Lower back pain when doing planks: the sag, the hip flexors, holding too long and what helps

What a plank asks of the back, and what no study has measuredEmerging evidence

A plank holds your trunk still while gravity tries to sag it towards the floor. In 18 adults planking on their toes, the six-pack and the side tummy muscles (the obliques) did most of the holding, at 34% to 40% of their maximum, the back muscles very little (about 5%), and the rectus femoris, a hip flexor that crosses the front of the hip and can tip the pelvis forward, about 20%. On the knees, every one of these muscles worked less, the hip flexor at 7%. In 30 adults no muscle passed 45% of its maximum in a plank, so it suits endurance more than strength.

Long holds usually end at the arms and legs: in 471 students holding a forearm plank as long as they could, the legs and the arms each ended about a third of holds, the abs about a fifth and the back about 6%. As the trunk tires, the hips can sink. In the only study we found that measured the spine during a plank, in 11 teenage basketball players over a minute, the lower back curve flattened a little in some and increased markedly in others. That growing arch is the usual reason a sore back complains (our reading: no study has linked it to pain).

No study has counted back pain caused by planks, and no study we opened measured the compression or shear on the lower spine in a front plank: the abstracts of the two spinal load studies of core exercises we found give no figure for it. One analysis found the same bending force on the lower back in a plank as in a v-sit, with the lower spine 28 degrees less bent in the plank. So this guide's map is built from muscle activity studies, plank tests, trials of exercise for back pain and NHS pages.

Sources 1, 2, 3, 4, 5, 6, 7

The usual cause: ordinary back pain that dislikes the archStrong evidence

For nearly all people with low back pain, no single cause can be found; only a small proportion have a well understood cause such as a broken bone, cancer or an infection. The NHS says back pain is very common and usually improves within a few weeks. An ache across the lower back that builds through a hold and eases when you drop to your knees is usually this kind of pain, showing up in a position it dislikes. Change the plank rather than stop moving: advice to stay active did slightly better than bed rest (2 trials, 401 people), and the NHS says don't stay in bed for long periods.

Sources 8, 9, 10, 11

A young athlete's back, and the facet jointsEmerging evidence

In young athletes, back pain is more often a stress fracture of the spine (spondylolysis) than in adults: 47 of 100 adolescent athletes with low back pain in one clinic had one, against 5 of 100 adults, and so did 30% of 1,025 non-elite adolescent athletes in another. Rates were highest in diving, cricket, baseball, rugby and weightlifting, and most young athletes return to sport without surgery. In adults, the same defect showed on computed tomography (CT) scans in 11.5% of 188 people and wasn't linked with back pain.

From mid-life, an ache worse leaning back and eased by bending forward is often put down to the facet joints. In 176 people with long-lasting back pain, 47% responded to a screening injection into these joints but only 15% to a confirming one, and no clinical feature picked them out; a review of 62 studies found only one specialist scan modestly helpful. No clinical test can confirm the facet joints.

Sources 12, 13, 14, 15, 16, 17

Nerves, discs and scansStrong evidence

Sciatica usually hurts in your bottom and the back of one leg, and the NHS says the back pain with it is not usually as bad as the leg pain; even if moving hurts, it's not harmful. Scans of people with no back pain show worn discs in 37% at 20 and 96% at 80, and disc bulges in 30% and 84%. The National Institute for Health and Care Excellence (NICE) says don't routinely offer imaging for low back pain outside specialist care, and in specialist care only if the result is likely to change management. Scans do matter for the warning signs and for a young athlete's suspected stress fracture.

Sources 18, 19, 20

The tummy wall: hernias, and a separation after pregnancyEmerging evidence

A lump in the groin, at the top of the thigh or near the belly button that is bigger when you strain and smaller lying down can be a hernia: the NHS says see a GP if you think you have one, and the warning signs say when it is urgent. In 14 students, the pressure inside the tummy stayed low in a forearm plank unless they forced their breath out.

A ridge or doming down the middle of the tummy during or after pregnancy is a separation of the six-pack muscle (diastasis recti). In 300 first-time mothers it was found in 60% at 6 weeks after the birth and 33% at a year, and it wasn't linked with lower back or pelvic pain. The NHS says the gap usually returns to normal by 8 weeks after the birth, to contact a GP if it is still obvious then, and to avoid sit-ups and planks to start with.

Sources 21, 22, 23, 24, 25

How it usually goesStrong evidence

No study follows back pain from planks. Across 33 groups of people with new back pain (11,166 people), average pain fell from 52 out of 100 to 23 at 6 weeks, 12 at 26 weeks and 6 at a year; in pain that had already lasted a while, from 51 to 23 at a year. The NHS says back pain usually improves within a few weeks but can sometimes last longer or keep coming back. No one can promise a date.

Sources 26, 10

Exercise helps back pain, but "core" exercise isn't specialStrong evidence

Across 249 trials, exercise eased long-lasting back pain by about 15 points out of 100 against no treatment, usual care or a placebo (moderate certainty), with mostly minor harms such as soreness. Motor control ("core") exercise beat minimal treatment but was no better than other exercise (29 trials, 2,431 people); stabilisation exercise was no better than other active exercise in the long term, and beat general exercise only in the short term (5 trials, 414 people). Two network analyses ranked Pilates highly and concluded that people should do the exercise they enjoy, and NICE suggests a group exercise programme of any type. A summary of a 2021 consensus for elite rowers goes further, saying static exercises such as planks are unlikely to help (expert opinion). So this plan uses planks as one part of a mix, not as the cure.

Sources 27, 28, 29, 30, 31, 32, 20, 33

The only plank trial for back pain is tinyEmerging evidence

One small trial randomised 38 people with long-lasting back pain to two kinds of bridge on the back or to the plank (a prone bridge): 5 holds of 30 seconds with 30 seconds' rest, 3 sets a day, 3 days a week for 8 weeks, supervised. Disability scores fell further with the plank (from 44 to 24 out of 100) than with the bridges (about 12 points), but the plank group started worse, nobody was blinded and there was no group without exercise. Its companion paper reported that after the 8 weeks, deep and side tummy muscle activity rose more with the plank than with the bridges (45 people enrolled, 38 analysed).

One NHS trust's back leaflet, now past its expiry date, uses five 30-second forearm planks, 3 to 4 times a week; NHS inform's back exercises include no plank.

Sources 34, 35, 36, 37

Planks aren't shown to harm the back, but no one has measured the loadEmerging evidence

The back muscles work at about 5% of their maximum in a plank, and trials of core exercise for back pain reported minor or no side effects. The lab team that measured muscle activity in planks suggests knee planks for people who need to keep spinal compression low, reasoning from muscle activity rather than a measured load. No study we opened measured the compression or shear on the lower spine in a front plank, and a high spinal load quoted for the full side plank is a figure we could only find second-hand. So the honest claim is that planks are not shown to harm the back, not that they are proven safe.

Sources 1, 28, 5, 6

Plank times and back pain don't line upMixed evidence

In a clinic sample, 32 people with long-lasting back pain held a plank for 28 seconds on average, against 73 seconds in 43 people without. In 34 active adults, those with mild back pain held planks about as long as those without (99 against 105 seconds, no real difference). In 117 adults, mostly students, those with back pain held longer, and plank time didn't predict back pain once other factors were counted. A plank test is not a test for back pain, and a long plank doesn't show a healthy back.

Sources 38, 39, 40

A weak core doesn't predict back painStrong evidence

A review of 26 cohort studies found strong evidence that trunk muscle endurance has no relationship with the risk of low back pain, and across 12 cohorts (5,459 people) neither tummy strength nor back endurance predicted it, though in one Danish cohort of 928 adults, weaker back-muscle endurance went with first back trouble in men. In 4,147 soldiers, core stabilisation exercise prevented no more back pain than sit-ups.

Sources 41, 42, 43, 44

The hip flexors: they work in a plank, but tightness isn't the causeEmerging evidence

The rectus femoris works at about 20% of its maximum in a plank on the toes, and in fine-wire recordings push-ups worked the deep hip flexor (the psoas) at up to 25% (a straight-arm plank is the top of a push-up; our inference). But in 118 adults, tight hip flexors weren't linked with back pain once body weight was counted, and in 600 adults hip flexor length wasn't linked with back pain while weaker hip flexors were. Strength comes before stretching.

Sources 1, 45, 46, 47

Hold length: shorter, repeated holds are sensible, not provenMixed evidence

No trial has compared short and long plank holds for back pain. Holds mostly end at the arms and legs, and in 30 adults holding a plank to failure, untrained people leaned more on their back muscles throughout than trained people (muscle activity only). Back-muscle oxygen fell during 30-second seated holds even at 2% of maximum effort in 8 adults, 6 weeks of hold-based training stiffened the trunk in healthy men, and in 15 trained adults a single 3-minute plank, but not a 1- or 2-minute one, caused short-term muscle swelling and loss of force.

Blood pressure passed a safety limit within about 50 seconds of a plank in 97% of 62 healthy adults, with no harm reported. So this guide uses short holds, repeated, and makes planks harder rather than longer.

Sources 3, 48, 49, 50, 51, 52

Plank cues: tuck, brace, and don't suck inEmerging evidence

In lab studies of healthy people, a slight backward tuck of the pelvis with the shoulder blades drawn together drew the most trunk muscle activity, including the back muscles, and felt hardest (15 adults); a long-lever plank with a pelvic tuck drew more ab activity than a standard plank (19 adults). Drawing the tummy in (hollowing) raised side and back muscle activity, not six-pack activity, and felt much harder (20 adults). In 8 men standing, bracing (tightening all round) improved modelled spine stability by 32% for a 15% rise in compression, and the authors saw no mechanical reason to hollow. None of these studies measured pain.

Sources 53, 54, 55, 56

After a birth, in pregnancy, with thin bones or a heart conditionEmerging evidence

The NHS says try to avoid sit-ups, planks and high-impact exercise to start with after a birth, and to begin with gentle tummy exercises. In pregnancy, it says it's usually fine to keep doing activities you already do, not to start a new vigorous activity, and to avoid lying flat on your back for long later on; neither the NHS nor the American College of Obstetricians and Gynecologists mentions planks.

With osteoporosis, a UK consensus advises changing or avoiding sustained, repeated or end-of-range forward bending, and engaging the tummy muscles; neither it nor the Royal Osteoporosis Society's exercise page mentions planks, so the view that planks suit thin bones is our reading of that guidance. With heart failure, the British Heart Foundation says avoid using your whole body as a weight, such as press-ups or planks.

Sources 25, 57, 58, 59, 60, 61

Your plank for now (written steps)

Our 3D figure can't show all of these. They come from lab studies in healthy people, NHS pages and our plank guides; where a step is our suggestion, it says so. No trial has tested a change to the plank for back pain.

• 1. Knees down, or hands on a bench or worktop (an incline plank), until a plank on your toes passes the pain rule. A knee plank drew much less trunk and hip flexor activity in a lab study.
• 2. Elbows under your shoulders, a slight tuck of the pelvis, glutes lightly squeezed, ribs down.
• 3. Brace (tighten all round, as if for a light poke in the stomach) rather than sucking your belly button in.
• 4. End each hold as soon as your hips start to sink or your back starts to arch, whatever the clock says, as plank tests do.
• 5. Short holds of about 10 seconds, repeated, rather than one long hold (our suggestion, as in our McGill big 3 routine and plank hold guide).
• 6. Breathe all the way through: count out loud. Blood pressure climbs quickly in a plank.
• 7. Head in line with your back, eyes on the floor just ahead of your hands (our plank guide).
• 8. Plank challenges and daily maximum-time tests wait until the end of stage 3 (our suggestion): long holds end at the arms and legs, and a plank time says little about your back.

Sources 1, 53, 54, 55, 56, 40, 3, 49, 51, 52

After a baby, with a hernia, in pregnancy or with a heart condition

After a birth, start with the NHS's gentle tummy exercise (a side-lying draw-in, held for a count of 10, up to 10 times) and bring planks back once a move no longer makes the middle of your tummy dome; the NHS says avoid planks to start with. With a hernia, or after a repair or other tummy surgery, follow your surgical team: our hernia page leaves planks out at first after a repair at or near the belly button.

In pregnancy, the NHS says it's usually fine to keep doing what you already do, and not to start a new vigorous activity. Move to your knees or an incline as your bump grows, and ease off if your tummy domes (our plank guide). With heart failure, the British Heart Foundation says avoid planks; with high blood pressure, keep holds short and keep breathing.

Sources 25, 57, 61, 52

Back to your usual training (written steps)

No study has tested how planks come back after back pain, so these steps are our suggestion unless a source is named.

• 1. One change at a time: from knees to toes, then longer holds up to about 45 seconds, then straight arms, then moving planks (walk-ups, shoulder taps, mountain climbers), then long-lever planks and rollouts.
• 2. Make it harder rather than longer once 3 holds of 45 seconds are steady, as our plank guide does.
• 3. Keep the dead bug, bird dog and side plank in your week: each trains a different job of the trunk.
• 4. Keep whatever activity you enjoy alongside, such as walking, Pilates or the gym: no exercise type is clearly best for back pain.
• 5. If a session flares it, go back a stage for a few days, not to nothing.

Sources 54, 31, 32, 20