In the spring of 2012, a short video profiling atrocities committed by a guerrilla leader named Joseph Kony was viewed by over 50 million people in less than three days. The rapidity with which that video spread was a testament to the power of digital networks and the unprecedented speed and scope with which information can move in the information age. Yet in medicine, arguably the most technologically advanced of the professional fields, our ability to spread information is stunningly slow and ineffective. The promulgation (or lack thereof) of evidence-based guidelines is a particularly blatant example. In a 2011 analysis of physician adherence to dyslipidemia prevention guidelines, only 36.9 percent were appropriately performing lipid profile screening, 27.6 percent were performing pharmacotherapy up-titration, and 21.0 percent pharmacotherapy initiation (Vashitz G, Meyer J, Parmet Y, Henkin Y, Peleg R, Gilutz H. Physician adherence to the dyslipidemia guidelines is as challenging an issue as patient adherence. Fam Pract 2011; 28: 524-31). Most surprising is that this poor performance is occurring nearly 10 years after the ATP III guidelines were published!
There have been many efforts to explain the ineffective dissemination of knowledge in medicine and in particular the failure to implement of evidence-based guidelines. A central tenet of this work is the belief that physicians are deeply affected and influenced by the peer group with whom they work. One of the earliest examples was the landmark study of physician-prescribing behavior by Coleman, Katz and Mentzel (Coleman J KE, Mentzel H. The diffusion of an innovation among physicians. Sociometry 1957; 20: 253-70) that demonstrated how relationships between physicians can predict the adoption of a new medication. Specifically, the adoption pattern begins with key opinion leaders, spreads to community physicians who have contact with those leaders, and then to physicians with social ties. Numerous studies, including those of Browman (Browman GP, Levine MN, Mohide EA, Hayward RS, Pritchard KI, Gafni A, et al. The practice guidelines development cycle: a conceptual tool for practice guidelines development and implementation. J Clin Oncol 1995; 13: 502-12), Parchman (Parchman ML, Scoglio CM, Schumm P. Understanding the implementation of evidence-based care: A structural network approach. Implement Sci 2011; 6: 14), and Wennberg (Wennberg JE. Unwarranted variations in healthcare delivery: implications for academic medical centres. BMJ 2002; 325: 961-4) have reinforced the importance of peer networks even in this era of electronic health records and digital communication.
That physicians are influenced by thought-leaders and peer practitioners is not surprising. This model is reinforced by the current training and practice paradigm that emphasizes the knowledge of thought-leaders and the practice habits of local communities. However, we are on the verge of a new era in healthcare with the emergence of healthcare-specific social and professional networks such as Doximity, Sermo, Healthtap, and DocBookMD. In creating a national community that leverages the inherent strengths of online social/professional networking, these networks have the potential to change the way that physicians interact and may disrupt the current communication paradigm.
We have partnered with Doximity, a private professional physician network, to take an early look at how peer-to-peer relationships are forming online. Generally speaking, our observations suggest that a physician’s online social/professional network differs across specialties both in the makeup of the network and geography of relationships. Even though these online communities are just beginning to develop, early trends suggest a shift away from locally based relationships that have heretofore driven dissemination of knowledge and best practices.
Doximity has over 130,000 United States verified physicians (over 20 percent of the U.S. physician workforce) as registered members (Bureau of Labor Statistics. Occupational Outlook Handbook: Physicians and Surgeons; 2012-2013). Doximity members are early adopters of peer-to-peer social networking. Rates of participation in Doximity vary by specialty from about 0.5 percent to 11 percent of members of a given specialty. There are notable differences in the size and make-up of individual's networks, depending on specialty. Generalists (Pediatrics, Family Medicine, and General Internal Medicine physicians) have greater proportion of connections with other generalists. Sub-specialtists such as Plastic Surgery, Thoracic Surgery, Endocrinology, Radiation Oncology, and Oncology have significantly fewer connections within their specialty, and in turn tend to be more connected with generalists. This trend may reflect typical referral patterns but could also be influenced by the higher rates of membership of generalists within Doximity. However, we observed less geographic preference than expected within the Doximity physician network. Individuals were relatively equal in their “local” (within 50 miles) and “non-local” (outside 50 miles) colleagues. This trend appears across both generalists and specialists, suggesting that online networks are not merely a reflection of local referral patterns.
These early observations reflect how the Facebook era is changing the relationship networks of physicians across the country. In a field where practice patterns have been based on local standards, we can hypothesize how these new, Web-based networks will allow innovation and information to propagate differently among physicians as a result of the differences in the structure of their professional networks. Generalists such as pediatricians, internists, etc. have more colleagues from the same specialty that can introduce and reinforce changes in practice. They act as “connectors” in what can be thought of as a “clustered network.” Specialists seem to demonstrate more of a “long ties” network structure with far fewer immediate colleagues to influence and be influenced by. Changing behaviors using “opinion leaders” might work well for generalists—two or three adopters in a highly connected, clustered network may be enough to seed and rapidly disseminate an idea or behavior. But for specialists, a different approach that leverages their “long ties” might be more effective.
As a field, we need to be cognizant of how social and professional networks are changing the information, education, and practice dynamics of healthcare. These tools have great potential to improve our system of healthcare and should be used early and often.
Christopher A. Longhurst, M.D., M.S. is an associate professor in the Department of Pediatrics at the Stanford School of Medicine, Palo Alto, Calif., and is CMIO at Lucile Packard Children’s Hospital, Stanford University.
Leslie Lenert, M.D., is a professor in the Department of Internal Medicine at the University of Utah, Salt Lake City, Utah.
Matthew J. Goldstein, M.D., Ph.D. is a resident physician in the Department of Internal Medicine, Brigham & Women’s Hospital, Boston, Mass.
Acknowledgements
Matthew J. Goldstein, M.D. is a research fellow with Doximity Inc., and has received both cash payment and equity compensation. Christopher A. Longhurst, M.D. and Leslie Lenert, M.D., are advisory board members to Doximity Inc. and have received non-cash, equity compensation.
