- Email: [email protected]
Professional Social Networking in Radiology: Who Is There and What Are They Doing?
ARTICLE IN PRESS
Original Investigation
Professional Social Networking in Radiology: Who Is There and What Are They Doing? Sumir S. Patel, MD1, C. Matthew Hawkins, MD, James V. Rawson, MD, Jenny K. Hoang, MBBS Rationale and Objectives: Although it is perceived that the use of social media professionally is increasing among radiologists, little is known about the habits and demographics of this subspecialty. This study aims to compare radiologists who use social networking for professional purposes to those who do not with regard to their characteristics, habits, and attitudes. Materials and Methods: Radiologists were invited by e-mail and through posts on social networks to participate in a survey on the use of social media platforms. Questions included type of user, pattern of use, and benefits and barriers. Professional users and professional nonusers were compared. Results: One hundred eighty-six radiologists responded. One hundred ten (59.1%) used social networking for professional purposes, 34 (18.2%) for personal-use only, and 42 (22.6%) denied using social media. LinkedIn was the most common platform among all professional users, and Twitter was the most commonly used platform among highly active professional users. Trainees comprised 52 out of 110 (47.3%) professional social networking users compared to 18 out of 76 (23.7%) nonusers (P < 0.01). A subgroup analysis on Twitter use for professional purposes revealed a significant gender difference: 15 out of 66 (22.7%) professional Twitter users were female compared to 48 out of 120 (40.0%) non-Twitter users (P < 0.05). The greatest barrier to professional social media use for nonusers was confidentiality. Conclusion: Nearly 60% of radiologist respondents use social networking for professional purposes. Radiology is likely to see growth in the role of social networking in the coming years as nearly half of professional users are radiology trainees. Twitter use for professional purposes among radiologists was disproportionately male. It is important to be cognizant of gender imbalance and to improve visibility of female leaders on social networking. Key Words: Social media; social network; radiology; survey; professional. © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
INTRODUCTION
S
ocial networking in healthcare is a growing phenomenon (1,2). Early users were patients interacting with other patients in the form of online forums and automated mailing lists, known as listservs. Later patients and healthcare professionals used dedicated platforms for online social interaction, such as Facebook, Twitter, and LinkedIn. Recently, physician-to-physician interaction on social networking is growing, with users finding benefits of obtaining education and news, following medical conferences, and sharing expertise, research, and opinions (3). The American Medical Association states that social networking can “support physicians’ Acad Radiol 2017; ■:■■–■■ From the Department of Radiology, Duke University Medical Center, Box 3808, Erwin Road, Durham, NC 27710 (S.S.P., J.K.H.); Department of Radiology and Imaging Sciences, Children’s Healthcare of Atlanta, Egleston Campus, Emory University, Atlanta (C.M.H.); Department of Radiology and Imaging, Medical College of Georgia, Augusta University, Augusta, Georgia (J.V.R.). Received July 21, 2015; revised March 30, 2016; accepted September 28, 2016. 1Present address: Department of Radiology and Imaging Sciences, Emory University, 1364 Clifton Road, NE Suite D112, Atlanta, GA 30322. Address correspondence to: J.K.H. e-mail: [email protected] © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.acra.2016.09.026
personal expression, enable individual physicians to have a professional presence online, foster collegiality and camaraderie within the profession, provide opportunity to widely disseminate public health messages and other health communication” (4). A 2011 survey of 4033 physicians by QuantiaMD found that physicians are highly engaged with social networks: 90% of physicians report personal use and 65% of physicians use social media for professional reasons (5). A later survey of Australian physicians by Brown et al. found that 74% used social media networks to some extent (6). None of these surveys queried use among specialties in medicine. There are no studies that describe the use of social networking among individual radiologists. Although professional social networking may offer advantages to all physicians, there are also specific reasons to understand the radiologists’ use of social networking. Radiologists are an important part of healthcare delivery, but are perceived as being invisible. Social media may present an opportunity for radiologists to interact with other health professionals and the public to provide information to patients, in line with the principles of Imaging 3.0 (7–9). It is particularly important for radiologists to be actively engaged on social networking when topics discussed pertain to radiology. A study of Twitter content posted in 2013 regarding 1
PATEL ET AL
radiation risk from computed tomography found that 59% of tweets were unfavorable (10). However, most articles were not peer-reviewed, and only 16% of unique users were physicians and only 3% were radiologists. Social networking for radiologists has also been highlighted in recent articles describing the advantages for professional development, Twitter trends at radiology conferences, and the impact of social media on readership of peer-reviewed radiology journals (11,12). Another study showed that dissemination of scientific material on a radiology blog promoted on Facebook can substantially augment the reach of more traditional publication venues (3). As more professional radiology organizations use social media to disseminate information and reach members, understanding which radiologists use social media for professional purposes is valuable for planning and predicting future trends in social media use in radiology (13–15). This study aims to compare the radiologists who use social networking for professional purposes to those who do not with regard to their characteristics, habits, and attitudes. METHODS This study was approved by our institutional review board. The need for written informed consent was waived due to the design of this survey study and anonymity of the survey respondents. Study Group
A survey was designed to query radiologists about their use of social networking. The multiple-choice survey was created using online survey software (Qualtrics, Provo, UT). Radiologists were invited to participate in the survey between December 21, 2014 and March 21, 2015 by e-mail solicitation from the authors’ personal radiologist networks and by posts on social media platforms, Twitter and Facebook. The e-mail networks included approximately 100 trainees and faculty in the senior author’s radiology department and 65 radiology colleagues from private and academic institutions in the United States. There were six Twitter and one Facebook social media posts from the senior author. The posts contained a message to participate in the survey and the web address for the survey. We estimated the target audience by Twitter “impressions.” An impression represents the tweet has been delivered to the Twitter stream of a particular account. The tweet may not have been read, so it is overestimating the potential audience. It was not possible to estimate audience reached by Facebook. Survey
The survey questions focused on four main categories of data: type of social media user, pattern of use (platforms, duration, frequency), benefits of use, and barriers to use (Table 1). The types of social network use were (1) professional use, (2) personal-only use, and (3) no social media use. Professional users were defined as radiologists who had social 2
Academic Radiology, Vol ■, No ■■, ■■ 2017
TABLE 1. Survey Questions Defining the user Do you use social media? Which of the following social media services do you use for networking with friends and family (personal account)? Which of the following social media services do you use for networking with colleagues or for educational reasons (professional account)? Which best describes your social media accounts? Pattern of use How long have you used social media in a professional capacity? How would you characterize your usage of social media services for professional purposes? How would you characterize the amount of time spent on social media services for professional purposes? Benefits How have you used or benefited from social media professionally? Barriers What are some barriers to using social media for professional purposes? Radiologist characteristics What is your gender? To which “generation” described below do you consider yourself to belong? Where are you in your radiology career? What is your radiology practice type? What is your radiology subspecialty by fellowship or practice?
networking accounts for which any use was related to radiology. This category is inclusive of those users who use social networking accounts for both professional and personal purposes. Personal-only users were defined as radiologists who had social networking accounts for which no radiologyrelated activities transpired. Radiologists with no social media use did not possess any type of social networking account. All respondents answered questions that characterized them by gender, age group, career position, and practice type. Personal-only and no social media user respondents did not answer questions about pattern and benefits of social networking (Table 1). Age groups were categorized by generations. Respondents were asked which generation they identify with. Baby Boomers were defined as being born from 1943 up to the early 1960s. Generation X was defined as being born from the early 1960s to the early 1980s. Generation Y was defined as being born from the late 1980s to the early 2000s. Additional categories included The Greatest Generation, defined as being born from around 1901 through 1924, and The Silent Generation, defined as being born from 1925 until 1942. Analysis of Survey Results
Radiologists who used social networking for professional purposes were compared to the professional nonusers (personal-only
Academic Radiology, Vol ■, No ■■, ■■ 2017
and no social media use). The characteristics of the respondents were compared for these two groups and the chi-square test was used to test differences. A P value of less than 0.05 was considered statistically significant. Th results from the online survey were exported into a spreadsheet for analysis (Excel, Microsoft, Redmond, WA). Statistical analyses were performed using SAS Enterprise (Version 4.2, SAS Institute, Cary, NC). RESULTS Study Group
Two hundred four responses were received, and 18 (8.8%) of the responses were excluded because they were incomplete. Of the 186 complete responses, 124 (66.7%) responded to the e-mailed survey and 62 (33.3%) responded to social media promotion. The response rate for the e-mailed survey was 75.2% (124 out of 165). From six tweets, the impressions, engagements, and link clicks were 4267, 258, and 72, respectively. Using impressions as potential audience reached, the response rate was 1.4% (62 out of 4267). Table 2 shows the characteristics of the respondents. There were 123 (66.1%) male respondents, 119 (64.0%) were in academic practices, and nearly half of the respondents (85, 45.7%) self-identified as belonging to Generation X, followed by Generation Y (62, 33.3%) and Baby Boomers (39, 21.0%). Regarding use of social networking, 110 (59.1%) were users for professional purposes, 34 (18.3%) were for personal-only, and 42 (22.6%) had no use. Of those respondents who use social media for professional purposes, 55 out of 110 (50.0%) stated that they kept the personal and professional accounts separate. Comparison of Users and Nonusers of Professional Social Networking
There were significant differences in the career position and generation between professional users and nonusers (Table 2). Fifty two out of 110 (47.3%) professional social networking users were in training as either residents or fellows compared to 18 out of 76 (23.7%) nonusers (P < 0.01). Generation Y comprised 46 out of 110 (41.8%) professional social network users compared to 16 out of 76 (21.1%) nonusers (P < 0.01). In contrast, Baby Boomers generation comprised 16 out of 110 (14.5%) professional social network users and 23 out of 76 (30.3%) nonusers (P < 0.01). There were no significant differences in the gender or practice type between professional users and nonusers. However, on a subgroup analysis of users of Twitter for professional purposes, there were gender differences: 15 out of 66 (22.7%) users were female compared to 48 out of 120 (40.0%) nonusers (P < 0.05). Habits of Radiologists and Benefits of Using Social Networks for Professional Purposes
The radiologists who used social networking for professional purposes describe their usage as “consuming and rarely
PROFESSIONAL SOCIAL NETWORKING IN RADIOLOGY
contributing” in 44.5% (49 out of 110), and “both consume and frequently contribute” in 30.9% (34 out of 110) and “only consume” in 24.5% (27 out of 110). The three most popular platforms for radiologists using social networking for professional purposes were LinkedIn (74 out of 110, 67.3%), Twitter (66 out of 110, 60.0%), and Doximity (44 out of 110, 40.0%). Most users had participated in social networking for 2–5 years (36 out of 110, 32.7%) or more than 5 years (24 out of 110, 21.8%), but there were also new users, 19 (17.2%) of them had been using social networking for less than 6 months, 8 (7.2%) for 6–12 months, and 23 (20.9%) for 1–2 years. Users were most commonly active several times daily (42 out of 110, 38.2%), followed by daily to several times per week (32 out of 110, 29.1%), weekly to several times per month (18 out of 110, 16.4%), and monthly or less (18 out of 110, 16.4%). Of the 42 radiologists who were active several times per day, 34 (81.0%) used Twitter, 28 (66.7%) used LinkedIn, and 14 (33.3%) used Facebook. Career positions of these highly active users were predominantly trainees (18 out of 42, 42.9%). Other highly active users included 7 (16.7%) who were in practice for <8 years, 5 (11.9%) in practice for 8–15 years, and 12 (28.6%) in practice for >15 years. Among these highly active users, only 6 (14.3%) were female, and 5 of the 6 females were trainees. Radiologists found many benefits to using social networks professionally, with 86 (78.2%) using their preferred platform to follow news from professional organizations. Radiologists also found value by communicating with their colleagues (69 out of 110, 62.3%), following events or news during scientific meetings (65 out of 110, 59.1%), and sharing and following radiology cases (63 out of 110, 57.3%). Barriers to Using Social Networks for Professional Purposes
The top four barriers to using social networks for professional purposes were the same for users and nonusers. These were not enough time, concerns over confidentiality, concerns over an unprofessional stigma, and unfamiliarity with the technical aspects of the medium (Table 2). The most commonly selected barrier differed between the users and nonusers. Not enough time was the most common barrier for professional users (57 out of 110, 51.8%). The most common barrier for nonusers was concerns over confidentiality (46 out of 76, 60.5%). DISCUSSION As radiologists become increasingly engaged in social networking platforms for professional purposes, it is necessary to consider the characteristics of both current users and nonusers. An understanding of trends in use and the benefits and barriers to use is key not only for all stakeholders, including the radiologists themselves, but also for professional radiology organizations, physicians of other specialties, and patients. 3
Academic Radiology, Vol ■, No ■■, ■■ 2017
PATEL ET AL
TABLE 2. Comparison of Users and Nonusers of Social Networking for Professional Purposes
All 186
Users of Social Media for Professional Purposes 110
Nonusers of Social Media for Professional Purposes 76
Gender
0.3045 Male Female
123 (66.1%) 63 (33.9%)
76 (69.1%) 34 (30.9%)
47 (61.8%) 29 (38.2%)
Academic Private Combined
119 (64.0%) 37 (19.9%) 30 (16.1%)
72 (65.5%) 23 (20.9%) 15 (13.6%)
47 (61.8%) 14 (18.4%) 15 (19.7%)
Practice type
0.5303
Generation
0.0034 “The Greatest Generation” “The Silent Generation” “Baby Boomers” “Generation X” “Millennials/Generation Y”
0 0 39 (21.0%) 85 (45.7%) 62 (33.3%)
16 (14.5%) 48 (43.6%) 46 (41.8%)
23 (30.3%) 37 (48.7%) 16 (21.1%)
Resident or Fellow In practice for <8 years In practice for 8–15 years In practice for >15 years
70 (37.6%) 39 (21.0%) 23 (12.4%) 54 (29.0%)
52 (47.3%) 17 (15.5%) 12 (10.9%) 29 (26.4%)
18 (23.7%) 22 (29.0%) 11 (14.5%) 25 (33.0%)
Career position
Social media networks used for professional purposes Benefits of professional social media activity
Barriers to professional social media activity
P Value
0.0086
LinkedIn Twitter Doximity
74 (67.3%) 66 (60.0%) 44 (40.0%)
News from professional organizations Communicate with colleagues Scientific meetings Education
86 (78.2%) 69 (62.3%) 65 (59.1%) 63 (57.3%)
Not enough time Concerns over confidentiality Concerns over unprofessional stigma Unfamiliarity with technical aspects
99 (53.2%) 78 (41.9%)
57 (51.8%) 32 (27.0%
42 (55.3%) 46 (60.5%)
56 (30.1%)
25 (21.0%)
31 (40.1%)
43 (23.1%)
15 (12.6%)
28 (36.8%)
Social media activity Consume only Consume and rarely contribute Consume and frequently contribute
This survey study finds that there is enormous potential for growth in social networking as nearly half of the users were radiology trainees. There are no significant gender differences in professional users and nonusers overall, but users of Twitter for professional social networking were disproportionately male. The higher proportion of trainees and Generation Y radiologists using social networking in our study is not surprising. Younger radiologists have grown up in a world of being immersed in online communities and are more likely to be technologically savvy. Our results agree with Brown et al.’s 4
27 (24.5%) 49 (44.5%) 34 (30.9%)
physician survey, which found a linear relationship between increasing age and decreasing social media use (6). In another study of Facebook users in the surgical specialty, 64% of residents had Facebook accounts compared to only 22% of faculty (16). Our study confirms and quantifies the differences in user age groups in radiology. Nearly half of the users in radiology were trainees and more than 85% of users were from Generation X or Y. In particular, Generation Y also brings the traits of being collaborative, creative, and having a positive “can-do” attitude (17). As these younger radiologists rise to leadership positions, social networking in the professional setting
Academic Radiology, Vol ■, No ■■, ■■ 2017
will only increase. There will be opportunities to change radiology education and engage with patients regarding imaging topics (10). Users of professional social networking describe numerous benefits, but it is the barriers among nonusers that are interesting to study because they have impact on future use. In our study, 61% of nonusers were concerned about confidentiality and regarded this as a factor preventing their use. Confidentiality concerns relate to maintaining patient confidentiality as well as the radiologist’s personal confidentiality. Brown et al. found similar results, with 66% of physicians being hesitant to immerse themselves in social media because of worries about public access and legal concerns (6). The American Medical Association specifically advises on professional conduct on social media and state that physicians should consider separating personal and professional content online (4). In our study, half of the professional users had this insight and did maintain separate professional and personal profiles. Ventola warns about damage to healthcare professional’s image with online behavior that could be construed as unprofessional, such as violations of patient privacy and inappropriate comments and photographs, but such cases are extremely rare (18). Radiologists who are cognizant of ethics and professionalism can use social networking with minimal risk to confidentiality, and experience the benefits highlighted by the users in this study. An important finding was differences in gender between the radiologists who use Twitter for professional purposes. Twitter is the second most common platform among all professional users and the most commonly used platform among highly active professional users. However, only one in four users was a female radiologist and only 14% of highly active users were female. This gender imbalance among Twitter users is likely multifactorial. Concerns about confidentiality may be more pronounced among female users of social networking, as shown in studies outside healthcare (19). Additionally, female radiologists may have less time to explore and participate on social networking platforms. A study of physicianresearchers has shown large gender differences in domestic and parenting responsibilities, which may contribute to different rates of career success (20). The impact of fewer female Twitter users has the potential to perpetuate the gender imbalance as there are fewer female radiology role models. In our study, only one of the six highly active female users was an attending radiologist. Awareness of gender imbalances has been raised by a Twitter-based application called Twee-Q (twee-q.com), which provides a score derived from how often a Twitter user retweets men or women. These gender differences suggest that different strategies may be needed to increase social media engagement among male and female radiologists. This study has several limitations. First, there is selection bias because the survey was sent to radiologists by e-mail and social media. E-mail distribution to the authors’ contacts leads to selection bias to include more academic radiologists and younger radiologists, and may not be generalizable to the current workforce of radiologists. Promotion on Facebook
PROFESSIONAL SOCIAL NETWORKING IN RADIOLOGY
and Twitter could overestimate the use of social networks and the use of these platforms. Additionally, the estimations of the response rate by Twitter impressions were crude and likely a gross underestimation given that impressions do not equate to the tweet being read. Finally, there is a nonresponse bias, which is inherent in survey studies. We predict that nonresponse bias would lead to overestimation of the proportion of professional social network users as users of social media are more likely to respond and share their experiences. Despite this, we saw similar proportions of social networking users compared to two other physician surveys on social media (5,6). CONCLUSION Nearly 60% of radiologist respondents use social networking for professional purposes. Radiology is likely to see growth in the number of users and in the role of social networking in the coming years as nearly half of professional users are radiology trainees. Twitter use for professional purposes among radiologists was disproportionately male, and especially among highly active users. It is important to be cognizant of gender imbalance and to improve visibility of female leaders on social networking. REFERENCES 1. AMN Healthcare, Inc. 2013 survey of social media and mobile usage by healthcare professionals. AMN Healthcare, Inc., 2013. 2. Antheunis ML, Tates K, Nieboer TE. Patients’ and health professionals’ use of social media in health care: motives, barriers and expectations. Patient Educ Couns 2013; 92:426–431. 3. Hoang JK, McCall J, Dixon AF, et al. Using social media to share your radiology research: how effective is a blog post? J Am Coll Radiol 2015; 12:760–765. 4. American Medical Association. Opinion 9.124 – professionalism in the use of social media. AMA Journal of Ethics 2015; 17:432–434. Available at: http://journalofethics.ama-assn.org/2015/05/coet1-1505.html. Accessed June 15, 2015. 5. Modahl M, Tompsett L, Moorhead T. Doctors, patients and social media. QuantiaMD. 2011. Available at: http://www.quantiamd.com/q-qcp/social _media.pdf. Accessed June 15, 2015. 6. Brown J, Ryan C, Harris A. How doctors view and use social media: a national survey. J Med Internet Res 2014; 16:e267. 7. Pathiraja F, Little D. Social media: the next frontier in radiology. Clin Radiol 2015; 70:585–587. 8. Kotsenas A. 7 reasons neuroradiologists should use Twitter. AJNR Blog, 2015. Available at: http://www.ajnrblog.org/2015/03/27/7-reasons -neuroradiologists-should-use-twitter/. Accessed June 15, 2015. 9. Matthews G. Why the evolution of radiology includes social media. WCG Common Sense, 2013. Available at: http://blog.wcgworld.com/2013/ 09/why-the-evolution-of-radiology-includes-social-media-sthash .MlJ6IqEM.dpuf. Accessed June 15, 2015. 10. Prabhu V, Rosenkrantz AB. Imbalance of opinions expressed on Twitter relating to CT radiation risk: an opportunity for increased radiologist representation. AJR Am J Roentgenol 2015; 204:W48–W51. 11. Hawkins CM, Duszak R, Rawson JV. Social media in radiology: early trends in twitter microblogging at radiology’s largest international meeting. J Am Coll Radiol 2014; 11:387–390. 12. Hawkins CM, Hillman BJ, Carlos RC, et al. The impact of social media on readership of a peer-reviewed medical journal. J Am Coll Radiol 2014; 11:1038–1043. 13. Naeger DM, Webb EM. Social media for professional purposes: introduction to the JACR “how to” video guide. J Am Coll Radiol 2013; 10:736–737. 14. Krishnaraj A. The history of the RFS. J Am Coll Radiol 2010; 7:382– 383.
5
PATEL ET AL
15. Krishnaraj A, Weinreb JC, Ellenbogen PH, et al. Impact of generational differences on the future of radiology: Proceedings of the 11th Annual ACR Forum. J Am Coll Radiol 2012; 9:104–107. 16. Landman MP, Shelton J, Kauffmann RM, et al. Guidelines for maintaining a professional compass in the era of social networking. J Surg Educ 2010; 67:381–386. 17. Slanetz PJ, Kung J, Eisenberg RL. Teaching radiology in the millennial era. Acad Radiol 2013; 20:387–389.
6
Academic Radiology, Vol ■, No ■■, ■■ 2017
18. Ventola CL. Social media and health care professionals: benefits, risks, and best practices. P T 2014; 39:491–520. 19. Hoy MG, Milne G. Gender differences in privacy-related measures for young adult Facebook users. J Interact Advert 2010; 10:28–45. 20. Jolly S, Griffith KA, DeCastro R, et al. Gender differences in time spent on parenting and domestic responsibilities by highachieving young physician-researchers. Ann Intern Med 2014; 160:344– 353.
Original Investigation
Professional Social Networking in Radiology: Who Is There and What Are They Doing? Sumir S. Patel, MD1, C. Matthew Hawkins, MD, James V. Rawson, MD, Jenny K. Hoang, MBBS Rationale and Objectives: Although it is perceived that the use of social media professionally is increasing among radiologists, little is known about the habits and demographics of this subspecialty. This study aims to compare radiologists who use social networking for professional purposes to those who do not with regard to their characteristics, habits, and attitudes. Materials and Methods: Radiologists were invited by e-mail and through posts on social networks to participate in a survey on the use of social media platforms. Questions included type of user, pattern of use, and benefits and barriers. Professional users and professional nonusers were compared. Results: One hundred eighty-six radiologists responded. One hundred ten (59.1%) used social networking for professional purposes, 34 (18.2%) for personal-use only, and 42 (22.6%) denied using social media. LinkedIn was the most common platform among all professional users, and Twitter was the most commonly used platform among highly active professional users. Trainees comprised 52 out of 110 (47.3%) professional social networking users compared to 18 out of 76 (23.7%) nonusers (P < 0.01). A subgroup analysis on Twitter use for professional purposes revealed a significant gender difference: 15 out of 66 (22.7%) professional Twitter users were female compared to 48 out of 120 (40.0%) non-Twitter users (P < 0.05). The greatest barrier to professional social media use for nonusers was confidentiality. Conclusion: Nearly 60% of radiologist respondents use social networking for professional purposes. Radiology is likely to see growth in the role of social networking in the coming years as nearly half of professional users are radiology trainees. Twitter use for professional purposes among radiologists was disproportionately male. It is important to be cognizant of gender imbalance and to improve visibility of female leaders on social networking. Key Words: Social media; social network; radiology; survey; professional. © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
INTRODUCTION
S
ocial networking in healthcare is a growing phenomenon (1,2). Early users were patients interacting with other patients in the form of online forums and automated mailing lists, known as listservs. Later patients and healthcare professionals used dedicated platforms for online social interaction, such as Facebook, Twitter, and LinkedIn. Recently, physician-to-physician interaction on social networking is growing, with users finding benefits of obtaining education and news, following medical conferences, and sharing expertise, research, and opinions (3). The American Medical Association states that social networking can “support physicians’ Acad Radiol 2017; ■:■■–■■ From the Department of Radiology, Duke University Medical Center, Box 3808, Erwin Road, Durham, NC 27710 (S.S.P., J.K.H.); Department of Radiology and Imaging Sciences, Children’s Healthcare of Atlanta, Egleston Campus, Emory University, Atlanta (C.M.H.); Department of Radiology and Imaging, Medical College of Georgia, Augusta University, Augusta, Georgia (J.V.R.). Received July 21, 2015; revised March 30, 2016; accepted September 28, 2016. 1Present address: Department of Radiology and Imaging Sciences, Emory University, 1364 Clifton Road, NE Suite D112, Atlanta, GA 30322. Address correspondence to: J.K.H. e-mail: [email protected] © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.acra.2016.09.026
personal expression, enable individual physicians to have a professional presence online, foster collegiality and camaraderie within the profession, provide opportunity to widely disseminate public health messages and other health communication” (4). A 2011 survey of 4033 physicians by QuantiaMD found that physicians are highly engaged with social networks: 90% of physicians report personal use and 65% of physicians use social media for professional reasons (5). A later survey of Australian physicians by Brown et al. found that 74% used social media networks to some extent (6). None of these surveys queried use among specialties in medicine. There are no studies that describe the use of social networking among individual radiologists. Although professional social networking may offer advantages to all physicians, there are also specific reasons to understand the radiologists’ use of social networking. Radiologists are an important part of healthcare delivery, but are perceived as being invisible. Social media may present an opportunity for radiologists to interact with other health professionals and the public to provide information to patients, in line with the principles of Imaging 3.0 (7–9). It is particularly important for radiologists to be actively engaged on social networking when topics discussed pertain to radiology. A study of Twitter content posted in 2013 regarding 1
PATEL ET AL
radiation risk from computed tomography found that 59% of tweets were unfavorable (10). However, most articles were not peer-reviewed, and only 16% of unique users were physicians and only 3% were radiologists. Social networking for radiologists has also been highlighted in recent articles describing the advantages for professional development, Twitter trends at radiology conferences, and the impact of social media on readership of peer-reviewed radiology journals (11,12). Another study showed that dissemination of scientific material on a radiology blog promoted on Facebook can substantially augment the reach of more traditional publication venues (3). As more professional radiology organizations use social media to disseminate information and reach members, understanding which radiologists use social media for professional purposes is valuable for planning and predicting future trends in social media use in radiology (13–15). This study aims to compare the radiologists who use social networking for professional purposes to those who do not with regard to their characteristics, habits, and attitudes. METHODS This study was approved by our institutional review board. The need for written informed consent was waived due to the design of this survey study and anonymity of the survey respondents. Study Group
A survey was designed to query radiologists about their use of social networking. The multiple-choice survey was created using online survey software (Qualtrics, Provo, UT). Radiologists were invited to participate in the survey between December 21, 2014 and March 21, 2015 by e-mail solicitation from the authors’ personal radiologist networks and by posts on social media platforms, Twitter and Facebook. The e-mail networks included approximately 100 trainees and faculty in the senior author’s radiology department and 65 radiology colleagues from private and academic institutions in the United States. There were six Twitter and one Facebook social media posts from the senior author. The posts contained a message to participate in the survey and the web address for the survey. We estimated the target audience by Twitter “impressions.” An impression represents the tweet has been delivered to the Twitter stream of a particular account. The tweet may not have been read, so it is overestimating the potential audience. It was not possible to estimate audience reached by Facebook. Survey
The survey questions focused on four main categories of data: type of social media user, pattern of use (platforms, duration, frequency), benefits of use, and barriers to use (Table 1). The types of social network use were (1) professional use, (2) personal-only use, and (3) no social media use. Professional users were defined as radiologists who had social 2
Academic Radiology, Vol ■, No ■■, ■■ 2017
TABLE 1. Survey Questions Defining the user Do you use social media? Which of the following social media services do you use for networking with friends and family (personal account)? Which of the following social media services do you use for networking with colleagues or for educational reasons (professional account)? Which best describes your social media accounts? Pattern of use How long have you used social media in a professional capacity? How would you characterize your usage of social media services for professional purposes? How would you characterize the amount of time spent on social media services for professional purposes? Benefits How have you used or benefited from social media professionally? Barriers What are some barriers to using social media for professional purposes? Radiologist characteristics What is your gender? To which “generation” described below do you consider yourself to belong? Where are you in your radiology career? What is your radiology practice type? What is your radiology subspecialty by fellowship or practice?
networking accounts for which any use was related to radiology. This category is inclusive of those users who use social networking accounts for both professional and personal purposes. Personal-only users were defined as radiologists who had social networking accounts for which no radiologyrelated activities transpired. Radiologists with no social media use did not possess any type of social networking account. All respondents answered questions that characterized them by gender, age group, career position, and practice type. Personal-only and no social media user respondents did not answer questions about pattern and benefits of social networking (Table 1). Age groups were categorized by generations. Respondents were asked which generation they identify with. Baby Boomers were defined as being born from 1943 up to the early 1960s. Generation X was defined as being born from the early 1960s to the early 1980s. Generation Y was defined as being born from the late 1980s to the early 2000s. Additional categories included The Greatest Generation, defined as being born from around 1901 through 1924, and The Silent Generation, defined as being born from 1925 until 1942. Analysis of Survey Results
Radiologists who used social networking for professional purposes were compared to the professional nonusers (personal-only
Academic Radiology, Vol ■, No ■■, ■■ 2017
and no social media use). The characteristics of the respondents were compared for these two groups and the chi-square test was used to test differences. A P value of less than 0.05 was considered statistically significant. Th results from the online survey were exported into a spreadsheet for analysis (Excel, Microsoft, Redmond, WA). Statistical analyses were performed using SAS Enterprise (Version 4.2, SAS Institute, Cary, NC). RESULTS Study Group
Two hundred four responses were received, and 18 (8.8%) of the responses were excluded because they were incomplete. Of the 186 complete responses, 124 (66.7%) responded to the e-mailed survey and 62 (33.3%) responded to social media promotion. The response rate for the e-mailed survey was 75.2% (124 out of 165). From six tweets, the impressions, engagements, and link clicks were 4267, 258, and 72, respectively. Using impressions as potential audience reached, the response rate was 1.4% (62 out of 4267). Table 2 shows the characteristics of the respondents. There were 123 (66.1%) male respondents, 119 (64.0%) were in academic practices, and nearly half of the respondents (85, 45.7%) self-identified as belonging to Generation X, followed by Generation Y (62, 33.3%) and Baby Boomers (39, 21.0%). Regarding use of social networking, 110 (59.1%) were users for professional purposes, 34 (18.3%) were for personal-only, and 42 (22.6%) had no use. Of those respondents who use social media for professional purposes, 55 out of 110 (50.0%) stated that they kept the personal and professional accounts separate. Comparison of Users and Nonusers of Professional Social Networking
There were significant differences in the career position and generation between professional users and nonusers (Table 2). Fifty two out of 110 (47.3%) professional social networking users were in training as either residents or fellows compared to 18 out of 76 (23.7%) nonusers (P < 0.01). Generation Y comprised 46 out of 110 (41.8%) professional social network users compared to 16 out of 76 (21.1%) nonusers (P < 0.01). In contrast, Baby Boomers generation comprised 16 out of 110 (14.5%) professional social network users and 23 out of 76 (30.3%) nonusers (P < 0.01). There were no significant differences in the gender or practice type between professional users and nonusers. However, on a subgroup analysis of users of Twitter for professional purposes, there were gender differences: 15 out of 66 (22.7%) users were female compared to 48 out of 120 (40.0%) nonusers (P < 0.05). Habits of Radiologists and Benefits of Using Social Networks for Professional Purposes
The radiologists who used social networking for professional purposes describe their usage as “consuming and rarely
PROFESSIONAL SOCIAL NETWORKING IN RADIOLOGY
contributing” in 44.5% (49 out of 110), and “both consume and frequently contribute” in 30.9% (34 out of 110) and “only consume” in 24.5% (27 out of 110). The three most popular platforms for radiologists using social networking for professional purposes were LinkedIn (74 out of 110, 67.3%), Twitter (66 out of 110, 60.0%), and Doximity (44 out of 110, 40.0%). Most users had participated in social networking for 2–5 years (36 out of 110, 32.7%) or more than 5 years (24 out of 110, 21.8%), but there were also new users, 19 (17.2%) of them had been using social networking for less than 6 months, 8 (7.2%) for 6–12 months, and 23 (20.9%) for 1–2 years. Users were most commonly active several times daily (42 out of 110, 38.2%), followed by daily to several times per week (32 out of 110, 29.1%), weekly to several times per month (18 out of 110, 16.4%), and monthly or less (18 out of 110, 16.4%). Of the 42 radiologists who were active several times per day, 34 (81.0%) used Twitter, 28 (66.7%) used LinkedIn, and 14 (33.3%) used Facebook. Career positions of these highly active users were predominantly trainees (18 out of 42, 42.9%). Other highly active users included 7 (16.7%) who were in practice for <8 years, 5 (11.9%) in practice for 8–15 years, and 12 (28.6%) in practice for >15 years. Among these highly active users, only 6 (14.3%) were female, and 5 of the 6 females were trainees. Radiologists found many benefits to using social networks professionally, with 86 (78.2%) using their preferred platform to follow news from professional organizations. Radiologists also found value by communicating with their colleagues (69 out of 110, 62.3%), following events or news during scientific meetings (65 out of 110, 59.1%), and sharing and following radiology cases (63 out of 110, 57.3%). Barriers to Using Social Networks for Professional Purposes
The top four barriers to using social networks for professional purposes were the same for users and nonusers. These were not enough time, concerns over confidentiality, concerns over an unprofessional stigma, and unfamiliarity with the technical aspects of the medium (Table 2). The most commonly selected barrier differed between the users and nonusers. Not enough time was the most common barrier for professional users (57 out of 110, 51.8%). The most common barrier for nonusers was concerns over confidentiality (46 out of 76, 60.5%). DISCUSSION As radiologists become increasingly engaged in social networking platforms for professional purposes, it is necessary to consider the characteristics of both current users and nonusers. An understanding of trends in use and the benefits and barriers to use is key not only for all stakeholders, including the radiologists themselves, but also for professional radiology organizations, physicians of other specialties, and patients. 3
Academic Radiology, Vol ■, No ■■, ■■ 2017
PATEL ET AL
TABLE 2. Comparison of Users and Nonusers of Social Networking for Professional Purposes
All 186
Users of Social Media for Professional Purposes 110
Nonusers of Social Media for Professional Purposes 76
Gender
0.3045 Male Female
123 (66.1%) 63 (33.9%)
76 (69.1%) 34 (30.9%)
47 (61.8%) 29 (38.2%)
Academic Private Combined
119 (64.0%) 37 (19.9%) 30 (16.1%)
72 (65.5%) 23 (20.9%) 15 (13.6%)
47 (61.8%) 14 (18.4%) 15 (19.7%)
Practice type
0.5303
Generation
0.0034 “The Greatest Generation” “The Silent Generation” “Baby Boomers” “Generation X” “Millennials/Generation Y”
0 0 39 (21.0%) 85 (45.7%) 62 (33.3%)
16 (14.5%) 48 (43.6%) 46 (41.8%)
23 (30.3%) 37 (48.7%) 16 (21.1%)
Resident or Fellow In practice for <8 years In practice for 8–15 years In practice for >15 years
70 (37.6%) 39 (21.0%) 23 (12.4%) 54 (29.0%)
52 (47.3%) 17 (15.5%) 12 (10.9%) 29 (26.4%)
18 (23.7%) 22 (29.0%) 11 (14.5%) 25 (33.0%)
Career position
Social media networks used for professional purposes Benefits of professional social media activity
Barriers to professional social media activity
P Value
0.0086
LinkedIn Twitter Doximity
74 (67.3%) 66 (60.0%) 44 (40.0%)
News from professional organizations Communicate with colleagues Scientific meetings Education
86 (78.2%) 69 (62.3%) 65 (59.1%) 63 (57.3%)
Not enough time Concerns over confidentiality Concerns over unprofessional stigma Unfamiliarity with technical aspects
99 (53.2%) 78 (41.9%)
57 (51.8%) 32 (27.0%
42 (55.3%) 46 (60.5%)
56 (30.1%)
25 (21.0%)
31 (40.1%)
43 (23.1%)
15 (12.6%)
28 (36.8%)
Social media activity Consume only Consume and rarely contribute Consume and frequently contribute
This survey study finds that there is enormous potential for growth in social networking as nearly half of the users were radiology trainees. There are no significant gender differences in professional users and nonusers overall, but users of Twitter for professional social networking were disproportionately male. The higher proportion of trainees and Generation Y radiologists using social networking in our study is not surprising. Younger radiologists have grown up in a world of being immersed in online communities and are more likely to be technologically savvy. Our results agree with Brown et al.’s 4
27 (24.5%) 49 (44.5%) 34 (30.9%)
physician survey, which found a linear relationship between increasing age and decreasing social media use (6). In another study of Facebook users in the surgical specialty, 64% of residents had Facebook accounts compared to only 22% of faculty (16). Our study confirms and quantifies the differences in user age groups in radiology. Nearly half of the users in radiology were trainees and more than 85% of users were from Generation X or Y. In particular, Generation Y also brings the traits of being collaborative, creative, and having a positive “can-do” attitude (17). As these younger radiologists rise to leadership positions, social networking in the professional setting
Academic Radiology, Vol ■, No ■■, ■■ 2017
will only increase. There will be opportunities to change radiology education and engage with patients regarding imaging topics (10). Users of professional social networking describe numerous benefits, but it is the barriers among nonusers that are interesting to study because they have impact on future use. In our study, 61% of nonusers were concerned about confidentiality and regarded this as a factor preventing their use. Confidentiality concerns relate to maintaining patient confidentiality as well as the radiologist’s personal confidentiality. Brown et al. found similar results, with 66% of physicians being hesitant to immerse themselves in social media because of worries about public access and legal concerns (6). The American Medical Association specifically advises on professional conduct on social media and state that physicians should consider separating personal and professional content online (4). In our study, half of the professional users had this insight and did maintain separate professional and personal profiles. Ventola warns about damage to healthcare professional’s image with online behavior that could be construed as unprofessional, such as violations of patient privacy and inappropriate comments and photographs, but such cases are extremely rare (18). Radiologists who are cognizant of ethics and professionalism can use social networking with minimal risk to confidentiality, and experience the benefits highlighted by the users in this study. An important finding was differences in gender between the radiologists who use Twitter for professional purposes. Twitter is the second most common platform among all professional users and the most commonly used platform among highly active professional users. However, only one in four users was a female radiologist and only 14% of highly active users were female. This gender imbalance among Twitter users is likely multifactorial. Concerns about confidentiality may be more pronounced among female users of social networking, as shown in studies outside healthcare (19). Additionally, female radiologists may have less time to explore and participate on social networking platforms. A study of physicianresearchers has shown large gender differences in domestic and parenting responsibilities, which may contribute to different rates of career success (20). The impact of fewer female Twitter users has the potential to perpetuate the gender imbalance as there are fewer female radiology role models. In our study, only one of the six highly active female users was an attending radiologist. Awareness of gender imbalances has been raised by a Twitter-based application called Twee-Q (twee-q.com), which provides a score derived from how often a Twitter user retweets men or women. These gender differences suggest that different strategies may be needed to increase social media engagement among male and female radiologists. This study has several limitations. First, there is selection bias because the survey was sent to radiologists by e-mail and social media. E-mail distribution to the authors’ contacts leads to selection bias to include more academic radiologists and younger radiologists, and may not be generalizable to the current workforce of radiologists. Promotion on Facebook
PROFESSIONAL SOCIAL NETWORKING IN RADIOLOGY
and Twitter could overestimate the use of social networks and the use of these platforms. Additionally, the estimations of the response rate by Twitter impressions were crude and likely a gross underestimation given that impressions do not equate to the tweet being read. Finally, there is a nonresponse bias, which is inherent in survey studies. We predict that nonresponse bias would lead to overestimation of the proportion of professional social network users as users of social media are more likely to respond and share their experiences. Despite this, we saw similar proportions of social networking users compared to two other physician surveys on social media (5,6). CONCLUSION Nearly 60% of radiologist respondents use social networking for professional purposes. Radiology is likely to see growth in the number of users and in the role of social networking in the coming years as nearly half of professional users are radiology trainees. Twitter use for professional purposes among radiologists was disproportionately male, and especially among highly active users. It is important to be cognizant of gender imbalance and to improve visibility of female leaders on social networking. REFERENCES 1. AMN Healthcare, Inc. 2013 survey of social media and mobile usage by healthcare professionals. AMN Healthcare, Inc., 2013. 2. Antheunis ML, Tates K, Nieboer TE. Patients’ and health professionals’ use of social media in health care: motives, barriers and expectations. Patient Educ Couns 2013; 92:426–431. 3. Hoang JK, McCall J, Dixon AF, et al. Using social media to share your radiology research: how effective is a blog post? J Am Coll Radiol 2015; 12:760–765. 4. American Medical Association. Opinion 9.124 – professionalism in the use of social media. AMA Journal of Ethics 2015; 17:432–434. Available at: http://journalofethics.ama-assn.org/2015/05/coet1-1505.html. Accessed June 15, 2015. 5. Modahl M, Tompsett L, Moorhead T. Doctors, patients and social media. QuantiaMD. 2011. Available at: http://www.quantiamd.com/q-qcp/social _media.pdf. Accessed June 15, 2015. 6. Brown J, Ryan C, Harris A. How doctors view and use social media: a national survey. J Med Internet Res 2014; 16:e267. 7. Pathiraja F, Little D. Social media: the next frontier in radiology. Clin Radiol 2015; 70:585–587. 8. Kotsenas A. 7 reasons neuroradiologists should use Twitter. AJNR Blog, 2015. Available at: http://www.ajnrblog.org/2015/03/27/7-reasons -neuroradiologists-should-use-twitter/. Accessed June 15, 2015. 9. Matthews G. Why the evolution of radiology includes social media. WCG Common Sense, 2013. Available at: http://blog.wcgworld.com/2013/ 09/why-the-evolution-of-radiology-includes-social-media-sthash .MlJ6IqEM.dpuf. Accessed June 15, 2015. 10. Prabhu V, Rosenkrantz AB. Imbalance of opinions expressed on Twitter relating to CT radiation risk: an opportunity for increased radiologist representation. AJR Am J Roentgenol 2015; 204:W48–W51. 11. Hawkins CM, Duszak R, Rawson JV. Social media in radiology: early trends in twitter microblogging at radiology’s largest international meeting. J Am Coll Radiol 2014; 11:387–390. 12. Hawkins CM, Hillman BJ, Carlos RC, et al. The impact of social media on readership of a peer-reviewed medical journal. J Am Coll Radiol 2014; 11:1038–1043. 13. Naeger DM, Webb EM. Social media for professional purposes: introduction to the JACR “how to” video guide. J Am Coll Radiol 2013; 10:736–737. 14. Krishnaraj A. The history of the RFS. J Am Coll Radiol 2010; 7:382– 383.
5
PATEL ET AL
15. Krishnaraj A, Weinreb JC, Ellenbogen PH, et al. Impact of generational differences on the future of radiology: Proceedings of the 11th Annual ACR Forum. J Am Coll Radiol 2012; 9:104–107. 16. Landman MP, Shelton J, Kauffmann RM, et al. Guidelines for maintaining a professional compass in the era of social networking. J Surg Educ 2010; 67:381–386. 17. Slanetz PJ, Kung J, Eisenberg RL. Teaching radiology in the millennial era. Acad Radiol 2013; 20:387–389.
6
Academic Radiology, Vol ■, No ■■, ■■ 2017
18. Ventola CL. Social media and health care professionals: benefits, risks, and best practices. P T 2014; 39:491–520. 19. Hoy MG, Milne G. Gender differences in privacy-related measures for young adult Facebook users. J Interact Advert 2010; 10:28–45. 20. Jolly S, Griffith KA, DeCastro R, et al. Gender differences in time spent on parenting and domestic responsibilities by highachieving young physician-researchers. Ann Intern Med 2014; 160:344– 353.