JMIR--Correlates of Health-Related Social Media Use Among Adults ...

Advertisement: Preregister now for the Medicine 2.0 Congress

Original Paper

Correlates of Health-Related Social Media Use Among Adults

Rosemary Thackeray^*, MPH, PhD; Benjamin T Crookston^*, MPH, PhD; Joshua H West^*, MPH, PhD

Brigham Young University, Department of Health Science, Provo, UT, United States
^*all authors contributed equally

Corresponding Author:
Rosemary Thackeray, MPH, PhD

Brigham Young University
Department of Health Science
221 Richards Building
Provo, UT, 84602
United States
Phone: 1 801 422 1704
Fax: 1 801 422 0273
Email:

---

ABSTRACT

Background: Sixty percent of Internet users report using the Internet to look for health information. Social media sites are emerging as a potential source for online health information. However, little is known about how people use social media for such purposes.
Objectives: The purpose of this study was two-fold: (1) to establish the frequency of various types of online health-seeking behaviors, and (2) to identify correlates of 2 health-related online activities, social networking sites (SNS) for health-related activities and consulting online user-generated content for answers about health care providers, health facilities, or medical treatment.
Methods: The study consisted of a telephone survey of 1745 adults who reported going online to look for health-related information. Four subscales were created to measure use of online resources for (1) using SNS for health-related activities; (2) consulting online rankings and reviews of doctors, hospitals or medical facilities, and drugs or medical treatments; (3) posting a review online of doctors, hospitals or medical facilities, and drugs or medical treatments, and (4) posting a comment or question about health or medical issues on various social media. Univariate and multivariate logistic regression analyses were performed.
Results: Respondents consulted online rankings or reviews (41.15%), used SNS for health (31.58%), posted reviews (9.9%1), and posted a comment, question, or information (15.19%). Respondents with a chronic disease were nearly twice as likely to consult online rankings (odds ratio [OR] 2.09, 95% CI 1.66-2.63, P<.001). Lower odds of consulting online reviews were associated with less formal education (OR 0.49, 95% CI 0.37-0.65, P<.001) and being male (OR 0.71, 95% CI 0.57-0.87, P<.001). Respondents with higher incomes were 1.5 times as likely to consult online rankings or reviews (OR 1.49, 95% CI 0.10-2.24, P=.05), than respondents with a regular provider (OR 2.05, 95% CI 1.52-2.78, P<.001), or living in an urban/suburban location (OR 1.61, 95% CI 1.17-2.22, P<.001). Older respondents were less likely to use SNS for health-related activities (OR 0.96, 95% CI 0.95-0.97, P<.001), as were males (OR 0.70, 95% CI 0.56-0.87, P<.001), whereas respondents with a regular provider had nearly twice the likelihood of using SNS for health-related activities (OR 1.89, 95% CI 1.43-2.52, P<.001).
Conclusions: People are using social media for seeking health information. However, individuals are more likely to consume information than they are to contribute to the dialog. The inherent value of ?social? in social media is not being captured with online health information seeking. People with a regular health care provider, chronic disease, and those in younger age groups are more likely to consult online rankings and reviews and use SNS for health-related activities.

(J Med Internet Res 2013;15(1):e21)
doi:10.2196/jmir.2297

KEYWORDS

social media; Internet; health information; consumer

The Internet is becoming an increasingly common source of health information. Approximately 60% of Internet users report using the Internet to look for health information [1,2]. In addition to seeking health information, Wen et al [3] found that 15% of Internet users also tracked personal health information on the Internet. Determinants of seeking health information online include education, gender, race, age, presence of children in the home, having a poor personal health condition, and geographic residence [1,4-7]. Similarly, predictors of using the Internet to track personal health information include gender, race, education, and having a health care provider [3].

Historically, online health seeking meant visiting an agency- or organization-sponsored website. Recently, social media sites are emerging as a potential source of online health information [8]. Social media refers to ?activities, practices, and behaviors among communities of people who gather online to share information, knowledge, and opinions using conversational media? [9]. These social media are broadly categorized as forums and message boards, review and opinion sites, social networks (eg, Facebook), blogging and microblogging (eg, Twitter), bookmarking, and media sharing (eg, YouTube) [10].

Individual use of social media is steadily increasing. Nearly two-thirds (65%) of adult Internet users in the United States are involved with a type of social media called social networking sites (SNS), such as MySpace, Facebook, or LinkedIn [11]. Technorati currently registers over 1.3 million blogs [12], 13 percent of Internet users (140 million people) have a Twitter account [13,14], and Facebook has 955 million active users [15]. By 2015, it is estimated that the number of individuals and corporations who have social networking accounts will reach over 3 billion [16].

Social media and SNS use varies by demographics. There are statistically significant differences in SNS use between younger and older ages and between males and females [11]. However, SNS are used fairly equally across education, income, race/ethnicity, and rural and urban locations [11]. Chou and colleagues [17] found that age and education were predictors of 3 forms of social media use (ie, participating in online support groups, blogging, and visiting a SNS).

In contrast to going online to seek health information, social media technologies allow online social media users to create, distribute, and share information independent of an organization. The level of use and involvement with social media technologies varies by individual. Bernoff and Anderson [18] and Li and Bernoff [19] classify individuals based on how they use social media. These classifications, although not mutually exclusive, include creators, conversationalists, critics, collectors, joiners, or spectators. Similarly, Hoffman and Novak [20] identify 4 goals for social media use: create, connect, consume, and control. The main conclusion from both typologies is that the range of social media activities that people engage in varies from consuming to creating content.

Despite the near ubiquity of social media use and the high prevalence of health information seeking on the Internet, there is a dearth of literature about the characteristics of people who use social media for seeking health information and how these people engage with social media. Thus, additional research is needed to determine whether social media users are primarily spectators, or if they are creators or critics. That is, are they looking for information or are they becoming part of the information creation and sharing process? Knowing the correlates of social media use for health information can allow health professionals to more accurately segment populations and tailor interventions accordingly. Therefore, the aim of this research was two-fold. First, to establish the frequency of various forms (eg, spectators, creators, or critics) of online health-seeking behaviors. Second, this research seeks to identify correlates of 2 health-related online activities: (1) using SNS for health-related activities, and (2) consulting online user-generated content for answers about health care providers, health facilities, or medical treatment.

Data Source and Sample

The data for this study were taken from the 2010 Health Tracking Survey conducted by Princeton Survey Research on behalf of the Pew Internet & American Life Project [21]. The data were collected during August to September 2010 through a telephone survey that included both cell phones and landlines. A random digit method was used to select participants who were US residents, aged 18 years and older, and who spoke English (n=3001). Data were weighted to the most recent US Census Bureau?s Current Population Survey. Data were stripped of identifying information and made available to the public. For the current study, inclusion criteria were adults who used the Internet at least occasionally (Pew question Q6a) and who reported going online to look for health-related information (Pew question healthseek). The final sample size was 1745.

Measures

Demographic, Socioeconomic, and Health Status

Demographic, socioeconomic, and health status covariates included ethnicity, education, income, gender, age, race, marital status, having a chronic health condition, geographic community type, health insurance status, and having a family doctor or health care professional. Response categories for race, education, and marital status were collapsed to account for small cell sizes.

To measure an individual?s level of social media health engagement, we created 4 subscales based on related survey items. Each of the response variables were dichotomous and coded as yes or no. We calculated Cronbach alpha to estimate internal reliability for each scale.

Used Social Networking Sites for Health-Related Activities

The 5 questions that focused on using SNS for health-related activities included (1) get health information, (2) start or join a health-related group, (3) follow your friend?s personal health experiences or health updates, (4) raise money or draw attention to a health-related issue or cause, and (5) remember or memorialize others who suffered from a certain health condition (Pew questions Q26a-e). The composite scale had an internal reliability of Cronbach alpha=.66

Consulted Online Rankings or Reviews

Three questions focused on consulting online rankings or reviews of (1) doctors or other providers, (2) hospitals or other medical facilities, and (3) particular drugs or medical treatments (Pew questions Q29a-c; Cronbach alpha=.69).

Posted a Review Online

Three questions focused on whether respondents had posted a review online of (1) a doctor, (2) a hospital, or (3) his/her experiences with a particular drug or medical treatment (Pew questions Q29d-f; Cronbach alpha=.61)

Posted a Comment or Question on Social Media

Five questions asked if respondents had posted comments, questions, or information about health or medical issues on various social media. These included (1) an online discussion, a listserv, or other online group forum, (2) a blog, (3) a social networking site, such as Facebook, MySpace, or LinkedIn, (4) Twitter or another status update site, and (5) a website of any kind, such as a health site or news site that allows comments and discussion (Pew questions Q25a-e; Cronbach alpha=.80)

Data Analysis

Unadjusted univariate analyses of demographics, socioeconomic, and health status variables with each social media?health engagement scale were computed. Variables that were significantly associated with the dependent variable were included in a multivariate regression model. Multivariate logistic regression analysis was performed with social media health engagement as the dependent variable and the demographics, socioeconomic, and health status variables as covariates. All analyses were conducted using IBM SPSS Statistics version 20 (IBM Corp, Armonk, NY, USA).

Demographic Characteristics

More than half of the study sample was female (56.16%, 980/1745) and white (79.20%, 1382/1745) (see Table 1). College graduates comprised 39.43% (689/1745) of the sample, 29.46% (514/1745) reported a household income between US $75,000 and $150,000, and 86.88% (1516/1745) reported having health insurance. Respondents reported consulting online rakings or reviews (41.15%, 718/1745) and using SNS for health (31.58%, 551/1745) more than they reported contributing content through posting reviews of doctors, hospitals, drugs, or medical treatments (9.91%, 173/1745), or posting a comment, question, or information about health or medical issues on a blog, SNS, Twitter, website, or online discussion or forum (15.19%, 265/1745).

Correlates of Social Media Health Engagement and Regression Analyses

Regression analyses revealed few correlates for posting reviews of a doctor, hospital, drug, or medical treatment (chronic disease, income, age, health insurance) and for posting a comment, question, or information on various social media sites (chronic disease, age, marital status). Therefore, the further analysis and data presented here are limited to using SNS for health and consulting online rankings or reviews.

An examination of correlates of consulting online rankings or reviews identifies several factors that are associated with higher use of online rankings and reviews (see Table 2). For example, approximately half (49.27%, 339/688) of those with a college degree reported using online rankings or reviews compared with 40.71% (204/501) of those with some college and 31.50% (172/546) of those with a high school education or less. Factors associated with use of SNS for health included income, gender, age, marital status, and having a personal or family doctor or health care provider.

Unadjusted Regression Analyses for Consulting Online Rankings

Unadjusted regression analyses revealed numerous factors associated with consulting online rankings or reviews of doctors, hospitals, drugs, or medical treatments (see Table 3). Having a chronic disease, reporting a higher annual income, living in an urban/suburban location, reporting health insurance coverage, and having a regular health care provider were each independently associated with increased odds of consulting online rankings. Decreased odds were observed among older respondents, those who were unmarried, those with lower levels of education, males, and those who were black/African American.

Unadjusted Regression Analyses for Using Social Networking Sites for Health

Table 4 presents the results of the unadjusted regression analyses for using SNS for health-related activities, such as getting information, joining a group, following friends? health experiences, raising money, increasing awareness, or remembering or memorializing others. Older respondents and males were each less likely to engage in such behaviors. Respondents who reported being unmarried or having a regular health care provider were more likely to use SNS for health-related purposes.

Adjusted Regression Analyses for Consulting Online Rankings

Results from adjusted regression analyses (see Table 3) revealed that respondents with a chronic disease were nearly twice as likely to consult online rankings as respondents who were free of chronic disease (OR 2.09, 95% CI 1.66-2.63, P<.001). For levels of education, high school or less (OR 0.49, 95% CI 0.37-0.66, P<.001) and some college (OR 0.70, 95% CI 0.54-0.91, P=.01) were each associated with lower odds of consulting online rankings than respondents who had at least obtained a college degree. With respect to income, respondents who reported an annual income of US $75,000 to $150,000 were 1.5 times as likely to consult online rankings (OR 1.49, 95% CI 0.10-2.24, P=.05) compared to those making less than US $20,000. Males were less likely than females (OR 0.71, 95% CI 0.57-0.87, P<.001), whereas respondents who have a regular provider were more than 2 times more likely to consult online rankings (OR 2.05, 95% CI 1.52-2.78, P<.001). Living in an urban/suburban location was associated with a 60% increased chance of consulting rankings (OR 1.61, 95% CI 1.17-2.22, P<.001). In the adjusted model, marital status, race, and insurance coverage were not significantly associated with consulting online rankings. Likewise, the influence of having a health care provider and income was attenuated.

Adjusted Regression Analyses for Using Social Networking Sites for Health

Adjusted odds ratios for using SNS for health-related purposes are presented in Table 4. As respondents? ages increased, their likelihood for using such sites decreased (OR 0.96, 95% CI 0.95-0.97, P<.001). With respect to gender, males had lower odds than females (OR 0.70, 95% CI 0.56-0.87, P<.001). Compared to respondents without a regular health care provider, respondents with a regular provider had nearly twice the likelihood of using SNS for health-related activities (OR 1.89, 95% CI 1.43-2.52, P<.001), a greater influence than in the unadjusted model. Marital status was not significantly associated with using SNS for health-related activities.

Source: http://www.jmir.org/2013/1/e21/

cleveland plain dealer barry sanders barry sanders john scott jimmie johnson juan pablo montoya crash chardon high school shooting