Text Analysis of “Work” and “Low Back Pain” on SNS: An Exploratory Study of Co-occurrence Structures Using Self-Organizing Maps and Network Analysis

Abstract

Objective: This study aimed to conduct an exploratory analysis of posts on X (formerly Twitter) concerning both “work” and “low back pain,” thereby clarifying trends and characteristics in order to examine potential applications for workplace environment improvements and low back pain countermeasures.

Methods: Using 2,850 tweets containing the keywords “work and low back pain” obtained from April to May 2023, we conducted analyses with text mining software.

Results: A self-organizing map identified eight clusters, revealing frequent co-occurrences of references to multiple symptoms such as abdominal pain or shoulder stiffness along with negative expressions related to workplace matters.

Conclusion: Posts on X concerning both “work” and “low back pain” suggest that physical burdens, including low back pain, may affect continued employment. Although these findings may inform occupational health policies and early interventions, limitations such as the short data collection period and lack of user attribute information indicate a need for more extensive and long-term investigations.