From The Ergoweb® Learning Center

We Told Them to Lift with Their Legs, But They Just Won’t Listen!

Peter Budnick, PhD, CPE

Note from study reviewer, Peter Budnick: I facilitated a 1-day workshop last week (Advanced Ergonomics Analysis – Using Quantitative Ergonomic Assessment Tools) at the Oregon Governor’s Occupational Safety & Health Conference, and the topic of ‘lifting with your knees’ came up, which reminded me of this article we published in 2010. So, why don’t people ‘lift with their knees’ even if they are trained to do so? Why is ‘lift with your knees’ a weak and ineffective approach to ergonomics? This study gives us some useful insight on those questions (also see another of our recent articles related to this topic, Research Review: Value of Back Injury Prevention Training Questioned).

In this study, the researchers Li and Zhang (2009) hypothesized that people whose back strength is greater than their total knee strength are more likely to adopt a back-lift strategy, and vice versa, and devised an experiment that considered the relationship between a person’s back and total knee strength, gender, self-knowledge of their strength, and the load weight on lifting strategy.

The authors review some of the scientific literature surrounding lifting strategies, such as bending, stooping or squatting (i.e., lifting with the legs). They note the following:

  • We don’t really know what factors drive a person’s lift strategy;
  • We focus primarily on back compressive force when evaluating the risk of a particular lifting strategy;
  • There’s no clear evidence that knowledge of the weight to be lifted influences a person’s choice of lifting strategy;
  • There is evidence that an individual’s perception and knowledge of his/her own strength can impact lifting strategy;
  • There is evidence that personal factors impact lifting strategy;
  • As muscle fatigue increases (during prolonged periods of lifting), lifting strategy shifts from leg lift to back lift strategy;
  • People with low back pain tend to try to limit back flexion and overall movement in an attempt to limit spinal loading, yet fail to do so because of muscle coactivation;
  • Stronger people tend to choose a heavier load per carry strategy, resulting in fewer carries per batch, and that those who are more aware of their strength capabilities “became more emboldened in load handling” (Li and Zhang, 2009);
  • Back strength is poorly correlated with back injuries;
  • The physiological/metabolic demands are much higher for leg lift strategies than for back lift strategies; and
  • The size of a load impacts strategy (i.e., larger items do not fit between the knees, making a leg lift strategy less likely);

Primary Findings

  • There was no significant difference in back strength between those found to have back-stronger-than-knee-strength (BSTK) and those found to have knee-stronger-than-back-strength (KSTB), but there was a significant difference between knee strengths between the groups;
  • Those with BSTK mainly used a back preferred lift strategy;
  • Those with KSTB used variable lift strategies;
  • However, the leg-preferred lift strategy was exclusively adopted by those with KSTB;
  • “It can be concluded that participants with back strength greater than their total knee strength tended to use a back preferred lift strategy, whereas those who used a leg-preferred lift strategy all had total knee strength greater than their back strength” (Li and Zhang, 2009);
  • There was a significant gender effect, with male subjects predominantly selecting back-preferred strategies, and female subjects predominantly selecting leg-preferred strategies; and
  •  Knowledge of strength feedback did not show a significant effect on lifting strategy.

The Bottom Line — How this Applies to Ergonomists

We’ve all heard it, and some may even be strong proponents of it: ‘lift with your legs, not with your back!’ The problem is, many or most people simply do not do so. This study looks deeper into why that is, and concludes that total (combined) knee strength vs. back strength is a significant predictor of how someone will choose to lift. Add this new knowledge to previous research that demonstrates that the physiological/metabolic demands of leg-lift strategies are far greater than those for back-lift strategies, that many items are too big/bulky to effectively use a leg-lift strategy, and consider the effects of knee function and injury, it’s no wonder few people follow the conventional wisdom advice.

The authors suggest that these results can be used to justify training programs that focus on increasing knee strength. While such an approach, under strict conditions, may produce a sustainable change in lifting behavior, there is still no clear evidence that such behavior changes will translate into a reduction in back injuries, which is yet another challenge to the “lift with your knees” approach to ergonomics.

However, rather than spending a great deal of time, money and resources trying to change people, which may or may not be successful, why don’t we focus on designing the system to better fit human capabilities, which will be successful? Why don’t we just get to the root of the problem? Why don’t we match the work to the people, vs. matching the people to the work? Why don’t we design our process so that lifting naturally occurs in the “power zone,” or the “strike zone,” meaning lifting close to the body near waist height? After all, this is also the most effective lift, reducing the time it takes to conduct the lift (reducing “movement waste” in Lean terminology) and reducing the likelihood of error (dropping, colliding with other equipment or people, etc. — “do it right the first time” in Lean terminology).

Now that’s ergonomics.

Study Limitations

The authors acknowledge four major limitations to their study:

  1. This study used a static postural index measurement rather than a dynamic measure, meaning that only the posture at the very start of the lift was analyzed and compared;
  2. This study only addresses the lifting kinematics, and had other variables, such as joint forces and torques been calculated, additional insights may have been gained;
  3. Unbalanced data used in the ANOVA statistical test may have led to some effects being overlooked;
  4. The study is limited in the number of lifting tasks, the simplicity of the tasks, the number of personal factors studied, and only young and healthy participants were included.


  • 32 subjects (16 male, 16 female, ages 20-24, mean age 24.28);
  • No reports of any musculoskeletal abnormality or discomfort at the time of the experiment;
  • 2 experimental sessions on 2 different days
  • The 1st session consisted of dynamic (isokinetic) strength tests of the shoulder, back, and knee joints;
  • Participants were then divided into 2 groups (8 men, 8 woman in each) with no statistical difference in strengths;
  • 1 group was provided with feedback regarding their strength capabilities, the other was given no feedback;
  • Prior to the 2nd session, subjects were asked to estimate their overall strengths relative to the general population;
  • In the 2nd session, all subjects performed simulated material handling tasks without knowledge of the weights being lifted (they chose how much to lift, chose their own lifting strategies, though were instructed to lift symmetrically in the sagittal plane, and were allowed and encouraged to rest to avoid fatigue effects);
  • Lifting postures where captured using reflective markers and a 5 camera video system.


Li, Kang; Zhang, Xudong, (2009). Can Relative Strength Between the Back and Knees Differentiate Lifting Strategy? Human Factors, Volume 51, Number 6, December 2009 , pp. 785-796(12). This article is available to HFES members and other subscribers to Sage Publications at:

This article is reprinted with permission from The Ergonomics Report™ Archives, where it originally appeared on November 10, 2010.