Social media use in healthcare: A systematic review of effects on patients and on their relationship with healthcare professionals

The aim of this paper is to gain insights in the benefits and challenges of the effects of social media use by patients within the healthcare system and especially the effects on patients and on their relationships with healthcare professionals. The effects we focus on in this paper can be both causal and reciprocal, but always start with the use of social media by patients.

Despite the popularity of social media, there is a confusion about what is exactly meant by the term social media. Therefore, in this paper we use the definition provided in the highly cited paper by Kaplan and Haenlein [9]. They describe social media as “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content”. The internet-based applications refer to the different categories of social media, which are blogs, content communities, social networking sites, collaborative projects, virtual game worlds and virtual social worlds. These types of social media are accessible to users to utilize for, among other things, health related reasons.

The term “users of social media in healthcare” in this paper refer to the patients and their family members. Patients are treated as any person who self-proclaims to be suffering from a certain condition, whether officially diagnosed by a healthcare professional or not. We define healthcare professionals as those who study, advise on or provide preventive, curative, rehabilitative and promotional health services based on an extensive body of theoretical and factual knowledge in diagnosis and treatment of conditions and other health problems [10].

For an article to be included in the study it had to meet several selection criteria as follows: (1) published in a peer-reviewed journal, (2) written in English, (3) full text available to the researcher, (4) contain primary empirical data, (5) the users of social media are patients, (6) the effects of patients using social media are clearly stated, (7) satisfy established quality criteria. The articles were assessed on their quality by using the standard quality assessment criteria as identified by [12].

Prior to final screening and selection of the papers, first and second author agreed to independently read 100 abstracts and select the articles that would be included in the study based on the selection criteria. Afterwards, the selected articles by the two authors were compared and there was complete concurrence on the category “yes, this one will be included”. For some of the articles that were marked as “maybe”, first and second author had a brief discussion to reach a consensus. This helped to reach higher reliability for the inclusion of the articles. Further in the process, the second author consulted the first author whenever there was a doubt whether to include or exclude the article. In addition, regular meetings with the third author also contributed to the overall process of the selection.

The resulting papers were characterized by the research aim and the type of research, which is reflected in the Table 1. The papers were further categorized according to the focus of the research question and data. Each paper’s empirical findings were categorized by looking at data and making first notes inductively. Following this, we looked at our notes on topics that emerged from analysed articles and compared them to earlier literature. In this way, concepts from prior literature helped us to make the sense of data from different articles and categorize them. A good example for that is the concept of social support, which we used to classify types of use. After analysing the articles in this way, we formulated propositions in the discussion section.

The searches were carried out in the period ending on March 17th, 2015. The application of the search strategy to the two search engines resulted initially in a total of 1,743 articles. Within the 1,743 articles many duplicates were found as well within the search engines as between the search engines. By removing duplicates the first found article was kept. In this way, we identified and removed 468 duplicates leaving us with 1,275 articles.

The remaining 1,275 articles were screened on title and abstract with regards to the selection criteria. Whenever we had doubts if an article is relevant or when title and abstract were not clear, we inspected the paper in more details by accessing full article. An article was removed when, for example, it became clear that the user of social media was not a patient but another user, like the hospital, a regular “healthy” person or healthcare professional. Additionally, several articles referred to internet use by patients for health related reasons and their effects, but did not specify the effects of social media. Therefore, such articles were removed. Moreover, articles that were written in a language other than English as well as articles that did not comprise primary data or did not elaborate on an effect of patients using social media. This left us with 22 articles that met our criteria. In addition, as a result of the referees’ suggestion to include term “client”, we identified one additional article, making the entire list of 23 articles for the quality assessment.

Quality of the articles was assessed by using the Standard Quality Assessment Criteria for Evaluating Primary Research Papers by [12] as presented in the Appendix C (See Additional file 3). This assessment tool distinguishes between qualitative and quantitative research and provides different quality assessment criteria for each type of research. The criteria are rated on their presence in the respective article and are either completely addressed in the article (resulting in 2 points), partly addressed (resulting in 1 point), or not addressed (resulting in 0 points). In case an article scored below the threshold of a 50 % score of the total amount of points possible, the article is assumed to be of low quality and removed from this paper. This cut-off point for inclusion is relatively liberal according to the authors of the assessment tool [12]. One article had a quality score below the 50 % cut-point and was excluded, which left us with the total of 22 articles for analysis.

The article selection process is shown in Fig. 1.

The Table 1 provides an overview of 22 articles included in the study. All studies except for three were published in or after 2010. Moreover, 19 articles were published in journals that are related to the medical field, whereas only three articles are published in journal that do not have a specific connection to medicine: Journal of Sociology, New Review of Hypermedia & Multimedia, and Information Research. Only two out of the 22 articles use a theory or a model to build their research on, namely the concept of masculinity [13] and the actant model [14]. The group of articles consists of nine quantitative, seven qualitative and six mixed methods studies.

The analysis of articles with regard to the type of social media and conditions is presented in the Appendix D (See Additional file 4), which shows that the 12 articles studied online support communities and most focused on chronic conditions. Other types of social media platforms and conditions were spread among the remaining articles.

This section presents findings from 22 articles we included in our study. First of all, an overview of the extracted findings is presented regarding the types of social media use by patients. Following this, we present the effects of social media use on patients. Subsequently, an overview of the extracted findings regarding effects of social media use by patients on the relationship between patients and healthcare professionals are presented, discussed, and categorized.

When patients are diagnosed with a certain condition that nobody in their close (offline) network has experienced before, patients can feel very lonely [27]. As a diabetic patient states “I literally felt like the only diabetic on the planet” [16]. However, social media provide an opportunity to easily connect with others and reduce this feeling of loneliness. Consequently, patients using social media for network support enhanced their psychological well-being. For example, social media provide means to connect with others in similar situations and this can break a patient’s loneliness [15]. This is in line with earlier studies that have shown how the existence of network support contributes to a better well-being of the patients [41, 42]. Interestingly, [41] suggest that the network support may not only benefit the patients themselves, but also their families who care for them. Yet, the relationship between the network support and psychological well-being may depend on the level of self-esteem. For example, college students with low self-esteem profited more from online social networking sites for bridging social capital and starting relationships than college students with high self-esteem [43]. In line with that, social networking sites provides the unique opportunity for patients to be able to talk about the sensitive aspects of the condition, as online communities provide the potential to reduce inhibitions felt in sharing experiences face to face [13]. Such an inhibition could reflect low self-esteem in terms of a reluctance to talk about the condition in face to face conversations.

Not all patients that make use of social media use it actively. Sometimes patients only use social media to read about other people’s stories, without actively contributing themselves. These people are called lurkers. The lurking behaviour may be related to the level of privacy concerns and computer anxiety [44]. In particular, anxiety leads to increase in lurking. Two articles in our sample were focused on the effects of patients using social media merely by reading other people’s stories. From the two articles, it becomes clear that the effects experienced by reading other people’s stories are being better informed [22, 26]. Additionally, by reading other people’s stories anxiety was found to significantly decrease [26]. Consequently, these findings suggest that reading other people’s stories on social media can lead to enhanced subjective well-being and improved self-management and control. However, [22] and [26] do not elaborate on the content of the stories read. Contrasting findings were found in other articles regarding how content affects the effects of reading other people’s stories. For example, cancer patients who read other people’s stories enhanced their subjective well-being [24]. Reading about success stories was found to enhance confidence to fight the condition, whereas reading about bad experiences prepared the patient mentally for difficult times ahead. On the other hand, the patients suffering from an inflammatory bowel disease who read other people’s stories about a bad experience suffered from diminished subjective well-being [25]. This is in line with earlier findings showing that the lack of sharing and feedback on this sharing may threaten the need for belonging [45]. Finally, patients suffering from infertility experienced diminished subjective well-being as the result of reading other people’s stories [35]. Reading stories about successful pregnancies led to increased feelings of jealousy, pain and a sense of alienation, whereas reading about bad experiences led to increased feelings of worry, anxiety and decreased optimism. Thus, this may lead to diminished subjective well-being. On the other hand, one study in our sample shows that this actually may enhance subjective-well-being [24]. In particular, this paper focused on blogs whereas other studies focused on online support groups [24]. Among other uses, blogs can be used as personals diaries to express thoughts, feelings, and stories [9]. Level of distress actually decreases when people blog about their emotional difficulties [46].

The effects of social media use by patients for health related reasons show that social media use by patients can lead to patient empowerment. Patient empowerment is an established concept in the medical research and has been promoted to foster patient autonomy [47]. As a result of the patient empowerment, patients may increasingly interact with their healthcare professional and get more involved in the decision making process [15]. In this case, social media can be seen as a “new” technology adopted by patients, which may shift the power balance between the healthcare professional and the patient. The use of new technologies in healthcare has been suggested as a way to empower end-consumers by enabling speed and convenience in accessing health related information [48]. In this line, the patients are able to actively participate in the interactions with healthcare professionals. On the other hand, the healthcare professionals may experience a decrease in power in the decision making process. According to the political variant of the interaction theory [49], “a product of the interaction of system features with the intra-organizational distribution of power, defined either objectively, in terms of horizontal or vertical power dimensions, or subjectively, in terms of symbolism can be resistance to the system”. Hence, redistribution of power between patients and healthcare professionals may cause the resistance from healthcare professionals. Yet, the role of health professionals has to change because embracing patient empowerment in healthcare means making a change, which sometimes seem difficult due to traditional approach, which is embedded in their current training [50].

However, increased patient involvement in the clinical interaction could potentially increase the risk placed on the healthcare professionals [13]. Healthcare professional may not be in complete control of the information used during decision making as the patient also has a voice, but the healthcare professional bears full responsibility for the decision taken. When patients bring in the information from social media to the consultation, this could lead to unnecessary processes of sorting relevant information from irrelevant information and can be experienced as challenging the healthcare professional’s expertise [1, 13]. Hence, based on these findings it is possible for healthcare professionals to resist this shift in the balance of power. However, increased equalization of the healthcare professional and patient communication can be a positive and desired effect. In particular, healthcare professionals may become more patient-centred, thus complementing the patient empowerment [51]. As a consequence of patient empowerment, we propose that the quality of clinical decision making may be enhanced.

According to the concept of bounded rationality [52], not all information can be gained on all available treatment options by healthcare professionals, as the human mind has a limited capacity to process the available information and often time is limited as well. Hence, healthcare professionals are unable to know all the information regarding treatment options and the newest developments, which affects their decision making. Thus, patients can extend this information base of the healthcare professional by specializing themselves in their own condition. This could provide an opportunity to increase the quality of the treatment decisions.

Notwithstanding the interesting results described above, this research has some limitations which, along with the three propositions, suggest opportunities for further research. It is possible that we missed some articles that could have used different terminology. Consequently, the results of this paper might not be generalizable for all social media platforms. For practical reasons, we excluded non-English papers. Finally, a limitation of every literature review is that the authors of the included articles will have had different objectives and used different methods and means of interpretation in reaching their conclusions. In this paper, we highlighted the most important findings on our topic of study and we categorized the key effects of social media use on patients and on their relationships with healthcare professionals.