23,25,31–33} (Table 2). The 9 quasi-experimental studies were pilot, and they were non-RCTs without control groups. Most of the social mediatools were the main component of interventions (n = 11). Four studies used social media as part of a larger intervention.^16,17,21,31 Three studies^27–29 were preliminary surveys that used Facebook to recruit participants for future intervention studies in cancer prevention. Duration of the interventions varied from a 1-time YouTube video playing in a breast cancer preventioncampaign¹⁸ to 4 years of comprehensive cancer prevention education.¹⁷ Median interventionduration was 12 months.

Participants

Most of the studies were conducted in western countries (14 in the United States, 1 in Italy, and 1 in Canada) and for white female participants (study n = 13). The age of participants ranged from 20.9 years in a study for all types of cancer²⁴ to 63.4 years in a study for lung cancer,²⁹except for 2 studies for young participants aged 18 or younger.^18,28 The mean age was younger than 45 years in 6 studies.^{22,24,26,27,30,32} Three studies did not report gender or age.^16,17,20 Most of the studies targeted healthy participants^{17,18,20,22,24–26,29} or patients with cancer.^{19,21,27,30–33}Two studies targeted both patients with cancer and their informal caregivers.^23,28 One study targeted patients with cancer, family caregivers, clinicians, and researchers.¹⁶ Most of the studies were conducted for all types of cancer^{16,17,21,22,24,32} or breast cancerspecifically.^{18,19,26,31,33} Two studies each were conducted for pediatric cancer^28,30 and gynecological cancer,^20,27 and only 1 study each was conducted for skin cancer,²⁵ colorectal cancer,²³ and lung cancer.²⁹

Settings and Investigators

Most studies recruited within community health centers or cancer centers.^{17–19,21,22,25,33} In other cases, the setting was not limited to a specific place.^{16,20,26,27,29,32} A few studies were conducted in a university setting²⁴ or large urban hospitals^23,30 or limited within the United States.^28,31 Four social media interventions were launched by national healthcare organizations (CDC,^20,26 National Cancer Institute,¹⁷ national nonprofit organization “LIVESTRONG”¹⁶). The remaining 14 intervention studies were conducted by researchers at public or private cancercenters or universities.

Primary Aims of the Social Media Interventions

Providing overall cancer-related knowledge was the most common primary aim of the social media interventions^16–26 (Table 2). The remaining intervention studies aimed to recruit individuals of a preliminary survey for future cancer intervention studies,^27–29 to provide social support among participants^30,31 and lifestyle modification,³² and to educate participants to create and use their own blogs.³³

Social Media Platforms Applied to the Interventions

Facebook was the most frequently used platform applied as a single channel or multichannels in 11 interventions,^{16,17,19,24–30,32} followed by blogs (n = 7), YouTube (n = 6), and Twitter (n = 5) (Table 2). Six of the 18 intervention studies used more than 1 tool (eg, Twitter + Facebook + YouTube).^{16,17,19,22,24,30} For example, an intervention may use multichannels embedding links to YouTube videos within Facebook or Twitter.²⁴ Heo et al²² developed 8 YouTube cancereducation videos and also created a blog to provide stationary materials and reminders. The remaining studies conducted interventions using only 1 social media tool (Facebook in 6 studies,^25–29,32 YouTube in 2 studies,^18,20 and blogs in 4 studies^21,23,31,33).

Feasibility and Usability of the Social Media Interventions

Feasibility

Social media interventions were feasible in cancer prevention and management across the non-RCT studies. The assessed feasibility outcomes were engagement rate,^19,25,26 cost,^20,28,29 and participant recruitment for the research projects.^27–29 The use of social media to assess engagement rate was feasible and showed higher engagement rate, calculated by the number of users who “liked,” “clicked,” “commented,” or “shared” divided by the number of users who received the posts via social media.^19,25,26 Instead of measuring engagement rate, a few studies assessed social media uses such as numbers of posts, total numbers, playing time, and views of posts.^{16,17,20,22,24,31,33}

To recruit participants for cancer intervention studies, announcements and brief descriptions of the research and preliminary surveys were posted on social media.^27–29 Three survey studies^27–29 developed Facebook pages and posted their research surveys on the Facebook pages. Akard et al²⁸ posted an invitation page and a survey link of a future Web-based intervention to recruit parents of children with cancer. Carter-Harris et al²⁹ posted a campaign and a survey link of lung cancer screening intervention to long-term smokers. Zaid et al²⁷conducted a feasibility study of future intervention studies for patients with cervical cancer, and an epidemiologic and quality-of-life survey was posted in Facebook daily newsfeed. Social media tools were cost-effective to conduct research surveys or recruit participants. Akard et al²⁸recruited 67 participants with greater geographic diversity and improved generalizability of the participants with a cost-effective approach. The cost per competed survey was one-fortieth, and recruited participants were 10-fold more in the intervention group than controls.²⁹ Carter-Harris et al²⁹ reported that the Facebook recruitment was more efficacious and cost-effective ($1.5 per survey, 18 participant recruitment per day) compared with a control group with newspapers ($40.8 per survey, 10 participant recruitment per day). Zaid et al²⁷ recruited 57 participants from 8 countries via Facebook at a low cost.

Usability

Social media usability was assessed through open-ended questionnaires^18,25,33 or 5-point Likert scales,^31,33 and participants reported high usability of the social media. Harris et al³³ measured the usability for both blog creators as target participants (mean, 2.9/5 at 1 month and 2.5/5 at 6 months) and blogs visitors (mean, 3.9/5). Technical difficulties in using blogs were reported by 80% of all 46 participants. Participants lacked time to contribute to blogs because of other obligations and stressors, or breast cancer–related medical treatment, lack of comfort with computers or lack of computer access, and physical limitations caused the loss of interest in blogs.³³ Lepore et al³¹ found that a Web site with blog functions was more useful (mean, 4.0/5) than a Web site without blog functions in the control group (mean, 3.7/5), but the difference was not statistically significant. Akard et al²⁸ assessed the technological capabilities of their project and found that the interventions using social media tools were preferred because of ease of use among the participants.

Components, Outcome Measures, and Effects of the Social Media Interventions

The summary of the components of the social media inteventions including outcomes measures were presented in Table 3. The Table, Supplemental Digital Content 1, http://links.lww.com/CN/A12, presented details of the outcome measures, effects of the social media interventions, and results and conclusions across the studies.

Cancer Prevention Eduction

In most of the studies, the social media interventions were designed with educational materials for cancer prevention. The main topics of cancer prevention were knowledge about risk factors of cancers^{16–20,24–26,29,33} or lifestyle modification (diet and physical activity).^21,25,32 Other educational components included behavioral changes related to excessive smoking^18,24,29 and alcohol consumption.²⁴ As part of the cancer prevention education, general cancer screening information such as types and reasons of cancer screening was addressed in all 18 interventionstudies. One study provided detailed instructions of bowel preparation for colorectal screening.²³

Three studies^19,23,24 showed a significant improvement of cancer-related basic knowledge and skills after the interventions of cancer prevention education. Lauckner and Whitten²⁴ conducted social media education using 4 different social media sites (Facebook, YouTube, Twitter, and blogs). Cancer-related knowledge improved in all participants, but YouTube was the most effective in delivering educational messages (higher recall and stronger attitude for reducing cancer risk) as an audiovisual form, compared with other text-based forms. Kang et al²³ showed improved bowel preparation and colorectal screening knowledge and skills in participants in the intervention group. One RCT study showed no statistically significant differences in change of knowledge about cancer and healthy lifestyles between the intervention and control groups.²¹Valle et al³² conducted an RCT of physical activity intervention via Facebook and showed increased mild exercise and self-efficacy for physical activity in the intervention group (P < .05). Pagoto and Baker²⁵ reported an RCT methodological article to examine the changes in general healthy lifestyle behaviors at baseline and postintervention, between the intervention and control groups.

Cancer Management

Eight studies^{17,19,21,22,27,30,31,33} provided interventions for cancer management including topics related to cancer survivorship, quality of life, and psychological distress. Cancer-related knowledge and management for cancer survivors (eg, chemotherapy regimen and adverse effects) were addressed in cancer education via social media in 6 studies.^{17,19,21,22,27,30} Zaid et al²⁷ and Gnagnarella et al²¹ specifically addressed quality of life for cancer survivors. Six studies^{19,21,27,30,31,33} measured psychological symptoms to examine the effects of the social media interventions on psychological distress or to collect preliminary data of mental health status for future cancer management interventions.

With regard to cancer survivorship examined in 2 studies,^19,30 breast cancer survivors reported increased knowledge about cancer survivorship (85.7% of the total participants) at postintervention. In a study for pediatric cancer survivors,³⁰ the positive or negative identity in cancer survivorship was surveyed with an open-ended questionnaire, and 60% of the total participants had a positive survivorship identity after the intervention. However, no pre-post within-subject or control group comparisons were conducted in either of the studies.^19,30 Two studies assessed quality-of-life scores using validated instruments.^21,27 Zaid et al²⁷ evaluated quality-of-life scores and showed low quality-of-life scores in patients with cancer, but this result was not compared with the control group. In Gnagnarella et al,²¹ quality-of-life scores improved in both the intervention and control groups but not to a statistically significant level.²¹

Psychological distress was assessed in several studies using validated instruments—depression,^30,33 anxiety,^19,27 and both depression and anxiety.^21,31 Harris et al's³³ RCT study found significant positive correlations between the use of negative content words and depressive symptoms in the intervention group. Similar to Harris et al,³³ Song et al³⁰ presented a positive relationship of depression with negative identity in cancer survivorship. Attai et al¹⁹showed decreased anxiety levels in 67% of the participants after the cancer preventionintervention using Twitter, blogs, and Facebook. Zaid et al²⁷ assessed anxiety levels with a Facebook preliminary survey, and the levels of anxiety were high in patients with gynecological tumors. No significant improvement in either depression or anxiety levels was observed in the intervention group, compared with the controls in Lepore et al³¹ and Gnagnarella et al.²¹

Social Support

Most social media interventions were multicomponent with combinations of cancer information and social support.^{19,21,22,26,27,30–33} The social support functions in social media interventions were designed as communication tools, through discussion boards, chatting rooms, or posting personal stories via photographs and videos. In Theiss et al²⁶ and Song et al,³⁰ participants created their own videos or photographs based on their own real-life experiences regarding cancer-related topics, posted them to social media, and shared them among participants. These communication tools embedded in social media (eg, message boards, discussion threads, and support groups) provided for sharing opinions, experiences, and cancer-related knowledge, as well as peer encouragement, among the participants.^{19,21,22,26,27,30–33}

Lepore et al's³¹ RCT assessed social support behaviors using a Word Count text analysis method. The intervention group showed increased blog postings regarding social support topics (74.4% of the total postings) compared with the control group (61.9% of the total postings, P < .004). In an RCT of physical activity intervention via Facebook,³² a self-monitoring Web site was embedded in the Facebook pages to increase interactions and social support among study administrators, cancer experts, and Facebook friends in the intervention group. The social support scores from Facebook friends positively mediated the Facebook intervention effects on moderate to vigorous physical activity (P = .0006). However, social support scores from Facebook friends increased in the control, but not in the intervention, group (P = .039).³²

Health Disparities

Health disparities were addressed in 2 interventions.^16,17 Alexander et al¹⁷ targeted ethnic minority groups (ie, African Americans, Hispanics, Asian Americans, American Indians, Alaska Natives) and multicultural audiences. They developed “Lifelines” as series of cancer education and disseminated social media outreach to minority ethnic groups through both traditional approaches such as print and radio and social media (primarily Twitter) avenues. Justice-Gardiner et al¹⁶ used Facebook and Twitter adding to the traditional print and radio cancerprevention campaigns. The intervention was developed addressing culturally relevant and linguistically appropriate cancer awareness to improve the reach and access to these social media campaigns for US Hispanics and Latinos.¹⁶ Outcome measures specific to health disparities were not assessed across the studies.

Users' Communications with Cancer Experts

The cancer experts provided feedback, advice, and consultation in response to participants' queries and communicated with them via discussion boards or text messages that were set up in the social media environment.^21,23,32 No relevant outcome measures such as satisfaction with experts' counseling were examined across the studies.

Discussion

Our study is the first to conduct a comprehensive systematic review of the contemporary social media application in interventions for cancer prevention and management. Although the effects of social media on the intervention outcomes varied with mixed results in this review, overall, social media tools have the potential to benefit cancer prevention and management. This review provides support for social media as acceptable, feasible, and potentially efficacious tools across the cancer continuum. In light of the increased adoption of contemporary social media by individuals, the public, and healthcare organizations,¹⁰ our findings may inform the design and development of cancer interventions in using various social media tools as the interventionplatform for cancer prevention and management.

We found that 3 studies did not provide gender or age of the participants.^16,17,20 One of the benefits in using social media is recruiting more participants by allowing participants to remain anonymous and protecting their privacy and personal information.⁷ However, the lack of demographics and other characteristics of participants precludes in-depth assessment of the social media intervention effects, which could differ by subgroups such as age and gender.¹⁰For example, participants older in age are less likely to use social media than younger participants, and adults 18 to 49 years old make up the largest users of contemporary social media.³⁴ Thus, the lack of participants' demographic information may limit the precise assessments of the social media effects on cancer prevention and management, and future studies should collect robust participant information. The quasi-experimental study design was common, and 6 of the 18 studies^{21,27,28,30,32,33} had a small sample size (n < 100) with insufficient statistical power. Half of the studies^{19,21,23,24,26,30–33} examined a statistical significance for their outcome variables, and the other half did not. Except for 6 studies,^{21,23,25,31–33} 12 studies did not examine the effects of social media interventions using rigorous study methods (eg, RCT, large sample size, validated instruments, studies informed by a power analysis to generate statistically significant findings, comparison with a control group).

We found that the main components of social media interventions were cancer preventioneducation and social support. Some social media interventions provided reliable and evidence-based health educational information, developed by cancer experts, as well as national health departments such as CDC and National Cancer Institute. However, there is a lack of consultation with, or supervision and quality assessment by, cancer experts about the content shared on social media, except for 3 studies.^21,23,32 Inaccurate health information may lead to users sharing incorrect information and managing themselves with erroneous or incomplete instructions, which can result in unexpected outcomes (eg, failing to follow physician recommendations, severe adverse effects).⁸ Therefore, a mechanism to audit the medical information posted and shared within social media applications may be needed in many cases to safeguard the quality and value of information exchange.

We found that the various cancer types and stages and target users (eg, patients, caregivers, clinicians, researchers, and healthy individuals or individuals at a high risk of cancer) were not fully examined across the studies. Because of the complexity of cancer prognosis, treatment, and survivorship, it is still unclear how social media tools are used for health information for individuals' unmet needs in different subgroups by cancer types and stages and target users.¹⁰Thus, various subgroups of participants should be further studied.

Although most of the studies were conducted with white populations, 2 studies^16,17 addressed health disparities and how they can be reduced via social media interventions for minority populations, demonstrating that social media interventions targeting specific races/ethnic groups were effective in reducing health disparities. Certain populations such as US Hispanics showed growing use of the social media more than any other racial/ethnic groups.¹⁶ Thus, social media may have the potential to reduce health disparities and to help the individuals easily access healthcare services via Internet-based approaches. However, people may also have limited digital literacy, which may contribute to disparities in health communication, in addition to the so-called digital divide—the lack of access to infrastructure that can facilitate access to online resources. This can be challenging for the underserved population with cancerincluding minority populations, individuals with low socioeconomic status, and individuals living in rural, medically underserved areas.^21,35 Furthermore, most of the social media interventions in 16 of the 18 studies were designed in English, which may further challenge access to populations with limited English proficiency.

There were limited outcomes measured pertaining to intervention components and clinical outcomes across the studies. Although symptom management is a major concern in patients with cancer and their caregivers,³⁶ only 6 of the 18 studies assessed psychological symptoms.^{19,21,27,30,31,33} None of the studies assessed the effect of social media on other major symptoms of cancer such as pain, fatigue, and sleep deprivation.³⁶ The diet and exercise lifestyle behaviors were addressed in 3 studies, but these studies did not measure the changes in dietary habits (eg, amount of healthy eating of proteins, vegetables) or body weight changes after the interventions.^21,25,32 The cancer-specific topics (eg, cancer symptom management, adverse effects of chemotherapy) and longitudinal patient outcomes in the final stages of cancersuch as palliative care and end-of-life issues were not addressed in many studies. Interventions connecting patients and health professionals were also less frequent compared with interventions connecting patients with each other. The cognitive behavioral therapy was effective in cancer prevention and management³⁷ but was not applied using social mediaplatforms. Finally, only 2 studies designed theory-based interventions informed by the social cognitive theory.^25,32

No effects of social media were found in 2 studies.^21,32 Gnagnarella et al²¹ provided an intervention using blogs for patients with cancer, but nonsignificant effects were observed. In this study, progressive cancer disease influenced the nonsignificant findings because of the lower interest in study participation, lower social media access, and lower social functioning to participate in the intervention than the general population. In addition, higher dropout rates resulted in the small sample size.²¹ In Valle et al,³² the control group showed a significant improvement of social support from Facebook friends and self-efficacy, but not in the intervention group. A self-monitoring function in Facebook was provided in the interventiongroup, and participants had interactions with study administrators and cancer experts. However, the control group was found to have more appropriate social support and self-efficacy from both offline and online family and friends, compared with the intervention group who had more interactions with study administrators and cancer experts online. Although there are no previous studies showing comparable results, these results point to consider the health conditions of patients with cancer and offline, in-person social interactions for social media.

A limitation of this review is that it included only studies published in the English language, potentially excluding other work. The selected studies in this review were subject to publication bias favoring literature to demonstrate benefits of social media use because nonsignificant or negative outcomes may not have been published in journals through the literature search process. Most of the participants included in this review were white women and healthy participants or patients with cancer in western countries. This limited the ability to generalize the findings for the application of social media interventions to the broad scope of target users from different cultural backgrounds. In addition, the selected studies did not always clearly define the social media applied to the cancer intervention, and thus, some valid studies may have not been included for this review. Finally, the interventions were heterogeneous, and the use of validated instruments was lacking across the studies. Thus, it was difficult to conduct meta-analyses and to have convergences of social media effects. In addition, most of the outcome measures were qualitative interviews or self-reported data. Clinically meaningful, quantitative outcome measures were lacking. Therefore, the effects of social media interventions on cancerprevention and management were not clear without clinical validation across the studies.

Implications for Nursing

This review provides opportunities for nurses to address gaps in nursing education, practice, and research in applying social media for cancer prevention and management. Today's nurses can maximize the benefits of social media by educating themselves to integrate nursing science, practice, social media, and technology to prevent and manage cancer, as coordinators and advocates of patient care. Nursing professionals are well positioned to use social media to improve the quality of cancer prevention and management. For example, nurses can use social media as communication tools to provide counseling and social and emotional support, impart cancer-related knowledge and skills, and monitor cancer-related symptoms for patients with cancer and family caregivers. In particular, nurses can deliver cancer interventions by social media tools for those who are underserved, living in rural areas, or with limited access to healthcare services. Furthermore, nursing researchers are involved in developing nurse-led, patient-centered social media interventions or in recruiting participants using social media for cancer research. Ultimately, the use of social media by nursing professionals may encourage self-management for cancer prevention and management and improve quality of life in patients with cancer and their caregivers. This effort will contribute to effective patient-centered nursing care in oncology.

Conclusions

Our systematic review highlights the potential benefits of applying contemporary social media to the individualized, patient-centered interventions for cancer prevention and management. The evidence of the impact of social media on cancer-related clinical outcomes is very limited from RCTs and longitudinal studies. Further research should consider quantitative clinical outcome measures with rigorous study methodologies and validated instruments to strengthen evidence for their efficacy/effectiveness and safety of social media as a tool for providing reliable online information. Second, various types of intervention modalities (eg, cognitive behavioral therapy), various health outcomes of interest, and various subgroups (eg, by target users, by ethnicity/race, by cancer types and stages) are suggested for future social media interventionstudies. Third, consideration of the variation in social media according to user-centered, culturally tailored parameters will be valuable in tailoring intervention to prevent and manage cancer. Finally, a theoretical framework should inform the design of the social mediainterventions in oncology. Harnessing social media tools and technology has the potential to deliver effective interventions and achieve positive outcomes in oncology.

References

1. Siegel R, Miller K, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 2017;67(1):7Y30.

2. Slev VN, Mistiaen P, Pasman HR, Verdonck-de Leeuw IM, van UdenKraan CF, Francke AL. Effects of eHealth for patients and informal caregivers confronted with cancer: a meta-review. Int J Med Inform. 2016;87:54Y67.

3. Kaplan AM, Haenlein M. Users of the world, unite! The challenges and opportunities of social media. Busi Horiz. 2010;53(1):59Y68.

4. Karrer T. Forums vs. social networks? e Learn Tech. 2008. http:// elearningtech.blogspot.com/2008/09/forums-vs-social-networks.html. Accessed January 16, 2017.

5. Virtual Worlds Review. What is a virtual world? http://www.virtualworlds review.com/info/whatis.shtml. Accessed January 17, 2017.

6. Chretien KC, Kind T. Social media and clinical care ethical, professional, and social implications. Circulation. 2013;127(13):1413Y1421.

7. Attai DJ, Sedrak MS, Katz MS, et al. Social media in cancer prevention and management: highlights, challenges & opportunities. Future Oncol. 2016;12(13):1549Y1552.

8. Patel R, Chang T, Greysen SR, Chopra V. Social media use in chronic disease: a systematic review and novel taxonomy. Am J Med. 2015; 128(12):1335Y1350.

9. McAlpine H, Joubert L, Martin-Sanchez F, Merolli M, Drummond KJ. A systematic review of types and efficacy of online interventions for cancer patients. Patient Educ Couns. 2015;98(3):283Y295.

10. Koskan A, Klasko L, Davis SN, et al. Use and taxonomy of social media in cancer-related research: a systematic review. Am J Public Health. 2014; 104(7):e20Ye37.

11. Di Cerbo A, Pezzuto F, Laurino C, Palmieri B. Web babel syndrome and false expectations during own multimedia oncological search. Clin Ter. 2014;165(3):e225Ye229.

12. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg. 2010;8(5):336Y341.

13. eBizMBA. Top 15 most popular social networking sites. http://www. ebizmba.com/articles/social-networking-websites. Updated May 2017. Accessed May 20, 2017.

14. Ellison Nicole B. Social network sites: definition, history, and scholarship. J Comput Mediat Commun. 2007;13(1):210Y230.

15. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and nonrandomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377Y384.

16. Justice-Gardiner H, Nutt S, Rechis R, McMillan B, Warf R. Using new media to reach Hispanic/Latino cancer survivors. J Cancer Educ. 2012;27(1): 100Y104.

17. Alexander J, Kwon HT, Strecher R, Bartholomew J. Multicultural media outreach: increasing cancer information coverage in minority communities. J Cancer Educ. 2013;28(4):744Y747.

18. Bottorff JL, Struik LL, Bissell LJ, Graham R, Stevens J, Richardson CG. A social media approach to inform youth about breast cancer and smoking: an exploratory descriptive study. Collegian. 2014;21(2):159Y168.

19. Attai DJ, Cowher MS, Al-Hamadani M, Schoger JM, Staley AC, Landercasper J. Twitter social media is an effective tool for breast cancer patient education and support: patient-reported outcomes by survey. J Med Internet Res. 2015;17(7):e188.

20. Cooper CP, Gelb CA, Chu J. Gynecologic cancer information on YouTube: will women watch advertisements to learn more? J Cancer Educ. 2016;31(3):602Y604.

21. Gnagnarella P, Misotti AM, Santoro L, et al. Nutritional online information for cancer patients: a randomized trial of an internet communication plus social media Intervention. J Cancer Educ. 2016;31(3):472Y480

22. Heo J, Chun M, Lee HW, Woo JH. Social media use for cancer education at a community-based cancer center in South Korea. J Cancer Edu. 2016:1Y15.

23. Kang X, Zhao L, Leung F, et al. Delivery of instructions via mobile social media app increases quality of bowel preparation. Clin Gastroenterol Hepatol. 2016;14(3):429.e3Y435.e3.

24. Lauckner C, Whitten P. The differential effects of social media sites for promoting cancer risk reduction. J Cancer Educ. 2016;31(3):449Y452.

25. Pagoto SL, Baker K. Engaging moms on teen indoor tanning through social media: protocol of a randomized controlled trial. JMIR Res Protoc. 2016;5(4):e228.

26. Theiss SK, Burke RM, Cory JL, Fairley TL. Getting beyond impressions: an evaluation of engagement with breast cancer-related Facebook content. Mhealth. 2016;2(41):1Y7.

27. Zaid T, Burzawa J, Basen-Engquist K, et al. Use of social media to conduct a cross-sectional epidemiologic and quality of life survey of patients with neuroendocrine carcinoma of the cervix: a feasibility study. Gynecol Oncol. 2014;132(1):149Y153.

28. Akard TF, Wray S, Gilmer MJ. Facebook advertisements recruit parents of children with cancer for an online survey of web-based research preferences. Cancer Nurs. 2015;38(2):155Y161.

29. Carter-Harris L, Bartlett Ellis R, Warrick A, Rawl S. Beyond traditional newspaper advertisement: leveraging facebook-targeted advertisement to recruit long-term smokers for research. J Med Internet Res. 2016;18(6):e117.

30. Song H, Nam Y, Gould J, et al. Cancer survivor identity shared in a social media intervention. J Pediatr Oncol Nurs. 2012;29(2):80Y91.

31. Lepore SJ, Buzaglo JS, Lieberman MA, Golant M, Davey A. Standard versus prosocial online support groups for distressed breast cancer survivors: a randomized controlled trial. BMC Cancer. 2011;11:379.

32. Valle CG, Tate DF, Mayer DK, Allicock M, Cai J. Exploring mediators of physical activity in young adult cancer survivors: evidence from a randomized trial of a Facebook-based physical activity intervention. J Adolesc Young Adult Oncol. 2015;4(1):26Y33.

33. Harris LN, Cleary EH, Stanton AL. Project connect online: user and visitor experiences of an Internet-based intervention for women with breast cancer. Psy Oncol. 2015;24(9):1145Y1151.

34. Pew Research Center. Internet/broadband fact sheet. 2017. http://www. pewinternet.org/fact-sheet/social-media. Accessed January 18, 2017.

35. Kim E, Han JY, Moon TJ, et al. The process and effect of supportive message expression and reception in online breast cancer support groups. Psych Oncol. 2012;21(5):531Y540.

36. Balogh EP, Ganz PA, Murphy SB, Nass SJ, Ferrell BR, Stovall E. Patientcentered cancer treatment planning: improving the quality of oncology care. Summary of an Institute of Medicine workshop. Oncol. 2011;16(12):1800Y1805.

37. Post KE, Flanagan J. Web based survivorship interventions for women with breast cancer: an integrative review. Eur J Oncol Nurs. 2016;25:90Y99.​​