Dead bug exercise: it is one of the most prescribed movements when back problems come up, and one of the most overrated in the promises that accompany it. It is presented as the exercise that "reactivates the deep core" and protects the spine. The literature tells a more sober story, and it is worth knowing before devoting months to the mat.
What the dead bug really teaches is a concrete skill: moving the arms and legs while the trunk stays still. It is an exercise with very low load, easy to grade, which almost everyone tolerates, including those who are afraid of moving their back. What has not been demonstrated is that it is superior to other ways of training, or that it works by correcting a specific defect of the deep muscles. In this guide you will find the technique, the mistakes that cancel out its purpose, the easier and advanced variations, and the honest distinction between what we know and what is told.
⚠ Important: read this before continuing
The exercises and advice in this article are general indications and may not be suitable for your situation. Every person has a unique clinical history: what works for others could be counterproductive for you.
Want a professional's opinion? Start from the questionnaire: you sign up and see who will follow you, a physiotherapist on the Italian professional register chosen for your case, whom you write to in chat. If you would rather hear their voice, the first consultation is free and we look at your situation together.
Before lying down on the mat: the signs that are not to be trained
The vast majority of back pain is benign and improves with movement. A minority is not, and no exercise resolves it. Before starting, check that you do not have any of these signs: they are the reasons to call a doctor instead of looking for an exercise.
- Loss of bladder or bowel control, difficulty urinating, or numbness in the area in contact with the saddle (perineum, inner thighs): this is a picture that requires urgent medical assessment, not a postponement until tomorrow.
- Real weakness in a leg or a foot that worsens over the days: a foot that flops, a knee that gives way, difficulty climbing stairs.
- Night pain that does not ease in any position, or that systematically wakes you.
- Fever, chills, general malaise together with back pain.
- Unexplained weight loss, or a personal history of cancer.
- Recent trauma (a fall, an accident) or known osteoporosis with pain that appeared suddenly.
In all other cases, which is the majority, moving is the right path. The American College of Physicians guidelines for chronic low back pain recommend starting with non-drug treatments, and place exercise at the top of the list together with approaches such as spinal manipulation, yoga, tai chi and cognitive behavioural therapy.
What the research really says
Here is the point that almost no article on this exercise addresses.
Motor control exercises work, but not better than the others. The Cochrane review on motor control exercises, the family to which the dead bug belongs, gathered 29 randomised studies and 2,431 participants with chronic non-specific low back pain. The verdict: compared with a minimal intervention the benefit exists and is clinically relevant (in the long term, a mean difference in pain of -12.97 points; 95% CI from -18.51 to -7.42), on evidence the authors judge to be of low to moderate quality. Compared with other forms of exercise, by contrast, no clinically important difference emerges, at any point in the follow-up. For completeness: the same review finds similar results compared with manual therapy (evidence of moderate to high quality), and an advantage in the comparison with "exercise plus instrumental physical therapies" (short-term pain, mean difference -30.18), but this last figure rests on evidence of very low to low quality. The authors' conclusion is explicit: since motor control is not superior to other forms of exercise, the choice should depend on the preferences of the patient and the therapist, the training of the latter, costs and safety.
A second meta-analysis reaches the same place. On stabilisation exercises: 29 studies included, of which 22 with pain data (2,258 patients) and 24 on disability (2,359). Compared with any alternative treatment or control, the long-term advantage exists statistically, pain -6.39 and disability -3.92 points on a 0-100 scale, but the difference between the groups is clinically insignificant. And in the direct comparison with other forms of active exercise, no difference remains, either statistical or clinical.
A third, on 5 studies and 414 participants, finds an advantage of core stability work over generic exercise only in the short term (mean difference in pain -1.29; in disability -7.14). On pain that advantage had already faded at 6 and at 12 months; on long-term disability the study provides no data.
One clarification that changes how all these numbers are read. None of these reviews studied the dead bug. They studied programmes of exercise, usually individualised, progressive, followed by a physiotherapist, of which the dead bug is at most one ingredient among many. There is no literature measuring this single movement done alone on your home mat. When it is read here that "it works but not more than the others", what is referred to is the family to which it belongs, not a trial on it.
And the "switched-off deep core"? It is the most widespread story: low back pain would arise from a transversus abdominis that does not switch on in time, and the exercise would work because it switches it back on. A suggestive model, which however the evidence does not confirm. In a randomised study of 109 people with chronic non-specific low back pain, eight weeks of exercises (low load for stability, high load in suspension, or generic) produced minimal or no changes in the activation timing of the deep abdominal muscles, measured with ultrasound in response to a rapid arm movement. The only difference between the groups was small (15 and 19 milliseconds), concerned one side only and favoured the high-load suspension exercises, not the low-load ones designed for the transversus. Above all: there was no association between the change in pain and the change in those activation times. If that were the mechanism, the exercises built specifically for it should beat the others: they do not.
What this means in practice. The dead bug remains a good exercise and it makes sense to keep proposing it. But if someone presents it to you as the exercise that protects the back, they are promising more than is known. The evidence says that what shifts the results is the fact of training, not which movement is chosen. Which, paradoxically, is good news: you can choose the exercise that suits you, that does not hurt and that you will be able to maintain for months. If that is the dead bug, excellent.
And prevention? Here the evidence is more solid, but it concerns exercise in general. A meta-analysis of 21 randomised studies and 30,850 people showed that exercise combined with education reduces the risk of an episode of low back pain (RR 0.55; 95% CI 0.41-0.74), with evidence of moderate quality; exercise alone shows an effect in the same direction, but on evidence of low to very low quality. In the same review, by contrast, lumbar belts and insoles are not found to be protective. No study isolates the dead bug.
Why it is called that
The name comes from the position: lying on your back with arms and legs in the air, you resemble an upturned insect that has lost its balance. It is a funny image that has helped the spread of the exercise more than any scientific article.
The name should not, however, make you think it is something trivial. The dead bug exercise entered spinal rehabilitation protocols for a very concrete reason: it is one of the few control tasks that can be offered to someone who is afraid of moving their back, because it is performed lying down and the spine does not have to bear the weight of the body.
What it trains, in practice
Beyond the promises, the dead bug exercise has precise technical features that make it useful in a programme.
Resisting extension. When you lower the arm and leg, gravity pulls the pelvis forward and tends to arch the lower back. The task is to prevent it. It is a control task, not a maximal strength task: that is why the effort is felt in an odd way, more as "holding" than as exertion.
Dissociating limbs and trunk. Moving the legs without the pelvis following them is a skill needed in walking, running, climbing stairs and picking something up from the floor. The dead bug trains it in very simplified conditions.
Resisting rotation. The right arm and left leg move together: the pair that is created tends to rotate the trunk, and the obliques work to prevent it.
Crossed pattern. It is the same crossed scheme as walking, so a familiar context for the nervous system.
Very low load. In the supine position the spine does not bear the weight of the body. It is the most concrete reason why it is used early in rehabilitation, and why it can be repeated often without tiring.
On how much it activates the muscles, the only direct figure is modest and should be reported for what it is: a study with surface electromyography on 10 healthy male volunteers compared four lumbar stabilisation exercises (curl-up, dead bug, superman and bird dog), each graded on five intensity levels. The general result is that increasing the level increases the activity of the trunk muscles; the significant peaks reported by the authors, however, concern the curl-up (rectus abdominis and external oblique) and the bird dog (multifidus and erector spinae), not the dead bug in particular. Note also a technical limit: surface electromyography does not reach the transversus abdominis, which is the muscle most often discussed and which that study did not measure.
Muscles involved
The dead bug isolates nothing: it distributes light work across many areas that must collaborate.
An important point: before trying these exercises, make sure you have received a correct diagnosis. People often think they have one problem, but the real cause of the pain is another. Doing the wrong exercises can make the situation worse instead of better.
Abdominal wall: rectus abdominis, internal and external obliques, transversus abdominis. They work in a stabilising way, that is to prevent a movement, not to produce it.
Hip flexors: iliopsoas and rectus femoris control the leg on the way down and bring it back up. They are also the reason why, if pelvic control is missing, the pull on the spine is felt straight away.
Scapular muscles and deltoid: they accompany the arm that goes down behind the head and keep the shoulder in position.
Back musculature of the trunk: erector spinae and multifidus take part in maintaining posture, in a task that in the dead bug is mainly anti-extension.
Correct technique step by step
Starting position. Lying on your back on a mat. Arms vertical towards the ceiling, perpendicular to the floor. Legs raised with hips and knees bent to 90 degrees, as if you were sitting on an upturned chair.
Back position. The most widespread convention is to slightly flatten the lower back against the floor, reducing the space between the back and the mat. It serves mainly as a reference: it gives you a clear and verifiable signal of when you are losing control. It is not a dogma: there are those who work keeping the natural lumbar curve and who do just as well, but for those starting out contact with the floor is the simplest way to notice mistakes without a mirror.
Preparation. Lightly activate the abdominal wall, as if you were about to receive a push on your stomach. "Lightly" is the keyword: a maximal contraction stiffens everything and stops you breathing.
Movement. Slowly lower the right arm behind your head and at the same time extend the left leg towards the floor, without resting it. The opposite arm and leg move together.
Depth. Lower only as far as you can maintain the chosen position. As soon as the lower back starts to lift off the floor, you have reached your limit: that is the stopping point, and it is individual. There is no prize for getting lower.
Return. Bring the arm and leg back to the starting position with the same control with which you lowered them. The way back up is where most people lose the position.
Alternation. Repeat on the other side: left arm, right leg.
Breathing. Breathe out as you lower, breathe in as you rise. If you could not talk during the set, you are gripping too hard.
Common mistakes
The back arching. It is the mistake that empties the exercise: if the lower back lifts off the floor, you are no longer controlling anything, you are only moving the limbs. Solution: reduce the range until control holds.
Going fast. Momentum does the work instead of the muscles. Slowing down is the simplest way to make the exercise harder without changing anything else.
Holding the breath. Apnoea is the shortcut by which the body creates stiffness without control. If the set works only by holding your breath, it is too difficult for now.
The ribs lifting. When the arm goes behind the head, the lower ribs tend to rise and the rib cage opens: it is the same movement as arching, seen from above. Keep them "anchored" downwards.
Tension in the neck. If you feel the back of the neck tight, rest the head on a thin pillow and keep the jaw relaxed.
Persisting through pain. A mild discomfort that does not worsen during the set and disappears afterwards is acceptable. A pain that grows repetition after repetition, or that leaves you worse the next day, is a signal to stop, not an obstacle to overcome.
Easier variations
If the full version makes you lose position, it falls apart. This is not a fallback: it is the correct way to start.
Feet on the floor. Knees bent, feet resting, only the arms move. It is the starting point for those who cannot keep the legs raised without arching.
Arms only. Legs still at 90 degrees, you lower only the arms, alternating them. Minimal demand on pelvic control.
Legs only. Arms still (towards the ceiling or along the sides), you extend one leg at a time. It is more demanding than the arms-only version, because the leverage of the leg is longer.
Reduced range. The full movement, but with a very small amplitude. It grows as control holds for longer.
Advanced variations
When the basic version is clean and boring, the bar is raised.
Isometric hold. In the low position, stay still for a few seconds before returning.
Straight leg. Instead of the bent knee, the leg lowers straight: the leverage lengthens and the anti-extension demand grows considerably. It is a jump in difficulty, not a detail.
With a resistance band. A mini band between hands and knees, kept under tension while you move: it adds a constant demand for counter-pressure.
With a ball. Opposite hand and knee press on a ball and stay still, while the other two limbs move. The pressure gives an immediate reference when the position is lost.
With a light load. A small weight in the hand that goes down increases the moment in extension. To be introduced only when the bodyweight technique is impeccable.
Dead bug or bird dog?
They are the two exercises that are always mentioned together, and in fact they complement each other.
Position. The dead bug is performed supine, the bird dog on all fours. Those with problems in the wrists or knees often tolerate the first better.
Direction of the disturbance. In both the main task is to resist lumbar extension, but gravity acts from different directions and the bird dog adds a balance component that the dead bug does not have, because lying down you cannot fall.
Feedback. The dead bug has a teaching advantage: the floor under your back tells you by itself when you are making a mistake. In the bird dog that reference is missing and a mirror, a phone placed on the back or an external eye is needed.
Which to choose. No evidence indicates that one is superior to the other. Using both is reasonable; using the one you can do well is more important than the choice itself.
How to fit it into the programme
In the warm-up. A few sets per side before training: it prepares the trunk without tiring it.
In a stability circuit. The dead bug exercise pairs well with the bird dog, the plank and side-supported exercises, which demand control in different directions.
As a maintenance routine. If you have a history of recurring episodes, a short and regular block is more sustainable than a long session now and then, and on prevention the evidence concerns exercise programmes, not single movements.
As active recovery. Between sets of heavy exercises, as an active break at minimal load.
How much and how often
It must be said honestly: there is no demonstrated optimal dose for this exercise. The formulas that circulate, three sets of ten every day, are practical conventions, not results of studies. What can reasonably be said:
- The load is low, so most people tolerate it even on close days.
- A few repetitions done well are worth more than many done badly: when the back starts to lift off the floor, the set is over, regardless of the number you had in mind.
- Slowing down is the simplest way to progress, even before adding bands or weights.
- The criterion for advancing is quality, not counting.
Particular situations
Chronic low back pain. It is the most common use and it makes sense: low load, easy gradation, good tolerability. But it should be placed within a broader programme of exercises for chronic low back pain, not used as the only exercise, and keeping in mind the starting point: the advantage over other forms of active exercise is not demonstrated.
Postpartum and abdominal diastasis. The dead bug exercise is often proposed at this stage, and that is understandable: it is low impact and can be graded very finely. On effectiveness, however, precision is needed. A systematic review with meta-analysis of 7 randomised studies and 381 women concluded that the evidence supporting specific exercise programmes to treat abdominal diastasis after childbirth is of very low quality: work on the transversus abdominis, on the only two studies that provided pooled data and compared with a minimal intervention, reduces the distance between the rectus muscles to a modest extent (mean difference -0.63 cm; 95% CI from -1.25 to -0.01), again with evidence of very low quality. Here too the dead bug as such has never been studied. Translated: it can be done, but nobody can promise that it will close the diastasis, and the path should be agreed with whoever is following you. And if alongside the exercises you were advised a support belt, the direct comparison does not favour it: how much the postpartum belt really holds.
After a spinal operation. It is one of the first control exercises reintroduced, precisely because of its minimal load. The timing and the modalities, however, are decided by whoever operated on you, not by an article.
For those doing physical work or sport. The realistic expectation is that regular exercise reduces the risk of episodes of low back pain, not that a single movement "shields" the back. The difference between the two sentences is the whole difference between an expectation that holds and one that sooner or later disappoints.
If after a few weeks nothing changes
It happens, and it is useful information, not a failure. If the pain does not move despite regular and well-performed work, the reasonable hypotheses are three: the exercise is too easy and needs progressing; the overall load of your day (work, sleep, stress, other activities) weighs more than the exercise session; or the starting problem is not the one you thought. In all three cases the answer is not to add repetitions, but to have someone look at the situation.
Telerehabilitation: why it can help
The limit of the dead bug done alone is that the mistakes that matter (the lower back lifting, the ribs rising, speed increasing when you get tired) are exactly the ones that cannot be seen from the inside.
With Ri-Hub you can consult qualified physiotherapists from home, for 29.90 euros per session. The physiotherapist watches how you perform the movement, corrects in real time, chooses the variation suited to your level and, if needed, tells you that for your case there is a more sensible exercise than this one.
Conclusion
The dead bug is a well-built exercise: low load, easy progression, natural feedback from the floor and a control demand that almost everyone can take on. These are excellent reasons to keep it in a programme.
It is not, as far as we know today, the exercise that shields the back, nor the correction of a deep defect. Systematic reviews say that exercises of this family beat a minimal intervention and are on a par with other forms of movement. That shifts the right question from "which is the best exercise" to "which exercise will I be able to do for the next six months". If the answer is the dead bug, you have made a good choice, and now you also know why.
Sources
- Motor control exercise for chronic non-specific low-back pain: Cochrane Database of Systematic Reviews, 2016
- An update of stabilisation exercises for low back pain: a systematic review with meta-analysis: BMC Musculoskeletal Disorders, 2014
- A meta-analysis of core stability exercise versus general exercise for chronic low back pain: PLoS One, 2012
- Prevention of Low Back Pain: A Systematic Review and Meta-analysis: JAMA Internal Medicine, 2016
- Effect of core stability exercises on feed-forward activation of deep abdominal muscles in chronic low back pain: a randomized controlled trial: Spine, 2012
- Electromyographic Changes in Trunk Muscles During Graded Lumbar Stabilization Exercises: PM&R, 2016
- What is the evidence for abdominal and pelvic floor muscle training to treat diastasis recti abdominis postpartum? A systematic review with meta-analysis: Brazilian Journal of Physical Therapy, 2021
- Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: A Clinical Practice Guideline From the American College of Physicians: Annals of Internal Medicine, 2017
Last source check: August 2026.
You might also be interested in:
Lumbar X-ray for Back Pain: When It Is Needed and What It Shows
Hip Pain: All the Causes, How to Recognise Them and How to Treat Them
Stay in touch with us
Join our dedicated Facebook group: we reply to all patients, give practical advice and clear up the most common doubts.
Need a consultation?
Start from the questionnaire on Ri-Hub: a few questions about your case, you sign up, and you see who will follow you: a physiotherapist on the Italian professional register, whom you write to in chat whenever you like. If you prefer to talk it through by voice, the first consultation is free, lasts twenty minutes and commits you to nothing: they arrive having already read what you wrote, and together you decide whether and how to continue.
