Globally, around 3.2-billion people are connected to the Internet. Online technologies are now ever-present in daily life, and they are influencing healthcare in new and exciting ways. The World Health Organisation has defined this phenomenon as “eHealth” – the practice of medicine and public health supported by electronic processes and communication. In an era where online technologies are so ubiquitous, the challenge for healthcare providers and researchers is to understand the utility, value, and limitations of these new technologies.
We decided to take up this challenge in a recent study  where we analysed almost three-quarters of a million Tweets about back pain. We were interested in finding out what people were tweeting just before they tweeted about a new episode of back pain. The rationale behind this was that we would identify some risk factors that might trigger a new onset of low back pain – risk factors that could eventually be targeted by online public health campaigns.
So, we identified 742 028 individuals across the world who tweeted about a new episode of back pain and searched their preceding tweets for ‘Risk Tweets’. A Risk Tweet was any tweet about a physical, psychological, or general health factor prior to their tweet about back pain. For example, a Physical Risk Tweet could be: “Just completed my first ½ marathon!”. Using a case-crossover study design , we were able to determine whether these Risk Tweets were associated with a subsequent tweet about low back pain (Figure 1). The advantage of this study design is that we have a very useful ‘self-matched’ control group. This allowed us to compare the frequencies of Risk Tweets in the same individual who was tweeting about their back pain to when they weren’t. By using this design, we controlled for a bunch of possible confounding variables that could have distorted our results.
Our analysis showed that tweeting about selected physical, psychological, and general health factors increased the odds of tweeting about back pain. Psychological factors had the highest odds, followed by physical, then general health factors. This means that people who tweeted about a psychological factor were most likely to tweet about their back pain within the following 48 hours.
Although it’s early days, we think that our findings provide useful information to guide future online interventions. In other areas of healthcare, epidemiologists have called for innovative approaches that use online social media to deliver preventive health interventions [1,2,6,7]. Although we expect some barriers and limitations to these online approaches to back pain, further work could be done to test whether informatics-based interventions can deliver targeted treatment via online media. Who knows… a tweet could save a back!
Tweeting back: predicting new cases of back pain with mass social media data is an open access paper and free to download.
About Hopin Lee
Hopin is part of the PREVENT team at Neuroscience Research Australia and is diving into the final stages of his PhD. Hopin’s research primarily focuses on understanding causal mechanisms that underlie the development of chronic pain, and the interventions that aim to treat it. On the side, Hopin is also interested in the use of social media data and the utility of mobile apps in musculoskeletal health.
Hopin catastrophizes about caffeine and chilli, loves football, tries to perfect the Spagetti alla Puttanesca, mindfully regulates his OCD for vinyl, and tries to attend as many gigs in Sydney while avoiding the heat. Oh and he tweets… sporadically… @hopinlee
 Berenbaum F. The social (media) side to rheumatology. Nat Rev Rheumatol 2014;10:314–8.
 Brownstein JS, Freifeld CC, Madoff LC. Digital disease detection – Harnessing the Web for public health surveillance. N Engl J Med 2009;360:2153–2155.
 Laranjo L, Arguel A, Neves AL, Gallagher AM, Kaplan R, Mortimer N, Mendes G a, Lau AYS. The influence of social networking sites on health behavior change: a systematic review and meta-analysis. J Am Med Inform Assoc 2014;0:1–10.
 Lee H, McAuley JH, Hübscher M, Allen HG, Kamper SJ, Moseley GL. Tweeting back: predicting new cases of back pain with mass social media data. J Am Med Informatics Assoc 2015:ocv168.
 Maclure M. The case-crossover design: a method for studying transient effects on the risk of acute events. Am J Epidemiol 1991;133:144–53.
 Milinovich GJ, Magalhães RJS, Hu W. Role of big data in the early detection of Ebola and other emerging infectious diseases. Lancet Glob Heal 2015;3:e20–1.
 Wehner MR, Chren MM, Shive ML, Resneck JS, Pagoto S, Seidenberg AB, Linos E. Twitter: an opportunity for public health campaigns. Lancet 2014;384:131–2.