The Need for Medical Professionals to Join Patients in the

Online Health Social Media Discourse

Hamman Samuel

, Fahim Hassan

and Osmar Za

ıane

Department of Computing Science, University of Alberta, Edmonton, Canada

School of Public Health, University of Alberta, Edmonton, Canada

Keywords:

Health Social Media, Trust, Misinformation, Privacy, Anonymity, COVID-19, Infodemic.

Abstract:

Health social media is frequently used by e-patients for seeking health information online for self-diagnosis,

self-treatment, and self-education. Health social media also provides various beneﬁts for patients and layper-

sons, such as allowing users to be part of virtual support groups, having quick access to advice, and the conve-

nience of access via the Internet. At the same time, it raises concerns about misinformation being propagated

by laypersons without professional medical expertise, especially during pandemics like COVID-19, leading

to an infodemic. There are only a handful of health social media websites that allow medical professionals to

participate in discussions with patients online. We postulate that the modern face of medicine and healthcare

needs medical professionals to be included in the online patient discourse so misinformation can be addressed

early on and head on. To this end, we propose a new and free health social network named Cardea that is

under development which aims to bring patients, laypersons, and medical professionals together on the world

wide web. Users can share experiences, ask questions, and get answers in three streamlined environments:

Patient to Patient (P2P), Patient to Medic (P2M), and Medic to Medic (M2M). While there are several forums

that cater speciﬁcally to patient-patient discussions or medic-medic connections, Cardea’s added value is in

providing a uniﬁed portal for both patients and medics, as well as enabling interactions between patients and

medics. Moreover, Cardea applies machine learning, information retrieval, and natural language processing

methods to promote credible health information and demote misinformation. At the same time, with enhanced

veracity, anonymity, and privacy controls, the vision of Cardea is for e-patients to conﬁdently share experi-

ences and opinions without being stigmatized or compromising their right to privacy. Our hope is to generate

discussion and gather insights from other researchers on developing Cardea.

1 INTRODUCTION

The Internet originally allowed a few people to gen-

erate content on websites. With the advent of social

media, anyone with access to the Internet was able to

have their say and generate content. Laypersons and

patients have used this opportunity to interact, col-

laborate, and share their personal health stories, ad-

vise, and opinions on Health Social Media (HSM).

However, this has led to varying degree of misinfor-

mation being propagated online, with severity rang-

ing from acute to chronic, depending on the nature of

the topics being discussed (Oliphant, 2009). Early on,

there were signs of the severity of this problem when

websites that promoted harmful cures for cancer us-

ing apricot pits were widely being read, despite being

banned by the U.S. Food and Drug Administration

(FDA) due to pits containing cyanide that would be

harmful when consumed in large doses. Also not too

long ago, the impact of health misinformation was felt

once more as anti-vaccination campaigns started af-

ter viral posts on Facebook linked autism and measles

vaccines (Nyhan et al., 2014). We are currently living

through the COVID-19 pandemic which has turned

into a full-blown infodemic with severe consequences

for misinformation being spread online about vari-

ous facets of the disease including origins, causes,

symptoms, prevention, and cures (Ferrara, 2020). In

the face of this new reality, laypersons crucially need

trusted information in HSM discourse.

Medical professionals have been using HSM as

well, albeit to a more limited extent. A few web-

sites enable medical professionals, such as doctors

and nurses, to consult with patients via video con-

ferencing or phone call. For instance, My Health

Alberta’s 811 HealthLink service

allows patients

HealthLink https://myhealth.alberta.ca/811

Samuel, H., Hassan, F. and Zaíane, O.

The Need for Medical Professionals to Join Patients in the Online Health Social Media Discourse.

DOI: 10.5220/0010325806370644

In Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2021) - Volume 5: HEALTHINF, pages 637-644

ISBN: 978-989-758-490-9

637

or caregivers to connect with registered nurses via

phone call to discuss or ask questions on non-severe

health issues (Sharun, 2012). Other paid telemedicine

services, such as Dialogue, offer virtual healthcare

services through web chat and video conferenc-

ing (Shecter, 2017). It has also become common for

healthcare organizations to have social media pres-

ence for brand recognition. Some websites are ded-

icated to enabling medical professionals exchange in-

formation with each other, such as DocCheck

. Other

forums like Doctors Lounge

enable patients to ask

questions of medical professionals.

The challenges of medical professionals dedicat-

ing time to HSM are understandable. With limited

human resources and busy schedules, it is not al-

ways feasible to engage patients online. However, it

is equally important to acknowledge that a new type

of patients, termed as an e-patients, are seeking self-

education for self-diagnosis and self-treatment on-

line (Fox, 2008). These e-patients seek information

about their health online and feel comfortable inter-

acting with health professionals online as well via

telemedicine. Getting involved in the online discourse

would be beneﬁcial at multiple levels of public health

and medical practice. Patients and laypersons who

resort to unproven cures often do so out of desper-

ation and lack of engagement and advise. Medical

professionals can remedy this situation by authorita-

tively clarifying misinformation and also understand-

ing latent needs of patients, essentially giving rise,

credence, and popularity to e-medics. There is deﬁ-

nitely growing awareness on the importance of know-

ing and being prepared for e-patients as part of medi-

cal education (Masters, 2017), but this may fall short

of expectations until medical professionals meet e-

patients where they are: online on the Internet.

Towards this goal of functional health literacy, the

vision of Cardea is to build a free-to-use HSM portal

that is suitable and attractive for medical profession-

als as well as patients and laypersons. At the core of

Cardea is the metaphor of the hospital building, rep-

resented in the online environment. A hospital has

different rooms, some are public and some are pri-

vate, some are meant for doctors or nurses only, oth-

ers for patients, and others for interactions between

doctors, nurses and patients. Similarly, Cardea pro-

vides Patient to Patient (P2P) secure online pages

where laypersons can interact exclusively with other

patients. Medical professionals can interact with pa-

tients in the Patient to Medic (P2M) online web pages,

while medical professionals can engage in private dis-

course on the Medic to Medic (M2M) pages. The in-

DocCheck https://www.doccheck.com

Doctors Lounge https://www.doctorslounge.com

teractions follow the traditional HSM format of asyn-

chronous textual conversations. Additionally, patients

and medics are able to chat synchronously in real

time via online chat on the P2M web pages. All dis-

cussions are indexed and grouped by health topics,

hence patients can join these virtual support groups

to get information by health topic. There are several

forums that cater speciﬁcally to only patient-patient

discussions, such as Patients Like Me, Doctissimo,

WebMD, Health Boards, Medical Sciences Stack Ex-

change, to mention a few. There are also some forums

that are meant for medic-medic connections, such as

the defunct BMJ’s Doc2Doc forums, and the afore-

mentioned DocCheck. Cardea’s added value is in pro-

viding a uniﬁed portal for both patients and medics.

In this paper, we present the functional design

of Cardea

and showcase how the needs of various

health and medical stakeholders can be served with

ease. In addition, we discuss how HSM can use ma-

chine learning, information retrieval, and natural lan-

guage processing to support information credibility,

right to privacy, and personal information security.

2 BENEFITS AND CAVEATS OF

HEALTH SOCIAL MEDIA

2.1 Fighting Misinformation

There are various deﬁnitions of trust among users, and

it is often used interchangeably with reliability and

credibility, but the general consensus is that trust in-

volves a willing interaction between two or more en-

tities. There is an implicit belief that the interaction

will at least be self-beneﬁcial in the worst case, and

mutually beneﬁcial to all entities involved in the best

case (Golbeck, 2005). There is no guarantee that this

belief is correct. Nevertheless, some level of trust,

however minuscule, is fundamentally essential for in-

teractions to happen, even when given limited or non-

existent knowledge about another entity or group of

entities. Despite the popularity of HSM and an im-

plicit sense of trustworthiness between users, untested

claims for cures, lack of contraindications, and false

claims about disease prevention are common features,

having life-threatening potential.

Social networks have presented enormous oppor-

tunities to connect across the world, but also inad-

vertently have allowed the rampant propagation of in-

accurate information, rumours, propaganda and con-

spiracy theories. Recently, the term “infodemic” has

Source code https://github.com/hwsamuel/Cardea

HEALTHINF 2021 - 14th International Conference on Health Informatics

638

gained popular attention as an umbrella term to re-

fer various types of inaccurate information related to

COVID-19 (Wang et al., 2019; Zarocostas, 2020).

Conjoining “information” and “epidemic”, this port-

manteau word uses disease or illness as a metaphor

and illustrates the pervasiveness of health misinfor-

mation. HSM like Cardea can help medical profes-

sionals to have a nuanced understanding of the detec-

tion and spread of health misinformation and to de-

sign effective intervention strategies against them.

Understanding human behaviours is a precursor

to design and implement such interventions. HSM

like Cardea therefore provide a unique opportunity to

closely examine the circumstances of infodemics and

to strategically design interventions that can go be-

yond simply raising awareness. Organic interaction

between patients and medical professionals can act

as a foundation for a comprehensive understanding of

how human behaviours are impacted by the infodemic

at a different individual, organizational, and commu-

nity level. This perspective can also help to explore

the different factors associated with a multi-faceted

problem like an infodemic.

Moreover, patients interacting with medical pro-

fessionals provides the laypersons with an authorita-

tive source of knowledge to gravitate towards when

dealing with actors who want to spread misinforma-

tion. The prevalent strategy in measuring information

quality in HSM presently is consensus-based “wis-

dom of the crowd” methods such as votes and likes.

When only patients and laypersons are interacting

with each other, these subjective metrics could easily

lead to creation of echo chambers. With medical pro-

fessionals getting involved in appraising online con-

tent through interactions on HSM, this also opens up

possibilities for objective trust metrics that are based

on machine learning models trained on the votes and

feedback of medical professionals.

2.2 Preventing Stigmatization

In addition to physical accessibility, there are various

cultural, social and psychological barriers to health

seeking behaviors (Stangl et al., 2019). Certain health

conditions and diseases, particularly related to addic-

tion, mental health, or sexually transmitted infections,

have social stigma associated with them. Stigmati-

zation of these health conditions often result in de-

lay in care and over-reliance on treatment based on

online research which can ultimately lead to possi-

ble harmful consequences, reinforce misinformation,

and overall poor health outcome of patients (Hatzen-

buehler et al., 2013). Even though there are sev-

eral health promotion initiatives on raising awareness,

the current healthcare system is primarily based on

an intervention-focused model which relies on self-

reports and does not account for public perception,

social norms, and other individual or social determi-

nants that inﬂuence both individual and population

health (Stangl et al., 2019).

HSM can mitigate the harmful consequences of

health-related stigma on both communicable and non-

communicable diseases. Peer networks and online

support circles can normalize reaching out for help

on health conditions which are perceived negatively

in our society. Moreover, such online platforms can

enhance the role of physicians and other healthcare

professionals as advocates of healthy lifestyles which

will facilitate a more prevention-based health mea-

sure (Luft, 2017).

At the same time, the personal identity of HSM

users usually needs to be hidden to avoid social

stigmatization online. Typically, the author of a post-

ing on a social media website is identiﬁable by their

registered user name or full name being displayed

next to the post. However, situations can arise in

which the user does not want to be identiﬁed at all.

For instance, if a user were to share a link with their

friends about sexual dysfunction or infertility, the user

may wish to do so anonymously to avoid any stigma-

tization resulting from the assumption that sharer suf-

fers from the condition. Despite the potential sever-

ity of online social stigma, options and controls to

anonymously post content are not well-supported in

most social media websites. Users on these web-

sites may hide their real identity by creating a new

account with a fake name or pseudonym, thereby du-

plicating the website’s user base. This is not ideal and

unnecessarily complicates the process of information

sharing. From the list of popular social media web-

sites such as Facebook, Twitter, LinkedIn, YouTube,

Stack Exchange and Quora, only the latter allows ask-

ing questions anonymously without needing to create

a new account. There are also potential drawbacks

with the approach to replace the user’s real identity

with the generic pseudonym “anonymous”. Firstly,

despite their name being hidden, users still may be

inadvertently revealing their identity because of the

similarities between the content they have posted in

the past. Phrases, wordings, topics and other nuances

about the writing style in the user’s past postings may

constitute a quasi-identiﬁer that can be associated

with a speciﬁc user. Quasi-identiﬁers are not unique

by themselves but can be correlated with an user’s

identity due to frequency of occurrence or other pat-

terns (Sweeney, 2000). Secondly, the generic anony-

mous pseudonym also eliminates the user’s associated

reputation, motivating the need for trust-preservation

The Need for Medical Professionals to Join Patients in the Online Health Social Media Discourse

639

Figure 1: Cardea homepage (screenshot from web application); 1 - Forums and support group folksonomies; 2 - Content

types; 3 - Authorship settings; 4 - Privacy controls.

during anonymization. Cardea addresses these issues

and enables free online discourse among patients and

medical professionals without fear of stigmatization.

2.3 Privacy Preservation

HSM offer insightful behavioural data which can im-

prove overall access to healthcare and can be used for

evidence-based policy-making (George et al., 2013).

However, for a successful implementation of such

networks, there are various factors that need critical

considerations, such as ensuring user privacy, ethical

use of technology, and understanding challenges and

limitations of the network (Grajales III et al., 2014).

Cardea generalizes the various viewpoints of pri-

vacy observed in HSM into two broad categories of

content visibility and user identity. Users are given

full control of who can see the content that is being

posted. Moreover, users also have control over their

how their identity is associated with the content they

post. Identity privacy is often not considered in HSM

and can lead to unwanted online social stigmatization.

A third category covers the need for analysis of

user-generated content for research purposes. To fa-

cilitate this while preserve the user’s right to privacy,

only aggregated content analysis will be supported af-

ter ﬁltering Personally-Identiﬁable Information (PII).

All Cardea users are required to consider the use of

HSM in the context of permission, privacy, informed

consent, security and the broader social impact. Fu-

ture extension and application of Cardea can be fur-

ther aligned with the existing legislation and regula-

tions on personal information protection and the ex-

change of electronic information.

3 FUNCTIONAL DESIGN

SPECIFICATIONS OF CARDEA

Cardea aims to address the various challenges out-

lined with trust, stigmatization, and privacy. From

a research perspective, Cardea has been conceptual-

ized as a sandbox for facilitating and understanding

HSM interactions. The Cardea homepage

is shown

in Figure 1 while various other features outlined in

this section are summarized in Figure 2.

3.1 User Types

Cardea allows two user roles: Patient or Medic. The

patient is a generic label for anyone who would be

seeking health information, while the medic label is

for identifying medical professionals. In addition, a

select group of users are assigned as Moderators for

housekeeping and administrative tasks related to the

website and content. Medics are manually veriﬁed

based on their institutional email and correspondence

with their organizations to ensure good standing. Un-

accredited users are assigned the patient role by de-

fault until veriﬁed.

Alpha version of Cardea demo website available on https:

//www.cardeahealth.ca

HEALTHINF 2021 - 14th International Conference on Health Informatics

640

Figure 2: Cardea features showcase (screenshot from web application); 1 - Search and exploration; 2 - Content sorting (default

by trust); 3 - External information integration and recommendation.

3.2 Content Types

Cardea allows registered users to post text or hyper-

links to websites, images, and videos. Content is

created within three categories: Questions, Discus-

sions, and Blogs. Questions are focused on getting

answers, while discussions are more open-ended con-

versations. Blogs provide an avenue for users to cu-

rate opinion pieces. Additionally, replies allow other

users to be able to respond, leading to conversation

threads. Replies to discussions and blog posts are re-

ferred to as Comments, while replies to questions con-

stitute Answers. A fourth content category of Chats

between medics and patients is also provided only

within the P2M forum.

3.3 Folksonomies

A folksonomy is a methodology for allowing users to

tag items, whereby the tags can be automatically orga-

nized as a classiﬁcation system based on tag frequen-

cies (Morrison, 2007). In Cardea, content is tagged

by Forums and Support Groups. There are 3 forums

based the metaphor of a hospital building for special-

ized and segregated online conversations: Patient to

Patient (P2P), Patient to Medic (P2M), and Medic to

Medic (M2M). The P2P forum allows only patients

to create new content, while the P2M forum allows

patients and medics to interact. The M2M forum is

meant exclusively for medics only. Depending on the

privacy settings, the content itself within the forums

may be visible to other members, but the ability to

create new content is restricted based on user types

associated with the forums. Support groups consti-

tute speciﬁc health-related topics that allow grouping

of questions, discussions, and blogs topically.

3.4 Privacy Controls

Users can specify who can view their content using

four levels of privacy, from broadest visibility to more

limited viewership: Public, Registered, Medics / Pa-

tients, and Connections. At the ﬁrst level, content can

be shared publicly and viewed by all visitors to the

website without needing an account. The next level

of privacy restricts content viewing only to users who

have registered on the website. Thirdly, the type of

user and the forum being browsed dictates viewer-

ship, either medic or patient. At the fourth level, users

can opt to share the content only with other users that

they have explicitly added as connections. Connec-

tions are required to accept requests, hence a connec-

tion is a two-way virtual relationship between users.

3.5 Authorship Settings

In addition to privacy control for visibility of con-

tent, users are also able to determine how their iden-

tity is displayed and associated with their own con-

tent. Three options are provided to label author-

ship: Myself, Pseudonym, or Anonymous. For the

ﬁrst option, the user’s registered name is shown

The Need for Medical Professionals to Join Patients in the Online Health Social Media Discourse

641

for authorship. For the second option, a trust-

preserving pseudonym will be automatically assigned

to the user. The pseudonym is assigned using a

two-stage approach outlined in the Iron Mask algo-

rithm (Samuel and Za

ıane, 2017): ﬁrstly, the content

to be posted is scrutinized to determine the probabil-

ity of de-anonymization using machine learning and

the whiteprint identiﬁcation approach (Keretna et al.,

2013) on the user’s prior historical content.

Whiteprint is deﬁned as the unique writing

style of an author based on grammatical and lexi-

cal patterns (Kayarkar and Ricchariaya, 2014; Ma-

sood et al., 2019). This approach enhances

anonymity by minimizing the risk of re-identiﬁcation

or de-anonymization. If there is no risk of de-

anonymization, then the user’s authorship label dis-

played to the user’s connections is “Your Connec-

tion”. This allows other connections to associate

some level of trustworthiness with this user, even

though their actual identity is hidden. For all other

users, the user’s type is displayed, either medic or

patient. The last option is to show the generic label

“Anonymous” as the author.

3.6 Trust Metrics

The veracity of content is established within Cardea

using subjective and objective metrics. Cardea allows

users to provide subjective feedback on the quality of

content by reacting with Likes or Dislikes. These spe-

ciﬁc reactions are limited to questions, discussions,

blogs, and comments. Answers to questions can be

Up-Voted or Down-Voted to reﬂect the extent to which

the answer addressed the original question, in addi-

tion to quality and correctness.

Objective metrics enable credible content to be

surfaced using established medical knowledge. This

is achieved by processing content containing medical

claims through MedFact (Samuel and Zaiane, 2018),

which uses automated information retrieval process

with natural language processing and machine learn-

ing to compare the claims against known facts ex-

tracted from publications in reputed medical journals.

At the end, a percentage score is calculated to repre-

sent Agreement of claims with known medical facts.

In addition, votes and likes by veriﬁed medics are

given more weight when scoring content quality and

ranking search results. Additionally, Cardea’s user

reputation system keeps track of positive and negative

feedback received on users’ content. Reputations are

displayed to other users, contextualized by topic. For

instance, a user may receive good feedback on topics

related to pediatrics, but negative feedback on other

topics.

3.7 Search and Recommender Results

Cardea provides standard search mechanisms, while

also allowing faceted ﬁltering within results by con-

tent topic and type. In addition, exploratory search

is available whereby new content can be discovered

from search results by displaying related concepts to

the user’s currently viewed content. For exploratory

search, Cardea incorporates BubbleNet (Mohajeri

et al., 2016), which presents an abstract and high-level

representation of major concepts, keywords, and top-

ics discussed in a set of conversation threads. The

relationships are visualized in the form of a network,

showing the topics as well as their inter-connections.

This network is built using an estimation of semantic

relationships between topics. The user then can navi-

gate through this network by either reﬁning or expan-

sion. The user can drill down from a given topic to see

other related concepts in a lower and more detailed

level. The user can also navigate to other related top-

ics and ﬁnally ﬁnd a set of documents talking about

their desired topics. This interface is shown in Fig-

ure 2 alongside other features.

Cardea facilitates discussions by recommending

external content that is relevant to existing conver-

sations, including real-time chats between patients

and medics using PubMedReco, a real-time recom-

mender to suggest citations from PubMed (Samuel

and Za

ıane, 2017). PubMedReco analyzes keywords

within synchronous chats to form search queries for

retrieving PubMed citations. The same approach is

applied to asynchronous conversations for suggesting

articles from Health Canada. Users can then directly

discuss these news items within Cardea.

4 DISCUSSION

The COVID-19 pandemic has revealed the widening

gap between health experts and the general public

in terms of following appropriate public health mea-

sures. To understand such gaps, it is important to criti-

cally examine the communication process itself: how

people seek health information, share their concerns

with medical professionals, interpret new information

or suggestions and follow the medical guidelines. The

current literature in public health and behavioral sci-

ence show how making decisions on personal health

is a subjective process embedded into personal expe-

rience, prior knowledge, cultural norms, and social

and physical environment (Glanz and Bishop, 2010).

In the absence of proper communication between e-

patients and medical professionals, laypersons can of-

ten be persuaded by anecdotal evidence and make cru-

HEALTHINF 2021 - 14th International Conference on Health Informatics

642

cial decisions based on prior belief or online misinfor-

mation (Swire-Thompson and Lazer, 2020).

On the other hand, there are various challenges

concerning patient-medic online interactions, such as

lack of ﬁnancial incentives for medics, regulations on

patient personal data transmission, and insurance cov-

erage on online telemedicine services. Current litera-

ture on health equity has also highlighted the fact that

although modern technology has increased the overall

accessibility of information, it is still far from ideal

and has inadvertently created a digital divide (Mul-

langi et al., 2019). What is even more concerning is

that people’s access to information now depends on

their online activities.

Search engines and social media inﬂuence peo-

ple’s information seeking behaviour and customize

results based on their online interaction (Mowshowitz

and Kawaguchi, 2002). Such selective exposure to in-

formation eventually creates “echo chambers” or dis-

jointed bubbles between experts and the public (Choi

et al., 2020). For example, Getman and colleagues

studied the network of online discussions on vacci-

nation and empirically proved the limited or rare in-

teractions between the experts and vaccine-hesitant or

anti-vaccination groups (Getman et al., 2018). These

gaps are further exacerbated by the spread of health

misinformation such as conspiracy theories of dis-

ease origins, ampliﬁed risks and false claims on health

beneﬁts. General social media platforms and other

pseudo-scientiﬁc websites are inadvertently becom-

ing hosts of such inaccurate information posted and

propagated by conspiracy theorists, political parties,

celebrities or autonomous programs (bots) (Vosoughi

et al., 2018). The convoluted online space makes it

difﬁcult for the public to track the source of news and

evaluate its credibility (Bridgman et al., 2020).

There is a perceived notion that these divides

emerge from the lack of education or informa-

tion (Mackert et al., 2016). While that might be true

to some degree, we argue that such divides are rooted

in more systemic issues within the practice of science

itself. Historically, health research, or the scholarship

of science as a whole, has been conducted in isolation

with limited engagement between researchers and the

public (Eagle et al., 2003). New discoveries are of-

ten conﬁned in academic circles or speciﬁc interest

groups. Timely access to health information based on

new discoveries of drugs or treatment is a major chal-

lenge for patients, particularly those in rural areas.

5 CONCLUSION

HSM can play a bridging role and address these chal-

lenges in various ways, but more speciﬁcally, by

empowering both medical professionals and patients

through inclusive online communication research; ad-

dressing health misinformation, creating a safe space

to discuss sensitive health issues; and improving the

overall access to healthcare. Grounding the health

communication process in equity, inclusion and im-

pact, such online networks can also help promote a

more user-centred design of technology as well as its

effective application to improve overall public health.

Additionally, through active participation in HSM,

medical professionals can gain valuable insights on

patient perspective and perception on emerging dis-

eases. Similarly, patients and caregivers can connect

with their peers and engage in knowledge exchange

which may lead to an increase in their awareness and

adherence to health guidelines. In this position paper,

we presented our ongoing development work on the

Cardea Health Social Network with various features

for trust-enforcement and privacy-preservation. The

ultimate goal of Cardea is to create an attractive en-

vironment for medical professionals to fully engage

with e-patients in online discourse as e-medics. From

a research perspective, our hope is to gather feedback

and generate discussion for enhancing Cardea.

REFERENCES

Bridgman, A., Merkley, E., Loewen, P. J., Owen, T., Ruths,

D., Teichmann, L., and Zhilin, O. (2020). The Causes

and Consequences of COVID-19 Misperceptions: Un-

derstanding the Role of News and Social Media. So-

cArXiv.

Choi, D., Chun, S., Oh, H., Han, J., et al. (2020). Rumor

Propagation is Ampliﬁed by Echo Chambers in Social

Media. Scientiﬁc Reports, 10(1):1–10.

Eagle, K. A., Garson Jr, A. J., Beller, G. A., and Sennett, C.

(2003). Closing the Gap between Science and Prac-

tice: The Need for Professional Leadership. Health

Affairs, 22(2):196–201.

Ferrara, E. (2020). What Types of COVID-19 Conspiracies

are Populated by Twitter Bots? First Monday, 25(6).

Fox, S. (2008). The Engaged E-Patient Population. Wash-

ington, DC: Pew Internet & American Life Project.

George, D. R., Rovniak, L. S., and Kraschnewski, J. L.

(2013). Dangers and Opportunities for Social Me-

dia in Medicine. Clinical obstetrics and gynecology,

56(3).

Getman, R., Helmi, M., Roberts, H., Yansane, A., Cut-

ler, D., and Seymour, B. (2018). Vaccine Hesitancy

and Online Information: The Inﬂuence of Digital Net-

works. Health Education & Behavior, 45(4):599–606.

The Need for Medical Professionals to Join Patients in the Online Health Social Media Discourse

643

Glanz, K. and Bishop, D. B. (2010). The Role of Behavioral

Science Theory in Development and Implementation

of Public Health Interventions. Annual review of pub-

lic health, 31:399–418.

Golbeck, J. A. (2005). Computing and Applying Trust in

Web-based Social Networks. PhD thesis, University

of Maryland at College Park.

Grajales III, F. J., Sheps, S., Ho, K., Novak-Lauscher, H.,

and Eysenbach, G. (2014). Social Media: A Review

and Tutorial of Applications in Medicine and Health

Care. Journal of medical Internet research, 16(2):e13.

Hatzenbuehler, M. L., Phelan, J. C., and Link, B. G.

(2013). Stigma as a Fundamental Cause of Popula-

tion Health Inequalities. American journal of public

health, 103(5):813–821.

Kayarkar, P. and Ricchariaya, P. (2014). An Enhanced Ap-

proach for Digital Forensics using Innovative GSP Al-

gorithm. International Journal of Computer Applica-

tions, 103(6).

Keretna, S., Hossny, A., and Creighton, D. (2013). Recog-

nising User Identity in Twitter Social Networks via

Text Mining. In IEEE International Conference on

Systems, Man, and Cybernetics, pages 3079–3082.

IEEE.

Luft, L. M. (2017). The Essential Role of Physician as Ad-

vocate: How and Why We Pass It On. Canadian med-

ical education journal, 8(3):e109.

Mackert, M., Mabry-Flynn, A., Champlin, S., Donovan,

E. E., and Pounders, K. (2016). Health Literacy and

Health Information Technology Adoption: The Poten-

tial for a New Digital Divide. Journal of medical In-

ternet research, 18(10):e264.

Masood, F., Almogren, A., Abbas, A., Khattak, H. A., Din,

I. U., Guizani, M., and Zuair, M. (2019). Spammer

Detection and Fake User Identiﬁcation on Social Net-

works. IEEE Access, 7:68140–68152.

Masters, K. (2017). Preparing Medical Students for the E-

Patient. Medical Teacher, 39(7):681–685.

Mohajeri, S., Samuel, H. W., Za

ıane, O., and Raﬁei,

D. (2016). BubbleNet: An Innovative Exploratory

Search and Summarization Interface with Applicabil-

ity in Health Social Media. In International Con-

ference on Digital Economy (ICDEc), pages 37–44.

IEEE.

Morrison, P. J. (2007). Folksonomies: Why Are They Tag-

ging, and Why Do We Want Them To? Bulletin of the

American society for information science and technol-

ogy, 34(1):12–15.

Mowshowitz, A. and Kawaguchi, A. (2002). Assessing Bias

in Search Engines. Information Processing & Man-

agement, 38(1):141–156.

Mullangi, S., Kaushal, R., and Ibrahim, S. A. (2019). Eq-

uity in the Age of Health Care Information Technol-

ogy and Innovation: Addressing the Digital Divide.

Medical care, 57:S106–S107.

Nyhan, B., Reiﬂer, J., Richey, S., and Freed, G. L. (2014).

Effective Messages in Vaccine Promotion: A Ran-

domized Trial. Pediatrics, 133(4):e835–e842.

Oliphant, T. (2009). “I Am Making My Decision on the

Basis of My Experience”: Constructing Authoritative

Knowledge about Treatments for Depression. Cana-

dian Journal of Information and Library Science / Re-

vue Canadienne Des Sciences De L’Information Et De

Bibliotheconomie, 33(3-4):215–232.

Samuel, H. and Za

ıane, O. (2017). Iron Mask: Trust-

Preserving Anonymity on the Face of Stigmatization

in Social Networking Sites. In International Confer-

ence on Trust and Privacy in Digital Business, pages

66–80. Springer.

Samuel, H. and Za

ıane, O. (2017). PubMedReco: A

PubMed Citations Recommender System for Real-

Time Chat. In 16th IMIA World Congress on Medical

and Health Informatics (MedInfo).

Samuel, H. and Zaiane, O. (2018). MedFact: Towards

Improving Veracity of Medical Information in Social

Media using Applied Machine Learning. In Canadian

Conference on Artiﬁcial Intelligence, pages 108–120.

Springer.

Sharun, M. (2012). MyHealth.Alberta.ca: Strategies and

Lessons Learned for a Provincial Health Portal. In

Medicine 2.0 Conference. JMIR Publications Inc.

Shecter, B. (2017). Health Startup Lets Workers See A Doc-

tor Without Skipping Work. Retrieved online from

https://ﬁnancialpost.com/technology/health-startup-

lets-workers-see-a-doctor-without-skipping-work on

2020-10-01.

Stangl, A. L., Earnshaw, V. A., Logie, C. H., van Brakel,

W., Simbayi, L. C., Barr

e, I., and Dovidio, J. F.

(2019). The Health Stigma and Discrimination Frame-

work: A Global, Crosscutting Framework to Inform

Research, Intervention Development, and Policy on

Health-Related Stigmas. BMC medicine, 17(1):31.

Sweeney, L. (2000). Simple Demographics Often Iden-

tify People Uniquely. Health (San Francisco),

671(2000):1–34.

Swire-Thompson, B. and Lazer, D. (2020). Public Health

and Online Misinformation: Challenges and Recom-

mendations. Annual Review of Public Health, 41:433–

451.

Vosoughi, S., Roy, D., and Aral, S. (2018). The

Spread of True and False News Online. Science,

359(6380):1146–1151.

Wang, Y., McKee, M., Torbica, A., and Stuckler, D.

(2019). Systematic Literature Review on the Spread

of Health-Related Misinformation on Social Media.

Social Science & Medicine, 240:112552.

Zarocostas, J. (2020). How to Fight an Infodemic. The

Lancet, 395(10225):676.

HEALTHINF 2021 - 14th International Conference on Health Informatics

644