Is Postural Analysis Actually Useful?

Why your client's uneven shoulders are probably not the problem — and what to look for instead

Walk into almost any sports massage clinic, physiotherapy practice or personal training studio in the UK and there is a reasonable chance your therapist will ask you to stand still while they look you up and down, scrutinising the height of your shoulders, the angle of your pelvis, the position of your knees. They will note any asymmetries, compare left to right, and then — more often than not — tell you that something is out of alignment and that this is contributing to, or even causing, your pain.

This is postural analysis. It is taught on almost every manual therapy course in the country. It feels clinical, systematic and authoritative. Clients find it compelling because it gives their pain a visible, tangible explanation.

There is just one significant problem: for most of what it claims to identify, the evidence base is remarkably weak.

What Is Postural Analysis?

Postural analysis is the systematic visual or physical assessment of a person's static body position, typically in standing. The practitioner looks for deviations from a theoretical "ideal" alignment — usually defined by a plumb line running through specific landmarks — and identifies asymmetries between the left and right sides of the body.

Common things assessed include shoulder height, pelvic tilt, spinal curves, knee alignment and foot position. Practitioners use bony landmarks as their reference points: the anterior superior iliac spine (ASIS) and posterior superior iliac spine (PSIS) for pelvic tilt, the acromion process for shoulder height, and so on.

The underlying assumption is that deviations from symmetrical alignment represent a dysfunction that places abnormal stress on tissues over time, eventually leading to pain or injury. Correct the posture, the theory goes, and you address the underlying cause.

It sounds entirely logical. And yet, when you examine the research, the foundations start to crumble fairly quickly.

Where Did Postural Analysis Actually Come From?

The story of modern postural analysis begins in 19th century Germany, and it is not quite the rigorous clinical origin story the profession tends to assume.

Christian Wilhelm Braune (1831–1892) was a German anatomist and professor of topographical anatomy at the University of Leipzig, and he is widely credited as one of the founding figures of the postural analysis model we still use today. In 1855, Braune brought the study of posture to medicine by dropping a plumb line down the backs of his research subjects as they stood to attention and measuring the results. According to historian William McNeill, this quantification of posture became the basis for discussions of the "straight line inside the body" that inform modern orthopaedics.

Braune's more influential work, however, was conducted not on living people but on frozen cadavers. He pioneered the use of frozen cadavers for anatomical investigations, working with physiologist Otto Fischer to research the position of the centre of gravity in the human body and its various segments. By first determining the planes of the gravitational centres of the longitudinal, sagittal and frontal axes of a frozen human cadaver in a given position, and then dissecting the cadaver with a saw, they were able to establish the centre of gravity of the body and its component parts.

The motivation behind this work is particularly telling. Braune and Fischer's aim was to improve the ergonomics of German infantry equipment, and to this end their primary goal was to find a natural initial attitude that would be appropriate for measurements and calculations. In other words, they were not trying to define a universal standard of healthy human posture for clinical practice — they were trying to solve a military engineering problem.

And yet the posture they defined for that purpose — a theoretical neutral alignment derived from frozen bodies positioned for the convenience of measurement — gradually became adopted as the reference standard against which living, moving, breathing people are still assessed today.

Several authors have attributed the authorship of the posture used as a reference for assessing postural deviations to Braune and Fischer. Signe Brunnstrom criticised the term "normalstellung", noting it could be interpreted as suggesting that a perpendicular posture is intrinsically desirable, writing: "Obviously, perpendicular posture does not coincide with Nature's way of balancing the body and, therefore, should not be used as a standard for good posture."

This is where the entire edifice of postural analysis begins to wobble. The "ideal" posture that practitioners are trained to assess against was defined by a 19th century anatomist working on cadavers that had been frozen solid and positioned for military biomechanics research — not derived from studies of healthy, pain-free living people. The standard was never a description of what normal human variation looks like. It was a measurement convention. And we have been using it as a diagnostic tool ever since.The Bony Landmark Problem — Your Skeleton Is Not Symmetrical

Here is something that tends to get overlooked in postural analysis training: the bony landmarks we use to assess alignment are not consistent from person to person. In fact, the natural variation between individuals — and even between the left and right sides of the same person — is so large that it fundamentally undermines the reliability of many standard postural assessments.

A particularly striking example comes from a study by Preece and colleagues (2008), published in the Journal of Manual and Manipulative Therapy. The researchers examined 30 cadaver pelves — real human bones, measured directly — and assessed the ASIS-PSIS angle, the standard measure used to identify anterior pelvic tilt in clinical practice.

What they found was remarkable. The ASIS-PSIS angle across the specimens ranged from 0 to 23 degrees, with a mean of 13 degrees and a standard deviation of 5 degrees. Side-to-side differences within the same pelvis reached up to 11 degrees in ASIS-PSIS tilt and up to 16 millimetres in innominate height.

To put that in plain terms: two people could stand in front of you with completely different ASIS-PSIS angles — one measuring 5 degrees, another measuring 20 degrees — and both could be in perfectly normal, neutral pelvic alignment for their own anatomy. Meanwhile, the same person's left pelvis could sit 16 millimetres higher than their right not because of any dysfunction, but simply because that is how their skeleton is shaped.

The conclusion from Preece et al. is important: variations in pelvic morphology may significantly influence measures of pelvic tilt and innominate rotational asymmetry. In other words, what looks like an anterior pelvic tilt or a pelvic asymmetry on your postural assessment may simply be the shape of your client's skeleton — not a sign of dysfunction at all.

This is not a minor methodological quibble. It strikes at the heart of the entire postural assessment process. If the landmarks we use to identify asymmetry vary this dramatically between individuals at a bony level, we cannot reliably distinguish between "this person has an anterior pelvic tilt" and "this person simply has a pelvis shaped this way." The measurement is confounded before we even begin.

Does "Bad" Posture Actually Cause Pain?

Even setting aside the measurement problem, we need to ask a more fundamental question: does postural deviation actually cause pain in the first place?

The research here is, to say the least, unsupportive of the postural analysis model.

Across a substantial body of research, the relationship between static posture and pain has proven stubbornly difficult to establish. Studies looking at forward head posture and neck pain have found no significant association. Research on lumbar lordosis and lower back pain shows no reliable correlation. Studies on scoliosis consistently find that many people with quite significant curves live without any pain at all, while others with minimal curvature experience significant discomfort.

Science writer and former massage therapist Paul Ingraham, whose work on PainScience.com has synthesised a large volume of this research, describes the field of "posturology" — the study of posture's relationship to pain — as largely pseudoscientific, characterised by researchers who assume their conclusion before they begin and then look for confirmation. He writes that poor posture is "hard to define, hard to fix, and hard to causally link to any other kind of pain or injury," and that the evidence that tissue pathology or biomechanical deviation explains chronic pain is consistently weak across common conditions including back pain, neck pain and knee osteoarthritis.

As Ingraham notes, this does not mean posture is completely irrelevant to pain — it is just dramatically less relevant than the therapy industry has made it out to be for decades.

Consider the everyday reality: billions of people sit, slouch, cross their legs and lean to one side throughout their lives without ever developing chronic pain as a result. If uneven shoulder height reliably caused problems, we would expect people with notably asymmetric posture to consistently report more pain than those who are perfectly symmetrical — and that is simply not what the evidence shows.

The Structuralism Problem

The broader issue here is what researchers call "structuralism" — an excessive focus on anatomical asymmetry and biomechanical imperfection as the primary explanation for pain.

Structuralism feels intuitive. It gives both practitioner and client something to point at. It generates a clear narrative: "your pelvis is tilted, your right shoulder is higher than your left, your arch has collapsed — this is why you are in pain." Clients often find this satisfying. Practitioners find it gives structure to their assessments and a clear rationale for their treatments.

But it rests on assumptions that the evidence does not support. Pain is not simply a mechanical output of structural deviation. It is a complex experience shaped by neurological, psychological, immunological and social factors that postural assessment cannot even begin to capture. The idea that slightly asymmetric shoulders reliably cause pain assumes a direct, linear relationship between structure and symptoms that the research repeatedly fails to confirm.

As Ingraham puts it, most people who have sought professional help for pain have been told — in one way or another — that the root cause is some kind of mechanical glitch. "Almost everyone who has ever sought professional help for a stubborn pain in their body has been told that they are deformed and fragile." The irony is that this message can itself worsen outcomes, by increasing fear, catastrophising and the belief that the body is damaged in ways that need correcting before it can heal.

So What IS Postural Observation Actually Good For?

None of this means we should abandon observational assessment entirely. It means we need to be much more precise about what we are actually able to identify — and much more cautious about the conclusions we draw.

There are things that postural and movement observation can genuinely and reliably tell us. They are just different from what is typically claimed.

Obvious structural or traumatic change is the clearest legitimate use case. A significant scoliosis, a visible leg length discrepancy following fracture, an obvious deformity following injury or surgery, or marked asymmetry resulting from neurological conditions — these are meaningful findings that postural observation can legitimately identify and that warrant clinical attention. The key distinction is that these are gross, objective changes, not the subtle asymmetries that most postural analysis focuses on.

Acute antalgic posture — the visible list or shift a person adopts when they are in acute pain — can be useful clinical information. Watching someone guard a body part, shift their weight away from a painful side or adopt a position of comfort tells you something real and clinically relevant about their current pain experience.

Movement quality in context offers more useful information than static posture. Watching how someone moves — how they get up from a chair, how they bend to pick something up, which movements seem effortful or guarded — gives you insight into their functional capacity and any movements that currently provoke or aggravate their symptoms. This is meaningfully different from noting that one hip is 3 millimetres higher than the other in standing.

Loading capacity and function is the genuinely important clinical question. Can this person do what they need to do? What movements are currently limited or painful? What is their capacity relative to what is required of them? These questions are far more clinically meaningful than whether their ASIS is level.

Red flag screening remains a legitimate and important use of clinical observation. Observation that prompts you to ask whether a presentation might involve serious pathology — an unusual pattern of deformity, a change in posture following trauma, signs of neurological involvement — is valuable. This is precisely the kind of observation that clinical training should prioritise.

A More Honest Way to Assess Your Clients

The shift we are advocating for is not from "assessment" to "no assessment." It is from an assessment framework built on unreliable measurements and unsupported theoretical models to one grounded in what the evidence actually supports.

That means being honest with your clients about what you are observing and what it means — and what it does not mean. One shoulder being slightly higher than the other is, in the vast majority of cases, simply normal human variation. It is not a diagnosis. It is not causing their pain. Telling a client their pelvis is tilted or their posture is "wrong" without a sound evidential basis for linking that to their symptoms is not just clinically unhelpful — it can actively make things worse by reinforcing the idea that they are structurally broken.

It means asking better questions. Rather than "what does your posture look like?", ask "what makes your pain better or worse?" "What can you not currently do that you want to be able to do?" "How long have you had this, and what has changed?" These questions tell you infinitely more about what is actually going on than comparing shoulder heights.

And it means prioritising movement, load and function as your primary assessment framework — and recognising that the treatment that most consistently helps people is the one that this blog has been building a case for throughout: progressive, meaningful, personalised exercise.

The Bottom Line

Postural analysis as it is typically practised rests on foundations that the research does not support. The bony landmarks we use to measure alignment vary enormously between individuals at a skeletal level — before muscle, fat or movement even enter the picture. The relationship between static postural deviation and pain is weak and inconsistent across the literature. And the broader habit of attributing pain to structural imperfection encourages a passive, diagnosis-centred model of care that is increasingly hard to defend.

Observation still has a place. Gross structural changes, acute pain behaviours, movement quality, functional capacity, and red flag screening all represent legitimate and valuable uses of clinical observation. What it does not support is the conclusion that your client's slightly uneven shoulders, mildly tilted pelvis or modestly hyperlordotic lumbar spine is the reason they are in pain.

The human body is not a perfectly symmetrical machine that breaks when it deviates from the ideal. It is a remarkably adaptable, resilient biological system that tolerates enormous variation. Our job as therapists is to build capacity, reduce fear and get people moving — not to identify and pathologise the infinite normal variations of human anatomy.

Sources:

Braune W, Fischer O. On the centre of gravity of the human body as related to the equipment of the German Infantry Soldier. Berlin: Springer-Verlag; 1889.

Ingraham P. Does Posture Matter? PainScience.com. Ingraham P. Alignment and Biomechanics Are Not Important Pain Causes. PainScience.com.

Preece SJ et al. Variation in pelvic morphology may prevent the identification of anterior pelvic tilt. Journal of Manual and Manipulative Therapy, 2008; 16(2):113-7.

This post is written for qualified sports massage therapists and allied health professionals and is intended to support evidence-informed clinical practice.