Military and Aviation Neck Research

Why air forces studied this and nobody else did

A fighter pilot wears a helmet weighing several kilograms, adds night vision equipment to the front of it, and then pulls sustained G, multiplying the effective mass of the whole assembly. Then they turn their head to look behind them.

The result is neck pain at rates that would be considered an epidemic in any other occupation:

The extremes of the fighter pilot range are both real studies: 83% among 58 Danish F-16 pilots, 18.9% among 90 Belgian F-16 pilots. The spread reflects genuine differences in airframe, mission profile, and how the question was asked.

Tang L, Zhang Y-H, Du S-H, Wang X-Q. Prevalence and related factors for neck pain in military personnel: a systematic review. EFORT Open Reviews, 2024. PMID 39087493; PMCID PMC11370721.

Operational readiness depends on those people being able to turn their heads. That is why this research got funded, and why it is more rigorous than anything the fitness industry has produced on the same subject.

What a neck actually does under load

The foundational paper is a 2004 review from the US Air Force School of Aerospace Medicine, and it produced two things that matter to anyone training a neck.

The angular thresholds. Joint reaction forces at the C7–T1 junction do not rise smoothly with angle. They rise slowly, then steeply, with inflection points beyond roughly 15° of flexion, 30° of extension, and 35° of axial rotation. “Checking six” — looking up and behind — is named the highest-risk head movement under load.

The intensity figures. Peak lateral neck muscle force in flight ranged from 50% to 257% of maximum voluntary contraction, mean 84.8%. The pilot who reached 257% sustained an acute injury that interrupted the flight. Cervical erector spinae at 4 G with rotation ranged from 28.2% to 189.7% MVC, mean 79.5%.

Values above 100% MVC are not measurement errors. A muscle resisting an external load eccentrically produces more force than it can generate voluntarily in a static test.

Coakwell MR, Bloswick DS, Moser R Jr. High-Risk Head and Neck Movements at High G and Interventions to Reduce Associated Neck Injury. USAF School of Aerospace Medicine / University of Utah, 2004.

The same paper made a recommendation that has aged well — that training should be both “specific,” meaning resistance applied to the movements the job actually demands, and “intensive,” meaning “multiple sets with multiple repetitions against high resistance levels,” with particular emphasis on the check-six position.

And it made an admission: as of 2004, no prospective controlled trial evaluating any intervention in fighter pilots had been conducted.

The twenty-year arc

That gap has since been filled, and the sequence makes a clean narrative.

2004 — Coakwell. Defines the problem, quantifies the loading, states that no controlled trial exists.

2004 — Alricsson. The most under-cited study in this literature, and the one with the most uncomfortable finding. Forty Swedish Air Force fighter pilots, all given the same programme — three sessions a week, 4 × 10 repetitions against progressive weights of 1, 2 and 4 kg, plus stretching and rubber-tube work — for six to eight months. Twenty received weekly supervision and encouragement. Twenty did not.

Alricsson M, Harms-Ringdahl K, Larsson B, Linder J, Werner S. Neck muscle strength and endurance in fighter pilots: effects of a supervised training program. Aviation, Space, and Environmental Medicine, 2004;75(1):23–28. PMID 14736129.

Read the right-hand column again. The unsupervised group had the programme and got substantially weaker anyway — extensor strength down 11.5 Nm, more than twice the magnitude of the supervised group’s gain in the opposite direction.

That is the strongest statement anywhere in this literature of something the site keeps returning to: the programme is not the hard part. Being seen doing it is.

The null, published here because it belongs here: neither group showed a measurable reduction in the frequency of neck complaints. Strength moved. Symptoms did not, over six to eight months, in forty people.

2007 — Netto. Matches gym exercise intensity to in-flight demand by electromyography. Establishes that elastic band resistance of any grade produces roughly 15% MVC — the equivalent of one G — while a machine at 50% of a three-repetition maximum reproduces the three-to-five-G range, and 70% exceeds it.

2020 — Murray. Twenty weeks, 108 Royal Danish Air Force helicopter aircrew, randomised controlled trial. Cervical extension strength rose in the training group and fell in the reference group (p=0.018). Rate of torque development likewise. And the study’s most useful result was accidental: only 29% of participants trained regularly, and those who did roughly tripled the headline effect.

This is also the one study in the entire neck-training literature that names its harness by brand and manufacturer — “The Original Neck Flex® Head Harness, Gonzo Companies, USA” — which makes it the only piece of consumer neck equipment with independent third-party documentation of this kind. What that establishes, and the several things it does not, are set out on Neck Flex in the published research. Gonza LLC, formerly The Gonzo Companies, publishes this site; see the disclosure.

2021 — Rausch. Twelve weeks, high-performance aircraft personnel, with MRI. Isometric strength up across all directions, muscle volume up 6.6–8.3%, and — the operationally critical finding — the muscle activity required to hold the head up while wearing the helmet fell significantly in the trained group and not in controls (p=0.01, and p<0.01 with night vision goggles added).

2024 — Chayrez. Eight weeks, USAF student fighter pilots, ten to fifteen minutes three times a week. Cervical endurance hold improved 33.6% overall, and 55.7% at one of the two bases (effect size 1.97).

2026 — Luo. Twelve weeks, aviation cadets, the only clean head-to-head between weighted headgear and elastic bands. Both improved to week six. Only the weighted group kept improving.

Twenty-two years from “we don’t know whether anything works” to a set of protocols with measured outcomes. That is unusually fast for applied exercise science, and it happened because someone had an operational reason to care.

The synthesis, as of 2025

The field now has its own scoping review — 54 peer-reviewed studies of neck training in military aviation personnel, covering isometric and dynamic resistance training, multimodal physiotherapy, machine and frame-based equipment, weighted helmets, trampoline work, virtual reality kinematic training, and deep neck muscle protocols.

Bruno J, Montoro-Bombú R, Sarmento H. Neck training in military pilots: A scoping review. International Journal of Sports Science & Coaching, 2026;21(1). Published online 25 October 2025. DOI 10.1177/17479541251388819.

Two conclusions matter, and they pull in different directions.

Targeted training protocols can reduce the incidence of neck pain, and improve cervical function. That is a stronger statement about symptoms than any individual study on this page supports, and it comes from pooling across 54 of them.

“No single approach emerged as universally superior.” The reviewers found integrated programmes — strength plus endurance plus coordination — worked best, and could not rank the modalities against each other. Which is the same conclusion this site reaches from the device side: no head-to-head equipment trial exists, and anyone claiming their category is proven better is describing a study nobody has run.

The gaps the reviewers name are worth recording, because they define what this whole literature still cannot tell you: sample sizes of 6 to 90, populations that are overwhelmingly male with almost no female pilots, durations of only six to twelve weeks, almost no long-term follow-up, inconsistent reporting of the actual G exposure and helmet weight the trainees were subject to, low adherence to self-administered protocols, and very few placebo-controlled designs.

Note how neatly that list converges with Alricsson’s 2004 finding twenty-one years earlier. Low adherence to self-administered protocols is still on the list of things holding the field back. It was the finding in 2004, it was the accidental finding in Murray 2020, and it is a named limitation in the 2025 synthesis.

The NATO task group, and what this archive has and has not read

NATO ran a multi-nation research task group on exactly this problem — HFM-252, Aircrew Neck Pain Prevention and Management, chaired by Dr Philip Farrell of Defence Research and Development Canada, reporting around 2017–18. Its final report is the most authoritative multi-nation synthesis in existence on this subject.

This archive has not read it. The full text is openly published at publications.sto.nato.int, and both that PDF and the NATO STO document page have returned 403 to every retrieval attempt. That is a retrieval failure, not a paywall, and it is recorded here rather than glossed over — see how studies are selected.

What can be reported is DRDC’s own public account of the group’s work, which is a different and weaker source than the report:

• Neck pain affects roughly 75% of CH-146 Griffon aircrew in the Canadian Armed Forces.
• The group evaluated ten candidate mitigations, including task posture and distribution, scheduling, helmet fit, exercise and education programmes, lower-profile helmets, helmet support devices, control and display concepts, and passive and active seat cushioning.
• Farrell’s summary: “there is no single silver bullet that will solve this problem, but rather carefully planned integration of two or more of these mitigating solutions will likely reduce the risk.”

Defence Research and Development Canada, Neck pain research yields solutions, canada.ca. Public communication about the task group, not the HFM-252 report itself.

That conclusion is worth sitting with, because an independent synthesis reached the same one eight years later: Bruno and colleagues found “no single approach emerged as universally superior” and recommended integrated programmes. A NATO task group looking at helmets, seats, scheduling and exercise, and a 2025 scoping review looking only at training protocols, converge on the same shape of answer.

Which is also the honest answer to give a buyer. Exercise is one mitigation among several, and it is the one an individual controls. It is not the whole solution and this site will not present it as one.

What transfers, and what doesn’t

The single most useful distinction in this literature is between maximum strength and relative effort under operational load.

Rausch’s group is the only one that measured both. Isometric strength gains were real but modest — extension +6.8%, flexion +17.7%, rotation ±22.9%. The change that actually mattered operationally was different: after training, holding the head up under helmet load consumed a significantly smaller proportion of the pilot’s available capacity.

That is the mechanism by which neck training helps anyone. Not that you can generate more force, but that the force your daily task demands becomes a smaller fraction of what you have. It generalises well beyond aviation — to the rugby player in a scrum, the rider in a helmet, and the person whose head is forward over a keyboard for eight hours.

Where rotary-wing work adds something different

A US Army Aeromedical Research Laboratory study instrumented UH-60 helicopter aviators and measured how much time they spent in severe cervical postures. Defining severe as flexion beyond 30°, twist beyond ±30°, and lateral bend beyond ±40°, the results were:

Williams ST, Madison AM, Chancey VC. Defining Normal Cervical Spine Range of Motion in Rotary-Wing Pilots (Part 2). USAARL-TECH-FR–2022-34, 2022.

Roughly a third of flight time in severe cervical rotation. It is a useful corrective to training programmes built almost entirely around flexion and extension, which is most of them.

Still outstanding

Three documents that would materially improve this page have not been obtained.

ADA571450, A Viper Pilot Neck Health and Conditioning Guide — an actual prescribed F-16 conditioning programme. Repeated retrieval attempts return HTTP 403.

NATO STO-TR-HFM-252, Aircrew Neck Pain Prevention and Management — likely the most authoritative multi-nation synthesis in existence on this subject. Also inaccessible.

Alricsson et al. 2004 (PMID 14736129), on a supervised training programme in Swedish Air Force fighter pilots — the classic study in this area, and a real gap.

These are listed rather than quietly omitted. A reference work should be explicit about the shape of its own holes.