Can Social Media Expand the Audience For Medical Research Articles

Take home point

The use of social media tools – Facebook, Twitter, LinkedIn, and ResearchBlogging.org – for the purpose of releasing an article in the clinical pain sciences increases the number of people who view the article and download it. Social media tools also have implications for the dissemination of medical research articles in other fields.

Problem

Generally, anyone interested in accessing research articles either utilizes a research database such as Pubmed or simply follows a few specific journals. As a result, many articles go unnoticed because end users who could benefit from them lack the time to find them. In addition, those who have time to search these databases may be overwhelmed with lots of articles that are not relevant to their needs.

Social media, on the other hand, does not depend on individuals accessing a database. Instead, end users have information sent to them often by individuals they have elected to follow based on shared interests or personal relationships. Researchers in Australia sought to determine if this unique aspect of social media could be used to increase the dissemination of research. Although this may seem obvious, little evidence exists to support this and researchers sought to contribute to an evidence base in this area.

Approach to Address Problem

The researchers selected 16 PLOS ONE articles using four inclusion criteria relevant to the clinical pain sciences, first published online between 2006 and 2011, of interest to readers of a research group blog (bodyinmind.org), and not previously mentioned in a blog post on bodyinmind.org. The articles were assigned randomly to four researchers who wrote blog posts on them, comprised of approximately 500 words and a link to the online version of the article. These blog posts were randomly assigned two dates: one date for a social media release and one date as a control. The control is not well explained by the researchers and appears to represent a period where nothing is being done to promote the article.

A better comparison would be the initial release of the article online because readers may be more prone to download articles in the latest issue instead of going to the archives to identify articles. In other words, the control may not be an adequate reflection of the amount of attention articles receive without social media releases.

On the social media release data, the researchers sent out their blog posts to researchblogging.org, Facebook, Twitter, and LinkedIn. Posts were released early in the morning (6 – 7 AM) or late at night (11 PM – 2 AM) Australian Eastern Summer Time. They do not clearly indicate the name of the Facebook, Twitter, and LinkedIn accounts that are used or how many people are linked to these accounts (such as number of Facebook friends). Also, given Twitter’s word limitations they are not clear about how they modified the 5oo word post for Twitter.

In addition, the researchers do not indicate whether the people receiving the articles are people who normally receive this type of information from them–perhaps through their bodyinmind.org blog. If so, these individuals may be more likely to uptake this information than others who do not regularly utilize social media tools for getting research information.

The researchers compared response to articles during the control and social media release dates. Their primary outcome variables were HTML views and PDF downloads a week before and after the social media release. They also looked at factors associated with these variables including reach (number of people receiving the blog posts), engagement (likes and other comments on blog posts), and virality (number of people who generated a secondary story from their blog posts–like this one on their research article). They assessed whether these factors led to more HTML views or PDF downloads. They also assessed whether the blog writer had some effect on its success.

Innovation

The key innovation in this research project was the use of social media to push research information to end users instead of waiting on the end users to “pull” the information from a database.

Key Results

The key result was a statistically significant increase in HTML views and PDF downloads one week after the blog posts when comparing the control date and the social media release date  (p < .05). However, none of the measures of social media reach, engagement, or virality related to the outcome variable. Hence, some other unknown factors are affecting HTML views and PDF downloads.

Implications for clinicians/health care system

Better dissemination of research can save clinicians time by improving the efficiency of information uptake by them. This is useful for health care systems because this is one of the many ways that a health system can improve its quality. One of the challenges is making sure that the information coming from social media is well catered to the needs of clinicians.

Implications for public health

Improved dissemination of research could also provide benefits to the public. In the case of patients or lay individuals, social media can be used to send not only research articles, but lay translations of the information in order to increase the probability of more people understanding the information. This can contribute to increasing health literacy of individuals.

Future research concerns/challenges

Moving forward, researchers in this field could seek to determine which types of health professionals are most likely to utilize social media for the purposes of receiving research information. In addition, they could determine which groups of patients are likely to utilize research information disseminated in this manner. Also, as stated, more clarity in the methods for these studies will help guide subsequent research and improve the ability of future researchers to replicate the findings of work in this area or refute the findings.

Article Link:

Author:

Perry Payne, MD/JD/MPP

View Comments

No comments yet.

Leave a Reply

Your email address will not be published. Required fields are marked *