A Mixed Methods Exploration of Social Media Use for Health Information

2. Materials and Methods

This mixed methods study included a quantitative assessment and data collection (N=256) followed by qualitative input (N=7) for triangulation. Procedures were approved by the Loma Linda University Institutional Review Board (#5240309).

2.2. Participants and Procedures

Inclusion criteria for both study portions were that participant be 1) 18 years or older and 2) a current resident of SBC. For the quantitative portion of the study, participants were recruited through social media stories, social media post advertisements and links shared by local organizations as well as in-person events (community health and resource distribution). After signing a virtual embedded informed consent, participants completed surveys over a 5-month period. Surveys were available in the participant’s preferred language (English or Spanish).

Because this survey was intended for individuals who already regularly use social media, it was digitally administered through Qualtrics. Besides demographic information, the survey explored perceptions, attitudes and practices of respondents’ general use of social media, as well as their use of social media for health information [27], and delays/barriers in accessing health care. For a subset of participants, we also obtained additional information about their social media use and health information in relation to the accounts they used. As compensation for their time and effort, those who completed the survey were given the option of entering a $25 gift card drawing.

The qualitative portion of our study was conducted to allow us to further explore and discuss issues identified in the survey. Participants were therefore recruited among those who completed the survey and were given the choice to join either a Spanish or English focus group (FG) discussion or be interviewed via Zoom, one month after the survey collection period. After being briefed about the purpose of the discussion and guided through the consent process, participants confirmed their consent to participate and be recorded via Zoom. They also completed a short demographic form. Two small focus groups (N=5) and 2 qualitative interviews were conducted in either English or Spanish. Each discussion lasted approximately one hour. All discussions were facilitated by a trained and IRB certified researcher - a native Spanish speaker in the case of the Spanish key informant interview. Interviews were conducted using a semi-structured guide, and were audio and video recorded. Questions explored social media use in general, social media use for health information, and feedback about two influenza (flu) related social media (Instagram) posts.

2.3. Measures, Data Management and Analysis

G*Power software was used to determine the necessary sample size for statistical power. The alpha level was set to .05 and desired power was set to 0.9. For a two-tailed, bivariate normal model correlation of 0.3 and a null hypothesis correlation of 0, the final sample size was sufficient for statistical power for all analyses performed, allowing for a deeper inquiry into different types of social media usage by participant group.

Data from the Spanish and English surveys were combined into one master data set, which was managed and cleaned in Google Spreadsheets and split into two data sets. The overall data set included all participants and was used for descriptive statistics. A subset of the data (N=130) was selected to further analyze the relationship between social media use and health information. Bivariate correlation analyses were conducted in R using the Julius AI platform. Variables were organized into several categories: demographic, personal healthcare, social media use, health information seeking behavior, and feedback on a specific social media message.

Demographic variables included age, gender, education (less than high school, completed high school or equivalent such as General Equivalency Diploma/GED, certificate or trade school, bachelor’s degree or graduate degree), employment (unemployed, employed part or full time, working 2 or more jobs), preferred language (Spanish or English) and whether they were a parent/caretaker (Par/Car) or not. Personal healthcare variables included healthcare insurance type - no insurance, public/government or private - and whether participants attended a doctor’s visit in the past year (Health Care), had a primary care physician (PCP), waited 2 or more weeks for a healthcare appointment (Appt wait), had a delay in seeing a healthcare provider due to financial reasons ($ Delay), and whether or not the participant had been diagnosed with a chronic condition (Chronic Disease). Response options for the last four variables were “yes”, “no” and “maybe”.

Social media use variables inquired about number of social media platforms used, frequency of use, types of social media platforms used (X, Snapchat, WhatsApp, Tik Tok, Instagram or Facebook) and purpose of social media use - whether to connect with family and friends, to view the news, for entertainment or for pop culture (trending fashion, music, movies, recipes, etc). Participants were also asked if they engaged most with or “followed” a person they considered a respected and knowledgeable health figure (“influencer”), a health organization, a health community or another type of health account.

For the qualitative data collection, health information seeking behaviors were assessed by asking participants if they used social media to search for health information, how easy it was to find health related information, and relatability, cultural relevance, and perceived ease of understanding of the content. Participants were also asked if they had made a health decision based on the information they saw on social media, and if they felt that it had helped them make health decisions, such as considering the use of a specific treatment or medication. Lastly participants were asked if they had spoken to their healthcare provider about the content of the health information they had seen on social media, and whether health information searches had occurred after a medical appointment.

Qualitative discussions were transcribed with participant consent– and translated into English if recorded in Spanish – and then verified against the audio-recordings to check for accuracy. High level thematic analyses were performed in MaxQDA [28]. Participants also completed a brief demographic form assessing their education level (no high school, high school graduation or GED, bachelor’s or graduate degree), their racial and ethnic identity (White, Black or African American, American-Indian or Alaska Native, Asian, Hispanic/Latino) and what social media platforms they used (Facebook, Instagram, Tik Tok, X, WhatsApp, Snapchat, YouTube, Reddit, Pinterest, LinkedIn or Threads). Each transcript was coded inductively, and codes were then categorized into themes. All codes and outputs were verified by the researcher to check for any errors or misinterpretation.

3. Results

Quantitative Results

3.1. Descriptive Results

A total of 256 participants completed the entire online survey. Baseline demographics and healthcare related information can be found in Table 1. While the participants’ age ranged between 18 and 78, the average age was 34 years old (standard deviation = 20.97). A majority (80.2%) chose English, more than half (61.9%) reported being a parent or caretaker, 39.7% had no more than a high school degree or equivalent and 31.91% were unemployed. Most (87.6%) had some form of health insurance coverage, and 37.7% reported having public health insurance. When asked about their health care, 82.1% reported visiting a doctor in the past year, 68.9% had a primary care physician and only 28% reported having a diagnosed chronic health condition. Regarding healthcare appointments, 74% of participants had to wait at least 2 weeks for an appointment and 43.2% attributed the delay to financial reasons.

The SM platforms used the most were Instagram and Facebook (74.6% and 69.2% of participants, respectively), followed by Tik Tok (66.2% of participants). Regarding frequency and purpose of use, 70% reported daily use and most (87.7%) reported using the platform to follow friends and family, while 58.5% used it to access the news. When asked about the types of accounts they used, most followed “influencer” accounts (51.5%), followed by 42.3%, who followed posts by organizations (see Table 2).

Participants were also asked about their experiences and behaviors around finding health information on social media (see Table 3). While only 17.9% reported using SM to search for health information (HI), 69.6% were not sure whether or not they had done so, while only 11.3% did so intentionally. Nearly half (45.9%) said they thought it was easy to find HI on SM in contrast to 19.5% who did not. Although 64.4% stated that the health information they saw was relatable and culturally relevant, the majority (61.1%) did not find it easy to understand. When asked about the timing of social media use in relation to healthcare visits, 51.8% reported looking up health information on social media after talking to their doctor but more than half (58.6%) reported having talked to their doctor about health information they had seen on social media. While only 14% admitted to having made a healthcare decision after watching health information on social media, as many as 52.9% said the information they saw helped them make a health decision. Even more (66.9%) said they had considered using a treatment or medication after viewing something on social media.

3.2. Bivariate Correlation Results

Table 4 describes the associations between social media use by platform, reason for use, types of accounts followed, and demographic variables. Language was statistically significantly associated with choice of platform: those who spoke Spanish were less likely to use X (Twitter), r(130) = -0.020, p=.021, but more likely to use WhatsApp, r(130) = 0.287, p=.001. Females were statistically significantly less likely to report the use of Instagram, r(130) = -0.207, p=.018, and reported being less likely to use SM for friend and family connection, r(130) = -0.207, p=.018, access news, r(130) = -0.209, p=.017, follow pop culture, r(130) = -0.196, p=.026 and entertainment, r(130) = -0.214, p=.014.). They were also less likely to follow a specific health figure or entity: health influencers, r(130) = -0.196, p=.026, health organizations, r(130) = -0.187, p=.033, health communities, r(130) = -0.186, p=.035, some other health account, r(130) = -0.198, p=.024. Those with a higher education level were more likely to use “X”, r(130) = 0.19, p=.031 while parents/caregivers were less likely to use “X”, r(130) = -0.034, p < .001. Furthermore, there was a very strong correlation between use of Instagram and likelihood of using social media to connect with family and friends, pop culture, news and entertainment, as well as following a health-related person/entity (influencer, organization, community other account) on social media. Lastly, those who used WhatsApp were more likely to use Snapchat and Facebook, r(130) = 0.218, p=.013 and r(130) = 0.287, p=.001, respectively.

The next set of correlation analyses (see Table 5) explored associations between healthcare and health information seeking patterns within certain demographics. Spanish speakers were less likely to consider a treatment or medicine, r(130) = -0.192, p=.028). No correlation was found between language or being a parent/caretaker and any social media platform or account. Older individuals were less likely to understand and relate with health information on social media r(130) = -0.210, p=.016 and r(130) = -0.211, p=.016 respectively. They were also less likely to report that health information from SM helped them to make health-related decisions, r(130) = -0.221, p=.016. Females were more likely to have a primary care physician r(130) = 0.281, p=.001, have annual healthcare appointments, r(130) = 0.249, p=.004, talk to the doctor about what they saw on social media, r(130) = 0.175, p=.046 and look up information after the doctor’s visit r(130) = 0.285, p=.001. They were also more likely to use SM to gain health information, r(130) = 0.207, p=.018, finding it relatable r(130) = 0.203, p=.021, and finding it helpful to make a health decision r(210) = 0.222, p=.011.

Those with higher education were more likely to have a primary care physician, r(130) = 0.176, p=.045 but also reported having to wait beyond 2 weeks for an appointment, r(130) = 0.274, p=.002 and to attribute the delay in making an appointment to financial reasons, r(130) = 0.224, p=.011. Employed participants were more likely to use social media for health information to make a healthcare decision, r(130) = 0.188, p=032, to delay an appointment due to finances, r(130) = 0.310, p<.001 and to have a chronic disease r(130) = 0.187, p=.033. Having healthcare insurance was associated with being more likely to attend annual medical visits, r(130) = 0.237, p=.007, having a PCP, r(130) = 0.340, p < .001 and to wait 2 weeks or more to receive an appointment, r(130) = 0.350, p < .001.

When exploring associations between health care variables, social media use for health information and demographic variables, we found a positive correlation between having a PCP and having seen a doctor in the past year, r(130) = 0.563, p < .001. In addition, those who had a PCP and had received care in the past year were more likely to use social media to seek health information, r(130) = 0.263, p=.003 and r(130) = 0.297, p=.001, respectively. Both groups were also more likely to find the social media health information content relatable, r(130) = 0.275, p=.002 and r(130) = 0.287, p=.001, respectively, to talk to their doctors during the visit about their social media search r(130) = 0.345, p < .001 and r(130) = 0.267, p = .002, respectively, and to seek information after the visit with a health care provider r(130) = 0.325, p< .001 and r(130) = 0.289, p = .001, respectively. They were also more likely to decide on treatment, r(130) = 0.206, p=.018 and r(130) = 0.173, p=.049, and to feel that social media information influenced their decision-making r(130) = 0.251, p=.004 and r(130) = 0.271, p=.002, respectively.

Furthermore, those who had a PCP and accessed care in the past year were more likely to make healthcare decisions based on the health information they saw on social media r(130) = 0.217, p=.013 and r(130) = 0.180, p=.040, respectively. They were also more likely to have to wait at least 2 weeks before getting an appointment r(130) = 0.379, p < .001 and r(130) = 0.352, p < .001, respectively. Participants with a PCP were more likely to understand health information found on social media r(130) = 0.218, p=.013 and to have a chronic disease r(130) = 0.209, p=.017. Those who had to wait at least two weeks for an appointment or delayed their appointment for financial reasons were more likely to also look up health information they might have heard about from their provider after this medical visit, r(130) =0.243, p=.005 and r(130) = 0.186, p=.034 respectively.

In the final correlation analyses (Table 6) we explored associations between social media platforms, types of accounts followed, and health information seeking behaviors. The only social media platform for which several statistically significant positive correlations with health information behavior was found is “Snapchat.” Its users were more likely to look up information on social media after a doctor’s visit, r(130) = 0.189, p=.031, to make health related decisions based on the content they saw, r(130) = 0.188, p=.032, and to be influenced to consider a medication or treatment r(130) = 0.199, p=.023. TikTok users were more likely to use health information from social media to decide on a specific treatment or medication r(130) =0.272, p=.002. Being an Instagram user was inversely correlated with easily finding health information on social media r(130) = -0.174, p=.048, finding it easy to understand r(130) = -0.255, p=.003, to relate to r(130) = -0.214, p=.015 and thinking that what was seen influenced their health decisions r(130) = -0.221, p=.012.

Lastly those who “followed” any of the 4 types of health accounts (health influencers, health organizations, health community or other health account) were less likely to understand social media health information (r(130) = -0.25, p=.004; r(130= -0.249, p=.004; r(130) = -0.248, p=.004; r(130) = -0.256, p=.003, to find it relatable (r (130)= -0.209, p=.017; r(130) = -0.193, p=.027; r(130) = -0.184, p=.036; r(130) = -0.217, p = .013) and to feel that it influenced their health decisions (r (130) = -0.219, p= .012; r(130) = -0.205, p = .020; r(130) = .019; r(130) = -0.205, r(130) = -0.196, p = .026).

3.3. Qualitative Results

All participants (N=7) in the qualitative discussions were female and the average age was 31.14 (ranging from 30 to 35 years old). Most (71.4%) identified themselves as Hispanic/Latino, followed equally by White and Asians (14.3%). All participants had at least a high school diploma or GED (see Table 7). Participants discussed their use of social media and what they experienced when looking for health information on social media. More specifically, they were asked if the information they found was easy to understand and relate to and what type of information they were more likely to implement or consider using. Themes (shown on Table 8) describe patterns of social media use (including choices of platforms and types of accounts), participants’ digital and health literacy - ability to find and understand health information, and use of social media for health decision-making.

3.4. Social Media Use

All participants used some type of SM account, with most (57.1%) using Instagram, TikTok, WhatsApp, and YouTube. Most agreed to using SM on a regular basis several hours a day but doing so in intervals spread across the day. Everyone initially said they used social media to follow friends and family but then also admitted to using it to access the news and other information. In regards to the format of the platform, there was a preference for TikTok because of the accessibility of the content. Participants also reported sharing the health information they found on social media with family members. The type of information they shared consisted of details that they found helpful in their own decision-making leading them to want to pass it along. In the context of sharing, the use of the WhatsApp platform as a means of sharing health information was referred to in a derisive manner.

3.5. Finding Health Information

Several participants said that they prefer searching for information on social media more than on Google. This was especially the case for TikTok users. They attributed their preference to ease of scrolling through the content of the messages. They also felt it was easier to ask questions on TikTok than on Google. However, a few still preferred using Google instead of social media platforms when searching for health information.

Participants also shared what types of accounts they preferred to follow and why. Everyone followed at least one health related influencer, and some also followed health organization. Others said they did not even realize health organizations had accounts, and one participant stated that she used TikTok the most. She also added that most organizations had an Instagram account but no TikTok accounts. In addition, almost every participant sought information for themselves and for family members. They were looking for more explanations about certain concepts and wanted to be able to make their own informed choice instead of being told what they should do.

3.6. Understanding Health Information

In understanding health information on social media, the problem wasn’t necessarily the content but rather the amount of information on a topic and then having to determine whether or not it is reliable and accurate information. Participants found the process overwhelming.

3.7. Using Health Information

Participants noted that they were more likely to use the information if there was a clear call to action. Another factor influencing use of information gained from SM seemed to be relatability. If the content they saw was relatable and they could see themselves engaging in the proposed behaviors, they were more likely to use the information. There was also an element of trust. If they trusted the source of information, they were more likely to use it. Trust and response to a call to action was also associated with relatability. The frequency with which participants heard a consistent message also increased their trust in the accuracy of the message. The effect of algorithms was evident, especially on TikTok’s “for you page” (FYP), as confirmed by one of the participants who accessed more baby health content after looking up remedies for colic.

4. Discussion

As society has become increasingly digitally dependent, the use of social media has reached such popularity that it is now an integral part of everyday life. The COVID-19 pandemic and the influx of information and misinformation online seemed to have been the catalysts that encouraged many organizations and institutions to establish a social media presence. The purpose of this study was to explore use of social media by an under-resourced population and the nature of their engagement with social media as a source of health information.

If there was ever a doubt that social media is widely used for health information in this population, participants (while overall on the younger side) in this study proved just that. Over 60% had accounts on three or more social media platforms and most of them used these daily. This was not limited to any particular type of account, as we found a strong, positive association of SM use for health information across all the categories of accounts (influencers, organizations and community). Although the use of the internet as a source of health information has been previously reported our findings support this to be true specifically in the context of social media [23,25,29]

As expected, having healthcare insurance was positively associated with access to health care but also with appointment delays, perhaps because of the current burden on the healthcare system, which affect any attempt at scheduling an appointment. Of note, after the peak of the COVID-19 pandemic crisis, the usual wait for visits has increased to 38 days [30,31].Thus, a delay in this post-COVID-19 era is mostly the default.

Considering that the topic of interest is social media, it not surprising that different generations and groups of people use SM in different ways. The younger the individual, the better their understanding of what was viewed on social media, and the more relatable and useful they found it in making health-related decisions. The average age of those agreeing to participate in the qualitative data collection (interviews) was slightly younger than that of those who participated in the quantitative data collection (31 years vs 34 years of age) and only females participated in further discussions after completing the survey. This is not surprising as the demographic variable that had the most significant associations across social media variables was gender. Although the correlation was weak, women (and not men) were less likely to use social media platforms to connect with friends, check the news, watch “pop culture” or entertainment content, and were less likely to have one or more of the listed accounts (influencer, health organizations...). They were however less likely to use Instagram.

Women were more likely to have a PCP and to attend their yearly visits. They were also more likely to share the information from social media and to search for health information after their medical visits. In other words, they combined social media findings with information garnered during the medical appointments to arrive at a decision. This is not too surprising considering women tend to be more invested in their health [27,32].

Another finding was the strong dependence on social media as a source of health information, seemingly of equal importance as that provided to them by their healthcare providers. The fact that individuals are communicating with their providers about content from social media and/or looking up health information on social media after seeing their provider makes the case for social media being a factor in health literacy and in influencing health behaviors. Those who had to wait at least two weeks for an appointment or experienced a delay due to financial reasons were more likely to go to social media for health information about something that was discussed with their health provider. The significant positive association with appointment delays and delays for financial reasons with talking to a doctor about HI on SM suggests that, even after waiting to see their healthcare providers, participants use social media to corroborate what they have been told during the visit. They may do so to ensure their wait and effort was worthwhile, which is in keeping with the increasing trends to self- and co-diagnosing [33,34]. This may also be an indication of lack of trust among those with lower incomes as there is a history of hesitancy among those most disenfranchised by society [35].

However, one would have expected that having a consistent primary care provider would have provided enough trust that patients would be less inclined to rely on social media to make decisions regarding their health, but this was not the case. Interestingly, patients who had a PCP and kept their annual medical appointments were more likely to use social media. Not only does this indicate a strong intersection between healthcare access and digital literacy in general (including navigation), but it also confirms the important place that social media information has in health decision-making of those who regularly visit their healthcare providers [33].

Those who had a PCP and attended their annual visits also reported more digital health literacy indicating that for them healthcare information on social media was easy to find, understand, and relate to. They also reported that they used what they learned on SM to inform their health decisions and to make treatment and medication-related decisions. While our data did not allow us to make assumption about the veracity about the information respondents get, it suggests that having access to social media may improve overall health literacy as individuals use social media to make their health decision and follow up on provider appointments. For those who worry about misinformation on social media, the fact that our respondents mainly used it to follow up on provider recommendations is encouraging. The fact that many also used SM to make health decisions, suggesting that the interplay between providers and patients about media information is important.

When looking at the descriptive information on overall social media use, more than half of participants followed influencers and to a degree “trusted” organizations that shared health information. In our qualitative discussions however, everyone looked up to influencers for health information, independently of whether or not they followed a “trusted” health organization. Again, further exploration could provide valuable insight in understanding if this is due to the format of messaging (less formal style of post, maybe a video which is not “overproduced”), since the only difference found between account types was in reach (number of individuals who saw the content).

According to our quantitative data, and consistent with previous reports, the Facebook platform had a high percentage of participants among social media users. Surprisingly, the platform with the highest number of account owners was Instagram, higher than the Facebook platform, which is unlike the average national trend [36].

The inverse correlation between Instagram use and the ability to find, relate with and understand information (health literacy) is somewhat unexpected and concerning, especially since a high percentage of our study participants use Instagram. Interestingly there was not necessarily a preference for any particular category of account (health organization or influencer). So, these results are not due to the type of account used to provide information. A possible reason may be due to the characteristics of individuals using Instagram such as low literacy. Or this could be the result of participants reaching a point of information overload, a phenomenon that was mentioned in our qualitative discussions and was a key component of the theme that relates to understanding of health information. Participants stated that they saw so much from so many sources about the same topic that they were overwhelmed. Since many have different types of accounts it could be that there is a point where access to information becomes too much and results in information overload [19,37]. Further exploration of these correlations is necessary to draw any meaningful conclusions.

Among those who participated in further discussions, Tik Tok, Instagram and WhatsApp had the most sub-accounts after Facebook, and TikTok seemed to be one of the favorite platforms. This is supported by data speaking to the increasing popularity of Tik Tok [38]. TikTok’s unique algorithm and the platform design seems to have been effective in not only attracting users but creating a dependency, with it competing with even Google in information searches.

This was validated in the focus group, with most participants who used TikTok admitting to it being the first stop when searching for information. Some preferred it over other platforms because of the anonymity it provided. In recent versions of the app, users must create an account to view content, but they can connect with other users and can remain relatively anonymous, even if they want to engage with content through comments.

Another stated reason for preferring TikTok was the format of the content. The platform has undergone several changes from its initial rule that content could only be viewed in 15 seconds and supported only “short form” content (short messages). It has since been adapted for “long form” content which lasts up to 10 minutes. In addition, it has recently also become an “influencer” platform.

Since participants in our qualitative discussions were either a parent or caretaker, they appreciated being able to scroll for a few minutes and go through several pieces of content in short periods of time. When searching specific information, they liked being able to scroll through videos from several accounts instead of having to click through websites and pause to read content, something that would be necessary on Google. Interestingly, those seeking health information on Tik Tok did so seeking explanations to questions they had to help them make informed decisions.

The ability to multitask and still gain knowledge during short periods of time is another perceived advantage of Tik Tok. Its change in algorithm and video experience resulted in its increasing popularity and, as a result, several other platforms made similar changes: Meta (another name for Facebook apps) overhauled its platform and algorithm to account for reels, YouTube introduced Shorts and even LinkedIn developed its own version of short interactions.

The most unexpected result was the positive association between Snapchat and various health information seeking behaviors, especially for decision-making. Snapchat was the only platform that was directly associated with decision-making related to health. The platform is not typically considered a resource for information but has evolved to incorporate that element with many organizations, news sites and podcasts using an informational voice. There are also several health influencers that have a considerable number of following on the platform and even the CDC regularly shares content there. Researchers are also considering the use of this platform as a health education tool especially for younger users[39]. As such it should be further explored.

Findings from this study confirm the widespread use of health information shared and consumed across all social media platforms, even though format, type of content and frequency may vary by audience. Based on the feedback from the discussion feedback about social media posts, it seems that users want information that is easily “digestible”, clear, using an “open” informative style rather than a “directive” style of communication (one that tells them what to do with no explanation of the reason behind the request). This points to a desire for social media to be a tool to empower individuals to make informed decisions that are best for themselves and their families. Participants clearly used SM for health guidance, and to ensure factual and unbiased information without being vulnerable, thus preserving their dignity and self-respect. Of note, few participants mentioned the use of SM from the usual trusted sources, a finding supported by Kwame and Petrucka [40]. This prompts the question of whether messages from trusted sources are truly created to engage this population and to explore how to best address this shortcoming.

On social media, health information can be accessed passively or actively. Active forms imply engagement from recipients and can include looking up a specific topic in a search on a platform or looking up a “creator” that shares information within a specific genre. Passive content is limited to posts that show up on a “feed” from accounts one already follows. Our paper did not distinguish between the two forms. This is important to understand this limitation as SM is driven by one’s own pattern of use: as one uses SM, more information of a similar type appears. In addition, a factor that was not explicitly shown in the data but is a clear mediating factor is the use of algorithms that SM platforms use to engage users. Depending on one’s searches and engagement with content, the algorithm usually offers similar types of health contents which is especially true on the TikTok “for you page” (FYP). Thus, it takes more than just producing a well-produced message by trusted sources. The complexity of the algorithm, which is partially influenced by the likelihood of viewer engagements, must be taken into account if the creator of the message is to reach its targeted audience.

Both the quantitative and qualitative results showed that users preferred content from influencers and were not necessarily consciously following health organizations or what would be considered a typical reputable source such as the Centers for Disease Control and Prevention (CDC) or a local county health department. Along those lines, often content from organizations tends to be very professional and “over-produced” with graphics that can seem out of touch (too “perfect”, formal and not applicable to the audience’s reality) in order to convey credibility. But it may be that relatability is what fosters trust and is more valued. FG participants said they liked hearing from people that they could relate with and that they could see themselves emulating. It is therefore important to understand that the appeal of the message may have as much to do with the way content is delivered as with the relatability of the influencer.

All of the findings from this study are important for health organizations to consider when creating content especially since these can influence the algorithm and thus determine the reach and the effectiveness of their messages. Engaging individual creators who understand what it takes to reach a specific audience when sharing social media content may be a valuable investment for health organizations that are serious about effectively communicating accurate health information to under-resourced populations

5. Strengths and Limitations

It is important to note the study’s limited sample size, the fact that it used a regional convenience sample to recruit lower income participants, its ratio of males to females and the relatively young age of our participants as a limitation. A larger, more balanced gender distribution may have lent itself to further insights, especially for males and our results cannot be generalized to other populations. The timing of the study may have also influenced recruitment as it took place during election season and during the change of the US government, a time with much uncertainty and a tense social and political climate. However, given the lack of available published studies about SM use and health information, we felt that our small mixed methods study was an acceptable way to begin to further explore this new area of research to better understand health information seeking patterns on social media in a region with under resourced populations.

6. Conclusion

It is evident that social media is widely used across many populations, including those living in an under-resourced region and that it has become an important source of health information for many users. However, there seems to be a “disconnect” between social media created by established credible sources of health information and the patterns of social media use among under-resourced populations, a group mainly exposed to non-organizational “influencers.” The results of this study could inform health organizations on the best ways to select platforms based on demographic characteristics, and more effectively connect with their audiences for improved health messaging to increase patient access to accurate health information.

Given that our respondents reported using SM after health care visits to deepen the information they received for health decision making, it is critically important for health organizations to provide engaging and accurate content to empower the consumers to help them make informed decisions and communicate more effectively with their healthcare providers. While our findings could provide critical information to healthcare entities to guide their social media reach, future research should extend this avenue of inquiry, especially considering the increasing burden on an understaffed health care system to provide quality care. This in turn could lead to more health literacy, a needed step towards implementing favorable health behaviors and ultimately result in better community health outcomes.