DOI: 10.1177/21925682261493432 ISSN: 2192-5682

Association of Arm Pain, Neck Pain, and Numbness Improvement with Patient Satisfaction During Pharmacological Treatment for Cervical Spine Disorders: A Prospective Multicenter Cohort Study

Hiroaki Nakashima, Norimitsu Wakao, Sadayuki Ito, Takeo Furuya, Toshitaka Yoshii, Hideyuki Arima, Yu Yamato, Shiro Imagama, Yasuaki Imajo, Hiroshi Miyamoto, Gen Inoue, Masayuki Miyagi, Shunsuke Kanbara, Yoichi Iizuka, Hirotaka Chikuda, Kei Watanabe, Kazuyoshi Kobayashi, Mikito Tsushima, Masashi Miyazaki, Mitsuru Yagi, Satoshi Suzuki, Masahiko Takahata, Michio Hongo, Masao Koda, Keiji Nagata, Kanji Mori, Akinobu Suzuki, Takashi Kaito, Naohisa Miyakoshi, Hiroshi Hashizume, Yukihiro Matsuyama, Mamoru Kawakami, Hirotaka Haro, Masaya Nakamura, Masahiko Watanabe, ,
Show PDF Cite

Study Design

Prospective multicenter cohort study.

Objectives

To examine associations between improvements in arm pain, neck pain, numbness, and patient satisfaction during pharmacological treatment of cervical spine disorders.

Methods

Of 256 patients with cervical spine disorders and upper extremity symptoms enrolled at 28 institutions, 220 patients (830 observations) were included. A generalized estimating equation with cluster-robust variance regressed satisfaction (5-point Likert; satisfied ≥4) on improvement in arm pain, neck pain, and numbness (NRS 0–10), adjusting for psychosocial burden (BS-POP), age, sex, and time point. An ordinal model was used as an alternative outcome specification, and sensitivity analyses were interpreted descriptively.

Results

Arm pain improvement was associated with satisfaction under both binary (OR 1.095 per point, P =.022; Holm-adjusted P =.044) and ordinal (OR 1.143, P =.019) coding. Neck pain improvement was associated with satisfaction under binary coding only (OR 1.121, P =.003; Holm-adjusted P =.009; ordinal P =.447). Numbness improvement showed the same direction but was not significant in either specification (OR 1.068, P =.106; ordinal P =.377). Confidence intervals overlapped across domains, and the arm pain estimate was attenuated under inverse-probability weighting and in diagnostic subsets. Approximately one-third of observations with persistent numbness had satisfaction scores ≥4.

Conclusions

Arm pain improvement was consistently associated with greater satisfaction, whereas the association with numbness improvement was less consistent. Residual numbness alone should not be considered treatment failure and should be interpreted alongside pain relief, neurological findings, and overall clinical condition. These findings support multidimensional assessment of treatment response rather than reliance on residual numbness alone in clinical practice.

More from our Archive