X, Formerly Twitter, Offers Valuable Insights Into Self-Reported Chronic Pain Using Machine Learning: Study

Social media platforms, including X, formerly known as Twitter, have emerged as valuable resources for understanding self-reported chronic pain, as outlined in a recent research effort conducted by multiple institutions.

The study showcases the automation of processes to establish a cohort focused on chronic pain, opening avenues for forthcoming data mining and causal association investigations, MedicalXpress reported.

Abeed Sarker, an associate professor at Emory University, articulated the study's goals, saying: "We aimed to detect large-scale discussions related to chronic pain on Twitter, develop methods for automatic detection of self-disclosures, and collect and analyze longitudinal data."

What Is Chronic Pain?

Chronic pain, often linked to opioid use, presents substantial challenges to public health and the economy. Aggregating public insights and experiences holds potential for guiding alternative therapies for specific types of pain.

The researchers gathered data through Twitter's academic API, followed by manual annotation and classification. For example, posts addressed therapies like meditation and chiropractic care, with sentiment analysis unveiling diverse public opinions on these treatments.

Sarker explained that social media can be a rich source of chronic pain-related information. Hence, Sarker noted that methods, such as NLP and machine learning, can extract key insights from such a large dataset.

The study's automation capabilities encompass identifying pertinent posts and discerning instances of self-reporting, facilitating the inclusion of individuals who may not typically be accessible through conventional research settings.

Cohorts established through social media channels may provide a distinctive perspective on chronic pain. This encompasses factors such as the social support subscribers receive, its influence on their quality of life, and the correlation between alternative therapies and the social dimensions of daily life.

"Our social media based approach will result in an automatically growing large cohort over time, and the data can be leveraged to identify effective opioid-alternative therapies for diverse chronic pain types," the study's abstract reads.

NLP and ML Framework

This publication stands as the first to construct an intricate NLP and machine-learning framework for accumulating chronic pain knowledge from patient-generated social media content.

The automated assembly of cohort members through classification, alongside the longitudinal analysis of their posts, is poised to generate a range of hypotheses regarding chronic pain management, offering an unprecedented resource for extensive studies in this domain.

Sarker conveyed the next steps, stating: "The next step will be to do a more comprehensive analysis of the data posted by our chronic pain cohort, particularly to generate hypotheses about effective opioid alternatives for specific chronic pain types."

The study concluded by highlighting the potential of the social media-based approach in creating a continuously expanding cohort, with the prospect of identifying effective opioid-alternative therapies for diverse chronic pain types.

The research team's findings were published in the journal Health Data Science.