Questioning the ideal sitting posture

By Dr Andrew Claus (the handsome fellow in the stripy shirt)

Andrew Claus 'sitting'

Sitting cops a lot of bad press, but the science to compare postures is pretty weak. One of our studies undertaken at the University of Queensland examined whether people could copy four sitting postures. Most people couldn’t sit with a curve-in at the low back (like standing, as thought to be ‘good’ sitting posture) unless they were taught how to do it. Perhaps people need to be taught how to adjust their back posture as well as their chair, or perhaps it is unrealistic to expect people to sit like this. More research is needed, but in the meantime – sit comfortably, and have a read of the ABC’s article on our research ‘Ideal’ sitting posture questioned.

Is ‘ideal’ sitting posture real?: Measurement of spinal curves in four sitting postures

Andrew P. Claus(a), Julie A. Hides(a), G. Lorimer Moseley(b), Paul W. Hodges(a)
(a) NHMRC Centre of Clinical Research Excellence in Spinal Pain, Injury and Health, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, St. Lucia, QLD 4072, Australia
(b) Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, UK

There is a lack of quantitative evidence for spinal postures that are advocated as ‘ideal’ in clinical ergonomics for sitting. This study quantified surface spinal curves and examined whether subjects could imitate clinically ‘ideal’ directions of spinal curve at thoraco-lumbar and lumbar regions: (i) flat – at both regions (ii) long lordosis – lordotic at both regions (iii) short lordosis – thoracic kyphosis and lumbar lordosis. Ten healthy male subjects had 3-D motion sensors adhered to the skin so that sagittal spinal curves were represented by angles at thoracic (lines between T1–T5 and T5–T10), thoraco-lumbar (T5–T10 and T10–L3) and lumbar regions (T10–L3 and L3–S2). Subjects attempted to imitate pictures of spinal curves for the flat, long lordosis, short lordosis and a slumped posture, and were then given feedback/manual facilitation to achieve the postures. Repeated measures analysis of variance was used to compare spinal angles between posture and facilitation conditions. Results show that although subjects imitated postures with the same curve direction at thoraco-lumbar and lumbar regions (slumped, flat or long lordosis), they required feedback/manual facilitation to differentiate the regional curves for the short lordosis posture. Further study is needed to determine whether the clinically proposed ‘ideal’ postures provide clinical advantages.

See full article at Manual Therapy 14,4, 404-8


  1. Andrew Claus says

    Thanks for the feedback comments. Indeed, a few good research teams around Australia and overseas are taking an interest in how the spine is used in sitting. A problem that we are working to overcome is that many past studies assumed that momentary measures with radiography or photography were an accurate representation of typical posture behaviour. New tools mean that we can (and should) study actual behaviour during functional tasks.

    The article was a foundation for more recent and on-going work measuring postural behaviour while using a computer over an extended period, and leading to clinical trials that will examine the merit of our clinical theories concerning posture, performance, comfort, discomfort and health.

  2. I find that guided position and movement after restoring functionality is very important.
    A common condition in my practice is psoas minor insufficiency due to T11/12 injury or subluxation. Though the range of lordosis kyphosis alterations can be found in both male and female; females more frequently develop anterior pelvic tilt with lower lumbar hyper lordosis and kyphosis starting at L1; males more frequently stabalise the pelvis with the rest of the core muscles and have a lordosis loss.
    Interestingly I had one elderly slender client that had lumbar kyphosis. Her mother had taught her to always pull her navel toward her spine. I suggested that her mother had the right idea but had stated it backward. When she began to pull the spine towards the navel her lordosis developed, her digestion improved and she had more stamina to do energy work with her clients,

  3. Excellent article — with prolonged computer-sitting, commute-driving and recreational computer use, I am seeing more people in my massage practice with probelms related to long-term static muscle postures. It’s quite possible, seems to me, we just aren’t designed to sit still! I’ve taken lately to asking people to use the ball for sitting, then switch to the chair, and throw in a lot of walk-breaks to water cooler, copy machine, or anywhere they can push away from the keyboard for a few minutes and use their muscles….

  4. I am thrilled to read news of this recent research from this much esteemed research group. As a physiotherapist of 15 years and a hatha yoga teacher of 25 years I am constantly faced with this phenomenon. One can speculate many reasons for it, such as hip or spinal stiffness, or even the prevalence of the so-called Scheurmanns Kyphosis (predominantly in the thoraco-lumbar region in men). It is only ongoing research that will give us the real answer. However, it is clearly observable in a clinical setting that men (more than most women) have extreme difficulty in achieving and maintaining a natural spinal curvature in sitting. If these postures are forced especially when sitting on the floor cross-legged (something many Western adults just can’t do) it appears to be a cause of knee problems, hip and trunk muscle spasm, breathing difficulties and a range of nervous system disorders that often manifest as stress and discomfort. I think the best general advice for most people is to sit comfortably as tall as you can but as relaxed as you can. Let the spine adopt the shape you feel most comfortable in with an ability to have your abdomen relaxed enough to be able to do relaxed diaphragmatic breathing. For most people to do this is usually best to sit with the knees slightly below the groin level, and for most people a comfortable sitting posture does not have the same spinal curvature that appears when standing.

    admin Reply:

    Thanks Simon — good to know the research is getting out there.