Published on in Vol 2, No 2 (2013): Jul-Dec

How Twitter Is Studied in the Medical Professions: A Classification of Twitter Papers Indexed in PubMed

How Twitter Is Studied in the Medical Professions: A Classification of Twitter Papers Indexed in PubMed

How Twitter Is Studied in the Medical Professions: A Classification of Twitter Papers Indexed in PubMed

Original Paper

1School of Systems Engineering, University of Reading, Reading, United Kingdom

2Department of Information Studies, University College London, London, United Kingdom

*all authors contributed equally

Corresponding Author:

Shirley Ann Williams, BSc, PhD

School of Systems Engineering

University of Reading


Reading, RG6 6AY

United Kingdom

Phone: 44 118 378 8613

Fax:44 118 975 1994


Background: Since their inception, Twitter and related microblogging systems have provided a rich source of information for researchers and have attracted interest in their affordances and use. Since 2009 PubMed has included 123 journal articles on medicine and Twitter, but no overview exists as to how the field uses Twitter in research.

Objective: This paper aims to identify published work relating to Twitter within the fields indexed by PubMed, and then to classify it. This classification will provide a framework in which future researchers will be able to position their work, and to provide an understanding of the current reach of research using Twitter in medical disciplines.

Methods: Papers on Twitter and related topics were identified and reviewed. The papers were then qualitatively classified based on the paper’s title and abstract to determine their focus. The work that was Twitter focused was studied in detail to determine what data, if any, it was based on, and from this a categorization of the data set size used in the studies was developed. Using open coded content analysis additional important categories were also identified, relating to the primary methodology, domain, and aspect.

Results: As of 2012, PubMed comprises more than 21 million citations from biomedical literature, and from these a corpus of 134 potentially Twitter related papers were identified, eleven of which were subsequently found not to be relevant. There were no papers prior to 2009 relating to microblogging, a term first used in 2006. Of the remaining 123 papers which mentioned Twitter, thirty were focused on Twitter (the others referring to it tangentially). The early Twitter focused papers introduced the topic and highlighted the potential, not carrying out any form of data analysis. The majority of published papers used analytic techniques to sort through thousands, if not millions, of individual tweets, often depending on automated tools to do so. Our analysis demonstrates that researchers are starting to use knowledge discovery methods and data mining techniques to understand vast quantities of tweets: the study of Twitter is becoming quantitative research.

Conclusions: This work is to the best of our knowledge the first overview study of medical related research based on Twitter and related microblogging. We have used 5 dimensions to categorize published medical related research on Twitter. This classification provides a framework within which researchers studying development and use of Twitter within medical related research, and those undertaking comparative studies of research, relating to Twitter in the area of medicine and beyond, can position and ground their work.

Med 2.0 2013;2(2):e2



Since their inception in 2006, Twitter and similar microblogging systems have provided data for research, with the first academic paper on the subject appearing in 2007 [1]. Articles in the popular news media highlight the potential of Twitter based research to meet a number of goals ranging from measuring public sentiment to spotting flu outbreaks [2]. However, there has been little work done beyond the headlines in understanding how or why people are using information gathered from Twitter systems for research, particularly around specific topic areas.

The terms microblog and Twitter are both widely used by authors, dating from the first paper on the subject [1]. The term microblogging is defined as:

A variant of blogging which allows users to quickly post short updates, providing an innovative communication method that can be seen as a hybrid of blogging, instant messaging, social networking and status notifications. The word’s origin suggests that it shares the majority of elements with blogging, therefore it can potentially be described using blogging’s three key concepts: the contents are short postings, these postings are kept together by a common content author who controls publication, and individual blog entries can be easily aggregated together.

Some writers hyphenate the term as “micro-blog” [5], while other do not [6]. We follow the majority and use the unhyphenated version, although while searching for papers on the topic we utilized both. Twitter is usually defined in terms as microblogging:

Twitter is a microblogging site, originally developed for mobile phones, designed to let people post short, 140-character text updates or “tweets” to a network of others. Twitter prompts users to answer the question “What are you doing?”, creating a constantly- updated timeline, or stream, of short messages that range from humor and musings on life to links and breaking news. Twitter has a directed friendship model: participants choose Twitter accounts to “follow” in their stream, and they each have their own group of “followers”.

PubMed is a free Web literature search service developed and maintained by the National Center for Biotechnology Information (NCBI) [8]. Since 1996, PubMed gives access to citation and abstracts of some 5400 biomedical journals covering the fields of medicine, nursing, dentistry, veterinary medicine, health care systems, and preclinical sciences. The intended users of PubMed are researchers, health care professionals, and the general public. For the intended users, PubMed serves as the primary tool for electronically searching and retrieving biomedical literature [9]. Fink [10] describes PubMed as “the best site for published medical and health research”. PubMed uses the Medical Subject Headings (MeSH) controlled vocabulary to supplement searches. MeSH pre-dates PubMed with its origins in the 1960s as a set of catalog headings across medicine composed by the US National Library of Medicine [11]. Entries to MeSH are regularly updated to match changes in medicine and technology.

In common with many other papers, we used the term Twitter to encompass all microblogging systems. The work was not a traditional literature review [10]. Instead, only papers indexed by PubMed were considered and only those related to Twitter were reviewed then classified.

This work will provide a framework with which researchers studying Twitter related topics and their applications in medical related areas will be able to position and ground their work. It will provide a single point where current work on the medical use of Twitter can be compared and contrasted. Additionally it will help to understand the scope and reach of using PubMed as a data source.

Our analysis shows that Twitter related research can be classified in a variety of ways: whether it is Twitter-focused or part of a wider social media related study; whether it is based on data, and if so, the quantity of data considered; the domain in which the work is based; the methods used; and the aspect–or characteristic–of Twitter considered. These dimensions of classification provide a framework in which Twitter-related medical research can be positioned and compared with other work within the area and beyond.

Data Collection

Researchers normally identify papers on a topic in a number of different ways such as chaining from existing papers and database searches [12,13]. There are many databases and search engines available to researchers wanting to find papers on a particular topic [10], some of which are freely available, while others are available via individual or institutional subscription [14]. Researchers in areas of emerging technologies sometimes limit themselves to groups of publications [15], single journal sources [16], or concentrate around conferences [17]. While many studies do not indicate their identification method, Cormode et al [18], for example, classify Twitter papers providing examples of “first studies” and the “next set of papers”. Within this work we wanted to investigate the area of Twitter based research in medicine, and for our data collection to be replicable we chose to make a structured search of journal articles.

Initial experimentation showed that for Google Scholar [19] the searches either had to be limited to searching the article’s title or it is full text. Searches limited to articles title would not return “OMG U got flu? Analysis of shared health messages for bio-surveillance” [5] as it does not contain any words related to Twitter. Full text searches returned articles which had “share this on Twitter” buttons on the page even though the article was nothing to do with microblogging. Using our institutional library’s facility to search freely available electronic resources for papers relating to Twitter in the biomedical field, we established that PubMed returned over 100 items while BioMed Central [20] returned around 20, and other databases returned very few papers, and almost all were already in the PubMed list.

Gold et al [21] faced a similar challenge when undertaking a systematic examination of the use of social networking sites for health promotion: from a systematic search of a range of databases they originally found 204 academic papers but closer investigation showed only one was relevant, a Web search revealed over 80 million electronic resources and an unknown number of social networking sites. Likewise Guse et al [22] investigated the use of digital media to improve adolescent sexual health searched a range of databases to identify 942 possible abstracts of which 10 met the inclusion criteria: while they do not indicate which databases they found each paper in, all the 10 studies can be found via PubMed.

It was determined for this study that a structured search using PubMed would be used to identify papers in journals. While this most certainly would not give an exhaustive list of papers on Twitter it does mean that the search is repeatable, by other researchers, allowing future studies to include papers added to PubMed. Using subscription based services (such as Scopus) would mean only some researchers could repeat the study limiting its usefulness as a benchmark.

The data collection was made for the papers that were first published between 2007 (the first year academic papers on Twitter appeared) and 2011 (the last complete year before this study); inclusive of papers available online as preprints ahead of the print version (epubs).

During 2010, the terms “Twitter messenging” and “Twitter messaging” were introduced into the MeSH controlled vocabulary under the headings Internet and Blogging respectively. There are no entries relating to the term microblog or its variants, although blogging is present. There are currently no papers within PubMed that are returned by searches on the MeSH terms: “Twitter messenging” or “Twitter messaging”. It should be noted that where papers have keywords, not indexed by MeSH terms, PubMed does not store these and so it is not possible to search PubMed for papers with keywords such as “Twitter” or “microblog”. Therefore, the terms Twitter, Tweet, Microblog, and Micro-blog were used as the basis for keyword searching across all fields in PubMed, and then cross-referenced and checked to remove spurious data. A total of 139 papers were initially identified which had used terms from the query in a medical context. Five of these were subsequently found to be only included in the results because one of the author’s surnames or usernames included “tweet”, and so a base corpus of 134 papers was created.

Data Classification

Previous research [23] showed that a number of dimensions could be identified and studied when Twitter-related academic papers and their abstracts are analyzed. These include:

  1. Focus. Papers can be predominantly about Twitter or related microblogging such as the use of the Chinese microblog site Sina Weibo [24], or they can be partially about Twitter but predominantly about other things, for example considering a number of different social networking sites of which Twitter is just one [25]. There are also unknowns where a paper has no abstract. Additionally there are papers where the term twitter is used with its conventional meaning such as a noise made by birds.
  2. Data. The data used in studies is varied, ranging from observations of small samples, through questionnaires, to collecting vast quantities of information via the Twitter API (an interface that allows technically skilled users to extract data). The date of the study also impacts on the timeliness, quantity and quality of data.
  3. Domain. Studies are undertaken from a number of different standpoints and often within a domain or a group of domains.
  4. Method. Researchers use a variety of methodological techniques when carrying out research into Twitter.
  5. Aspect. The aspect or characteristic of Twitter considered. Many studies concentrate on looking at the message (tweets), while others study the user (tweeter), with smaller numbers look at the underlying technology and how it can be developed. A number of papers consider the concept of Twitter without any detail of its use.

The overarching approach to classification was based on the approach used in a study of research on microblogging in education [15], with independent coding and then discussion until consensus was reached. For each paper in our corpus, the focus was identified, based on close reading of the title and abstract. Those papers identified as Twitter-focused were subject to a qualitative classification on the title, abstract and full paper using open coded analysis to determine groupings for the data used in the work described. Corbin and Strauss [26] have shown how this methodology facilitates the breaking of corpora data into delineated concepts as well as featuring in grounded theory [27] where initial and focused line by line coding produces label variables from within the data itself. The approach has been previously used successfully to classify Twitter posts [4]. The grouping of method, domain and aspect was initially identified from the paper’s title and abstract and verified by consulting the full paper.


Multimedia Appendix 1 summarizes the flow of selection of papers from our base corpus of 134 papers. From this corpus thirty [5,6,28-55] were Twitter-focused. The papers had a significant proportion that was related to some aspect of microblogging. For example Chew and Eysenbach [31] in their paper entitled “Pandemics in the age of Twitter: content analysis of Tweets during the 2009 H1N1 outbreak” study how Twitter was used in relation to the spread of infection in a pandemic.

There were 57 corpora [21,56-111] that mentioned Twitter but were primarily about another topic. For example Turner-McGrievy and Tate [105] in their paper, “Tweets, Apps, and Pods: Results of the 6-month Mobile Pounds Off Digitally (Mobile POD) randomized weight-loss intervention among adults” study a combination of podcasts and other techniques including using Twitter in relation to weight loss.

Out of 134 papers, 36 [112-147] had no abstract, for example the article “Are you using Twitter for your next survey?” by Pattillo [127]. Further investigation showed that this is a news article within the publication. Papers without abstracts are therefore not considered in any further detail, given that they were news reports rather than academic articles per se. News stories have been shown to be rated differently by medical professionals according to their authorship [148]. Wilson et al [149] took a similar decision to concentrate on academic papers when reviewing papers related to Facebook, and highlighting that while unpublished manuscripts, dissertations, position papers, and popular press articles offer thoughtful insights, their quality is variable.

Out of 134 articles, there were 11 [150-160] not related to microblogging, with 10 of these the term “twitter” being used with original, non-microblogging meanings. For example “Why do shrews twitter? Communication or simple echo-based orientation” [156] is about the noise made by shrews. Exceptional was a paper entitled “Plant twitter: ligands under 140 amino acids enforcing stomatal patterning” [159], as the paper is not about microblogging but in the area of plant research. The MeSH terms used to classify the paper support this, but interestingly the only appearance of “twitter” is in the title; a form of pun. These non-microblogging papers are not considered in any further detail.

Table 1 shows the number of Twitter-focused papers and the number of papers mentioning Twitter published each year between 2007 and 2011, and compares them with the numbers for general journals [23], found by searching Scopus [161] and Web of Science within Web of Knowledge [162]. Note there were no such papers published in medical fields in 2007 and 2008, although they were appearing in other disciplines. Since 2009 the number of papers has increased each year. This analysis suggests that although the use of Twitter in medical research came later than in some other disciplines, its use is growing and its importance is increasing as time progresses. Initial indications for 2012 suggest that the number of papers published both in the area of medicine and more generally will be greater than the numbers published in 2011.

The 2 papers in the corpus published in 2009 [28,29] and 3/8 published in 2010 [30,32,36] discussed the merits of Twitter and whether it should be used by medical professionals. The study of Twitter content for medical related terms was first seen within the corpus in 2010 papers [31,35], while general examination of terms was first presented in 2007 [1].

In the following we consider only the Twitter-focused papers in medical related disciplines. Those papers that use Twitter or other microblogs as a primary source and topic for research as identified via PubMed. Multimedia Appendix 2 combines the information presented in Tables 1-5 for all the Twitter-focused papers.


Across the papers a number of different types of data sources were reported including surveys, user profiles, tweets (posts), and individual words in tweets. The size of data set examined ranged from small, with a few items, to large scale, with billions of individual data points. Some papers were not based on data, particularly those early papers that were introducing the concept of Twitter.

For some papers the abstracts indicated the data studied, for example in a paper “Use of Twitter to encourage interaction in a multi-campus pharmacy management course” [41] the abstract includes the following:

More than eighteen hundred tweets were made by students, guests, and the instructor... One hundred thirty-one students completed an optional evaluation survey.

Indicating the type of data and quantities, the full paper shows that the students posted 1775 tweets over 6 days, as well as indicating the use by other participants. The Twitter data was collected by graduate teaching assistants using a Twitter list in preference to hashtags, which the students are reported to have found cumbersome. In other papers, the abstract provides only partial information about the dataset. For example in a paper “Social media & stem cell science: examining the discourse” [38], the abstract indicates that Twitter posts are analyzed. But the full paper needs to be consulted to identify that the researchers used TweetDeck to collect 2 sets of tweets, one group of 35 using the term “DeGette” over a 6 day period, and a group of 50 using “trachea stem cells” over a 4 day period. Similarly, the paper “Diurnal and seasonal mood vary with work, sleep, and day length across diverse cultures” [43] indicates in the abstract that millions of Twitter messages are considered, the full paper provides more details:

Using’s data access protocol, we collected up to 400 public messages from each user in the sample, excluding users with fewer than 25 messages. The resulting corpus contained about 2.4 million individuals from across the globe and 509 million messages authored between February 2008 and January 2010.

The paper “Implementing Twitter in a health sciences library” [32] is a report on the establishing of a Twitter presence by the communications team within the library. The work is not based on data although in the evaluation section the authors do report on the number of followers (66) the account has gathered and classifying these in relationship to the library.

Stratifying across the different descriptions of data we identified 4 categories which can be used to describe the datasets used to study Twitter in a medical context.

  1. Large. Studies looking at vast amounts of data that would require a team of researchers and the use of automated tools if the data is to be analyzed in a timely manner. Typically considering over a million tweets and/or a million accounts. The term “big data” is often used to describe the quantity of data in such studies
  2. Medium. Studies using quantities of data that could realistically be analyzed manually by a dedicated researcher or a small team with limited tool support. Typically considering thousands of tweets or accounts.
  3. Small. The data handled could be reasonably handled by a researcher alongside other tasks. Typically considering surveys, groups, tweets, and user profiles, with up to a thousand items.
  4. Not data based. Papers not based on data collection and analysis.

Table 2 shows the categorization of data in the Twitter related papers by year published. The early papers (2009 and 2010) were predominantly not based on data, typically explaining the affordances of Twitter. In 2011 all papers had a data element, while there were a range of papers using large, medium, and small scale datasets. There is an increase in large scale analysis of Twitter from 1 study in 2010 to 6 in 2011, indicating that computational analysis of large scale datasets of Twitter data are becoming more common.


All the papers in this study are from PubMed and so the broad domain is medical, however the researchers have a number of different standpoints. Consideration was given to the selection of domains from sub-area and disciplines of medicines, but typically there are only a few papers in each sub-area, see Table 3.

Based on an analysis of the contents of full papers we have identified the following broader topic, or domain, areas. Some papers are allocated to more than one of these domains:

  • Academic. Seven papers in total [30,32,34,37,40,41,48] have an academic perspective ranging through education for professions, libraries, and scholarly publications, to an experimental use of Twitter with groups of students.
  • General Communication. Fourteen papers [5,6,31,35,39,43-46,50-54] examine the general Twitter interface, and do not in any ways select individuals. These include all the papers which analyze large scale datasets.
  • Medical Professional Communication. Nine papers [32,33,36,38,40,42,47,48,55] consider use by professionals within an area, both among themselves and with patients, as well as one way communication to the more general public (including marketing).
  • Targeted Communication. Two papers [38,49] involve other identifiable groups not related to medical professionals. one was an analysis of accounts that were identified as related to quitting smoking [49].
  • Guides. Four of the papers [28-30,36] are written primarily as guides: all of these concentrated on explaining the concept and purpose of Twitter.
Table 1. Number of Twitter related papers published per year.
YearMentions Twitter (Medical)Twitter-focused (Medical)Mentions Twitter (General)Twitter-focused (General)
Table 2. Data categorization of Twitter papers by year.
YearLargeMediumSmallNot data based

Table 3. Sub-areas and number of papers.
Sub-areaNumber of papers
Health education1
Information science1
Natural science1

Methods and Aspects

Initially, the papers’ titles and abstracts were read to try to identify the methodological approach use by the researchers. For the papers with structured abstracts and some others this clearly indicated the approach taken. For example a paper entitled “'What's happening?' A content analysis of concussion-related traffic on Twitter” [54] clearly used a content analysis approach. Following this initial pass, all papers were examined for details of methods used. An open coding approach was used to capture the diversity of approaches. This resulted in across the 30 papers 53 methods identified, and not all of which were distinct, see Table 4.

These methods were then stratified into 3 broad categories:

  1. Analytic. Where the researchers had performed some type of analysis, which may be quantitative or qualitative. Sometimes these methods are supported by existing or new techniques from artificial intelligence, mathematics and statistics to facilitate knowledge discovery and mining of information. Many of the papers use the techniques of content analysis: for example in “Pandemics in the age of Twitter: content analysis of Tweets during the 2009 H1N1 outbreak” [31], while in “OMG U got flu? Analysis of shared health messages for bio-surveillance” [5] machine learning techniques are used alongside content analysis. Social network analysis is used in the paper “Modeling users' activity on twitter networks: validation of Dunbar's number” [44] to extract and analyze 25 million conversations from some 380 million tweets.
  2. Design and Development. Where systems are proposed or built, to interact with Twitter, such systems are often demonstrators used by the authors within their own context. For example, in a paper entitled, “A new support system using a mobile device (smartphone) for diagnostic image display and treatment of stroke” [55], the method of the work is presented as the creation of a communication system that was piloted in the author’s hospital, the system includes the capability to tweet to other professionals. While in “Machine intelligence for health information: capturing concepts and trends in social media via query expansion” [52], the authors develop information retrieval techniques to facilitate working with their Twitter corpus, and in “A visual backchannel for large-scale events” [33] they describe a system they have developed and trials that allows the tweets related to an event to be presented graphically.s
  3. Examination. Where the authors had undertaken review and survey type works, including approaches such as: case studies, categorizations, essays, ethnographic studies, interviews, and investigation. For example in a paper entitled, “Twitter as a communication tool for orthopedic surgery” [42], they identified, categorized, and reviewed Twitter profiles of over 400 orthopedic professionals. While in a paper entitled “Should you be tweeting?” [28], interviews with scientists who use Twitter are presented. This paper would itself be classed as an examination paper.

Alongside the methods the aspect of Twitter primarily considered in research was identified according to the 4 categories:

  1. The messages (tweets).
  2. The users (tweeter).
  3. The underlying technology and how it can be developed.
  4. The concept of Twitter without any detail of its use.

For all medical related papers it was possible to identify a primary method and primary aspect considered by the researchers and these are summarized in Table 5. Some papers also were identified as having secondary aspects, as shown in Multimedia Appendix 2.

It is interesting to note that the majority of the papers report research using analytic methods, and the majority of this group look at the contents of the tweets sent, rather than the users. The 6 papers using examination methods such as reviews considering the concept of Twitter are the same as the 6 papers in Table 2 that are not based on data. A similar classification of general papers [23] identified proportionally many more papers using the design and development methods. The general papers 154 of the total 575 papers primarily using a design and development method on the message aspect. None of the PubMed papers took this approach. Otherwise the PubMed papers do have a similar spread to the general papers.

Table 4. Methodological approaches initially identified.
Methods identifiedNumber of papers
Content analysis12
Machine intelligence2
System development2
System implementation2
Classification analysis1
Comparative analysis1
Correlation analysis1
System design1
Text analysis1
Text mining1
Table 5. Number of papers with primary method and aspect.

Design and development00426

Principal Results

Across PubMed 123 papers were identified that were Twitter related; this is a very tiny proportion of the more than 21 million citations held in the database. The first papers indexed by PubMed were published in 2009, 3 years after the launch of Twitter and 2 years after the first Twitter papers appeared in other disciplines. The early Twitter focused papers introduced the topic and highlighted the potential, not carrying out any form of data analysis. However subsequent studies analyzed quantities of Twitter data and one of the principal findings of this study is the size of studies that are now possible based on Twitter in the medical field. The first of the large studies of over a million pieces of data was published in November 2010 [31]. Researchers are now reporting collecting billions of items of data over almost 3 years [6]. Collecting large quantities of data is challenging, as explained,

Our research material of tweets was gathered by using the Twitter4J … an open-source Java library for the Twitter Application Programming Interface (API). The tweets were stored locally as Twitter limits online search to one week. This strategy allowed an increased sample size improving the likelihood of detecting trends. Twitter API provided approximately one per cent of all real-time tweets. Our tweet corpus included English tweets over fourteen days. The data was gathered during 4 Jan 2011 at 13:36–20:10 EST with 300,000 tweets and 582,975 words.

The Edinburgh Twitter corpus of 97 million tweets was used in one paper [5], however that corpus is no longer available due changes to Twitter’s current terms and conditions [163]. This means researchers are no longer able to share corpuses of Twitter data and so the handling of large sets of data need teams to include the expertise and capacity to extract, store and manipulate large quantities of information. Teams also need to be aware of limitations placed by Twitter on developer’s access to Twitter data and the possibilities of changes during the lifetime of a project. Likewise the methods for understanding the data collected are moving on from what can be undertaken by lone researchers using qualitative approaches, and while the methods used are still broadly analytic they are using techniques from knowledge discovery and mining of information [40].


Limiting the papers examined in this study to those indexed in PubMed between 2007 and 2011 means that there is a body of work published since the start of 2012 that is not considered. While PubMed indexes some 5400 journals there are journals not indexed, including those not in English. A lot of papers published on the subject of Twitter are in conference proceedings. For instance, the Scopus database [161] returns approximately twice as many conference papers as journal papers on the subject (across all fields not just medicine), and there are many conferences that are not indexed. Over and above papers there are many blog posts reporting medical use of Twitter. For example, Bottles [164] describes his personal use of Twitter, and Neylon [165] discusses links shared by nurses. However there is no reliable way of identifying all such posts, nor is it possible to guarantee the posts will remain available. The selection of a single data source does mean that the study is reproducible, and based on published, peer-reviewed research rather than accounts and reflections by individuals. Future comparison can be done on a year by year basis to trace the changing use of Twitter in the medical domain.

Searching on the MeSH terms did not prove useful in highlighting relevant papers. Given the terms “Twitter messaging” and Twitter messenging” were only added to the vocabulary during 2010 this is not totally surprising, although we did expect to see some use of these terms in the most recent publications. This indicates that the MeSH vocabulary system is not being adequately used by authors and publications writing about Twitter, which is problematic given that it is the only faceted search available in PubMed.

The word “twitter” is sometimes used in medical related research with its original meaning. Papers that did this were discounted from this study. Potentially papers may be incorrectly excluded, for example a paper that related both patients with twitters and who used microblogging. We do not believe this was the case in the papers considered here but it is certainly a potential limitation with the approach.

Given that this paper covers only the first few years of academic research in the area of Twitter, it is likely that some of the approaches reported upon are fledgling and that over the next years the methods applied will reach a degree of maturity that will impact on the broad methodological classification presented here.

Analysis of Papers’ Findings

The papers reviewed and categorized here were diverse in their finding and conclusions. Of the findings many were closely linked to the domain of study rather than the use of Twitter or social media in general. For example, the findings and conclusions of Golder and Macy [43] all relate to mood change and day patterns. There was no discussion as to the use of Twitter as a source of data.

In the papers in the domain of professional communications, where usually papers concentrate on the concept of Twitter, rather than findings extrapolated from Twitter data, the approach was usually a review or other method classified above as examination. These tended to conclude that they had introduced Twitter and highlighted its potential. Although some were less enthusiastic.

Despite the growing popularity of social media across multiple disciplines, the majority of pharmacy preceptors surveyed were not willing to use these venues in professional practice.

Papers looking at medium and large data sets often included indications that their work illustrated the potential for studies in medical related area to use Twitter and other social media data.

The study adds to evidence supporting a high degree of correlation between pre-diagnostic social media signals and diagnostic influenza case data, pointing the way towards low cost sensor networks.

Also among these studies authors indicate that the abundance of data will change the way in which researchers approach their studies [6].


This work is to the best of our knowledge the first broad study of medical related research based on Twitter and related microblogging. We have identified that medical related research in this area was first published in 2009 and that the number of papers has increased in both the following years.

From the some 5400 journals indexed by PubMed, we have identified thirty papers that focus on Twitter and 57 that mention it. There are also a number of papers in which the term twitter is used with its original meaning and not at all related to microblogging. There are some papers indexed that appear to relate to Twitter but do not have abstracts further investigations shows these to be editorial or news type items as opposed to academic oriented papers. Further work will need to be undertaken to identify and classify work beyond the academic papers indexed by PubMed, this would include diverse sources such as book chapters, conference proceedings, and blog posts.

While the early Twitter-focused papers were predominantly introductory explaining to the readership what Twitter was about and considering its potential, we are now seeing work reported were researchers have examined large quantities of Twitter data, using these large data sets to obtain better understanding of topics within medicine. We have classified this usage of data into 4 categories: large, medium, small, and no data. This access to large amount of data stemming from individual tweets coupled with metadata of location, time of day, networks of followers holds potential for many future studies building on existing work such as identification of the spread of infectious diseases but it has also potential for the identification of previously impossible studies based on personal thoughts put into a public space. While most studies use methods that can be broadly classed as analytic, the large quantities of data mean that analysis techniques that facilitate knowledge discovery and mining of information are starting to be used. As the number of research papers grows, the dimension of domain will need to be revisited as other stratifications may become possible.

The results presented here will provide researchers with an insight into the medical domain and Twitter use, where there is work in related sub-areas that can be used to inform new studies and those that have still to be studied rigorously. The large data studies that have completed certainly have information on techniques for data collection and method for analysis that will be useful in other domains. Identifying areas where further research is needed is difficult, but we would suggest that the following are neglected areas within the realms of twitter and medicine:

  • Outreach and investigating the reach and scope of Twitter messages. Although Prochaska et al [49] have reviewed the content of accounts related to Quitting Smoking, none of the studies have investigated the reach of such accounts, or the best ways to use them.
  • Public engagement. While Adams et al [38] have investigated what is said about their subjects, there are no investigations where discussion is invited or prompted surrounding medical areas.
  • Legal and ethical issues. While a number of papers (particularly the early ones [28,29]) discuss the general use there are no academic studies of the ethical issues of medical professionals using Twitter, nor any detailed studies of the legal implications of using Twitter in a medical context.

This study provides a framework within which researchers studying the development and use of Twitter within medical related research will be able to position their work and against those undertaking comparative studies of research relating to Twitter in the area of medicine and beyond will be able to ground their work. We have provided an analysis of the use and usefulness of microblogging within medical fields at a time when social media is being increasingly used for research purposes across many domain and in a reproducible manner, which can be built upon in future as more studies are published.


We are grateful to Anne Welsh, UCL Department of Information Studies, for her advice on the selection of databases. The anonymous reviewers provided valuable suggestions that improved the paper.

Conflicts of Interest

None declared.

Multimedia Appendix 1

Flow diagram of search strategy.

PDF File (Adobe PDF File), 37KB

Multimedia Appendix 2

Overview table.

PDF File (Adobe PDF File), 97KB

  1. Java A, Song X, Finin T, Tseng B. Why we twitter: understanding microblogging usage and communities. : ACM; 2007 Presented at: Joint 9th WebKDD and 1st SNA-KDD Workshop '07; August 12, 2007; San Jose, California , USA p. 56-65.
  2. Savage N. Twitter as medium and message. Commun. ACM 2011 Mar 01;54(3):18-20. [CrossRef]
  3. Karger DR, Quan D. What would it mean to blog on the semantic web? In: McIlraith SA, Plexousakis D, van Harmelen F, editors. The Semantic Web – ISWC 2004, LNCS 3298. Berlin: Springer-Verlag; 2004:214-228.
  4. Ross C, Terras M, Warwick C, Welsh A. Enabled backchannel: conference Twitter use by digital humanists. Journal of Documentation 2011;67(2):214-237. [CrossRef]
  5. Collier N, Son NT, Nguyen NM. OMG U got flu? Analysis of shared health messages for bio-surveillance. J Biomed Semantics 2011;2 Suppl 5:S9 [FREE Full text] [CrossRef] [Medline]
  6. Dodds PS, Harris KD, Kloumann IM, Bliss CA, Danforth CM. Temporal patterns of happiness and information in a global social network: hedonometrics and Twitter. PLoS One 2011;6(12):e26752 [FREE Full text] [CrossRef] [Medline]
  7. Marwick AE, boyd D. I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society 2010 Jul 07;13(1):114-133. [CrossRef]
  8. National Center for Biotechnology Information. PubMed. 2012.   URL: [accessed 2012-06-06] [WebCite Cache]
  9. Lu Z. PubMed and beyond: a survey of web tools for searching biomedical literature. Database (Oxford) 2011;2011:baq036 [FREE Full text] [CrossRef] [Medline]
  10. Fink A. Conducting Research Literature Reviews: From Internet to paper. In: Conducting Research Literature Reviews: From the Internet to Paper. Thousand Oaks, California: Sage Publications, Inc; 2010.
  11. Lipscomb CE. Medical Subject Headings (MeSH). Bull Med Libr Assoc 2000 Jul;88(3):265-266 [FREE Full text] [Medline]
  12. Ellis D. A Behavioural Approach to Information Retrieval System Design. Journal of Documentation 1989;45(3):171-212.
  13. Talja S, Maula H. Reasons for the use and non-use of electronic journals and databases: A domain analytic study in four scholarly disciplines. Journal of Documentation 2003;59(6):673-691.
  14. Falagas ME, Pitsouni EI, Malietzis GA, Pappas G. Comparison of PubMed, Scopus, Web of Science, and Google Scholar: strengths and weaknesses. FASEB J 2008 Feb;22(2):338-342 [FREE Full text] [CrossRef] [Medline]
  15. Gao F, Luo T, Zhang K. Tweeting for learning: A critical analysis of research on microblogging in education published in 2008-2011. Br J Educ Technol 2012 Aug 24;43(5):783-801. [CrossRef]
  16. Barbera E, Gros B, Kirschner P. Temporal issues in e-learning research: A literature review. British Journal of Educational Technology 2012;43(2):E53-E55. [CrossRef]
  17. Cheong M, Ray S. A Literature Review of Recent Microblogging Developments. In. Victoria, Australia: Clayton School of Information Technology, Monash University; 2011.
  18. Cormode G, Krishnamurthy B, Willinger W. A manifesto for modeling and measurement in social media. [online social networks; modeling; measurement. First Monday 2010;15(9) [FREE Full text] [CrossRef]
  19. Google. Google Scholar. 2011.   URL: [accessed 2012-06-06] [WebCite Cache]
  20. BioMed. BioMed Central. 2012.   URL: [accessed 2012-06-06] [WebCite Cache]
  21. Gold J, Pedrana AE, Sacks-Davis R, Hellard ME, Chang S, Howard S, et al. A systematic examination of the use of online social networking sites for sexual health promotion. BMC Public Health 2011;11:583 [FREE Full text] [CrossRef] [Medline]
  22. Guse K, Levine D, Martins S, Lira A, Gaarde J, Westmorland W, et al. Interventions Using New Digital Media to Improve Adolescent Sexual Health: A Systematic Review. Journal of Adolescent Health 2012;51(6):535-543. [CrossRef]
  23. Williams SA, Terras M, Warwick C. What do people study when they study Twitter? Classifying Twitter related academic papers. Journal of Documentation 2013;69(3):384-410. [CrossRef]
  24. Bamman D, O'Connor B, Smith NA. Censorship and deletion practices in Chinese social media. FM 2012 Mar 02;17(3). [CrossRef]
  25. boyd D, Ellison NB. Social Network Sites: Definition, History, and Scholarship. Journal of Computer-Mediated Communication 2007;13(1):210-230. [CrossRef]
  26. Corbin J, Strauss AL. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Thousand Oaks, CA: Sage Publications, Inc; 2008.
  27. Glaser BG, Strauss AL. The discovery of grounded theory; strategies for qualitative research. Chicago: Aldine Pub. Co; 1967.
  28. Bonetta L. Should you be tweeting? Cell 2009 Oct 30;139(3):452-453. [CrossRef] [Medline]
  29. Bush H. Time to tweet? Hosp Health Netw 2009 Jun;83(6):46-51. [Medline]
  30. Bristol TJ. Twitter: consider the possibilities for continuing nursing education. J Contin Educ Nurs 2010 May;41(5):199-200. [CrossRef] [Medline]
  31. Chew C, Eysenbach G. Pandemics in the age of Twitter: content analysis of Tweets during the 2009 H1N1 outbreak. PLoS One 2010;5(11):e14118 [FREE Full text] [CrossRef] [Medline]
  32. Cuddy C, Graham J, Morton-Owens EG. Implementing Twitter in a health sciences library. Med Ref Serv Q 2010 Oct;29(4):320-330. [CrossRef] [Medline]
  33. Dörk M, Gruen D, Williamson C, Carpendale S. A visual backchannel for large-scale events. IEEE Trans Vis Comput Graph 2010;16(6):1129-1138. [CrossRef] [Medline]
  34. Qiu L, Leung AK, Ho JH, Yeung QM, Francis KJ, Chua PF. Understanding the psychological motives behind microblogging. Stud Health Technol Inform 2010;154:140-144. [Medline]
  35. Scanfeld D, Scanfeld V, Larson EL. Dissemination of health information through social networks: twitter and antibiotics. Am J Infect Control 2010 Apr;38(3):182-188 [FREE Full text] [CrossRef] [Medline]
  36. Schneider A, Jackson R, Baum N. Social media networking: Facebook and Twitter. J Med Pract Manage 2010;26(3):156-157. [Medline]
  37. Stieger S, Burger C. Let's go formative: continuous student ratings with Web 2.0 application Twitter. Cyberpsychol Behav Soc Netw 2010 Apr;13(2):163-167. [Medline]
  38. Adams A, Lomax G, Santarini A. Social media & stem cell science: examining the discourse. Regen Med 2011 Nov;6(6 Suppl):121-124. [CrossRef] [Medline]
  39. Bollen J, Gonçalves B, Ruan G, Mao H. Happiness is assortative in online social networks. Artif Life 2011;17(3):237-251. [CrossRef] [Medline]
  40. Eysenbach G. Can tweets predict citations? Metrics of social impact based on Twitter and correlation with traditional metrics of scientific impact. J Med Internet Res 2011;13(4):e123 [FREE Full text] [CrossRef] [Medline]
  41. Fox BI, Varadarajan R. Use of Twitter to encourage interaction in a multi-campus pharmacy management course. Am J Pharm Educ 2011 Jun 10;75(5):88 [FREE Full text] [CrossRef] [Medline]
  42. Franko OI. Twitter as a communication tool for orthopedic surgery. Orthopedics 2011 Nov;34(11):873-876. [CrossRef] [Medline]
  43. Golder SA, Macy MW. Diurnal and seasonal mood vary with work, sleep, and daylength across diverse cultures. Science 2011 Sep 30;333(6051):1878-1881 [FREE Full text] [CrossRef] [Medline]
  44. Gonçalves B, Perra N, Vespignani A. Modeling users' activity on twitter networks: validation of Dunbar's number. PLoS One 2011;6(8):e22656 [FREE Full text] [CrossRef] [Medline]
  45. González-Bailón S, Borge-Holthoefer J, Rivero A, Moreno Y. The dynamics of protest recruitment through an online network. Sci Rep 2011;1:197 [FREE Full text] [CrossRef] [Medline]
  46. Heaivilin N, Gerbert B, Page JE, Gibbs JL. Public health surveillance of dental pain via Twitter. J Dent Res 2011 Sep;90(9):1047-1051 [FREE Full text] [CrossRef] [Medline]
  47. Kukreja P, Heck Sheehan A, Riggins J. Use of social media by pharmacy preceptors. Am J Pharm Educ 2011 Nov 10;75(9):176 [FREE Full text] [CrossRef] [Medline]
  48. Mistry V. Critical care training: using Twitter as a teaching tool. Br J Nurs 2011;20(20):1292-1296. [Medline]
  49. Prochaska JJ, Pechmann C, Kim R, Leonhardt JM. Twitter=quitter? An analysis of Twitter quit smoking social networks. Tob Control 2012 Jul;21(4):447-449 [FREE Full text] [CrossRef] [Medline]
  50. Reips UD, Garaizar P. Mining twitter: a source for psychological wisdom of the crowds. Behav Res Methods 2011 Sep;43(3):635-642. [CrossRef] [Medline]
  51. Signorini A, Segre AM, Polgreen PM. The use of Twitter to track levels of disease activity and public concern in the U.S. during the influenza A H1N1 pandemic. PLoS One 2011;6(5):e19467 [FREE Full text] [CrossRef] [Medline]
  52. Su XY, Suominen H, Hanlen L. Machine intelligence for health information: capturing concepts and trends in social media via query expansion. Stud Health Technol Inform 2011;168:150-157. [Medline]
  53. McNeil K, Brna PM, Gordon KE. Epilepsy in the Twitter era: a need to re-tweet the way we think about seizures. Epilepsy Behav 2012 Feb;23(2):127-130. [CrossRef] [Medline]
  54. Sullivan SJ, Schneiders AG, Cheang CW, Kitto E, Lee H, Redhead J, et al. 'What's happening?' A content analysis of concussion-related traffic on Twitter. Br J Sports Med 2012 Mar;46(4):258-263. [CrossRef] [Medline]
  55. Takao H, Murayama Y, Ishibashi T, Karagiozov KL, Abe T. A new support system using a mobile device (smartphone) for diagnostic image display and treatment of stroke. Stroke 2012 Jan;43(1):236-239 [FREE Full text] [CrossRef] [Medline]
  56. Conde C. Medbloggers beware: watch what you say on the web. Tex Med 2009 May;105(5):29-32. [Medline]
  57. Eysenbach G. Infodemiology and infoveillance: framework for an emerging set of public health informatics methods to analyze search, communication and publication behavior on the Internet. J Med Internet Res 2009;11(1):e11 [FREE Full text] [CrossRef] [Medline]
  58. Gamble KH. Just a tweet away. Healthc Inform 2009 May;26(5):30, 32, 34 passim. [Medline]
  59. Hawn C. Take two aspirin and tweet me in the morning: how Twitter, Facebook, and other social media are reshaping health care. Health Aff (Millwood) 2009;28(2):361-368 [FREE Full text] [CrossRef] [Medline]
  60. Lefebvre C. Integrating cell phones and mobile technologies into public health practice: a social marketing perspective. Health Promot Pract 2009 Oct;10(4):490-494. [CrossRef] [Medline]
  61. Vance K, Howe W, Dellavalle RP. Social internet sites as a source of public health information. Dermatol Clin 2009 Apr;27(2):133-136, vi. [CrossRef] [Medline]
  62. Cain J, Fink JL. Legal and ethical issues regarding social media and pharmacy education. Am J Pharm Educ 2010 Dec 15;74(10):184 [FREE Full text] [Medline]
  63. Cain J, Romanelli F, Fox B. Pharmacy, social media, and health: opportunity for impact. J Am Pharm Assoc 2010;50(6):745-751. [CrossRef] [Medline]
  64. Clauson KA, Ekins J, Goncz CE. Use of blogs by pharmacists. Am J Health Syst Pharm 2010 Dec 1;67(23):2043-2048. [CrossRef] [Medline]
  65. Cooper PS, Lipshultz D, Matten WT, McGinnis SD, Pechous S, Romiti ML, et al. Education resources of the National Center for Biotechnology Information. Brief Bioinform 2010 Nov;11(6):563-569 [FREE Full text] [CrossRef] [Medline]
  66. Daniel D. Engaging the masses. Aust Fam Physician 2010 Sep;39(9):615 [FREE Full text] [Medline]
  67. Donahue A. Google wave: have CTSA-minded institutions caught it? Evid Based Libr Inf Pract 2010 Jan 1;5(4):70-82 [FREE Full text] [Medline]
  68. Duffin C. Community service. Nurs Stand 2010;24(28):20-21. [Medline]
  69. Garven JJ. Social media: the word of mouth revolution. Northwest Dent 2010;89(6):33-35, 68. [Medline]
  70. Greysen SR, Kind T, Chretien KC. Online professionalism and the mirror of social media. J Gen Intern Med 2010 Nov;25(11):1227-1229 [FREE Full text] [CrossRef] [Medline]
  71. Hawker MD. Social networking in the National Health Service in England: a quantitative analysis of the online identities of 152 primary care trusts. Stud Health Technol Inform 2010;160(Pt 1):356-360. [Medline]
  72. Kaldy J. The social pharmacist: tweeting and posting the way to success. Consult Pharm 2010 Jan;25(1):26-30, 32. [CrossRef] [Medline]
  73. Kind T, Genrich G, Sodhi A, Chretien KC. Social media policies at US medical schools. Med Educ Online 2010;15 [FREE Full text] [CrossRef] [Medline]
  74. Landman MP, Shelton J, Kauffmann RM, Dattilo JB. Guidelines for maintaining a professional compass in the era of social networking. J Surg Educ 2010;67(6):381-386. [CrossRef] [Medline]
  75. Miller RJ. Internet marketing 101. Facial Plast Surg Clin North Am 2010 Nov;18(4):509-516. [CrossRef] [Medline]
  76. Pujazon-Zazik M, Park MJ. To tweet, or not to tweet: gender differences and potential positive and negative health outcomes of adolescents' social internet use. Am J Mens Health 2010 Mar;4(1):77-85. [CrossRef] [Medline]
  77. Rozental TD, George TM, Chacko AT. Social networking among upper extremity patients. J Hand Surg Am 2010 May;35(5):819-823 e1. [CrossRef] [Medline]
  78. van Manen M. The pedagogy of Momus technologies: Facebook, privacy, and online intimacy. Qual Health Res 2010 Aug;20(8):1023-1032. [CrossRef] [Medline]
  79. Angelle D, Rose CL. Conversations with the community: the Methodist Hospital System's experience with social media. Front Health Serv Manage 2011;28(2):15-21. [Medline]
  80. Baptist AP, Thompson M, Grossman KS, Mohammed L, Sy A, Sanders GM. Social media, text messaging, and email-preferences of asthma patients between 12 and 40 years old. J Asthma 2011 Oct;48(8):824-830. [CrossRef] [Medline]
  81. Basak P. Development of an online tool for public health: the European Public Health Law Network. Public Health 2011 Sep;125(9):600-603. [CrossRef] [Medline]
  82. Chatterjee P, Biswas T. Blogs and Twitter in medical publications: too unreliable to quote, or a change waiting to happen? S Afr Med J 2011 Oct;101(10):712-714. [Medline]
  83. Dreesman J, Denecke K. Challenges for signal generation from medical social media data. Stud Health Technol Inform 2011;169:639-643. [Medline]
  84. Dubose C. The social media revolution. Radiol Technol 2011;83(2):112-119. [Medline]
  85. Evans P, Krauthammer M. Exploring the use of social media to measure journal article impact. AMIA Annu Symp Proc 2011;2011:374-381 [FREE Full text] [Medline]
  86. Folkestad L, Brodersen JB, Hallas P, Brabrand M. Laypersons can seek help from their Facebook friends regarding medical diagnosis. Ugeskr Laeger 2011 Dec 5;173(49):3174-3177. [Medline]
  87. Fortinsky KJ, Fournier MR, Benchimol EI. Internet and electronic resources for inflammatory bowel disease: a primer for providers and patients. Inflamm Bowel Dis 2012 Jun;18(6):1156-1163. [CrossRef] [Medline]
  88. Galloro V. Status update. Hospitals are finding ways to use the social media revolution to raise money, engage patients and connect with their communities. Mod Healthc 2011 Mar 14;41(11):6-7, 16, 1. [Medline]
  89. George DR, Dellasega C. Use of social media in graduate-level medical humanities education: two pilot studies from Penn State College of Medicine. Med Teach 2011;33(8):e429-e434. [CrossRef] [Medline]
  90. Giordano C, Giordano C. Health professions students' use of social media. J Allied Health 2011;40(2):78-81. [Medline]
  91. Juárez Giménez JC, Puyal González C, Valdivia Vadell C, Palacio Lacambra ME, Vidal Otero J, Cerqueira Dapena MJ. Application of the Technology Web 2.0 in a drug information centre. Farm Hosp 2011;35(6):315.e1-315315. [CrossRef] [Medline]
  92. Kamel Boulos MN, Resch B, Crowley DN, Breslin JG, Sohn G, Burtner R, et al. Crowdsourcing, citizen sensing and sensor web technologies for public and environmental health surveillance and crisis management: trends, OGC standards and application examples. Int J Health Geogr 2011;10:67 [FREE Full text] [CrossRef] [Medline]
  93. Keim ME, Noji E. Emergent use of social media: a new age of opportunity for disaster resilience. Am J Disaster Med 2011;6(1):47-54. [Medline]
  94. Kishimoto K, Fukushima N. Use of anonymous Web communities and websites by medical consumers in Japan to research drug information. Yakugaku Zasshi 2011;131(5):685-695 [FREE Full text] [Medline]
  95. Lee JY, Kang DH, Moon HS, Kim YT, Yoo TK, Choi HY, et al. Analysis of content legibility for smartphones of websites of the korean urological association and other urological societies in Korea. Korean J Urol 2011 Feb;52(2):142-146 [FREE Full text] [CrossRef] [Medline]
  96. Liang BA, Mackey TK. Prevalence and Global Health implications of social media in direct-to-consumer drug advertising. J Med Internet Res 2011;13(3):e64 [FREE Full text] [CrossRef] [Medline]
  97. Locke PA. Communication of radiation benefits and risks in decision making: some lessons learned. Health Phys 2011 Nov;101(5):626-629. [CrossRef] [Medline]
  98. Lyon A, Nunn M, Grossel G, Burgman M. Comparison of web-based biosecurity intelligence systems: BioCaster, EpiSPIDER and HealthMap. Transbound Emerg Dis 2012 Jun;59(3):223-232. [CrossRef] [Medline]
  99. Melamud A, Otero P. [Facebook and Twitter, are they already in the pediatrician's office? Survey on the use of social networks]. Arch Argent Pediatr 2011 Oct;109(5):437-444 [FREE Full text] [CrossRef] [Medline]
  100. Nierenberg K, Hollenbeck J, Fleming LE, Stephan W, Reich A, Backer LC, et al. Frontiers in Outreach and Education: The Florida Red Tide Experience. Harmful Algae 2011 May 1;10(4):374-380 [FREE Full text] [CrossRef] [Medline]
  101. O'Keeffe GS, Clarke-Pearson K, Council on Communications and Media. The impact of social media on children, adolescents, and families. Pediatrics 2011 Apr;127(4):800-804 [FREE Full text] [CrossRef] [Medline]
  102. Pletneva N, Cruchet S, Simonet MA, Kajiwara M, Boyer C. Results of the 10 HON survey on health and medical internet use. Stud Health Technol Inform 2011;169:73-77. [Medline]
  103. Sajadi KP, Goldman HB. Social networks lack useful content for incontinence. Urology 2011 Oct;78(4):764-767. [CrossRef] [Medline]
  104. Tobias E. Using Twitter and other social media platforms to provide situational awareness during an incident. J Bus Contin Emer Plan 2011 Oct;5(3):208-223. [Medline]
  105. Turner-McGrievy G, Tate D. Tweets, Apps, and Pods: Results of the 6-month Mobile Pounds Off Digitally (Mobile POD) randomized weight-loss intervention among adults. J Med Internet Res 2011;13(4):e120 [FREE Full text] [CrossRef] [Medline]
  106. Wolpert Barraza E. [Physician retirement from clinical practice: how and when]. Gac Med Mex 2011;147(3):262-265. [Medline]
  107. Wong WW, Gupta SC. Plastic surgery marketing in a generation of "tweeting". Aesthet Surg J 2011 Nov;31(8):972-976. [CrossRef] [Medline]
  108. Zhou T, Medo M, Cimini G, Zhang ZK, Zhang YC. Emergence of scale-free leadership structure in social recommender systems. PLoS One 2011;6(7):e20648 [FREE Full text] [CrossRef] [Medline]
  109. Omurtag K, Jimenez PT, Ratts V, Odem R, Cooper AR. The ART of social networking: how SART member clinics are connecting with patients online. Fertil Steril 2012 Jan;97(1):88-94. [CrossRef] [Medline]
  110. Kamel Boulos MN, Sanfilippo AP, Corley CD, Wheeler S. Social Web mining and exploitation for serious applications: Technosocial Predictive Analytics and related technologies for public health, environmental and national security surveillance. Comput Methods Programs Biomed 2010 Oct;100(1):16-23. [CrossRef] [Medline]
  111. Robinson M, Robertson S. Young men's health promotion and new information communication technologies: illuminating the issues and research agendas. Health Promot Int 2010 Sep;25(3):363-370 [FREE Full text] [CrossRef] [Medline]
  112. Skiba DJ. Nursing education 2.0: Twitter & tweets. Can you post a nugget of knowledge in 140 characters or less? Nurs Educ Perspect 2008;29(2):110-112. [Medline]
  113. Chhanabhai P. A tweet a day keeps the doctor away. N Z Med J 2009 Apr 3;122(1292):97-98. [Medline]
  114. Coffsky J. Twitter, the lost years and the geeks. J Med Assoc Ga 2009;98(4):28. [Medline]
  115. Hamilton JA. It's Twitter time. Hosp Health Netw 2009 Aug;83(8):6. [Medline]
  116. Parslow GR. Commentary: Twitter for educational networking. Biochem Mol Biol Educ 2009 Jul;37(4):255-256. [CrossRef] [Medline]
  117. Reisman M. Will pharma twitter? P T 2009 Aug;34(8):421 [FREE Full text] [Medline]
  118. Thames G. Twitter as an educational tool. J Child Adolesc Psychiatr Nurs 2009 Nov;22(4):235. [CrossRef] [Medline]
  119. van den Broek WW. The journal on Twitter. Tijdschr Psychiatr 2009;51(10):711-714 [FREE Full text] [Medline]
  120. Weissmann G. Walter Benjamin and Biz Stone: the scientific paper in the age of Twitter. FASEB J 2009 Jul;23(7):2015-2018 [FREE Full text] [CrossRef] [Medline]
  121. Williams N. High-level twitter. Curr Biol 2009 May 12;19(9):R350-R351. [Medline]
  122. anonymous. All a-Twitter about chemistry. Nat Chem 2010 Jul;2(7):511. [CrossRef] [Medline]
  123. Atkinson L. Do you twitter and tweet? Iowa Med 2010;100(6):24-25. [Medline]
  124. Martens E. Twitter for scientists. ACS Chem Biol 2010 Feb 19;5(2):149. [CrossRef] [Medline]
  125. Orsini M. Meet the Web 2010: interactive and informative. Caring 2010 May;29(5):44-45. [Medline]
  126. Parfitt T. Plant science. Pavlovsk's hopes hang on a tweet. Science 2010 Aug 20;329(5994):899. [CrossRef] [Medline]
  127. Pattillo R. Are you using Twitter for your next survey? Nurse Educ 2010;35(5):207. [CrossRef] [Medline]
  128. Williams B. Twitter, facebook and youtube: the TMA turns to social media to engage members. Tenn Med 2010 Feb;103(2):27-28. [Medline]
  129. Ali FR, Finlayson AE. Tweet to collaborate with poorer nations. Nature 2011 Jul 28;475(7357):455. [CrossRef] [Medline]
  130. anonymous. Tweet me. Nat Methods 2011 Apr;8(4):273. [Medline]
  131. anonymous. Do you use Facebook or Twitter? MGMA Connex 2011 Oct;11(9):18. [Medline]
  132. Bottles K. Twitter: an essential tool for every physician leader. Physician Exec 2011;37(3):80-82. [Medline]
  133. Chretien KC, Azar J, Kind T. Physicians on Twitter. JAMA 2011 Feb 9;305(6):566-568. [CrossRef] [Medline]
  134. Curioso WH, Alvarado-Vásquez E, Calderón-Anyosa R. Using Twitter to promote continuous education and health research in Peru. Rev Peru Med Exp Salud Publica 2011 Mar;28(1):163-164 [FREE Full text] [Medline]
  135. Duffy M. iNurse: Facebook, Twitter, and LinkedIn, oh my!. Am J Nurs 2011 Apr;111(4):56-59. [CrossRef] [Medline]
  136. Kubben PL. Twitter for neurosurgeons. Surg Neurol Int 2011;2:28 [FREE Full text] [CrossRef] [Medline]
  137. Leow JJ, Groen RS, Sadasivam V, Kushner AL. Twitter and mobile technology as diagnostic aids in the Democratic Republic of Congo. Am Surg 2011 Nov;77(11):E242-E243. [Medline]
  138. Mandavilli A. Peer review: trial by Twitter. Nature 2011 Jan 20;469(7330):286-287. [CrossRef] [Medline]
  139. McKee M, Cole K, Hurst L, Aldridge RW, Horton R. The other Twitter revolution: how social media are helping to monitor the NHS reforms. BMJ 2011;342:d948. [Medline]
  140. Miller G. Sociology. Social scientists wade into the tweet stream. Science 2011 Sep 30;333(6051):1814-1815. [CrossRef] [Medline]
  141. Morris K. Tweet, post, share: a new school of health communication. Lancet Infect Dis 2011 Jul;11(7):500-501. [Medline]
  142. Peregrin T. Time to tweet: social networking for surgeons. Bull Am Coll Surg 2011 Feb;96(2):46-48. [Medline]
  143. Rajani R, Berman DS, Rozanski A. Social networks: are they good for your health? The era of Facebook and Twitter. QJM 2011 Sep;104(9):819-820 [FREE Full text] [CrossRef] [Medline]
  144. Reich ES. Researchers tweet technical talk. Nature 2011 Jun 23;474(7352):431. [CrossRef] [Medline]
  145. Spence D. All a-Twitter. BMJ 2011;343:d8122. [Medline]
  146. Trueman MS, Miles DG. Twitter in the classroom: twenty-first century flash cards. Nurse Educ 2011;36(5):183-186. [CrossRef] [Medline]
  147. Weinberger S. Spies to use Twitter as crystal ball. Nature 2011 Oct 20;478(7369):301. [CrossRef] [Medline]
  148. Wilson A, Robertson J, McElduff P, Jones A, Henry D. Does it matter who writes medical news stories? PLoS Med 2010 Sep;7(9) [FREE Full text] [CrossRef] [Medline]
  149. Wilson RE, Gosling SD, Graham LT. A Review of Facebook Research in the Social Sciences. Perspectives on Psychological Science 2012 May 16;7(3):203-220. [CrossRef]
  150. Atencio CA, Blake DT, Strata F, Cheung SW, Merzenich MM, Schreiner CE. Frequency-modulation encoding in the primary auditory cortex of the awake owl monkey. J Neurophysiol 2007 Oct;98(4):2182-2195 [FREE Full text] [CrossRef] [Medline]
  151. Arai YC, Sakakibara S, Ito A, Ohshima K, Sakakibara T, Nishi T, et al. Intra-operative natural sound decreases salivary amylase activity of patients undergoing inguinal hernia repair under epidural anesthesia. Acta Anaesthesiol Scand 2008 Aug;52(7):987-990. [CrossRef] [Medline]
  152. Kajikawa Y, de la Mothe LA, Blumell S, Sterbing-D'Angelo SJ, D'Angelo W, Camalier CR, et al. Coding of FM sweep trains and twitter calls in area CM of marmoset auditory cortex. Hear Res 2008 May;239(1-2):107-125 [FREE Full text] [CrossRef] [Medline]
  153. Rochefort C. The FoxP2 gene makes humans speak and birds twitter. Med Sci (Paris) 2008 Nov;24(11):906-907 [FREE Full text] [CrossRef] [Medline]
  154. Walker KM, Ahmed B, Schnupp JW. Linking cortical spike pattern codes to auditory perception. J Cogn Neurosci 2008 Jan;20(1):135-152. [CrossRef] [Medline]
  155. Chen HC, Kaplan G, Rogers LJ. Contact calls of common marmosets (Callithrix jacchus): influence of age of caller on antiphonal calling and other vocal responses. Am J Primatol 2009 Feb;71(2):165-170. [CrossRef] [Medline]
  156. Siemers BM, Schauermann G, Turni H, von Merten S. Why do shrews twitter? Communication or simple echo-based orientation. Biol Lett 2009 Oct 23;5(5):593-596 [FREE Full text] [CrossRef] [Medline]
  157. Bolhuis JJ, Okanoya K, Scharff C. Twitter evolution: converging mechanisms in birdsong and human speech. Nat Rev Neurosci 2010 Nov;11(11):747-759. [CrossRef] [Medline]
  158. Jin T. Near-infrared fluorescence detection of acetylcholine in aqueous solution using a complex of rhodamine 800 and p-sulfonatocalix[8]arene. Sensors (Basel) 2010;10(3):2438-2449 [FREE Full text] [CrossRef] [Medline]
  159. Rychel AL, Peterson KM, Torii KU. Plant twitter: ligands under 140 amino acids enforcing stomatal patterning. J Plant Res 2010 May;123(3):275-280. [CrossRef] [Medline]
  160. Watson CF, Caldwell CA. Neighbor effects in marmosets: social contagion of agonism and affiliation in captive Callithrix jacchus. Am J Primatol 2010 Jun;72(6):549-558. [CrossRef] [Medline]
  161. Elsevier. SciVerse Scopus. 2012.   URL: [accessed 2012-06-06] [WebCite Cache]
  162. MIMAS. Web of Knowledge. 2012.   URL: [accessed 2012-06-06] [WebCite Cache]
  163. Twitter. API Terms of Service: Archive. 2012.   URL: [accessed 2013-01-27] [WebCite Cache]
  164. Bottles K. 2011. How Twitter changed the life of this physician executive consultant   URL: [accessed 2012-07-11] [WebCite Cache]
  165. Neylon C. Science in the open: the online home of Cameron Neylon. 2012 Jun 11. Tracking research into practice: Are nurses on twitter a good case study?   URL: http:/​/cameronneylon.​net/​blog/​tracking-research-into-practice-are-nurses-on-twitter-a-good-case-study/​ [accessed 2012-07-11] [WebCite Cache]

NCBI: National Center for Biotechnology Information
MeSH: Medical Subject Headings

Edited by G Eysenbach; submitted 16.07.12; peer-reviewed by B McGowan, A Pedrana; comments to author 09.10.12; revised version received 27.01.13; accepted 12.05.13; published 18.07.13


©Shirley Ann Williams, Melissa Terras, Claire Warwick. Originally published in Medicine 2.0 (, 18.07.2013.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in Medicine 2.0, is properly cited. The complete bibliographic information, a link to the original publication on, as well as this copyright and license information must be included.