The need to integrate interprofessional education (IPE) activities into health care education is a topic of growing importance. Members of the Interprofessional Education Collaborative, including national associations such as the American Association of Colleges of Nursing and Association of American Medical Colleges, value IPE and encourage institutions to create opportunities for health care trainees and practitioners from different disciplines to learn with, from, and about each other [1]. Although it is important to train health care professionals to work together effectively, many training programs have difficulty implementing IPE activities owing to schedule and location conflicts, lack of faculty support, and financial constraints [2].

The use of social media may provide a solution to many of the challenges inherent to IPE encountered by health care educators. Commonly used for personal social networking, social media tools provide a unique way for IPE learners to broaden their professional networks without the time and location constraints inherent to in-person networking [3]. In addition, these tools encourage a workforce prepared to adopt—with ease—new technologies for communication and care coordination [4]. Health care educators need to know the best ways to use social media in IPE to implement rigorous multidisciplinary learning. To address this gap in the literature, this review aims to identify papers describing studies of the use of social media in health care IPE and outcomes of social media use in IPE.

We searched the literature in 5 key databases (PubMed, Cumulative Index to Nursing & Allied Health Literature, Education Source, Education Full Text, and Academic Source Complete) from each database’s beginning coverage date through our search date (May 27, 2017) for papers on the use of social media in health care IPE. The databases were selected because of their comprehensive representation of available peer-reviewed journals, as well as their dedicated scope in the relevant disciplines to this topic (eg, medicine, nursing, allied health, and education). Key search concepts were selected with their Medical Subject Headings terms and other controlled vocabulary equivalents, as well as related keywords—“social media” OR “social networking” AND “interprofessional education (IPE)” OR “interprofessional learning” (Multimedia Appendix 1).

We then transferred titles and abstracts identified by this search into a reference management system. Two authors (ANR and MSC) independently evaluated each abstract to determine whether the paper potentially met the following a priori-determined inclusion criteria: accessible full-text, English language, peer-reviewed journal, IPE learners, social media educational intervention evaluation, and focus on IPE. If either reviewer found the abstract to meet the inclusion criteria, the paper advanced to full-text review and its PDF version retrieved and attached within the reference manager. Two authors (ANR and MSC) then independently evaluated the PDF of each full-text paper for inclusion. If the paper did not meet the inclusion criteria, each reviewer specified the reason for exclusion. A third reviewer (MT) resolved discordant reviewer responses.

Two authors (ANR and MSC) independently extracted relevant data from the included paper. The original version of the Kirkpatrick Model was used to assess each study’s outcome measures. Reviewers identified each study’s Kirkpatrick hierarchy level [5] and, on the Web-based software platform Covidence.org, assessed the quality of each study using the validated Medical Education Research Quality Instrument (MERSQI) for Quantitative Studies [6]. Discordant reviewer responses for data extraction, Kirkpatrick hierarchy level, and MERSQI scores were resolved by consensus of the authors. Kirkpatrick’s hierarchy was used to identify learning outcomes ranging from learner participation to changes in patient outcomes. In addition, Kirkpatrick’s Learning Model evaluates training programs’ impact on a hierarchy of 4 levels—perceptions (level 1) to outcomes (level 4). MERSQI scores were based on the study design, sampling, type of data, validity of an evaluation instrument, data analysis, and outcomes; scores could range from 5 to 18.

Our initial database search identified 48 unique titles, of which we selected 24 for full-text review (Figure 1). After full-text review, we determined that 8 papers met our inclusion criteria [2-4,6-11]. Of 8 studies identified by this review, 4 were single-group, posttest-only in design [3,4,10,11], 3 were single-group, pre- and posttest [7,8,9], and 1 was a nonrandomized 3-group study [2]. Multimedia Appendix 2 summarizes the results.

Study participants encompassed students from a broad range of professions as follows: nursing (7 studies), medicine (5 studies), pharmacy (5 studies), occupational therapy (2 studies), dentistry (2 studies), dental hygiene (1 study), physical therapy (1 study), radiography (1 study), and public health (1 study). A variety of social media tools and software were used as part of the educational intervention, with some studies using multiple tools as follows: blogs (2 studies), Wikis (2 studies), discussion boards (2 studies), Ning (1 study), and Second Life (1 study). Studies used quantitative (1 study), qualitative (4 studies), and mixed-method (3 studies) outcome measures. Most measured outcomes at the level of learner perceptions (5 studies), a few at the level of learner social media use behavior (3 studies), and none at the level of skills or patient outcomes. The included studies MERSQI scores ranged from 6 to 11.5.

The one single-group, posttest-only study that reported quantitative results found that 68% of students deposited information on (ie, contributed to) the Wiki, 42% edited the Wiki, and 20% visited the Wiki only to view content [10]. All remaining single-group, posttest studies reported qualitative results in the form of learner reflections. One study reported that students found the intervention to include respectful, engaging interaction among team members [3]. One study reported student reflections on learning, improvement, and innovations. These covered a range of topics, including culture, impact on practice, and module delivery [11], all of which received mixed student reviews. Furthermore, one study reported that the majority of student reviews were positive [4].

Of 3 single-group, pre- and posttest studies, only one compared learners’ pre- and posttest on validated quantitative measures [Interdisciplinary Education Perception Scale (IEPS) and Readiness for Interprofessional Learning Scale (RIPLS)] [9]. We found no significant difference before or after the intervention (P>.10). Of the other 2 single-group, pre- and poststudies, one compared medical, nursing, and pharmacy students on quantitative social media use metrics and found significant differences in the duration of use in weeks 1, 2, and 4 between medicine and both nursing and pharmacy students (P<.05) [7]. In addition, 2 of the 3 single-group, pre- and posttest studies reported qualitative results [8,9]. One had all students respond to a postintervention survey instrument and found that they were fairly split on whether the intervention should be used again, although many students had positive remarks on the flexibility the intervention provided and the impact on their learning [9]. The other reported extremely positive feedback with all participants wanting to continue paired learning in their professional development [8].

Only one study, which received the highest MERSQI score, was designed to compare multiple interventions delivered to interprofessional groups of learners and found no significant differences in IEPS and RIPLS scores among the 3 groups, although no P value was reported [2]. In addition, these 3 groups did not significantly differ in their social media use metrics (the number of posts and time spent on site) showed (P=.47).