unhealthy work logo, unhealthywork.org



Paul A. Landsbergis, EdD, MPH
Cornell University Medical College
Janet Cahill, PhD
Rowan College of New Jersey
Peter Schnall, MD, MPH
University of California at Irvine
International Congress on Occupational Health
Stockholm, Sweden
September 17, 1996


We live in a period of time in which new systems of work organization are being introduced throughout the industrialized world — raising the question of whether these systems impact on the prevalence of job strain (that is, work combining high demands & low control). We are intrigued by this question due to our own work, and that of our colleagues (for example, Bob Karasek, Tores Theorell, Tage Kristensen, and Jeff Johnson), linking job strain with hypertension and heart disease.

In Bob and Tores’ 1990 book “Healthy Work” they not only discuss the evidence linking job strain and illness, but also experiments in work organization which might possibly create healthier jobs — for example, QWL programs, Scandinavian socio-technical systems, or Japanese management techniques, also known as “lean production”. They described the debate around these systems, and posed a question about lean production.

They said: “The unresolved question is the amount of control really afforded” to workers in Japanese-owned plants in the U.S. In other words, do workers have greater or lesser job decision latitude, or more or less job strain and risk of illness under lean production than under other systems?

The 1990 text from MIT which had assessed lean production in auto manufacturing, “The Machine that Changed the World”, had argued that, in the best Japanese auto companies, “multiskilled” workers can solve quality problems at their source and boost productivity. The “freedom to control one’s work” replaces the “mind numbing stress” of mass production. Armed with “the skills they need to control their environment,” workers in a lean plant have the opportunity “to think actively, indeed proactively,” to solve workplace problems. This “creative tension” makes work “humanly fulfilling”.

Much has subsequently been written about this subject but, until recently, very few empirical studies had been conducted. Today, we have some data than can help to answer this “unresolved” question about lean production’ impact on job dimensions and health.

Lean production is an attempt to reduce impediments to the smooth flow of production through various techniques, including continuous improvement in productivity and quality (“kaizen”), “just-in-time” (JIT) inventory systems, and training workers to conduct quality control through Quality Circles.

Here we provide, from earlier case studies, models of the extent of job decision latitude in lean production, in socio-techncial systems, and under the U.S. human resources approach.

Some common models of “New Work Systems”

Lean Production Human Resources Socio-Technical Systems
ORIGIN (Japan) (U.S.) (England, Scandinavia)
*WORKER TEAMS Supervised No Self-directed
*CYCLE TIME Short Short Long
VERTICAL TASKS Just QC process No Yes
KNOWLEDGE DEPTH Trade off depth for greater breath Limited Team characteristic, not individual
UNIONS Company unions Anti-union Strong unions

Source: Appelbaum E, Batt R. The new American workplace. Ithica, NY: ILR Press, 1994

Lean production does seem to provide for more job enlargement, cross-training, and problem-solving opportunities than traditional manufacturing job design. However, quality circles are not autonomous work teams nor are they empowered to make managerial decisions. Cycle-time in lean production is still very short (1 min. in auto assembly). Following highly standardized steps at each narrowly defined task is mandatory.

Reliance is placed on time studies and standards to ensure maximum workloads (e.g., working 57 sec. out of a min. vs 40-45 sec. in a traditional plant). “Just-in-time” (JIT) inventory systems remove the stock between operations as the buffers in the system. Instead, workers’ personal time and flexibility become buffers — long work hours and involuntary overtime are common.

However, many earlier reports promoting lean work have not discussed the potential for work stress (due to high demands and limited latitude), or injuries and illness. While the debate continues on lean pro-duction and important issues such as productivity, quality, labor relations, worker solidarity, the spread of low-wage, supplier jobs, we focus today on a critical but relatively neglected area — that of employee injuries, such as upper extremity musculoskeletal disorders (UEMDs), and on job characteristics related to job strain (job demands, control, social support).

We also focus on this issue since lean work and other team-based methods are spreading to many industries. Thus, a careful empirical assessment of the impact of new work systems on job characteristics is critical in understanding trends in worker health not just in auto manufacturing production but throughout Western economies.


The surveys and case studies we review today are primarily from North America and include several very recently conducted surveys. We do not claim that this is an exhaustive list on this topic, rather major surveys and case studies are included. We carefully reviewed each study and rated associations between lean work, job characteristics and health outcomes. We are expecting to receive additional studies soon. We would also appreciate learning of any additional studies that you are aware of. So, what do these studies indicate?

Lean production in auto manufacturing, upper extremity musculoskeletal disorders (UEMDs) and stress • 3 surveys by Canadian Auto Workers, 1993-1996
• 4 U.S. case studies, 1993-1996 Lean production in auto manufacturing and job characteristics • 3 surveys by Canadian Auto Workers, 1993-1996
• 1 survey by United Steelworkers in Canada, 1995
• 1 survey by United Auto Workers in Michigan, 1993
• 2 U.S. case studies, 1991-1995 New work systems in other industries and job characteristics

• 5 surveys and 3 case studies from electronics, garment, health care and other industries in England and the U.S


Associations between lean production in auto manufacturing and:

Sample N UEMDs fatigue, stress, tension
Canadian surveys:
16 Auto Parts Suppliers 1,670 0 ?
CAMI (GM-Suzuki) 100 (4x) +
CAMI, GM, Ford, Chrysler 2,424 ? ?
U.S. case studies:
Auto Alliance (Mazda-Ford) +
6 Japanese “transplants” + +
NUMMI (GM-Toyota) +
Subaru-Isuzu + +

0 = no association, ? = equivocal association, + = modest association, ++ = strong association

Authors of case studies of Japanese-owned auto assembly plants in the U.S. (including the 1990 visit to 6 such “transplants” by Christian Berggren and his colleagues here at the Swedish Royal Institute of Technology) reported high levels of perceived stress and of musculoskeletal disorders, such as tendinitis and carpal tunnel syndrome — due, in their opinion, to the fast work pace, long work hours, highly repetitive work, and limited rest breaks. Other factors mentioned were lack of early treatment due to peer pressure to “work in pain” and not report injuries, and the denial of work-relatedness by management — the comment that “there are weak and there are strong people”.

While we could locate no studies which specifically assessed the prevalence or incidence of musculoskeletal disorders in lean workplaces, the 3 surveys conducted by the Canadian Auto Workers Union suggest a worsening problem. Among the auto parts supplier workers, survey data indicated that increases in job tension and fatigue over the past 2 years were significantly greater in lean companies than in traditional companies.

67% of workers at GM (the most lean of the 4 companies surveyed) reported working in pain at least half of the days in the last month. At the Japanese joint venture, CAMI, 42% of workers reported working in pain. Across the 4 companies, a majority reported more job tension and fatigue than several years ago.

Associations between lean production in auto manufacturing and :

High job demands Low decision authority Poor skill development
Canadian surveys:
16 Auto Parts Suppliers ++ ? 0
CAMI (GM-Suzuki) ? + +
CAMI, GM, Ford, Chrysler + +
Walker Exhaust 0 0 0
U.S. survey:
Auto Alliance (Mazda-Ford) + + +
U.S. Case studies:
6 Japanese “transplants” +
Subaru-Isuzu (Indiana) + +

0 = no association, ? = equivocal association, + = modest association, ++ = strong association

More detailed information is available from these surveys on the job characteristics which comprise “job strain”. Job demands were fairly consistently reported to be elevated. Among the auto parts workers, those in lean companies reported significantly heavier workload, and “too few people”, than in traditional companies. 79% of GM workers and 54% of Ford workers said they could “not keep up without working as fast as you can”. 73% of Mazda workers reported “I will likely be injured or worn out before I retire”.

Low or decreasing decision authority was also consistently reported. At CAMI, while 51% reported at Survey 1 being “actively involved in making decisions at work”, 1.5 years later at Survey 4 this proportion fell to 33%. The role of teams in providing a place for “influence over the job” similarly fell from 79% to 45%. During early start-up periods, “participation” and job rotation was practiced, but during production, dissat-isfaction rose. At survey 4, 60% reported that teams “get us all to pressure one another” compared to only 19% at Survey 1. At Mazda, participation was also very limited. As the local union president put it, “They promised us a rose garden. They gave us a desert.”

The promise of lean production in producing highly trained “multi-skilled” workers was also challenged by the survey data. At CAMI, the proportion of workers who felt that “as as a result of training, I’m no more skilled since all the jobs are about the same” rose from 53% at Survey 1 to 72% at Survey 4. At Mazda, 89% of respondents said that “the actual training I have received is fair or poor” — only 6% reported it as excellent.

Associations between new work systems in other industries and:

Sample N Psych strain Low Decision authority Job demands
Cross-sectional surveys:
123 companies (Pennsylvania) 537 0 0
British prospective surveys:
Car seat manufacturers 35 +1 0 +1
Electronics assemblers 56 0 (increase in control) ?
Electronics assemblers 44 0 ? ?
U.S. Case studies:
Patient-focused care (health care) ?2 (increased authority for aides – if staffing reasonable ?2
Modular manufacturing (garment) + (some increase in authority) 0

1if no worker participation in implementation of system
if understaffing
0 = no association, ? = equivocal association, + = modest association, ++ = strong association

Some additional small-scale surveys have also been conducted in other industries. The British surveys show similar (although not always consistent) effects of JIT work systems as the North American auto plant surveys.

In the U.S. health care industry, experiments in team-based decentralized work have been called “patient-focused care”. Surveys show some increase in decision-making authority and job satisfaction, mainly for nurse aides — if staffing levels are kept reasonable. Professional nurses (RNs) are concerned since some of their work is being transfered to poorly trained lower-paid aides — and since such programs are often designed to save health care costs by eliminating professional nurse positions — causing understaffing and greater stress for nurses.

Responses of workers/unions to lean production

• 5-week strike (Canadian Auto Workers – CAMI, 1992)
• 2-hour strike (UAW – NUMMI, 1994)
• Strike threat (UAW – Mazda, 1991)
• OSHA inspection and settlement (UAW – NUMMI, 1992-4, as well as GM, Ford and Chrysler)
• Surveys to document conditions
• Collective bargaining

It was apparent that, in unionized facilities, it was impossible to study a “pure” form of lean production. In most of the cases we examined, collective worker resistance to the stress of lean production occurred — through strikes, strike threats, collective bargaining, and requesting government safety inspections.

Modification of lean production through union efforts

• Improved staffing through a Temporary Assignment Pool of workers to fill in for absent or injured workers. (Undermines 1 purpose of teams: peer pressure to discourage absenteeism, encourage working while injured.)
• Fairer access to training
• Increased transfers between departments
• Joint committees on health and safety, ergonomics, training
• The right to elect or recall team leaders
• Team leaders’ duties in contract
• Increasing work loads due to absenteeism is forbidden
• Temporary assignments offered to workers on the basis of seniority
• Some limits on line speeds and job standards
• Ergonomics programs:
Job analysis, worksite evaluations
Engineering controls, Task rotation, Rest periods
Review models before production
Better medical management
Reduce line speed
No reprisals for reporting injuries

These efforts have “humanized” lean work to some extent — through moderated work demands (more staff, and control over line speed and job standards), increased latitude (electing team leaders, ability to transfer, joint committees), fairer access to training, and ergonomics programs.


Recent surveys tend to confirm some earlier case studies that lean production does create an intensified work pace and work demands, and that promised increases in decision authority and skill levels are very modest. Decision latitude remains low. Thus, such work can be considered to have high job strain.

In jobs with physical ergonomic stressors (such as auto manufacturing), this intensification of labor appears to lead to increased musculoskeletal injuries.

Since fast-paced repetitive short-cycle work with few rest breaks, long work hours and low authority, are risk factors not only for musculoskeletal injuries — they also, in effect, define job strain — we suggest that high rates of musculoskeletal injuries in lean workplaces may be a warning sign of chronic job strain. Therefore, they may well be predictors of illnesses with longer latencies — such as hypertension and heart disease.


Lean production leads to increased job demands and limited latitude — a formula for “job strain”
• In occupations with physical ergonomic stressors, this formula increases risk of musculoskeletal disorders
• In lean workplaces, musculoskeletal disorders may also be a warning sign of future hypertension
• Forms of lean production are being promoted throughout the economy
• Unions are a countervailing force which can moderate the stress of lean production
• Alternatives to lean production exist
• Much additional research is needed

If we are correct in our analysis, the potential for dramatically increased rates of hypertension and heart disease exists if lean methods (including outsourcing, longer work hours, understaffing and insecure work) are widely implemented in the labor market. Forms of lean production are by no means limited to Japanese-owned companies nor just to manufacturing, but are promoted in many sectors, including health care.

We have also provided some evidence today that unions are a countervailing force to this trend — that they have modified, humanized to some extent, the lean production system.

Alternatives to lean production exist. Christian Berggren has written persuasively about socio-technical efforts in Scandinavia and the increased latitude provided by autonomous work groups. The Saturn auto plant in the U.S. also appears to be modeled more on Scandinavia than lean production.

However, many questions remain. We need better data from prospective studies. We need to know the impact of lean production on:

1) Worker skills. Is it memorizing standardized company-specific procedures or true skill development? Who controls access to training or the content?
2) Co-worker support and worker solidarity? Is peer pressure used increase work pace?
3) On health in non-union workplaces, especially supplier firms.
4) On workforces that are older and not so highly selected.
5) What is the time period of the effect of lean methods? In some cases, employees were very satisfied with their participation in planning production. However, once the system was in operation, they felt betrayed by the speed-up and lack of participation.

We also need better data on alternatives, such as socio-technical systems:

1) Are there alternatives which promote healthy work — which reduce job strain, injuries, or blood pressure?
2) What about the health impact of related work systems such as TQM, re-engineering, modular or cellular manufacturing, or agile manufacturing?

Studies reviewed

Surveys co-sponsored by the Canadian Auto Workers

Lewchuk W, Robertson D. The quality of worklife under lean production: A human centered benchmarking study (unpublished manuscript). Hamilton, Ontario: McMaster University, 1995.

Lewchuk W, Roberts B, McDonald C, Robertson D. Working conditions study: Benchmarking auto assembly plants. Willowdale, Ontario: Canadian Auto Workers, 1996.

Rinehart J, Huxley C, Robertson D. Team concept at CAMI. In: Babson S (ed.) Lean Work: Empowerment and Exploitation in the Global Auto Industry (pp. 220-233). Detroit: Wayne State University Press, 1995.

Robertson D, Rinehart J, Huxley C, Wareham J, Rosenfeld H, McGough A, Benedict S. The CAMI report: Lean production in a unionized auto plant. Willowdale, Ontario: Canadian Auto Workers, 1993.

Survey by the United Steelworkers in Canada

Murphy C, Olthuis D. The impact of work reorganization on employee attitudes towards work, the company and the union. In: Schenk C, Anderson J (eds.) Re-shaping Work: Union Responses to Technological Change (pp 76-102). Don Mills, Ontario: Ontario Federation of Labor, Technological Adjustment Research Programme, 1995.

U.S. survey of autoworkers

Babson S. Lean or mean: The MIT model and lean production at Mazda. Labor Studies Journal 1993;18:3-24.

U.S. case studies of autoworkers

Adler P. “Democratic Taylorism”: The Toyota Production System at NUMMI. In: Babson S (ed.) Lean Work: Empowerment and Exploitation in the Global Auto Industry (pp. 207-219). Detroit: Wayne State University Press, 1995.

Adler P. Ergonomics, employee involvement, and the Toyota Production System: A case study of NUMMI’s 1993 model introduction. Los Angeles: University of Southern California, 1996 (manuscript).

Berggren C, Bjorkman T, Hollander E. Are they unbeatable? Report from a field trip to study transplants, the Japanese owned auto plants in North America. Stockholm: Royal Institute of Technology, 1991.

Division of Occupational Safety and Health. Citation and Notification of Penalty (Citation R1D4-4014). Department of Industrial Relations, State of California, January 6, 1993.

Division of Occupational Safety and Health. Special Order (NUMMI inspection). Department of Industrial Relations, State of California, January 18, 1994.

Graham L. On the Line at Subaru-Isuzu: The Japanese Model and the American Worker. Ithaca, New York: ILR Press, 1995.

Surveys of other industries

Jackson PR, Martin R. Impact of just-in-time on job content, employee attitudes and well-being: a longitudinal study. Ergonomics 1996;39:1-16.

Mullarkey S, Jackson PR, Parker SK. Employee reaction to JIT manufacturing practices: A two-phase investigation. International Journal of Operations and Productions Management (in press).

Parker SK. How do modern manufacturing systems affect shopfloor jobs and well-being? Presented at the APA/NIOSH Conference on Occupational Stress, Washington, DC, September 13-16, 1995.

Parker SK, Jackson PR, Wall TD. Autonomous group working within integrated manufacturing: A longitudinal investigation of employee role orientations (pp 44-49). In: Salvendy G, Smith MJ (eds.) Human-Computer Interaction: Application and Case Studies. Amsterdam: Elsevier, 1995.

Parker SK, Wall TD. Job design and modern manufacturing. In: Warr PB (ed.) Psychology at Work. London: Penguin Books (in press).

Parker SK, Wall TD, Myers C. The effects of a manufacturing initiative on employee jobs and strain. In: Robertson SA (ed.) Contemporary Ergonomics 1995. London: Taylor and Francis (in press).

Case studies of other industries

Greiner A. Cost and quality matters: Workplace innovations in the health care industry. Washington, DC: Economic Policy Institute, 1995.

Richardson T. Reengineering the hospital: Patient-focused care. In: Parker M, Slaughter J. Working smart (pp. 113-120). Detroit: Labor Education and Research Project, 1994.

Wunderlich GS, Sloan FA, Davis CK (Eds.). Nursing staff in hospitals and nursing homes: Is it adequate? Washington, DC: National Academy Press, 1996.

Comments are closed.

© 2023 Unhealthy Work  |  For more information regarding this site, please contact us