Patient-friendly summary
If you read nothing else
Bottom line
There is moderate-certainty evidence that craniocervical flexion training (alone or with resistance training) can produce measurable neuromuscular changes in the neck muscles of people with chronic non-specific neck pain, with adaptations largely specific to the type of exercise performed.
Moderate evidencePublished
Evidence hierarchy
Study participants
Adults (>=18 years) with chronic non-specific neck pain
Study Summary
This systematic review and meta-analysis examined how exercise targeting the neck muscles (performed for at least two weeks) changes neuromuscular function in adults with chronic non-specific neck pain. Fourteen studies (published 1999–2023) were included after screening 2,110+ citations, covering 17 different exercise approaches assessed mainly by electromyography (EMG). The authors found moderate certainty of evidence that craniocervical flexion (CCF) training, alone or combined with resistance training, can induce neural adaptations in the neck muscles, and a meta-analysis of SCM EMG data (6 studies, 9 interventions) showed a statistically significant reduction in sternocleidomastoid muscle activity after neck exercise versus controls. A recurring theme was that neuromuscular adaptations tended to be specific to the type of training performed.
Key Findings
| Finding | Detail | Impact |
|---|---|---|
| Neck exercise reduced superficial sternocleidomastoid muscle activity | A meta-analysis of SCM EMG amplitude pooling 6 of the 14 studies (9 interventions) found an overall effect favouring reduced SCM activity after exercise versus control, with an effect size estimate of -1.67 (SE 0.73), Z = 4.47, p < 0.00001, and a 95% CI of -2.40 to -0.94 that did not cross the no-effect line. Note: substantial heterogeneity was present (reported I-square = 0.96). | High |
| Moderate certainty of evidence for craniocervical flexion (CCF) training inducing neural adaptations | Per GRADE appraisal, there is moderate certainty that CCF training, used in isolation or combined with resistance training, can induce neural adaptations within the neck muscles in people with chronic non-specific neck pain. Resistance (strength/endurance) training and neck-specific exercise programs were also rated moderate; global postural re-education was rated high (single study); several single-study approaches (postural, suspension, mixed physiotherapy, Feldenkrais, stretching plus strength) were rated very low. | High |
| Neuromuscular adaptations were largely specific to the type of training performed | The type of EMG change generally matched the task and muscles trained. For example, Jull 2009 reported improved control of deep and superficial neck flexors after 6 weeks of CCF training but not after neck flexor strength training, even though pain decreased comparably in both groups; Falla 2006 found reduced SCM/anterior scalene activity during isometric flexion after resistance training but not after low-load CCF training. Some exceptions (faster muscle onset after training) were noted. | High |
| Cervical neuromuscular adaptations can occur rapidly with training | Two studies using short programs (Beer 2012, 2 weeks; Yan 2022, 4 weeks) showed decreased SCM and upper trapezius EMG during the craniocervical flexion test, suggesting adaptations occur early, consistent with literature on early neural adaptations. The benefit of longer versus shorter training durations was not established. | Medium |
| Most studies reported improvements in pain and disability alongside neuromuscular changes | All included studies reported positive improvement in pain and disability after training except Lundblad 1999 (possible floor effect from its disability measure) and Beer 2012 (only a 2-week program, possibly too short for pain relief despite early neural adaptation). The review could not establish whether neuromuscular changes explain the symptom improvements. | Medium |
| No study used transcranial magnetic stimulation; deeper neck muscles were rarely assessed | Although corticospinal excitability via transcranial magnetic stimulation was an eligible outcome, none of the 14 articles used it. Only one study (Jull 2009) assessed the deep cervical flexors (longus capitis and longus colli), reporting increased deep flexor EMG during the CCFT after CCF training but no change after strength training. | Medium |
A meta-analysis of SCM EMG amplitude pooling 6 of the 14 studies (9 interventions) found an overall effect favouring reduced SCM activity after exercise versus control, with an effect size estimate of -1.67 (SE 0.73), Z = 4.47, p < 0.00001, and a 95% CI of -2.40 to -0.94 that did not cross the no-effect line. Note: substantial heterogeneity was present (reported I-square = 0.96).
Per GRADE appraisal, there is moderate certainty that CCF training, used in isolation or combined with resistance training, can induce neural adaptations within the neck muscles in people with chronic non-specific neck pain. Resistance (strength/endurance) training and neck-specific exercise programs were also rated moderate; global postural re-education was rated high (single study); several single-study approaches (postural, suspension, mixed physiotherapy, Feldenkrais, stretching plus strength) were rated very low.
The type of EMG change generally matched the task and muscles trained. For example, Jull 2009 reported improved control of deep and superficial neck flexors after 6 weeks of CCF training but not after neck flexor strength training, even though pain decreased comparably in both groups; Falla 2006 found reduced SCM/anterior scalene activity during isometric flexion after resistance training but not after low-load CCF training. Some exceptions (faster muscle onset after training) were noted.
Two studies using short programs (Beer 2012, 2 weeks; Yan 2022, 4 weeks) showed decreased SCM and upper trapezius EMG during the craniocervical flexion test, suggesting adaptations occur early, consistent with literature on early neural adaptations. The benefit of longer versus shorter training durations was not established.
All included studies reported positive improvement in pain and disability after training except Lundblad 1999 (possible floor effect from its disability measure) and Beer 2012 (only a 2-week program, possibly too short for pain relief despite early neural adaptation). The review could not establish whether neuromuscular changes explain the symptom improvements.
Although corticospinal excitability via transcranial magnetic stimulation was an eligible outcome, none of the 14 articles used it. Only one study (Jull 2009) assessed the deep cervical flexors (longus capitis and longus colli), reporting increased deep flexor EMG during the CCFT after CCF training but no change after strength training.
Strengths
- First systematic review to comprehensively examine the effect of all forms of neck-targeted exercise on neuromuscular function (not just CCF training) in chronic non-specific neck pain
- Conducted to a registered, published protocol (PROSPERO CRD42021298831) and reported per PRISMA, following Cochrane back and neck group guidelines
- Used the Cochrane risk-of-bias tool and GRADE to appraise the certainty of evidence, with two independent reviewers and 100% reviewer agreement at all stages
- Searched eight databases including grey literature (Open Grey, Zetoc) from inception
- Quantitative meta-analysis of sternocleidomastoid EMG where data permitted, supplemented by structured narrative synthesis
Limitations
- Low sample sizes across included studies (evidence downrated for imprecision when total sample size was under 24)
- Substantial heterogeneity in the meta-analysis (reported I-square = 0.96) and wide variation in exercise dosage, duration, and quality of exercise description
- Difficulty comparing EMG data across studies due to the range of assessment tasks (e.g., craniocervical flexion test, typing, shoulder MVC, upper-limb tasks)
- Only sternocleidomastoid EMG could be pooled; deeper neck muscles were assessed in only one study and no study used transcranial magnetic stimulation
- Most studies enrolled only women and a limited age range (~25–50 years), limiting generalizability
- English-language-only restriction, raising potential for language/selection bias; the review could not determine whether neuromuscular changes explain improvements in pain and disability
Key Takeaways for Patients
What This Means for You
- 01Exercise that targets the neck muscles can change how those muscles work, not just how they feel — studies measured these changes using electromyography (muscle activity recordings).
- 02Craniocervical flexion training (gentle deep-neck-flexor exercises, often done lying down nodding the head) has moderate-quality evidence for retraining neck muscle control in people with long-standing, non-specific neck pain.
- 03The benefits of exercise tend to be specific to what you practice — so the type of neck exercise your clinician prescribes matters and should match your goals.
- 04Most studies in this review also showed reductions in neck pain and disability after exercise, though the review could not prove that the muscle changes are what caused the pain relief.
- 05Some changes appeared after as little as 2 to 4 weeks of training, suggesting the neck muscles can begin adapting fairly quickly.