Most neck and shoulder pain is mechanical and settles with movement and time. The best-supported treatment is exercise that loads the neck, shoulder and upper-back muscles; a stiff, painful shoulder responds to targeted exercise, with a steroid injection to calm it enough to begin. Three common beliefs do not hold up: that a disc bulge on a scan explains the pain, that a forward head or poor posture brings it on, and that keyhole shoulder decompression beats a placebo operation.
After whiplash, staying active recovers better than resting in a collar. A small set of warning signs needs a doctor, above all the slow picture of clumsy hands and an unsteady walk that points to the spinal cord being squeezed.
Practice Ranking
Every practice we track for Neck & Shoulder Pain, ranked by how well the evidence supports it for this condition. Strength describes the evidence, not our endorsement.
5 practices · 1 to start with
| # | Practice | Evidence | Type | Cost | Effort | Results In | Add to plan |
|---|---|---|---|---|---|---|---|
| 1 | Resistance Training Strengthening the neck, upper back and shoulder is the strongest self-directed lever for both neck pain and rotator-cuff shoulder pain. | Moderate | Self-Directed | Free to $$ | Moderate to Hard | Weeks to Months | |
| 2 | Mobility & Stretching After a whiplash-type strain, gentle movement recovers better than immobilizing the neck. | Moderate | Self-Directed | Free | Easy | Days to Weeks | |
| 3 | Yoga Helps chronic neck pain and mobility. | Moderate | Self-Directed | Free to $ | Moderate | Weeks to Months | |
| 4 | Acupressure & Self-Massage Massage made little measurable difference to neck pain against a placebo, though it is low-risk and pleasant. | Preliminary | Self-Directed | Free | Easy | Days to Weeks | |
| 5 | Tai Chi & Qi Gong Tai chi eased chronic neck pain in early trials. | Emerging | Self-Directed | Free to $$ | Easy to Moderate | Weeks to Months | |
Default order puts the best-supported practices first, with self-directed care ahead of clinical options. Click any row to open the practice.
What It Is
Neck and shoulder pain is one of the most common aches there is, and most of it is mechanical. It comes from the ordinary load, posture and movement of everyday life, and it eases over weeks with movement and time. It is not a sign of damage or disease.
Knowing which kind you have points to what helps:
- Non-specific mechanical neck pain has no single structure to blame. It follows long hours at a desk, a poor night, a sudden strain or a period of stress; it aches and stiffens, and it loosens as you move. This is the common kind, and it responds to the movement and strengthening covered next.
- Cervical radiculopathy is nerve-related pain. A nerve root in the neck is irritated or pinched, sending pain, tingling or numbness down the arm along the path of one nerve, sometimes with weakness in a specific muscle. Most cases settle on their own over weeks to months without surgery.
- Rotator cuff and subacromial shoulder pain hurts with reaching and overhead work, when the tendons and the small bursa under the tip of the shoulder blade are loaded.
- Frozen shoulder is the capsule around the joint tightening until the shoulder stiffens and range of motion itself is lost. It takes patience to recover.
- Whiplash-associated neck pain follows a sudden jolt, usually a car accident, with pain and stiffness that come on over the following hours and days.
Pain can also be referred, felt in the neck or shoulder while it arises elsewhere, which is why a few whole-body warning signs sit at the end of this page. Scans are more misleading than helpful here: bulging discs and age-related wear show up on the neck scans of most people who have no neck pain at all, so imaging is reserved for when a warning sign is present.
What Helps
The best-supported treatments load and move the painful area, and they follow an order: what you can do yourself first, the shorter-term aids next, and the treatments that do not hold up last.
Movement and strengthening come first. Exercise that loads the neck, shoulder and upper-back muscles is the best-supported treatment for lasting neck pain, drawn from a Cochrane review of 27 trials, and the benefit holds as long as you keep it up; general fitness or stretching on its own does much less. For a rotator cuff or subacromial shoulder, a targeted exercise program is the mainstay in the same way, and shoulder-specific exercises work better than generic ones. After whiplash, people told to carry on with normal activity did better months later than those given time off and a soft collar.
Resting a sore neck is the wrong instinct. The strongest evidence is for the treatments you do yourself: move early, load the muscles, and treat rest as the exception.
A steroid injection has a limited role for a stubborn shoulder. It settles subacromial pain in the short term, and an ultrasound-guided injection places it more accurately, but it has not been shown to beat a good exercise program. It is used to calm a shoulder enough to begin the exercise that carries the longer benefit. For frozen shoulder, exercise and time together, sometimes with an injection, do better than waiting alone.
Several gentler options add to that base, and each suits a different reader:
- Tai chi lowered chronic neck pain about as much as conventional neck exercises did, a route for someone who would rather move that way.
- Acupuncture eases mechanical neck pain in the weeks after treatment, doing a little better than sham needling and clearly better than waiting it out.
- Manual therapy, spinal manipulation or gentler mobilization, can ease neck pain in the short term and make it easier to start moving again, with the clearest signal for manipulation applied to the upper back. Manipulation and mobilization work about equally well, so the gentler option is a reasonable choice.
- Heat gives short-term relief that makes moving easier.
- For nerve-related pain, the course itself is reassuring: most cervical radiculopathy improves without surgery over weeks to months.
Not everything sold for a sore neck holds up. The ladder below shows each treatment at its true strength, including the ones that matched a placebo.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Pain
Keyhole shoulder decompression was no better than placebo surgery (32.7 vs 34.2)
Keyhole surgery to decompress a painful shoulder worked no better than a sham operation, and both were only marginally ahead of doing nothing, by less than the amount that counts as meaningful.
CSAW was a multicenter, placebo-controlled, three-group randomized surgical trial. Patients with at least three months of subacromial pain, intact rotator cuff tendons, and prior non-operative treatment including exercise and at least one steroid injection were randomized (1:1:1) to arthroscopic subacromial decompression, investigational arthroscopy only (a placebo omitting the bone and soft-tissue removal), or no treatment. The primary outcome, the Oxford Shoulder Score at 6 months (0 worst to 48 best), was 32.7 (SD 11.6) after decompression versus 34.2 (SD 9.2) after placebo arthroscopy (mean difference -1.3, 95% CI -3.9 to 1.3, p = 0.31). Both surgical groups exceeded no treatment by a margin below the trial's target difference of 4.5 points.
The study · 1
Beard et al., Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a placebo-controlled randomised surgical trial · Lancet 2018;391(10118):329-338
Neck and upper-back strength training lowered chronic neck pain (pooled SMD -0.71)
Exercises that actually load the neck, shoulder and upper-back muscles, done regularly, are the best-supported way to bring down long-running neck pain. General fitness or stretching on its own does much less.
The Cochrane review of exercises for mechanical neck disorders pooled 27 randomized trials (2,485 of 3,005 randomized participants analyzed) and graded the evidence with GRADE. For chronic neck pain, moderate-quality evidence supported cervico-scapulothoracic and upper-extremity strength training (pooled SMD -0.71, 95% CI -1.33 to -0.10 immediately post-treatment) and combined cervical, shoulder and scapulothoracic strengthening plus stretching (SMD -0.33, -0.55 to -0.10 for pain, -0.45, -0.72 to -0.18 for function). Breathing exercises, general fitness training and stretching alone carried only low-quality evidence and may not change pain or function. For acute neck pain the review found no clear sign in either direction.
The study · 1
Gross et al., Exercises for mechanical neck disorders · Cochrane Database Syst Rev 2015;(1):CD004250
Acupuncture eased mechanical neck pain more than sham needling in the short term
Acupuncture eases mechanical neck pain in the weeks after treatment, doing a little better than placebo needling and clearly better than waiting it out.
The 2016 Cochrane review of acupuncture for neck disorders included 27 trials spanning whiplash-associated disorders, myofascial neck pain, arthritic neck pain, non-specific neck pain, radicular signs and subacute or chronic mechanical neck pain. For mechanical neck pain, acupuncture was beneficial at immediate-term follow-up versus sham for pain intensity, at short-term follow-up versus sham or inactive treatment for pain intensity, at short-term versus sham for disability, and at short-term versus wait-list for pain and disability improvement. Pooling against sham was statistically appropriate given homogeneity (P = 0.83, I2 = 20%), and the pooled result favored acupuncture. The review did not establish long-term effects.
The study · 1
Trinh et al., Acupuncture for neck disorders · Cochrane Database Syst Rev 2016;(5):CD004870
Exercise beat no treatment for rotator cuff shoulder pain (SMD -0.94)
For rotator cuff and subacromial shoulder pain, a targeted exercise program is the mainstay, and shoulder-specific exercises work better than generic ones.
This BJSM systematic review and meta-analysis of conservative interventions for shoulder impingement found exercise superior to non-exercise controls for pain (SMD -0.94, 95% CI -1.69 to -0.19) and specific exercises superior to generic ones (SMD -0.65, -0.99 to -0.32). Manual therapy added to exercise was superior to exercise alone at the shortest follow-up only (SMD -0.32, -0.62 to -0.01). The authors concluded that exercise should be considered first, while noting the primary trials were of very low quality, so effect sizes are uncertain even where the direction is consistent, and it remains unclear how exercise compares directly with injections.
The study · 1
Steuri et al., Effectiveness of conservative interventions in adults with shoulder impingement: a systematic review and meta-analysis of RCTs · Br J Sports Med 2017;51(18):1340-1347
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
A steroid injection beat no treatment for shoulder pain short-term (SMD -0.65)
A steroid injection can settle a painful shoulder in the short term, but it has not been shown to beat a good exercise program, so it works best to calm things enough to get moving, not as the whole answer.
Within the BJSM meta-analysis, corticosteroid injection was superior to no treatment (SMD -0.65, 95% CI -1.04 to -0.26) and ultrasound-guided injection was superior to unguided injection (SMD -0.51, -0.89 to -0.13). The comparison was against no treatment rather than against exercise, and the review stated it was unclear how injections compare with exercise. NSAIDs also beat placebo, with a smaller effect (SMD -0.29, -0.53 to -0.05).
Use an injection to bring pain down far enough to start the exercise that carries the longer-term benefit, rather than as a standalone fix, and a guided injection places the steroid more accurately.
The study · 1
Steuri et al., Effectiveness of conservative interventions in adults with shoulder impingement: a systematic review and meta-analysis of RCTs · Br J Sports Med 2017;51(18):1340-1347
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
Most cervical radiculopathy improves without surgery over weeks to months
Nerve-related neck pain that shoots into the arm usually gets better on its own over weeks to months, and most people never need surgery.
Narrative and systematic reviews of cervical radiculopathy describe a natural history in which the majority of patients improve with nonoperative management, including relative rest with early return to activity, physical therapy, analgesia and selective nerve-root corticosteroid injection. Surgical options (anterior cervical decompression and fusion, cervical disc arthroplasty, posterior foraminotomy) are considered mainly for progressive neurological deficit or persistent disabling pain, but the reviews note that precise surgical indications and optimal timing have not been fully defined.
The studies · 2
Iyer & Kim, Cervical radiculopathy · Curr Rev Musculoskelet Med 2016;9(3):272-280
Childress MA, Becker BA, Nonoperative Management of Cervical Radiculopathy (referenced review) · Am Fam Physician 2016;93(9):746-754
Staying active after whiplash beat a collar and rest at 6 months
After a whiplash-type neck injury, resting up in a soft collar leaves people worse off months later than simply carrying on with normal activity.
This single-blinded randomized trial recruited 201 patients with neck sprain from a car accident at a Norwegian emergency clinic. During the first 14 days one group was encouraged to act as usual and continue pre-injury activities; the other was given sick leave and immobilized with a soft neck collar. Both improved from intake to 24 weeks, but the act-as-usual group had significantly better outcomes at six months across subjective symptoms, including pain localization, pain during daily activities, neck stiffness, memory and concentration, and visual-analogue neck pain and headache.
After an ordinary whiplash, keep the neck moving and return to normal activity early rather than resting it in a collar, while watching for the warning signs listed below after any significant accident.
The study · 1
Borchgrevink et al., Acute treatment of whiplash neck sprain injuries: a randomized trial of treatment during the first 14 days after a car accident · Spine 1998;23(1):25-31
Tai chi lowered chronic neck pain by 10.5 mm versus a wait list
Twelve weeks of tai chi lowered chronic neck pain about as much as conventional neck exercises did, so it is a reasonable option for someone who would rather move that way.
In this German randomized controlled trial, adults with chronic nonspecific neck pain were assigned to 12 weeks of group tai chi, conventional neck exercises, or a wait-list control (75 to 90 minute weekly sessions). The primary outcome, pain intensity on a visual analogue scale, favored tai chi over the wait list by 10.5 mm (95% CI -20.3 to -0.9, P = 0.033), with group differences also for pain on movement, functional disability and quality of life. There was no meaningful difference between tai chi and conventional neck exercises, and only minor side effects were reported.
If conventional neck exercises are not appealing, tai chi is a comparable route to the same benefit; the significant result was against no treatment, so do one or the other rather than nothing.
The study · 1
Lauche et al., The Effects of Tai Chi and Neck Exercises in the Treatment of Chronic Nonspecific Neck Pain: A Randomized Controlled Trial · J Pain 2016;17(9):1013-1027
Frozen shoulder usually improves a lot but often does not fully resolve on its own
Frozen shoulder was long taught to run a set course and then fully recover by itself. The evidence does not support that: it usually improves a lot, but often not completely, and treatment plus time does better than time alone.
The review examined the natural-history theory of frozen shoulder, that it progresses through painful, stiff and recovery phases to full recovery without treatment. From 508 citations, seven studies with no-treatment comparison groups met inclusion. Low-quality evidence indicated some but not complete improvement in range of motion after one to four years, and no clear sign supported progression through recovery phases to full resolution without treatment. Moderate-quality data from three randomized trials with longitudinal follow-up showed most improvement occurred early rather than late, contradicting the classical late-recovery phase.
The study · 1
Wong et al., Natural history of frozen shoulder: fact or fiction? A systematic review · Physiotherapy 2017;103(1):40-47
Manual therapy eased neck pain short-term, strongest for upper-back manipulation (SMD -1.26)
Hands-on treatment, spinal manipulation or gentler mobilization, can ease neck pain in the short term, with the clearest signal for manipulation applied to the upper back. Manipulation and mobilization work about equally well, so the gentler option is a reasonable choice.
The 2015 Cochrane review of manipulation and mobilization for neck pain pooled 51 randomized trials (2,920 participants) covering acute, subacute and chronic neck pain with or without cervicogenic headache or radicular symptoms. Thoracic manipulation versus an inactive control improved pain at short-term follow-up (five trials, 346 participants, moderate quality, pooled SMD -1.26, 95% CI -1.86 to -0.66) and function (four trials, 258 participants, moderate quality, SMD -1.40, 95% CI -2.24 to -0.55), with a funnel plot suggesting publication bias. Cervical manipulation and mobilization gave similar results to one another at immediate, short and intermediate follow-up, multiple cervical manipulation sessions were more effective than certain medications, and mobilization as a stand-alone intervention was not clearly better than an inactive control. Results for cervical manipulation versus control were few and diverse.
Who this may not transfer to:The trials pooled acute, subacute and chronic neck pain together with cervicogenic headache, so the numbers do not separate out how much a given person with ordinary chronic neck pain should expect, and the strongest result is for upper-back rather than neck manipulation.
Use hands-on treatment as a short-term step that makes it easier to move and to start the strengthening that carries the longer benefit. Since manipulation and mobilization work about the same, the gentler mobilization is a reasonable choice, especially given the rare but serious risk of neck manipulation set out in the cautions.
The study · 1
Gross et al., Manipulation and mobilisation for neck pain contrasted against an inactive control or another active treatment · Cochrane Database Syst Rev 2015;(9):CD004249
Massage made little difference to neck pain versus placebo (about 3.4 points on 100)
A one-off rubdown does about as much for neck pain as a placebo, but a proper course of longer, repeated sessions may help.
The 2024 Cochrane review of massage for neck pain included 33 trials (1,994 participants analyzed, aged 18 to 70, 70% female, mostly non-specific subacute-to-chronic neck pain). Against placebo, low-certainty evidence indicated little to no difference in pain (mean 20.55 with placebo, improving 3.43 points with massage, 95% CI 8.16 better to 1.29 worse, on a 0 to 100 scale), function-disability or quality of life at up to 12 weeks. A dose subgroup (at least eight sessions over four weeks of at least 30 minutes) showed a clinically important difference favoring massage. Certainty was downgraded for indirectness because most placebo-comparison trials used single, suboptimal sessions.
If you try massage, a short course of substantial sessions is more likely to do something than a single brief treatment.
The study · 1
Gross et al., Massage for neck pain · Cochrane Database Syst Rev 2024;(2):CD004871
Measurement And Diagnosis
Disc bulges showed up on the neck scans of 87.6% of people with no pain
Bulging discs turn up on the neck scans of most people who have no neck pain at all, even in their twenties, so a bulge on a scan usually is not the reason a neck hurts.
This cross-sectional study imaged 1,211 healthy volunteers, roughly 100 per decade per sex from ages 20 to 70. Disc bulging was present in 87.6% overall and increased with age in frequency, severity and number of levels, yet was already present in 73.3% of men and 78.0% of women in their twenties. In contrast, spinal cord compression (5.3%) and increased cord signal intensity (2.3%) were uncommon and rose mainly after age 50, most often at C5-C6 and C6-C7.
The study · 1
Nakashima et al., Abnormal findings on magnetic resonance images of the cervical spines in 1211 asymptomatic subjects · Spine 2015;40(6):392-398
Forward head posture differed by just 4.84 degrees with neck pain, and not at all in teenagers
In adults there is a small link between a forward-head, rounded posture and neck pain, but it does not appear in teenagers and is tangled up with age, so posture is one modest factor rather than the single cause it is often made out to be.
The systematic review and meta-analysis pooled 15 cross-sectional studies. Ten compared forward head posture between people with neck pain and pain-free controls, giving an overall mean difference of 4.84 degrees (95% CI 0.14 to 9.54) in adults, but a non-significant difference in adolescents (-1.05; -4.23 to 2.12). Eight studies showed negative correlations between forward head posture and neck pain intensity (r = -0.55) and disability (r = -0.42) in adults and older adults, while in adolescents only lifetime prevalence and doctor visits predicted posture. The authors concluded age acts as a confounding factor in the posture-pain relationship.
The study · 1
Mahmoud et al., The Relationship Between Forward Head Posture and Neck Pain: a Systematic Review and Meta-Analysis · Curr Rev Musculoskelet Med 2019;12(4):562-577
What The Beliefs Get Wrong
A few widely repeated ideas about neck and shoulder pain point people in the wrong direction. Each one is contradicted by the evidence.
A disc bulge on a scan is usually not the cause. Among 1,211 healthy volunteers with no neck pain, 87.6% had a bulging disc on a neck MRI, including roughly three-quarters of people in their twenties, and the finding became more common with age. These changes are so ordinary in pain-free necks that a bulge seen alongside pain is often incidental, which is why a scan taken for routine neck pain is often misleading.
Bad posture and a forward head are a smaller factor than the story suggests. Across 15 studies, adults with neck pain held their heads slightly more forward than pain-free adults, by about 4.84 degrees, with a modest correlation to pain intensity. There was no such link in teenagers, and age itself was a confounder. Posture is one modest factor, and chasing a perfect posture is not the cure it is sold as.
Keyhole shoulder decompression matched a placebo operation. In the CSAW trial of 313 patients, the Oxford Shoulder Score six months after real arthroscopic decompression was no different from the score after a placebo arthroscopy, and both surgical groups were only marginally ahead of no treatment, by less than the amount that counts as meaningful. For ordinary subacromial shoulder pain, this operation does not help.
Frozen shoulder does not always fully resolve on its own. It was long taught to pass through set phases and recover completely without treatment. A systematic review found that the shoulder usually improves a great deal but often not completely over one to four years, that most of the improvement happens early, and that exercise and time together do better than time alone.
How It Works
Mechanical neck and shoulder pain is a signal from irritated, sensitized tissues and their nerves. It does not track structural damage, which is why the pain and the picture on a scan match so poorly. Sustained load, awkward positions, a poor night and stress all raise the sensitivity of the local muscles and joints, and the pain eases as the tissues calm and normal movement returns.
Targeted strengthening works with this process. It builds the capacity of the neck, shoulder and upper back to carry everyday load, and it lowers the sensitivity that keeps the pain going, which is why using the neck helps it.
The same applies to the shoulder. The rotator cuff tendons and the bursa under the tip of the shoulder blade are loaded with reaching and overhead work, and a graded exercise program restores their tolerance for that load. A steroid injection lowers the local inflammation enough to move, so it works alongside exercise. Nerve-related pain comes from an irritated nerve root, and as the irritation settles the arm symptoms usually recede, which is the recovery that most cervical radiculopathy makes on its own.
Go Deeper
- Arthritis and joint pain: the wear-and-inflammation picture behind many stiff, aching joints, and the movement and strength that help most.
- Low back pain: the same active-recovery pattern for the other end of the spine, where scans are misleading in exactly the same way.
- Resistance training: building the strength around the neck, shoulder and upper back, the most consistently supported treatment for lasting neck pain.
- Tai chi and qi gong: the gentle mind-body movement that matched conventional neck exercises for chronic neck pain.
- Dizziness and vertigo: when neck symptoms come with dizziness or unsteadiness, where that overlaps and where it needs its own assessment.
The Chinese Medicine View
The Chinese Medicine View
Chinese medicine reads neck and shoulder pain as obstruction in the channels that cross the area, not as a worn structure, and treats it by pattern. The Small Intestine, Gallbladder and Bladder channels all run across the neck and shoulder, so the pathway maps closely onto where the pain sits. A classic stiff neck from a draft is read as Wind-Cold lodging in those channels; a fixed, stabbing pain after a strain is read as Qi and Blood stagnation. Acupuncture and tui na (Chinese therapeutic massage) are long-standing strengths here, and the research puts acupuncture for mechanical neck pain at a modest strength, better than sham needling in the short term. Read the patterns below as an interpretive lens on the presentation, not a reading of a scan. Strength of tradition is a reason to take a practice seriously; it is not a claim that the practice outperforms the treatments above.
Stiff, aching pain that comes on or worsens with cold, wind or a draft and eases with warmth and movement, the classic stiff neck after a cold night by an open window. The direction is to dispel Wind and Cold, warm the area and free the channels, which fits a presentation that loosens as the neck warms and moves.
Sharp, fixed, stabbing pain, often after a strain, an awkward night or an injury, worse with pressure and with staying still. The direction is to move Qi and Blood and unblock the channels crossing the neck and shoulder, which acupuncture and tui na aim to do directly.
Chronic, nagging weakness and stiffness in an older or run-down person, worse with tiredness and better with rest and warmth. The direction is to tonify the Liver and Kidney and nourish the sinews alongside the gentle strengthening that carries the physical benefit.
Cautions
Everything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Vertebral-artery stroke was linked to chiropractic and family-doctor visits alike, about threefold under 45
This population-based case-control and case-crossover study used Ontario health billing records for 818 vertebrobasilar artery (VBA) strokes over a population exceeding 100 million person-years, with four age- and sex-matched controls per case. In people under 45, cases were roughly three times more likely than controls to have seen a chiropractor in the prior year, but they were also more likely to have seen a primary care physician, and practitioner visits billed for headache and neck complaints were highly associated with subsequent VBA stroke. There was no increased chiropractic association above age 45, and no excess risk from chiropractic care compared with primary care.Cassidy et al., Risk of vertebrobasilar stroke and chiropractic care: a population-based case-control and case-crossover study
Decompression surgery improved function in cervical cord compression (479 patients)
This prospective multicenter AOSpine International study enrolled 479 patients with symptomatic degenerative cervical myelopathy who underwent surgical decompression. At follow-up, functional status (modified Japanese Orthopaedic Association score), disability (Neck Disability Index) and quality of life (SF-36) all improved significantly from baseline. As a single-arm outcome study it shows improvement after surgery rather than a controlled comparison, but combined with the tendency of untreated myelopathy to progress, it underlines why the early neurological signs, hand clumsiness, gait and balance change, warrant prompt assessment.Fehlings et al., A global perspective on the outcomes of surgical decompression in patients with cervical spondylotic myelopathy: the prospective multicenter AOSpine international study on 479 patients
Neck manipulation and a rare artery tear
Neck manipulation has been linked to a rare stroke from a tear in a vertebral artery. A population study of 818 such strokes found people under 45 were about three times more likely to have seen a chiropractor beforehand, but they were equally more likely to have seen a family doctor. That pattern points to the tear itself causing the neck pain and headache that send a person to seek care, not the treatment causing the tear. The risk is very small and cannot be fully ruled out. Gentler mobilization works about as well as manipulation, so it is a reasonable choice. A sudden severe new headache, or neck pain with dizziness or slurred speech, is a reason to seek urgent care.
A shoulder injection is a short-term aid, not a cure
A steroid injection settles a painful shoulder in the short term, but it has not been shown to beat a good exercise program. Its role is to calm the shoulder enough to start exercising, not a standing schedule of repeated injections.
Do not wait out worsening arm or leg signs
Ordinary neck pain settles with movement and time. Arm or hand weakness or numbness that is getting worse, or the slow picture of clumsy hands and an unsteady walk, are reasons to see a doctor promptly, not to keep treating the neck. These are set out in the warning signs below.
Most neck and shoulder pain is safe to stay active with. Move sensibly, start gently, and consult a licensed practitioner if you have questions, or promptly if any of the warning signs below fit you.
When to See Someone
Most neck and shoulder pain is not dangerous, and this is the short list of exceptions. See a doctor promptly, and seek urgent care for the first three, if you have:
- Worsening clumsy hands, trouble with buttons or handwriting, an unsteady walk, or weakness or numbness in the arms or legs, which can mean the spinal cord is being squeezed (degenerative cervical myelopathy). This tends to progress, and delay costs recovery, so it needs prompt specialist referral.(seek urgent care)
- Loss of bladder or bowel control alongside neck symptoms.(seek urgent care)
- A sudden severe new headache or neck pain with dizziness, slurred speech, double vision or unsteadiness, which can signal a tear in an artery in the neck.(seek urgent care)
- Neck pain after a significant fall or car accident.
- Fever, night sweats, or feeling generally unwell alongside the neck pain, which can point to infection.
- Unexplained weight loss, or a history of cancer.
- Arm or hand weakness or numbness that is steadily getting worse.
- Severe pain that does not ease with rest or a change of position, or pain that steadily wakes you at night.
None of this is meant to alarm you. The great majority of neck and shoulder pain settles with movement and time, and the signs above are the rare exceptions to watch for so you can act on them.
Common Questions
What is the best treatment for neck pain?
Targeted strengthening of the neck, shoulder and upper back is the best-supported treatment for lasting neck pain. In a Cochrane review of 27 trials, exercises that actually load those muscles lowered chronic neck pain by a moderate to large amount, while general fitness or stretching alone did much less, and the benefit holds as long as you keep it up (Gross, Cochrane 2015).
Does a disc bulge on a scan mean that is what is causing my pain?
Usually not. Among 1,211 people with no neck pain, 87.6% had a bulging disc on a neck MRI, including about three-quarters of those in their twenties, and the finding grew more common with age (Nakashima, Spine 2015). Because bulges are so ordinary in pain-free necks, one seen alongside pain is often incidental, which is why scans are unhelpful for routine neck pain.
Does bad posture cause neck pain?
Less than the story suggests. Across 15 studies, adults with neck pain held their heads about 4.84 degrees more forward than pain-free adults, a small difference with a modest link to pain, but there was no such association in teenagers and age was a confounder (Mahmoud, Curr Rev Musculoskelet Med 2019). Posture is one modest factor, and chasing perfect posture is not a cure.
Should I rest in a collar after whiplash?
Staying active is the better choice. In a trial of 201 people after a car-accident neck sprain, those told to carry on with normal activity had less pain and stiffness at six months than those given time off and a soft collar (Borchgrevink, Spine 1998). Keep the neck moving, while watching for the warning signs after any significant accident.
Is keyhole surgery worth it for a painful shoulder?
For ordinary subacromial pain, no. In the CSAW trial of 313 patients, shoulder scores six months after real keyhole decompression were no different from scores after a placebo operation, and both were only marginally ahead of no treatment, by less than a meaningful amount (Beard, Lancet 2018). Targeted exercise, with an injection to settle a stubborn shoulder, is the mainstay.
Does frozen shoulder go away on its own?
It usually improves a great deal, but often not completely. A systematic review of the old idea that frozen shoulder runs a set course to full recovery found the shoulder improves over one to four years with most of the gain coming early, yet recovery is frequently incomplete, and exercise and time do better than time alone (Wong, Physiotherapy 2017).
Is it safe to have my neck manipulated?
The serious risk is very small. Neck manipulation has been linked to a rare stroke from a torn artery, but a study of 818 such strokes found the same association with ordinary family-doctor visits, which points to the tear causing the neck pain that sends people to seek care, not the treatment causing the tear (Cassidy, Spine 2008). Gentler mobilization works about as well, so it is a reasonable choice.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 16 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 9, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.