Lifting lessons and back belts don’t prevent back pain at work

On the first day of a warehouse or nursing-home job, the safety talk often comes down to a few words: bend your knees, keep your back straight, lift with your legs. In some workplaces a back belt hangs by the time clock.

Both ideas have been tested at scale. Randomized trials covering more than 17,000 workers found that training in lifting technique did not prevent back pain. Federal researchers declined to recommend back belts for uninjured workers in 1994, and their own study drawing on nearly 14,000 retail material handlers later found belts made no difference to injury claims or back pain.

Meanwhile, overexertion and repetitive motion cause more serious workplace injuries than any other kind of event. The federal rule written to address them lasted about two months.

This is ergonomics away from the desk, where the load is a heavy box or a patient who cannot stand. Chairs and monitors belong to our workstation article.

Overexertion causes more serious work injuries than falls

The Bureau of Labor Statistics (BLS) tracks injuries serious enough to keep someone home, restrict their duties or move them to another job. It calls these DART cases, for days away, restricted or transferred. Its January 2026 release, covering private industry in 2023 and 2024 combined, counted 2,983,110 of them.

The largest single cause was a category BLS calls overexertion, repetitive motion and bodily conditions: 946,290 cases, close to one in three. That outnumbered contact incidents (860,050) and falls, slips and trips (721,720). These cases also ran longer, with a median of 24 days against 14 for all DART cases.

Two cautions apply. That event category is not the same as musculoskeletal disorders, the strains, sprains and nerve problems ergonomics targets, and BLS’s national tables specific to those disorders stop at 2020. BLS also revised its injury coding for 2023-2024, adding among other things a category for the novel coronavirus, and says the change creates a break in series. Comparing these figures with earlier years would mislead.

Health care has long carried a heavy share. A 2013 fact sheet from the Occupational Safety and Health Administration (OSHA), using 2011 data, reported 6.8 work-related injuries and illnesses per 100 full-time hospital employees, almost twice the private-industry rate. Overexertion or bodily reaction caused 48% of hospital injuries involving days away, and 16,680 days-away musculoskeletal cases were tied to interactions with patients. The figures are old but explain why much of the evidence below comes from health care.

So the subject is not chairs. The industries OSHA has published ergonomics guidance for make the point: nursing homes, poultry processing, retail grocery stores, shipyards, foundries and meatpacking.

A federal standard that lasted two months

OSHA did write an ergonomics standard. It was published on 14 November 2000 and took effect on 16 January 2001.

Congress then used the Congressional Review Act, a law that lets it cancel a newly issued federal rule by passing a joint resolution of disapproval. The Senate passed one on 6 March 2001, the House followed on 7 March, and the President signed it on 20 March as Public Law 107-5. Its operative text is a single sentence: Congress disapproves the Labor Department’s ergonomics rule, and the rule shall have no force or effect. OSHA removed the standard from the Code of Federal Regulations in a notice effective 23 April 2001.

The disapproval had a long tail. Under the act, a rule struck down this way may not be reissued in substantially the same form, and no substantially similar new rule may be issued unless a later law specifically authorizes it. OSHA’s ergonomics FAQ acknowledges that bar. Without new legislation, the agency cannot simply try again with a lightly edited version.

How OSHA pursues ergonomic hazards without a rule

The general duty clause

What remains is section 5(a)(1) of the Occupational Safety and Health Act, known as the general duty clause. It requires each employer to provide employment and a workplace “free from recognized hazards that are causing or are likely to cause death or serious physical harm…” OSHA’s ergonomics FAQ says the agency will use that clause to cite employers for ergonomic hazards, and its nursing-home enforcement page says serious ergonomics hazards in nursing and residential care will be handled the same way. The clause does the same job for workplace heat, as the library’s heat standard article explains.

According to the FAQ, OSHA weighs four questions: whether an ergonomic hazard exists, whether it is recognized, whether it is causing or likely to cause serious physical harm, and whether a feasible way to reduce it exists. Each must be established workplace by workplace, which takes more than pointing to a violated line in a standard. OSHA’s industry guidelines describe good practice, but they are not standards.

What a recent case looked like

In January and February 2023, OSHA announced citations at six warehouses run by a major e-commerce company. The first announcement described workers lifting heavy items at high frequency, twisting, bending and reaching far while lifting, and working long hours. Both said the company was cited under the general duty clause. The two announcements listed proposed penalties of $60,269 and $46,875.

In December 2024, the Labor Department announced a corporate-wide settlement of the ergonomics cases arising from inspections at 10 of the company’s facilities beginning in summer 2022, cases that had been headed for trial before the Occupational Safety and Health Review Commission. That announcement does not name the legal provision behind the citations. The agreement covers all of the company’s fulfillment centers, sortation centers and delivery stations, among other facilities, under federal OSHA jurisdiction.

Its commitments are about process rather than equipment: a corporate ergonomics team, a named ergonomics lead at every site, yearly site risk assessments with feasible controls, training, anonymous ways to raise concerns, compliance tracking, and OSHA monitoring inspections. The term is two years, and either side could seek to end it after one. The penalty was $145,000, which the department said was more than 90% of the amount assessed. Set against a national network, the fine is small; the substance is in the commitments and OSHA’s ability to check them.

States that went further

California’s repetitive motion rule, operative since July 1997, is narrow. It applies only after more than one employee doing identical repetitive work has a repetitive motion injury that a licensed physician has objectively diagnosed, that the job predominantly caused (50% or more), and that was reported within the past 12 months. Then the employer must evaluate the work, control the exposures and train workers.

Minnesota went broader in 2023, with amendments in 2024. Its statute requires a written ergonomics program at hospitals, outpatient surgical centers and nursing homes, and at warehouse distribution centers and meatpacking sites with 100 or more employees. The program must assess risks, train workers, ensure early reporting of symptoms and be evaluated yearly, with feasible engineering or administrative controls where risk exists.

What trials found for lifting training and back belts

Training in technique

The most direct test of lift-with-your-legs training comes from the Finnish Institute of Occupational Health. A 2008 systematic review in The BMJ gathered six randomized trials with 17,720 participants, plus five cohort studies. Eight of the 11 studies involved lifting and moving patients; three involved baggage handlers or postal workers. None of the trial comparisons found a significant difference in back pain or the disability and sick leave that follow it. The authors wrote: “The findings challenge current widespread practice of advising workers on correct lifting technique.”

The same group’s 2011 Cochrane review widened the net to nine trials with 20,101 employees. It found moderate-quality evidence that training, with or without lifting aids, did not prevent back pain or related disability compared with no intervention or alternatives. Its searches ended in February 2011, so it is not recent.

A 2016 meta-analysis in JAMA Internal Medicine, pooling 21 trials with 30,850 people, reached the same verdict on education alone: no effect on the risk of a back pain episode. In the library’s terms, technique training as a way to prevent back pain has been tested and not supported. In a 2020 trial among aged-care workers, which our posture article describes, a posture-feedback device did not even change how often they bent forward past a set limit.

Back belts

In 1994, a working group at NIOSH, the federal institute for workplace safety research, concluded that belts remained unproven at lowering back injury risk in uninjured workers, did not recommend them for that purpose, and did not consider them personal protective equipment. NIOSH’s October 1996 fact sheet repeated the position for workers who have never been injured, and raised two concerns: a false sense of security, and wearers trying to lift more than they otherwise would. Those were stated as risks, not measured outcomes.

NIOSH then ran the large test. Its study, published in JAMA in 2000, drew on 13,873 material handlers in 160 new retail stores across 30 states, 89 of which required belts while 71 left them optional. Neither wearing a belt often nor working in a store that required one was linked to fewer back injury claims or less self-reported back pain. The authors called it the largest prospective cohort study of back belt use. The 2016 meta-analysis agreed: belts did not reduce the risk of a back pain episode, on evidence rated low to very low quality.

One point of history deserves an honest note. NIOSH’s 1994 group recommended a full ergonomics program instead of belts, and its list included ongoing training in lifting mechanics alongside job assessments, symptom surveillance and medical management. The trials since suggest the training piece, on its own, does little to prevent back pain. The job assessment piece is where the engineering lives.

Engineering the lift instead of the lifter

NIOSH’s answer to how heavy is too heavy is the Revised NIOSH Lifting Equation, developed in 1991 and set out in an applications manual revised in September 2021. It starts from a load constant of 51 pounds (23 kilograms), the most the equation allows under ideal conditions, then shrinks that figure for real ones. Some multipliers describe where the load sits: how far the hands are from the body, how high or low they are, and how far the load travels. Others describe the worker’s motion, namely twisting and how often and how long lifting goes on, and one rates the grip.

The output is the Recommended Weight Limit, which the manual defines as a load nearly all healthy workers could handle over a substantial period without added risk of lifting-related low back pain. Divide the actual load by that limit to get the Lifting Index. Above 1.0, the manual says, some fraction of workers faces increased risk; above 3.0, nearly all do.

The multipliers do the real work. Under the formula, holding a load 20 inches from the body instead of 10 halves the recommended limit, and past 25 inches the limit falls to zero. One worked example in the manual, loading stock into a punch press, yields a limit of about 16 pounds at the start of the lift, under a third of the load constant, mainly because of the reach and a low starting height.

Three cautions apply. The 51-pound figure is not a legal limit or an OSHA rule. NIOSH’s founding paper, in the journal Ergonomics in 1993, acknowledged the equation had not been fully validated, while arguing its limits matched or undercut those in the literature. And it does not apply to many real tasks: lifting with one hand, while seated or kneeling, in cramped spaces, at high speed, while carrying, pushing or pulling, or with an unstable load.

What the equation does is shift attention from the worker’s technique to the task. NIOSH’s 1996 fact sheet sketches the redesigns that follow: keep loads close to the body and between knuckle and shoulder height, take the twist out of the lift, bring in slides, hoists and hand trucks, and cut weight as far as feasible.

Patient handling: where laws have been tested

People are the unstable load the equation leaves out. In a 2007 review in the American Journal of Nursing, NIOSH researcher Waters applied it to the narrow case of a cooperative patient unlikely to move suddenly. In general, the paper concluded, the equation yields a maximum of about 35 pounds for patient-handling tasks, with assistive devices for anything heavier. That is one researcher’s application, not formal NIOSH policy, but the gap between 35 pounds and an adult patient is the core argument for lift equipment.

Several states have written that argument into law. A 2026 policy study in the journal New Solutions found that since 2006, 11 states have enacted safe patient handling and mobility policies, and two later repealed them. The policies leaned toward higher-acuity settings over long-term residential care, and toward training and committees over equipment; the authors urged policymakers to put engineering controls first. One author works for a medical device company.

Two before-and-after studies suggest the laws help:

• California. A law that took effect in 2012 requires general acute care hospitals to have a plan to prevent patient-handling injuries. A University of California, San Francisco analysis of 199,547 workers’ compensation claims for musculoskeletal disorders, published in 2022, found that from 2011 to 2016 the patient-handling claim rate among hospital workers fell 7.3% a year. Nursing and residential care facilities, which the law did not cover, saw no significant change.
• Minnesota. After the state’s 2007 Safe Patient Handling Act, nursing-home claims for lost wages from patient-handling injuries fell 25% in years four to six and 38% in years seven to nine, according to a University of Minnesota study published in 2021. Homes that kept at least 75% of their staff each year had 17% lower claim rates than those keeping under 65%.

Neither is a randomized trial, and other things may have changed over the same years. The Cochrane trials, meanwhile, found that training plus lifting aids did not prevent self-reported back pain. The two bodies of evidence measured different things, compensation claims in one case and reported pain in the other, under different programs. Together they lean toward equipment backed by enforceable programs over technique lessons alone, with less certainty than anyone would like.

Questions to ask about a job’s lifting and repetition

These questions are this article’s own, built from the NIOSH equation and the programs above. They are meant for understanding a job, not as legal advice.

1. What are the heaviest and the most common loads, and where do they start: the floor, waist height, overhead?
2. How far from my body will I hold them, and will I twist or reach while holding them?
3. How many lifts or repetitions in an hour, and for how many hours a shift?
4. What lifting aids are on site, such as carts, hoists, slides or patient lifts, and are they close enough and staffed well enough to use?
5. Has anyone assessed this job’s lifting risk, and when was that last updated?
6. How do I report early aches or numbness, and is there an anonymous route?
7. Is lifting training the whole program, or one part of one?

When to get help

Some symptoms point past ordinary soreness. The National Institute of Arthritis and Musculoskeletal and Skin Diseases says to see a doctor for back pain that does not improve, or that comes with tingling or numbness, weakness or numbness in the legs, trouble urinating, fever, unexplained weight loss, or follows a fall or injury. Hands that go numb or grow clumsy deserve attention too; our neck pain article explains how that pattern can start in the spine rather than the wrists. New loss of bladder or bowel control, or numbness in the groin and inner thighs, alongside back pain calls for same-day emergency care, as the back pain article sets out.

A clinician can weigh whether symptoms fit the work you do; a plain description of tasks, loads and hours helps. Whether an injury is reported as work-related, and what that means for a claim, is a separate question for your state’s workers’ compensation agency. Changes to a job for someone with a lasting condition are taken up in a later article on accommodations.

The most useful thing to know

The weight of the evidence sits with the job, not the worker’s spine. Technique training and back belts have been tested on tens of thousands of workers without preventing back pain. What the engineering points to is the lift itself: how far the load sits from the body, how low it starts, how much twisting it takes, how often it happens, and whether a machine can do it instead.

If you have a say in one change, make it the reach. Bringing a load from 20 inches out to 10 doubles the weight the NIOSH equation treats as reasonable, and that gain does not depend on anyone remembering a lesson.

Keep exploring: posture and back pain and physical therapy setting.

Sources

1. U.S. Bureau of Labor Statistics. “Employer-Reported Workplace Injuries and Illnesses, 2023-2024 (news release USDL-26-0101, January 22, 2026; Table 2, private industry).” bls.gov
2. U.S. Bureau of Labor Statistics. “Survey of Occupational Injuries and Illnesses: case and demographic characteristics, historical data (MSD tables 2016-2020).” Www.Bls.gov
3. Occupational Safety and Health Administration. “Worker Safety in Your Hospital (OSHA 3689, September 2013; 2011 data).” osha.gov
4. Occupational Safety and Health Administration. “Ergonomics (industry-specific guidelines listed).” osha.gov
5. U.S. Government Publishing Office (govinfo). “Public Law 107-5 (S.J. Res. 6), approved March 20, 2001: disapproval of the Department of Labor ergonomics rule.” govinfo.gov
6. Occupational Safety and Health Administration (Federal Register). “Ergonomics Program; final rule, removal (66 FR 20403, effective April 23, 2001).” osha.gov
7. Legal Information Institute, Cornell Law School. “5 U.S. Code 801, Congressional review (subsection (b)).” law.cornell.edu
8. Legal Information Institute, Cornell Law School. “29 U.S. Code 654, Duties of employers and employees (section 5(a)(1) of the OSH Act, the general duty clause).” law.cornell.edu
9. Occupational Safety and Health Administration. “Ergonomics: Standards and Enforcement FAQs.” osha.gov
10. Occupational Safety and Health Administration. “Nursing Homes and Personal Care Facilities: OSHA Enforcement.” osha.gov
11. U.S. Department of Labor, Occupational Safety and Health Administration. “National News Release 23-63-NAT, January 18, 2023: ergonomic-hazard citations at three warehouses under the general duty clause.” dol.gov
12. U.S. Department of Labor, Occupational Safety and Health Administration. “National News Release, February 1, 2023: general duty clause citations at three more warehouses.” dol.gov
13. U.S. Department of Labor, Occupational Safety and Health Administration. “National News Release 24-2583-NAT, December 19, 2024: corporate-wide ergonomics settlement.” dol.gov
14. Legal Information Institute, Cornell Law School (California Code of Regulations). “Cal. Code Regs. tit. 8, section 5110, Repetitive Motion Injuries (operative July 3, 1997).” law.cornell.edu
15. Minnesota Office of the Revisor of Statutes. “Minnesota Statutes 182.677, Ergonomics (2023 c 53 art 1 s 21; 2024 c 110 art 3 s 7,8).” revisor.mn.gov
16. The BMJ. “Martimo KP, Verbeek J, Karppinen J, et al. "Effect of training and lifting equipment for preventing back pain in lifting and handling: systematic review." BMJ, 2008. PubMed. DOI: 10.1136/bmj.39463.418380.BE.” pubmed.ncbi.nlm.nih.gov
17. Cochrane Database of Systematic Reviews. “Verbeek JH, Martimo KP, Karppinen J, Kuijer PP, Viikari-Juntura E, Takala EP. "Manual material handling advice and assistive devices for preventing and treating back pain in workers." Cochrane Database of Systematic Reviews, 2011. PubMed. DOI: 10.1002/14651858.CD005958.pub3.” pubmed.ncbi.nlm.nih.gov
18. JAMA Internal Medicine. “Steffens D, Maher CG, Pereira LS, et al. "Prevention of Low Back Pain: A Systematic Review and Meta-analysis." JAMA Internal Medicine, 2016. PubMed. DOI: 10.1001/jamainternmed.2015.7431.” pubmed.ncbi.nlm.nih.gov
19. Occupational and Environmental Medicine. “Ribeiro DC, Milosavljevic S, Terry J, Abbott JH. "Effectiveness of a lumbopelvic monitor and feedback device to change postural behaviour: the ELF cluster randomised controlled trial." Occupational and Environmental Medicine, 2020. PubMed. DOI: 10.1136/oemed-2019-106293.” pubmed.ncbi.nlm.nih.gov
20. National Institute for Occupational Safety and Health (CDC Stacks). “Workplace Use of Back Belts: Review and Recommendations, DHHS (NIOSH) Publication No. 94-122 (May/July 1994).” stacks.cdc.gov
21. National Institute for Occupational Safety and Health. “Back Belts: Do They Prevent Injury? DHHS (NIOSH) Publication No. 94-127 (October 1996).” cdc.gov
22. JAMA. “Wassell JT, Gardner LI, Landsittel DP, Johnston JJ, Johnston JM. "A prospective study of back belts for prevention of back pain and injury." JAMA, 2000. PubMed. DOI: 10.1001/jama.284.21.2727.” pubmed.ncbi.nlm.nih.gov
23. National Institute for Occupational Safety and Health (CDC Stacks). “Applications Manual for the Revised NIOSH Lifting Equation, DHHS (NIOSH) Publication No. 94-110 (revised September 2021).” stacks.cdc.gov
24. Ergonomics. “Waters TR, Putz-Anderson V, Garg A, Fine LJ. "Revised NIOSH equation for the design and evaluation of manual lifting tasks." Ergonomics, 1993. PubMed. DOI: 10.1080/00140139308967940.” pubmed.ncbi.nlm.nih.gov
25. American Journal of Nursing. “Waters TR. "When is it safe to manually lift a patient?" American Journal of Nursing, 2007. PubMed. DOI: 10.1097/01.NAJ.0000282296.18688.b1.” pubmed.ncbi.nlm.nih.gov
26. New Solutions: A Journal of Environmental and Occupational Health Policy. “Moskowitz AF, Doran E, Ramirez M, Tschida B, Wiggermann N, Schofield K. "Policy Surveillance of Safe Patient Handling and Mobility Laws to Reduce Injury among Healthcare Workers." New Solutions, 2026. PubMed. DOI: 10.1177/10482911261447546.” pubmed.ncbi.nlm.nih.gov
27. American Journal of Industrial Medicine. “Lee SJ, Lee JH, Harrison R. "Safe patient handling legislation and musculoskeletal disorders among California healthcare workers: Analysis of workers' compensation data, 2007-2016." American Journal of Industrial Medicine, 2022. PubMed. DOI: 10.1002/ajim.23366.” pubmed.ncbi.nlm.nih.gov
28. Occupational and Environmental Medicine. “Rosebush CE, Zaidman B, Schofield KE, et al. "Evaluation of the Minnesota Safe Patient Handling Act: trends in workers' compensation indemnity claims in nursing home workers before and after enactment of the law." Occupational and Environmental Medicine, 2021 (online 2020). PubMed. DOI: 10.1136/oemed-2020-106575.” pubmed.ncbi.nlm.nih.gov
29. National Institute of Arthritis and Musculoskeletal and Skin Diseases (National Institutes of Health). “Living With Back Pain: Health Information Basics for You and Your Family.” niams.nih.gov