Embed Size (px)
Citation preview
Variability in Facebook posting across health conditions Robert J Smith BS1; Patrick Crutchley, BSE1,2; H. Andrew Schwartz, PhD1,2,3 ; Lyle H. Ungar PhD 1,2,3; Frances Shofer, PhD4; Kevin A. Padrez BS1; Raina M Merchant MD MSHP1,5 1Penn Medicine Social Media and Health Innovation Lab, University of Pennsylvania; 2Positive Psychology Center, University of Pennsylvania; 3Department of Computer and Information Science, University of Pennsylvania; 4Department of Emergency Medicine, University of Pennsylvania Perelman School of Medicine
Results
Discussion & Conclusions
§ Over two billion individuals worldwide have social media accounts; >71% of American adults use Facebook.
§ Social media has been used to glean meaningful insights about human health and behavior (i.e. predicting influenza outbreaks, monitoring public sentiment around vaccines, and supporting smoking cessation programs).
§ Social media data may contribute to the development of “automated hovering” initiatives to follow patients’ routine, everyday behaviors (such as diet, exercise, and medication adherence) in a manner that is welcomed and convenient for the purpose of improving health outcomes.
Background
1. To describe variability in social media use across a cohort of social media users within a healthcare setting.
2. To explore language topics correlated with frequency of social media use.
3. To evaluate the differences in the quantity of social media postings across individuals with different disease diagnoses.
4. To determine if patients could accurately predict their own levels of social media engagement.
Objectives
Demographic N Mean 95% CI p-‐value
Sex Female 514 27 24, 32 0.0059 Male 181 18 15, 23
Race African American
489 27 24, 32 0.0449
White 141 19 14, 25 Other race 65 20 13, 31
Age 18-‐29 437 28 24, 33 0.0216 30-‐39 173 20 16, 26 40-‐49 60 23 15, 35 >49 25 12 6, 23
Facebook Friend Count
Q1 (874-‐4800) 173 40 31, 51 <.0001 Q2 (483-‐873) 175 22 17, 28 Q3 (295-‐482) 175 26 21, 34 Q4 (13-‐294) 171 16 12, 21
Perceived PosDng frequency
3+ Emes daily 154 51 40, 66 <.0001 1-‐3 Emes daily 161 42 33, 53 Every few days 176 25 20, 32 Once per week or less
204 9 7, 11
Health CondiDon n Unadjusted Mean Posts Adjusted Mean Posts
Mean 95% CI p-‐value mean 95% CI p-‐value
Depression screen PosiEve 120 38 27 52 0.0029 37 27 49 0.0046
NegaEve 575 23 20 26 23 20 26
Depression Yes 134 38 29 50 0.0013 38 28 50 0.0014
No 561 22 19 26 22 19 26
Asthma Yes 134 34 25 45 0.0163 31 23 41 0.1092
No 561 23 20 26 23 20 27
Headaches Yes 272 29 24 35 0.0517 28 23 34 0.6924
No 423 22 19 26 23 19 27
Anemia Yes 177 23 18 30 0.5656 22 17 29 0.3614
No 518 25 22 29 26 22 30
Diabetes Yes 66 26 17 40 0.7286 29 19 43 0.4628
No 629 25 21 28 24 21 28
Hypertension Yes 98 26 18 37 0.7786 30 21 42 0.2675
No 597 25 21 28 24 21 27
Neoplasm Yes 108 24 17 34 0.9416 26 19 36 0.6924
No 587 25 22 28 24 21 28
Back pain Yes 243 25 20 31 0.9632 25 20 31 0.9569
No 452 25 21 29 25 21 29
Table 2: Unadjusted & Adjusted (Sex, Race, Age) Mean Facebook Posts in 6 months prior to enrollment by presence or absence of
highly prevalent ICD-9 codes in the EMR
Figure 2: Language Topics of High and Low Frequency Facebook Posters.
Methods Study design • All adult patients in a single, urban adult emergency department
(ED) were approached from March to October 2014. • Patients were asked if they used Facebook. Patients responding
affirmatively were asked to participate in a study about social media. Consenting sharers provided access to: (1) their Facebook accounts and (2) their historical electronic medical record (EMR).
Assessing variability in social media use across a cohort • A Facebook application created by the study team was used to
extract data from each participant’s Facebook account. • The following variables were extracted from each user: Number
and content of status updates ; number of friends.; and number of Facebook posts in the six months preceding enrolment.
Assessing language of social media • To distill the language of our sample into a smaller feature spaces,
we used a natural language processing technique called Latent Dirichlet Allocation (LDA).
Assessing accuracy of perception of social media posting • During enrolment, participants were surveyed about their perceived
frequency of posting on social media. Patients were grouped into four categories based on their reported frequency of posting (3+ times daily, 1-3 times daily, every few days, about once per week or less).
Scatterplot demonstrates mean actual number of posts on Facebook in 6-months prior to enrollment by patient “perceived
category of posting frequency” compared to projected number of posts on Facebook. A 1:1 relationship is demonstrated by the
reference line Y=X. The trend line is logarithmic.
Total study participants n=1432 (53%)
Non-sharers n=424 (29%)
Declined any participation in the study
n= 1285 (47%)
EMR and social media sharers n=1008 (71%)
Asked to participate in the study n=2717 (52%)
Did not use Facebook or Twitter (exclusion criteria)
n=2539 (48%)
Screened and approached for enrollment n=5256
Table 1: Facebook posts in the 6-month period prior to enrollment by demographics
3+ times daily
1-3 times/ day
Every few days
Once per week or less
y = 13.166ln(x) - 33.267 R² = 0.99091
0
10
20
30
40
50
60
0 100 200 300 400 500 600
Mean actual number of posts by perceived posting
frequency category
Mean projected number of posts by perceived posting frquency category
Figure 3: Actual vs Projected Mean Number of Posts By Perceived Posting Frequency
Perceived Posting Frequency Category Reference (Y=x)
Patients with sufficient Facebook data for LDA analysis
n=695
Figure 1: Enrollment flowchart
Word clouds represent language topics most positively (blue) and negatively (green) correlated with the highest quartile of Facebook
posters in the sample. Increased word size indicates higher prevalence of the word within the topic. All correlations are significant
to p<0.05.
r = 0.240 r=0.214 r=0.183
Language topics most strongly
positively correlated with
posting frequency
r= -0.223 r= -0.174 r= -0.173
Language topics most strongly
negatively correlated with
posting frequency
Our study had four major findings: (1) there was significant variation in Facebook posting frequency within
our patient sample. (2) high-frequency posters wrote about topics related to health. (3) when controlling for demographic variables, there were significant
differences in Facebook posting quantities between individuals with and without a clinical diagnosis of depression.
(4) patients are good relative predictors of their Facebook posting frequency over time.
Future Research and Implications: § Facebook and depression. The directionality of the depression-posting
relationship is unclear: (1) Individuals in the developing stages of depression may be posting on Facebook with greater frequency as a means to reach out to a social network, escape a sense of isolation, or maintain connectivity; (2) Conversely, high frequency activity on Facebook or any other virtual connectivity platform may actually contribute to underlying depression. Either way, knowing that a user is a high-frequency poster may be a useful signal to indicate appropriate screening for depression.
• Patients prediction of social media use. Accuracy of patient self-reported activity is often either unreliable or unknown in other domains of medicine. If social media usage is ever deemed to be a risk factor for the development or presentation of an illness, our findings suggest that patients may provide reliable information regarding their own usage
Acknowledgements & Funding • We thank our RAs for patient recruitment and data collection. • This study was funded by an Innovation Grant through the Leonard
Davis Institute of Health Economics at the University of Pennsylvania. Raina Merchant is supported via grants from NIH, K23 Grant 10714038 and NHLBI R01 Grant HL122457. The authors have no conflicts of interest to disclose.
