McKenzie Exercises for Back Pain: Who They Help and Who They Don't

The McKenzie method is one of the most widely used approaches for back pain, and one of the most frequently misapplied. In clinical settings, it is a systematic assessment that identifies which direction of movement helps a specific patient. On the internet, it has been reduced to “do press-ups for your back pain.” The assessment is the valuable part. The press-up is only one possible outcome of it.

I went through this myself. I had disk pain, found a McKenzie video, and started doing press-ups because that was what the video said. It happened to help, because extension was my directional preference. But for the person in the next office with facet joint irritation, the same exercise would have made things worse. The method is named after the assessment, not the exercise.

What is happening

The McKenzie method, formally called Mechanical Diagnosis and Therapy (MDT), is built on a specific observation: for many back pain presentations, repeated movement in one direction reduces symptoms while movement in the opposite direction worsens them. This is called a directional preference [1].

The logic depends on what structure is generating the pain. For a person with a posterior disk herniation, extension (arching backward) may move the disk material anteriorly, reducing nerve compression. Their pain centralizes: it moves from the leg toward the spine, which is a positive prognostic sign [5]. Extension is their directional preference.

For a person with facet joint irritation, extension typically worsens the pain because extension loads the facet joints. Their directional preference may be flexion or a lateral shift correction.

For a person with lateral disk herniation, the directional preference may be a lateral movement rather than pure extension or flexion. The assessment must test all three planes.

The concept of centralization is central to understanding why the method works when it works. Centralization means that during repeated movements, the pain moves from a peripheral location (the leg, the buttock) toward the midline of the spine. This shift in pain location correlates with better outcomes across multiple studies [5]. It suggests that the mechanical loading is moving the source of irritation away from the nerve root. When pain peripheralizes instead, moving further into the leg during testing, it indicates that the movement is worsening the nerve compression. This is not a subtle distinction: centralization is the green light, peripheralization is the red light.

The evidence

A systematic review of 11 randomized controlled trials found the McKenzie method superior to passive treatments (rest, modalities) and comparable to other active interventions for low back pain [2]. An earlier systematic review found it slightly better than passive therapy for short-term pain reduction [4].

The strongest evidence supports matching the exercise to the patient’s directional preference. Long, Donelson and Fung randomized 312 patients to receive exercises matching their directional preference, exercises opposite to their preference, or evidence-based advice. Patients matched to their directional preference had significantly better outcomes. Patients given the wrong direction did worse [3].

This is the critical finding. The method works when the direction is right. It fails, and can cause harm, when the direction is wrong. The implication is straightforward: any McKenzie protocol that skips the assessment and jumps to extension exercises is not the McKenzie method. It is a guess that happens to be right about 60 to 70% of the time, because most disk presentations have an extension preference. For the remaining 30 to 40%, it is a guess that makes things worse.

The evidence also suggests that the benefit is strongest in the acute and sub-acute phases: pain that has been present for less than about 12 weeks. For chronic back pain, the directional preference approach shows less consistent benefit compared to multidisciplinary programs that combine exercise with behavioral and cognitive strategies.

How to find your directional preference

Extension test (prone press-up): Lie face-down on a firm surface. Place your hands under your shoulders and push your upper body up, keeping your hips on the floor. Your lower back arches into extension. Perform 5 to 10 repetitions slowly. Pause at the top of each repetition for a second before lowering back down.

• If your pain centralizes (moves from the leg or buttock toward the midline of the spine) or decreases: extension is likely your direction.
• If your pain increases or peripheralizes (moves further into the leg): stop. Extension is not your direction.

Flexion test (supine knee-to-chest): Lie on your back. Bring both knees to your chest and hold for a few seconds. Repeat 5 to 10 times.

• If this centralizes or reduces your pain, flexion may be your direction. This is less common for disk herniations but relevant for spinal stenosis and some facet presentations.

Lateral test (side glide): Stand with the painful side away from a wall. Lean your hips toward the wall while keeping your shoulders level. If this reduces or centralizes the pain, a lateral component is part of your directional preference. Lateral shift corrections are common when pain presents asymmetrically and the patient is visibly shifted to one side.

If neither direction helps: not all back pain has a directional preference. The McKenzie method’s classification includes a “non-mechanical” category for these presentations, which require a different approach. If repeated movements in any direction fail to change the pain location or intensity, the MDT system does not apply, and other assessment approaches are more appropriate.

What to try

If extension is your direction:

1. Prone lying. Lie face-down for 5 to 10 minutes. This is a gentle starting position that introduces extension without forcing end-range. If this position alone reduces symptoms, it confirms that extension is the right direction.
2. Prone on elbows. Prop up on your forearms, keeping your hips on the floor. Hold for 2 to 3 minutes. Your lower back should be gently arched. If this is comfortable and your symptoms centralize or reduce, progress to the next step.
3. Prone press-up. Full extension with hands under shoulders, hips on the floor. Push up until your arms are straight or until you reach the end of comfortable range. Hold for 1 to 2 seconds at the top, then lower. Perform 10 to 15 repetitions.
4. Standing extension. Place your hands on your lower back and lean backward. This is the version you can do throughout the day when lying down is not practical. 5 to 10 repetitions.
5. Frequency. Every 2 to 3 hours during the acute phase. The protocol is designed for frequent, short bouts throughout the day, not a single long session. Reduce frequency as symptoms improve: as centralization becomes consistent and the pain diminishes, twice a day is usually sufficient.

If flexion is your direction, the exercises are the reverse: knee-to-chest stretches, seated forward bending, and child’s pose. The same principles apply: frequent repetitions, monitoring for centralization, and stopping if symptoms peripheralize. Flexion-preference patients are less common in the disk herniation population, but they are well represented among people with spinal stenosis and certain facet-mediated pain patterns.

A common mistake is to progress too quickly through these stages. Each step should be comfortable before moving to the next. If prone lying reproduces the centralization response but prone press-ups cause peripheralization, stay with prone lying. The progression is not a fixed timeline: it is governed by symptom response. Some people progress from prone lying to full press-ups in a single day. Others take a week. Both are normal.

Contraindication: Do not perform extension exercises if they peripheralize your pain. Do not perform them in the presence of cauda equina symptoms (bladder or bowel dysfunction, saddle numbness, progressive leg weakness). If you are unsure of your directional preference, a McKenzie-trained clinician (credentialed through the McKenzie Institute) can perform a full MDT assessment, which typically takes one session and gives a clear answer.

Red flags: when to see a doctor

Back pain accompanied by bladder or bowel changes, progressive leg weakness, saddle-area numbness, or bilateral leg symptoms requires prompt medical assessment. These are cauda equina symptoms and they are a surgical emergency. McKenzie exercises are for mechanical back pain. They are not a substitute for medical evaluation when neurological symptoms are present.

What the evidence does not support

Prescribing extension exercises to everyone with back pain without first assessing directional preference has no support. The Long trial showed that mismatched exercises worsen outcomes [3]. The internet’s reduction of McKenzie to “do press-ups” skips the most important step.

The method has less evidence for chronic back pain than for acute or sub-acute presentations. For long-standing pain, multidisciplinary approaches (exercise plus cognitive-behavioral strategies) have stronger overall evidence. This does not mean the directional preference concept is irrelevant for chronic pain, but it does mean that directional exercises alone are unlikely to be sufficient.

The idea that more repetitions or more forceful end-range loading produces faster results is also not supported. The protocol calls for repeated movements to end range, not forced movements past end range. Pushing harder does not accelerate centralization, and it can provoke a flare that sets the process back. The goal is consistent, moderate-intensity repetition over the course of the day.

When to see a doctor. If your back pain is severe, getting worse, or accompanied by bladder or bowel changes, leg weakness, or numbness in the groin or inner thighs, seek medical care promptly. A GP or physical therapist can assess whether imaging or specialist referral is appropriate.