Accurate and Efficient Assessments of Working Posture

Posted on by Brian D. Lowe, Ph.D.


Musculoskeletal disorders (MSDs) account for roughly one-third of workplace illness and injuries in the United States. According to Liberty Mutual Insurance Company, overexertion injuries, which include those from work-related lifting, pushing, pulling, holding, carrying, or throwing, cost U.S. businesses $14.2 billion in direct costs in 2013. To address the burden of preventable musculoskeletal and overexertion injuries, practitioners in the fields of Occupational Safety, Industrial Hygiene and Ergonomics need workplace assessment methods that are both accurate (valid) and time efficient. This need was highlighted in a 2008 NIOSH exposure assessment workshop in which a health/safety/ergonomics practitioner said:

“I’m a consultant, and ….we have to do 450 (facilities) in four years. So speed is essential, and we are using the risk assessment checklist, semi-quantitative measures to estimate risk.”

This practitioner was indicating two important points. First, because he was responsible for conducting a large number of worksite ergonomic assessments, an efficient risk-assessment method was important. Second, like other OS&H practitioners conducting risk assessments, he relies on semi-quantitative observation-based methods to evaluate risk factors, such as postural stress. Even though technology for assessing body kinematics (motion) continues to advance, use of observation-based approaches is commonplace among practitioners conducting workplace assessments of risk factors for musculoskeletal and overexertion injury. While often relied upon, these methods are not standardized, nor were they systematically developed.

These points were part of the motivation for the recent publication Observation-Based Posture Assessment: Review of Current Practice and Recommendations for Improvement; developed by NIOSH and colleagues at the University of Windsor affiliated with the Canadian Centre of Research Expertise for the Prevention of Musculoskeletal Disorders, CRE-MSD. This new document, intended to assist practitioners, is based on a number of recent studies that systematically determined an optimal observational framework for classifying postural stresses in the workplace, among multiple levels of severity. This optimal framework considers the accuracy (validity) of the posture assessment: that is, how many errors are made, and the size of the errors, in visually estimating posture. The optimal framework also considers efficiency: that is, how fast the observer can visually estimate and classify posture among multiple levels. The document distills several scientific publications into a more accessible format for the practitioner.

This blog is an opportunity to re-emphasize the most important information in the document– the recommended framework for optimally classifying working posture of the trunk, shoulder, and elbow depicted in the figure below. The recommended optimal framework is to classify trunk flexion (forward bend) in four categories, trunk lateral (side) bend in three categories, shoulder flexion in five categories, shoulder abduction in five categories, and elbow flexion in four categories, as shown below:

posture 1

One safety resource blogger commented on the new Observation-Based Posture Assessment stating that “This guide from NIOSH looks to be a bit higher level, yet looks suspiciously like the Rapid Upper Body Assessment (RULA) tool.”   The NIOSH/CRE-MSD guide actually looked at RULA and seven other common job analysis methods and the accompanying number of posture levels. For most postures the number of posture levels is equal to, or within one unit of, the optimal number of categories recommended:

number of posture levels (average of 8 previous studies) optimal number of posture levels – DHHS (NIOSH) Pub 2014-131
trunk flexion 4 4
trunk lateral bend 2 3
shoulder flexion/extension 4 5
shoulder abduction/adduction 3 5
elbow flexion 3 4

It turns out that common traditional approaches, on average, have been close to what recent evidence shows to be optimal. In fact, the optimal method presented in Observation-Based Posture Assessment agrees perfectly with RULA1 for shoulder flexion posture (both use five levels), but less so for shoulder abduction/adduction posture (two versus five levels).

We hope that this document helps practitioners decide on the most accurate and efficient posture assessment method for their needs.


Brian D. Lowe, Ph.D.

Research Industrial Engineer, NIOSH, Division of Applied Research and Technology


(Note: The blog author wishes to express gratitude and acknowledgement to Drs. Patricia Weir and David Andrews, University of Windsor, Windsor, Ontario, collaborators in the development of the above-mentioned document.)


1McAtamney L, Corlett EN [1993]. RULA: a survey method for the investigation of work-related upper limb disorders. Appl Ergon 24(2):91–99.

Posted on by Brian D. Lowe, Ph.D.

9 comments on “Accurate and Efficient Assessments of Working Posture”

Comments listed below are posted by individuals not associated with CDC, unless otherwise stated. These comments do not represent the official views of CDC, and CDC does not guarantee that any information posted by individuals on this site is correct, and disclaims any liability for any loss or damage resulting from reliance on any such information. Read more about our comment policy ».

    The use of a backbrace is beneficial for injury protection, bit only when used for unusual amount of weight/strain

    Very useful postural evaluation system. I particularly appreciate that trunk rotation is not addressed because it is probably very difficult for individuals unaccustomed to evaluating trunk posture to differentiate thoracic rotation from lumbar rotation especially when observation is obscured by clothing. It is probably best not to attempt to assess rotation. I wonder if the author left out rotation for this reason?

    I do not think there is sufficient evidence to support a claim about the efficacy of back belts for back injury prevention. The literature suggests that back belts may reduce lumbar motion when cinched uncomfortably tight. If one, nonetheless, makes the assumption that back belts reduce intersegmental motion of the lumbar vertebrae (as many back belt manufacturers claim) the problem of motion at L5-S1 probably remains. For example, if a worker lifts an object from the floor with a back belt on and that back belt reduces lumbar motion, the motion reduction will probably be realized primarily at L1-L5. As currently designed, back belts are probably incapable of restraining L5-S1 motion. They typically do not cross that joint. Since the amount of bending required to pick up an object from the floor does not change, any reduction in L1-L5 motion will need to be “made up” by increased motion above and below the back belt. This could increase the risk of back injury since L5-S1 is the most frequently injured spinal unit.

    Thanks for the great comment. We did not include trunk rotation in the document because the scientific studies did not report optimal posture categories for trunk rotation similarly to trunk flexion and lateral bend. You correctly note that trunk postures are a result of motion occurring at individual motion segments of the spine, including the multiple lumbar and thoracic motion segments. Observational-based methods for assessing trunk posture typically categorize posture by simplifying the trunk to that of a single body segment. We attempted to describe this biomechanical simplification on page 8 of the document, stating that “The definition of postures simplifies the representation of joint position for the purpose of characterizing postural stress….” and “ …segment motions throughout the lumbar and thoracic spine are typically simplified in the definition of trunk postures to include trunk flexion/extension (or sometimes trunk inclination), trunk lateral bend, and trunk rotation.” We agree that observational estimation of trunk rotation angle, in general, would be most difficult because a view that is perpendicular to the axial rotation plane would be from somewhere above the worker.

    Amazing post and working posture is very very important while working..Great job for publishing such a beneficial article.

Post a Comment

Your email address will not be published. Required fields are marked *

All comments posted become a part of the public domain, and users are responsible for their comments. This is a moderated site and your comments will be reviewed before they are posted. Read more about our comment policy »

Page last reviewed: August 25, 2017
Page last updated: August 25, 2017