Scholarlink

Background <p>Forward head posture is commonly quantified using the cervicovertebral angle (CVA), a simple static postural measure used in clinical and occupational settings. Although reduced CVA has been associated with nonspecific neck pain (NSNP), its discriminative utility as a single postural metric remains uncertain because NSNP is multifactorial. This study aimed to examine the association between CVA and NSNP in female manufacturing workers and to identify exploratory CVA cutoff values using logistic regression (LR) and decision tree (DT) models.</p> Methods <p>This cross-sectional study included 135 female manufacturing workers, classified into an NSNP group (<i>n</i> = 64) and a non-pain group (<i>n</i> = 71). CVA was measured using standardized sagittal-view digital photography and ImageJ software. Unadjusted and adjusted binary LR analyses were performed to examine the association between CVA and NSNP. Receiver operating characteristic curve analysis was used to determine the optimal LR-based cutoff value using the Youden index. A DT algorithm was applied to identify hierarchical cutoff values. Model performance was evaluated using accuracy, sensitivity, specificity, and area under the curve (AUC).</p> Results <p>The NSNP group showed a significantly smaller CVA than the non-pain group (51.16° ± 5.47° vs. 53.21° ± 4.11°, <i>p</i> = 0.015). In the unadjusted LR model, CVA was significantly associated with NSNP (odds ratio [OR] = 0.914, 95% confidence interval [CI]: 0.849–0.984, <i>p</i> = 0.017). This association remained significant after adjustment for age, height, weight, and body mass index (OR = 0.911, 95% CI: 0.830–0.999, <i>p</i> = 0.047). The LR model identified an optimal cutoff value of 51.40°, with an AUC of 0.623. The DT model identified a primary cutoff of 51.54° and a secondary threshold of 42.27°, with an AUC of 0.645.</p> Conclusions <p>CVA was significantly associated with NSNP in female manufacturing workers, and both models identified a consistent primary cutoff of approximately 51.5°. However, CVA alone showed limited classification ability. Therefore, the identified cutoff values should be interpreted as supplementary screening references.</p> Trial registration <p>This study was registered by the Korea Clinical Research Information Service (KCT0011501).</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The cervicovertebral angle as an associated factor of neck pain in female manufacturing workers: a dual-model cutoff analysis using logistic regression and decision trees

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月23日

查看全文
  • Seung-yoon Han,
  • Ui-jae Hwang,
  • Yoon-ji Lee,
  • Il-kyu Ahn,
  • Yul-ji Won,
  • Ji-hoo Moon,
  • Oh-yun Kwon

摘要

Background

Forward head posture is commonly quantified using the cervicovertebral angle (CVA), a simple static postural measure used in clinical and occupational settings. Although reduced CVA has been associated with nonspecific neck pain (NSNP), its discriminative utility as a single postural metric remains uncertain because NSNP is multifactorial. This study aimed to examine the association between CVA and NSNP in female manufacturing workers and to identify exploratory CVA cutoff values using logistic regression (LR) and decision tree (DT) models.

Methods

This cross-sectional study included 135 female manufacturing workers, classified into an NSNP group (n = 64) and a non-pain group (n = 71). CVA was measured using standardized sagittal-view digital photography and ImageJ software. Unadjusted and adjusted binary LR analyses were performed to examine the association between CVA and NSNP. Receiver operating characteristic curve analysis was used to determine the optimal LR-based cutoff value using the Youden index. A DT algorithm was applied to identify hierarchical cutoff values. Model performance was evaluated using accuracy, sensitivity, specificity, and area under the curve (AUC).

Results

The NSNP group showed a significantly smaller CVA than the non-pain group (51.16° ± 5.47° vs. 53.21° ± 4.11°, p = 0.015). In the unadjusted LR model, CVA was significantly associated with NSNP (odds ratio [OR] = 0.914, 95% confidence interval [CI]: 0.849–0.984, p = 0.017). This association remained significant after adjustment for age, height, weight, and body mass index (OR = 0.911, 95% CI: 0.830–0.999, p = 0.047). The LR model identified an optimal cutoff value of 51.40°, with an AUC of 0.623. The DT model identified a primary cutoff of 51.54° and a secondary threshold of 42.27°, with an AUC of 0.645.

Conclusions

CVA was significantly associated with NSNP in female manufacturing workers, and both models identified a consistent primary cutoff of approximately 51.5°. However, CVA alone showed limited classification ability. Therefore, the identified cutoff values should be interpreted as supplementary screening references.

Trial registration

This study was registered by the Korea Clinical Research Information Service (KCT0011501).