#TwitterHomework During Pathology Electives: Transforming Pathology Pedagogy

This is an institutional review board–approved, retrospective, survey-based study analyzing the use of Twitter by rotating medical students who participated in “Twitter Homework” in the Department of Pathology and Laboratory Medicine at Loyola University Health System, Maywood, Illinois, from 2017 to 2018.

Medical students on their 2- or 4-week pathology elective were asked to tweet a pathology-related tweet every day during their elective. They had the option of either retweeting (or sharing) an already existing tweet, or composing a pathology-related tweet themselves, such as an interesting case or commentary on their daily elective-related experience. Regardless of the content, tweets were indexed by using the hashtags #TwitterHomework and/or #PathElective. This was not a mandatory component of the elective, but was strongly encouraged. The choice of using Twitter or not during the elective/observership did not influence the final grade of the student.

Students were asked if they thought Twitter had an effect on their awareness and opinion of pathology as a specialty, and if it was useful for the attainment of medical knowledge, networking, and their professional development. A second survey was sent to those who did not use the social media platform to see if they thought the project could be beneficial despite choosing not to participate. All data collection concluded in April 2019, with 18 of the 37 students who used Twitter responding to the survey. Survey results were analyzed qualitatively. The χ² (Question 1 in Figure 1) and Fisher exact tests (Questions 2, 3, and 4 in Figure 1) were used to see if there was a statistically significant difference in the subjective responses between the Twitter users (students who used Twitter during their elective) and the Twitter nonusers (students who did not use Twitter during their elective) regarding their experiences of the pathology elective.

In addition to the surveys, we retrospectively analyzed the distribution of tweets marked with #TwitterHomework and #PathElective hashtags with the social media analytics program Symplur Signals (Symplur LLC).²⁰  Included in the analysis were tweets and retweets by the students participating in “#TwitterHomework” and nonstudent users who used the hashtags in the context of pathology. The number of impressions the tweets generated was also included in the analysis. The number of impressions was calculated by summing the number of tweets multiplied by the number of followers for every participant, which means impressions correspond to the number of potential views and should not be misinterpreted as the number of actual views or interactions. User profiles that used the hashtags in tweets not related to pathology were excluded. On the basis of the user biography, users were put into different health care stakeholder categories that are defined by Symplur Signals.^21,22  In addition to the categories assigned by Symplur Signals, the doctor category was further divided into “Pathologists,” “Pathology residents,” and “Doctor other” (specialties other than pathology) on the basis of the user biography by manual sorting. Manual corrections were also used in instances where it was assessed that the Symplur Signals machine learning failed to put the users into the appropriate health care stakeholder category. All of the Twitter analytics data were collected in February–March 2019 after all of electives in the mentioned period concluded.

Thirty-four students responded to the survey; 18 of the responders did use Twitter during their pathology electives, 16 of the responders did not. Some of the responders only answered a few questions, leaving most of the questions with fewer responders than the numbers mentioned above.

There was no statistically significant difference (P = .78) in the interest in pursuing pathology as a career choice between the Twitter users and Twitter nonusers: 6 of 17 Twitter users (35%) and 6 of 15 Twitter nonusers (40%) (Figure 1, Question 1). Twitter users described a more positive overall pathology elective experience (16 of 17 [94%] excellent/good versus 11 of 15 [73%] excellent/good in Twitter nonusers; Figure 1, Question 2) and described their understanding of the field of pathology as slightly better (15 of 16 [94%] excellent/good versus 14 of 16 [88%] excellent/good in Twitter nonusers (Figure 1, Question 3); however, the difference was not statistically significant (P = .24 and P > .99, respectively). Twitter users had a slightly higher rate of perceived usefulness of the pathology rotation, which was statistically significant (P = .01) (14 of 15 [93%] strongly agreed/agreed versus 13 of 15 Twitter nonusers [87%] strongly agreed/agreed; Figure 1, Question 4). Most Twitter users felt #TwitterHomework increased their medical knowledge (14 of 17 users [82%]; Figure 2, Question 3), reported #TwitterHomework was overall a strongly positive/positive experience (14 of 17 users [82%], 2 users described their experience as neutral [12%], 1 user [6%] felt it was negative, and 0 users felt it was strongly negative; Figure 5, Question 10), and experienced no negative consequences (16 of 17 users [94%]; Figure 5, Question 2). A perceived improved personal responsibility for accurate diagnosis was reported in 9 of 17 Twitter users (53%) (Figure 5, Question 7), and 12 of 17 users (70%) reported #TwitterHomework improved their professional development, networking, and communication (Figure 5, Question 9). A positive change to their perspective of pathology was reported in 7 of 17 users (41%) (Figure 5, Question 3). Both Twitter users and nonusers thought #TwitterHomework would be useful for future students to increase their medical knowledge (15 of 18 Twitter users [83%] versus 12 of 15 Twitter nonusers [80%]; Figure 4, Question 7).

Of the Twitter nonusers, most did not have a Twitter account before the rotation (11 of 15 [73%] versus 6 of 17 [35%] of Twitter users did not have an account prior to the elective; Figure 2, Question 1). Six of 15 Twitter nonusers (40%) felt that daily posting would have been a burden (only 1 of 18 Twitter users [6%] felt that daily posting was a burden, Figure 3, Question 5; and 16 of 17 users [94%] did not think that #TwitterHomework took valuable time away from other educational activities, Figure 3, Question 6).

Symplur Signals identified 580 tweets (averaging 34.1 per month) by searching the hashtags #PathElective and or #TwitterHomework from the 37 student participants between August 1, 2017 (the first time a student participant used one of the hashtags) and January 2, 2019 (the last time a student participant used one of the hashtags in the study period). Of the 580 tweets, most were original tweets (527 [90.86%] versus 53 retweets of preexisting content [9.14%]; Figures 6 and 7). While all of the 37 student participants generated at least 1 original tweet during the period that was studied (average, 14.2 original tweets per user; mean, 14.2; median, 10), Twitter activity varied: 16 students (43.24%) composed fewer than 10 original tweets, 13 students (35.14%) composed between 10 and 19 original tweets, 6 students (16.22%) composed between 20 and 29 original tweets, and 2 students (5.41%) composed at least 30 original tweets (one composed 33 tweets and the other composed 104 tweets) (Figure 8). The original tweets resulted in a total of 78 422 impressions during this period.

A total of 922 nonstudent users were identified through their use of #PathElective or #TwitterHomework or their retweeting of existing tweets containing these hashtags. #TwitterHomework was first used in the context of pathology on March 17, 2017, by one of us (C.A.A).²³  Data were collected from the nonstudent users from March 17, 2017, to March 17, 2019. During this time, the nonstudent users composed 4191 tweets (averaging 174.63 tweets per month) and 4.55 tweets per user (mean, 4.55; median, 1): 213 original tweets (5.08%) and 3978 retweets (94.92%), which generated 8 891 461 impressions.

Among the 922 nonstudent users, 810 retweeted the students at least once and collectively a total of 3399 times, which resulted in 6 360 731 impressions (Figures 9 and 10; Table 1). On the basis of the user biography, the 810 nonstudent users were put into different health care stakeholder categories that are defined by Symplur Signals. Their user identities (health care stakeholder categories) are summarized in Table 2. Pathologists composed most of the retweets with a total of 296 pathologists retweeting the students 2061 times (60.6% of the retweets). From the location manually entered into the Twitter profile by users, Symplur Signals mapped the location of the users (Figures 11 and 12). Most of the identifiable users were located in the United States (177 of 810 users, 21.9%), India (55 of 810 users, 6.79%), Spain (43 of 810 users, 5.31%), Mexico (37 of 810 users, 4.57%), and the United Kingdom (27 of 810 users, 3.33%). Within the United States, most users were situated in Illinois (20 of 177 users, 11.3%), Texas (17 of 177 users, 9.6%), New York (16 of 177 users, 9.04%), Pennsylvania (12 of 177 users, 6.78%), Michigan (10 of 177 users, 5.65%), Massachusetts (9 of 177 users, 5.08%), Indiana (8 of 177 users, 4.52%), California (8 of 177 users, 4.52%), Maryland (6 of 177 users, 3.39%), and Missouri (5 of 177 users, 2.82%).

Medical student education is changing, transforming to better adapt to the learning style of the next generation of doctors. With the expansion of social media use and reach, pathologists have the opportunity to lead the field of medicine in creating innovative teaching styles to improve the medical student experience. We suggest that the implementation of social media into the medical education process can potentially improve the student's learning experience.

Building on prior publications that show Twitter use is beneficial for networking and career development for aspiring pathologists, this is the first report that has studied pathology elective students and the consumers of their tweets by using impressions and retweets as the “surrogate endpoints” to determine the network extent. Our relatively small dataset allowed us to create our own health care categories (pathologists, pathology resident, doctor other) that added further accuracy when trying to determine if the content reached the target audience for networking within pathology.

Students participating in our study had overall positive attitudes toward using Twitter for pathology educational and professional purposes. Twitter student users reported an overall better elective experience and described their understanding of the field of pathology as slightly better than Twitter nonusers; however, the difference was not statistically significant. They reported overwhelmingly that its use increased their medical knowledge, that they were able to find reliable pathology material on the platform, and that it helped with their professional development, networking, and for some it improved their opinion of pathology.

The weaknesses of the study relate to the small sample size of students and the qualitative nature of the survey, which make it difficult to extrapolate the findings to medical students in general. Furthermore, of the 37 students who participated in the #TwitterHomework project only 18 chose to reply to the survey, leaving the survey with self-selection bias. A more robust study would have a larger sample size and include all of the students who used Twitter in the survey. Also, to answer the question of whether using Twitter helped the students attain more knowledge, a prospective study with a control group while providing a pre–pathology rotation and post–pathology rotation examination/assessment would be more appropriate.

In this study we were able to demonstrate that the students who tweeted about pathology, using #Twitterhomework and #PathElective, in the pathology department at Loyola University Medical Center had an overall positive experience doing so. The students thought that it helped them increase their medical knowledge, and that the pathology material that was found on the platform was reliable. The students also thought the project helped them in their networking and professional development. With the use of Symplur Signals we were able to confirm the networking potential that the use of Twitter has both within the United States and internationally and that a great number of pathologists engaged with the tweets that the students made.