SICE Journal of Control, Measurement, and System Integration (Dec 2023)

Factors in development of low back pain during anteflexion in children identified by a multiple logistic regression model

  • Yushin Yoshizato,
  • Kiyohisa Natsume

DOI
https://doi.org/10.1080/18824889.2023.2223404
Journal volume & issue
Vol. 16, no. 1
pp. 227 – 236

Abstract

Read online

The number of children with low back pain (LBP) has increased. However, the underlying cause of LBP remains unclear. LBP is classified on the basis of whether anteflexion or postflexion induces pain. We focused on paediatric LBP induced by anteflexion (LBPAF) and studied its causes using a multiple logistic regression (MLR) model. Children aged 4–15 years participated in the experiments. About 6% of children had LBPAF and all were older than 10 years of age. Participants answered a questionnaire on sports and daily life and underwent physical assessments. Thirty-six MLR models were constructed using this data. Among them, the best model with the largest mean area under the curve (AUC) for the four test datasets had the explanatory variables of sport time, finger floor distance score, and Ely test score. These variables were present in 21–33 models in all thirty-six models. The MLR model was also able to detect approximately 70% of non-LBPAFs. These three selected variables suggest that an imbalance of the anterior and posterior thigh muscles may have caused paediatric LBPAF. This imbalance has not been reported in adults with LBPAF; hence, it is a characteristic of paediatric LBPAF.

Keywords