Do Chin Tucks Actually Work? What Has to Hold for the Effect to Last

Chin tucks work, on the outcomes they were designed to change, at the dose the trials used, for as long as the training continues. In a 54-person randomised controlled trial run at Khon Kaen University in Thailand and published in the Journal of Pain Research in 2019, Suvarnnato and colleagues found that six weeks of deep cervical flexor training lowered Neck Disability Index scores by 5.25 points more than usual care (95% CI 1.52–8.98, p=0.003) and opened the craniovertebral angle by 2.59 degrees more (95% CI 0.85–4.33, p=0.002). Both advantages were still measurable three months after training stopped. The pain advantage was not. It measured 0.63 points on an 11-point scale immediately after the six weeks (p=0.029) and had vanished by the one-month follow-up.

Two different exercises are called a chin tuck, and they train different muscles

The chin tuck studied in neck pain research is craniocervical flexion: a low-load nodding movement performed lying on the back, guided by an inflatable pressure sensor behind the neck. Jull, O'Leary and Falla described the standardised version in the Journal of Manipulative and Physiological Therapeutics in 2008, naming its targets as the longus capitis and longus colli — muscles on the front surface of the cervical vertebrae, behind the throat, well underneath anything visible in a mirror.

That the movement actually reaches those muscles is not an inference. Falla, Jull and Hodges verified it directly in Spine in 2004 by threading custom electrodes through the nose and fixing them by suction to the back of the oropharynx in 10 patients with chronic neck pain and 10 controls. Deep flexor activity rose in a straight line across the five stages of the test (p=0.002), and was significantly lower in the neck pain group at the harder stages (p<0.05).

A second exercise carries the same name. Chin tuck against resistance, or CTAR, is a swallowing-rehabilitation technique in which the chin presses down against a ball. Huang and colleagues, reporting in Dysphagia on 90 healthy adults aged 65 to 90, found suprahyoid activity — the muscles in the soft triangle under the jaw — significantly higher than sternocleidomastoid activity during the hold. Same two words, different anatomy, different literature. Most of the jawline confusion starts here.

What the trial actually prescribed

Suvarnnato's team screened 57 patients with chronic mechanical neck pain and enrolled 54, allocating 18 to each of three arms: deep cervical flexor training, semispinalis cervicis training, and usual care. The trial is registered as NCT02656030.

The deep flexor arm attended 12 supervised sessions, twice weekly across six weeks, and practised twice daily at home. Price and colleagues, extracting the protocol for their 2020 systematic review in PLoS ONE, recorded the intensity as three sets of 10 repetitions, each held 10 seconds, with 30 seconds of rest between sets. The pressure sensor was inflated to a 20 mmHg baseline, and each participant worked at the highest of five increments — 22, 24, 26, 28 or 30 mmHg — they could reach without recruiting the superficial neck muscles, which the researcher monitored by palpating the sternocleidomastoid.

Craniovertebral angle was photographed from 200 cm, measuring the line from the tragus of the ear to the seventh cervical spinous process against the horizontal, a technique with a reported intraclass correlation coefficient of 0.94.

| Outcome at week 6 (baseline-adjusted mean) | Deep cervical flexor training | Semispinalis cervicis training | Usual care | |---|---|---|---| | Neck Disability Index (%) | 14.99 | 13.29 | 20.24 | | Numeric pain scale (0–10) | 2.86 | 2.30 | 3.49 | | Craniovertebral angle (degrees) | 50.78 | 50.41 | 48.18 | | Deep neck-flexor strength (kg) | 2.07 | 1.89 | 1.77 |

At the three-month follow-up, the deep flexor group still held a 5.07-point disability advantage over usual care (p=0.004) and a 2.46-degree angle advantage (p=0.009). Its pain advantage by then measured 0.10 points, p=1.

The size of the change deserves a straight reading

Pool and colleagues established thresholds for these instruments in Spine in 2007, from 183 patients with non-specific neck pain. They reported a minimal detectable change of 10.5 points on the Neck Disability Index — scored 0 to 50 — and an optimal ROC cutoff of 3.5 points. Suvarnnato's team reported the index as a percentage, so its 5.25-point between-group difference converts to roughly 2.6 raw points. That sits below both of Pool's thresholds. On pain, Pool reported an NRS minimal detectable change of 4.3 points against Suvarnnato's 0.63.

The trial's own within-group movement was larger: the deep flexor arm entered at a mean disability score of 27.3% and finished six weeks at an adjusted 15.0%. The usual-care arm also improved. What the exercise adds over ordinary physiotherapy is statistically real and clinically small.

Whether the posture change is the mechanism is a separate question, and the correlation evidence is weaker than the enthusiasm around it. Yip, Chiu and Poon compared 62 people with neck pain against 52 without in Manual Therapy in 2008. The craniovertebral angle was smaller in the neck pain group, and correlated negatively with disability at r=−0.31 (p=0.015). Once age was accounted for, the correlation with pain intensity lost significance entirely (r=−0.185, p=0.154).

What has to hold: the adherence figure decides the result

The most direct answer comes from a 2025 randomised trial in Musculoskeletal Science and Practice. Villanueva-Ruiz and colleagues gave 65 people with chronic non-specific neck pain four weekly sessions of either manual therapy or tailored neck-specific exercise, and followed them for 12 weeks after treatment ended. Manual therapy produced more responders at every measured point — odds ratios of 0.14 at two weeks, 0.31 at completion and 0.19 at 12 weeks, all favouring manual therapy.

Then the authors reran the analysis on participants whose exercise adherence reached 95% or better. Every difference disappeared, and the exercise group's craniocervical flexion performance overtook manual therapy. Their conclusion reads: "Manual therapy may only be superior to specific-exercise when high exercise adherence cannot be assured." Among the paper's authors is Deborah Falla, Chair in Rehabilitation Science and Physiotherapy at the University of Birmingham and director of its Centre of Precision Rehabilitation for Spinal Pain.

Ninety-five per cent is a demanding number. Johnston and colleagues measured what adherence looks like without unusual care, in a 12-month cluster-randomised trial of 740 office workers across 14 Brisbane organisations published in BMC Musculoskeletal Disorders in 2021. Participants in the exercise arm — three supervised 20-minute sessions a week for 12 weeks, opening each session with an upper neck flexion warm-up — attended a mean 67.4% of their 13 sessions. Neck pain fell further in that arm than in the health-promotion comparison at 12 weeks: −0.53 points versus −0.17 across all workers (p=0.019), and −2.32 versus −1.75 among those who started with pain of 3 or more (p=0.04).

At 12 months, the gap was gone. The exercise arm sat at −0.18 and the comparison arm at −0.14 (p=0.535). This was not simple abandonment: 62.5% of the exercise group reported still training regularly, against 15.4% of the comparison group. The authors wrote that their result suggests "the potential need for exercise interventions to be long-term or continuous to maintain benefits."

Sources disagree about where this leaves the exercise. Price and colleagues, reviewing 26 trials covering 2,288 participants and 15 exercise programmes, concluded that motor control plus segmental exercises are the most effective option for short-term pain and disability — and stated flatly that no high-quality trial has investigated long-term outcomes at all. Villanueva-Ruiz's group, working with a directly comparable population, put manual therapy ahead unless adherence is near-perfect. Both conclusions can be true; neither has been tested past a year.

Deep neck-flexor rehabilitation versus the double-chin claim

These are separate interventions aimed at separate tissue, and the search queries that combine them are combining two literatures that never overlap.

| | Craniocervical flexion (supine chin tuck) | Chin tuck against resistance (CTAR) | Deoxycholic acid injection | |---|---|---|---| | Tissue targeted | Longus capitis and longus colli, on the front of the cervical vertebrae | Suprahyoid muscles, under the jaw | Subcutaneous submental fat | | Clinical field | Neck pain rehabilitation | Swallowing rehabilitation | Aesthetic dermatology | | Outcomes measured in trials | Disability, pain, craniovertebral angle, flexor strength | Tongue pressure, dysphagia scales, muscle activation | Clinician and patient fat grading, MRI fat volume | | Effect on submental fat | Not measured in any trial cited here | Not measured | 43% achieved ≥10% volume reduction versus 5% on placebo |

No exercise trial has demonstrated that training a muscle reduces the fat lying over it. Vispute and colleagues tested the proposition directly in the Journal of Strength and Conditioning Research in 2011: 24 sedentary adults, seven abdominal exercises, two sets of 10 repetitions, five days a week for six weeks, on an isocaloric diet. Curl-up endurance rose from 32 ± 9 repetitions in controls to 47 ± 13 in the training group. Body fat percentage, android fat, abdominal circumference and skinfold measurements did not move.

The US Food and Drug Administration's prescribing information for deoxycholic acid injection — approved for "improvement in the appearance of moderate to severe convexity or fullness associated with submental fat in adults" — describes two trials with 514 treated subjects and 508 on placebo. A composite clinician-and-patient one-grade improvement occurred in 70.0% and 66.5% of treated subjects across the two trials, against 18.6% and 22.2% on placebo. In an MRI subgroup of 449, 43% of treated subjects reached at least a 10% reduction in submental fat volume, against 5% on placebo.

The screening that comes before the first repetition

Suvarnnato's team excluded three of the 57 people they screened: two for positive neurological signs and one for severe neck pain arising from spinal infection. Their exclusion criteria also removed anyone with diagnosed cervical radiculopathy or myelopathy, a history of cervical or thoracic fracture, dislocation or surgery, spinal osteoporosis, fibromyalgia, or a whiplash or head injury. Those groups were never in the trial, so it says nothing about them.

Two published test clusters describe what clinicians look for. Cook and colleagues, analysing 249 patients in the Journal of Manual and Manipulative Therapy in 2010, identified five findings for cervical myelopathy — gait deviation, a positive Hoffmann's test, an inverted supinator sign, a positive Babinski test, and age over 45. None of the five present gives a negative likelihood ratio of 0.18; three of five gives a positive likelihood ratio of 30.9. For nerve root involvement, Wainner and colleagues reported in Spine in 2003, from 82 patients, a four-item cluster with a positive likelihood ratio of 30.3. The American Physical Therapy Association's 2017 neck pain guideline, written by Blanpied and colleagues, puts screening for serious pathology ahead of any exercise prescription.

Frequently asked questions

Will chin tucks improve the jawline?

No trial has measured jawline appearance after chin tucks. The supine version trains the longus capitis and longus colli, which lie behind the throat against the cervical vertebrae, not under the chin skin. Trials of the exercise measure neck disability, pain and craniovertebral angle. Appearance has never been an endpoint.

Is it okay to do chin tucks every day?

Daily practice is what the trials used. Suvarnnato's 2019 protocol prescribed twice-daily home repetitions across six weeks, alongside 12 supervised sessions. Price and colleagues concluded in 2020 that increased frequency of motor control exercise may improve effectiveness. The load is low, with no external resistance, which is why the frequency is tolerable.

When doing chin tucks, where should you feel it?

Deep at the front of the neck, not along the rope-like sternocleidomastoid running from behind the ear to the collarbone. Falla, Jull and Hodges confirmed by direct electrode recording in 2004 that the deep flexors carry the movement. In Suvarnnato's trial, a researcher palpated the sternocleidomastoid to catch superficial substitution.

Do chin tucks work for a double chin?

No published evidence supports it. Vispute and colleagues found in 2011 that six weeks of abdominal training raised muscular endurance without changing any measure of overlying fat. Submental fullness is subcutaneous fat, which is why the FDA-approved treatment is an injected fat-destroying drug rather than an exercise.

Are chin tucks dangerous?

The exercise is low-load and performed lying down, and the neck pain trials report it as tolerated. Safety in specific groups is untested, because those groups were excluded: Suvarnnato's team screened out people with radiculopathy, myelopathy, prior cervical fracture, dislocation or surgery, spinal infection, osteoporosis, or whiplash history.

Which neck symptoms should a clinician check before you start chin tucks?

Cook and colleagues list gait deviation, a positive Hoffmann's test, an inverted supinator sign and a positive Babinski test as myelopathy findings. Arm pain with weakness, numbness or reflex loss points toward nerve root involvement, per Wainner's cluster. The APTA's 2017 guideline places this screening before exercise prescription.