Article Text


Short report
Heavy manual work throughout the working lifetime and muscle strength among men at retirement age
  1. K Walker-Bone1,2,
  2. S D'Angelo1,
  3. H E Syddall1,
  4. K T Palmer1,2,
  5. C Cooper1,2,3,4,
  6. D Coggon1,2,
  7. A A Sayer1,3,5,6,7
  1. 1MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton, UK
  2. 2Arthritis Research UK/MRC Centre for Musculoskeletal Health and Work, University of Southampton, Southampton, UK
  3. 3NIHR Southampton Biomedical Research Centre, University of Southampton and University Hospital Southampton NHS Foundation Trust, Southampton, UK
  4. 4NIHR Musculoskeletal Biomedical Research Unit, University of Oxford, Southampton, UK
  5. 5Academic Geriatric Medicine, Faculty of Medicine, University of Southampton, Southampton, UK
  6. 6NIHR Collaboration for Leadership in Applied Health Research and Care, Wessex, UK
  7. 7Newcastle University Institute for Ageing and Institute of Health & Society, Newcastle University, Newcastle upon Tyne, UK
  1. Correspondence to Dr K Walker-Bone, MRC Lifecourse Epidemiology Unit, Southampton General Hospital, Tremona Road, Southampton SO16 6YD, UK; kwb{at}


Introduction Reductions in heavy manual work as a consequence of mechanisation might adversely impact muscle strength at older ages. We investigated the association between grip strength at retirement age and lifetime occupational exposure to physically demanding activities. Grip strength is an important predictor of long-term health and physical function in older people.

Methods Grip strength (maximum of three readings in each hand) was measured in men from the Hertfordshire Cohort Study at a single examination when their mean age was 65.8 (SD 2.9) years. Associations with lifetime occupational exposure (ascertained by questionnaire) to three activities (standing/walking ≥4 h/day; lifting ≥25 kg; and energetic work sufficient to induce sweating) were assessed by multivariable linear regression with adjustment for various potential confounders.

Results Complete data were available from 1418 men who had worked for at least 20 years. After adjustment for age, height and weight, those with longer exposures to walking/standing and heavy lifting had lower grip strength, but the relationship disappeared after further adjustment for confounders. Working at physical intensity sufficient to induce sweating was not significantly associated with grip strength.

Conclusions We found no evidence that physically demanding occupational activities increase hand grip strength at normal retirement age. Any advantages of regular physical occupational activity may have been obscured by unmeasured socioeconomic confounders.

  • Muscle strength

Statistics from

What this paper adds

  • Stronger grip strength in later life is associated with less disability, morbidity and mortality.

  • Participation in leisure-time physical activity is positively associated with grip strength.

  • It is currently unknown if lifetime physical workplace activities convey similar benefit.

  • We have shown that men undertaking heavy lifting, prolonged walking/standing and/or energetic activities sufficient to induce sweating over prolonged periods of their working life had lower grip strength than men doing less of these activities.

  • Heavy work is not associated with greater grip strength at retirement age, perhaps because men working in these jobs are disadvantaged in other ways.


As workplaces modernise, exposure to heavy manual work is reducing, but it is unclear what, if any, impact this might have on the long-term health of workers. One measure of health and longevity in older people is grip strength. Epidemiological studies have shown that stronger grip strength in later life is associated with less disability, morbidity and mortality.1 ,2 Leisure time physical activity is an important positive determinant of muscle strength3–6 and, therefore, it could be that intensive occupational physical activity might also convey benefit for muscle strength in later life.

The aim of the current study was to test the hypothesis that cumulative lifetime manual work has a measureable benefit on grip strength in men at normal retirement age.


Most aspects of the study design have been described elsewhere.7 Briefly, during the period 1998–2003, we surveyed 1675 men born in Hertfordshire, UK, 1931–1939, and still living there. Data were collected by questionnaire on (among other things) diet, smoking, age left full-time education, social class, physical activity (the Dallosso activity inventory)8 and all jobs held for >1 year. For each job, participants were asked whether an average working day entailed standing/walking for ≥4 h, lifting weights ≥25 kg or energetic activity sufficient to cause sweating.

Grip strength was measured using a Jamar hand-held dynamometer according to a standardised protocol. Each participant made three attempts using each hand and the maximum of the six measurements was determined.

Statistical analysis was performed with Stata V.13.1 (StataCorp, College Station, Texas, USA). We focused on the duration of exposure (years) to three occupational risk factors: standing/walking for ≥4 h/day, lifting weights ≥25 kg and energetic physical work sufficient to cause sweating. Each variable was derived by summing the times spent in relevant jobs (with a correction factor of 0.5 for part-time employment), and non-zero values were classified to approximate thirds of their distribution in the study sample. Associations of maximum grip strength with these and other risk factors were assessed by linear regression and characterised by estimates of mean difference in grip strength with 95% CIs from that in a reference category of low exposure. For each occupational risk factor, estimates were derived with no adjustment (model 1); adjusted for age, height and weight-adjusted-for-height residuals, which are known to be importantly associated with grip strength (model 2); adjusted for age, height, weight-adjusted-for-height residuals and self-rated health (model 3); and adjusted for age, anthropometry, social class (at birth and current), smoking, diet, leisure-time physical activity and age left full-time education (model 4).

We repeated the same process to derive the number of years spent doing each of the activities when aged <40, 41–54 and >55 years. These variables were not normally distributed, so Spearman correlation coefficients were computed to assess the inter-relation between occupational activities across the working lifetime.

Ethical approval was provided by the East and North Hertfordshire NHS Research Ethics Committees, and all patients gave written informed consent.


Analysis was based on 1418 men who provided complete information on grip strength, exposures during ≥20 years at work and potential confounders. A further 257 men with incomplete data had similar grip strength but were slightly younger and more likely to rate their health ‘fair/poor’ (20.4% vs 11.9%, p<0.001). Those analysed had a mean age of 65.8 (SD 2.9) years, and 66.5% were current or past cigarette smokers. Approximately one-fifth (19.5%) had left full-time education before 14 years of age, and only 27 (1.9%) were still in full-time work. Their mean hand grip strength was 44.0 (SD 7.6) kg.

Most men (93%) had held jobs entailing ≥4 h standing/walking/day, the reported durations of such employment ranging from 2 to 47 years. However, 20.2% reported that they had never lifted weights ≥25 kg at work, and 26.6% reported their work was never sufficiently energetic to cause sweating. We found that men whose work involved an activity between ages 41 and 54 years were significantly more likely to be doing the same activity at younger and older ages (all p<0.001). Men reporting the longest durations of physically demanding activities were more likely to have left full-time education age <14 years and were considerably over-represented in social classes IV–V. They tended also to be more active outside work.

In unadjusted analyses, the longest durations of exposure (standing/walking ≥4 h/day for >47 years; (β=−1.48, 95% CI −2.47 to −0.49 kg), lifting weights ≥25 kg for >46 years (β=−1.43, 95% CI −2.53 to −0.32 kg) and doing energetic work sufficient to induce sweating for >44 years (β=−1.17, 95% CI −2.30 to −0.04 kg)) were negatively associated with grip strength when compared with the shortest non-zero durations of exposure. Trends suggested an exposure–response relationship, although only standing/walking showed a statistically significant negative relationship to grip strength when the group with moderate exposure (36.5–47 years) was compared with those with the lowest non-zero levels of exposure (β=−1.0, 95% CI −1.99 to −0.01 kg).

After adjustment for age and anthropometry, the crude associations were attenuated (figure 1). However, men with higher exposures to walking/standing and heavy lifting again tended to have lower grip strength, and for moderate exposures, the differences from the reference were statistically significant (β=−0.92, 95% CI −1.82 to −0.01 kg, and β=−1.22, 95% CI −2.17 to −0.26 kg, respectively). After additional adjustment for self-rated health, the associations disappeared. Similarly, adjustment for socioeconomic variables instead of self-rated health eliminated the associations. Thus, in model 4 (adjusted for age, anthropometry, social class (at birth and current), smoking, diet, leisure-time physical activity and age left full-time education), there were no statistically significant associations with grip strength.

Figure 1

Mean differences in grip strength (kg) according to cumulative lifetime exposure to physically demanding occupational activities.


We found that grip strength at retirement age was not positively associated with cumulative lifetime exposure to heavy physical work, and if anything, there was some suggestion of the converse.

Grip strength is a simple, cheap, reproducible and widely used clinical measure of muscle strength. It is the strength of the current study that we used an evidence-based, validated measurement protocol. However, the limitations of our data should be acknowledged. The ascertainment of occupational exposures, although more complete and detailed than in other studies, and encompassing jobs held across the whole working lifetime, was dependent on recall. There is no reason to expect that participants with higher grip strength would systematically remember activities less completely than those with weaker grip strength, but non-differential errors of recall may have biased risk estimates towards the null.

Our findings are consistent with those reported by others.6 ,9–11 Although Russo et al9 measured grip strength considerably later in life (mean age 86 years) and classified work as manual or non-manual according to recall of the occupation held for the longest duration during working life, they reported significantly poorer grip strength among manual workers (28.2 kg) than non-manual workers (32.5 kg). Another study including 89 male and female Finnish workers (mean age 51.9 years) found that those with lowest perceived physical workload had stronger mean grip strength than those who perceived their workload as physically demanding.10 Furthermore, when some of these workers were followed-up over 16 years, mean grip strength declined, but the grip strengths of those with higher perceived baseline workload tended to remain weaker.10 Among 902 metal workers, recruited in 1973 and followed-up in 2000 (when mean age was 62.6 years), follow-up grip strength was not measured but significantly lower scores for physical performance on the SF-36 physical functioning scale were found among those reporting the most strenuous work at baseline, with an exposure–response relationship.6 These findings were robust after adjustment for baseline grip strength. Furthermore, better physical function at follow-up was associated with more vigorous leisure-time physical activity at baseline, and this protective effect was independent of the negative effect of work strenuousness. A large Danish study reported no association between grip strength at age 59 years and occupational heavy lifting or duration of standing but reported a weakly positive association with years of kneeling among men, which they hypothesised might be explained by use of hand-held tools in occupations involving kneeling (eg, carpenters and plumbers).11 Our study did not collect data on kneeling at work.

The apparent paradoxical difference in associations with leisure-time physical activity as compared with work might be explained if people who are very physically active at work participate less often in vigorous leisure-time physical activity. However, we found no evidence to suggest that the two were inversely associated. Occupational physical demands are different from leisure-time physical activities. The intensity, nature, duration and type of physical activity in leisure time is under the control of the individual, whereas occupational demands are often of longer duration, may be monotonous and repetitive, and are often outwith the individual's control. Moreover, such demands may be accompanied by other disadvantages of blue-collar employment such as working in adverse weather conditions, receiving lower pay and having less autonomy.

We found strong associations of physically demanding work with manual social class and leaving education before 14 years of age, and other researchers have found evidence suggesting a social gradient in grip strength.12 ,13 It may be that manual occupations are associated with poorer health through effects of social inequalities, poor housing and diet/lifestyle choices, such that any benefit of occupational activity is outweighed by a greater negative impact of these other factors.12 Our finding that adjustment for both social class and self-rated health reduced the inverse associations between grip strength and occupational exposures lends support to this hypothesis.

In conclusion, grip strength at normal retirement age is not positively associated with cumulative lifetime exposure to manual work and may be inversely associated. Mechanisation is unlikely to negatively affect workers’ strength at retirement.


The Hertfordshire Cohort Study was supported by the Medical Research Council of Great Britain and Arthritis Research UK. The authors thank all of the men and women who took part in the Hertfordshire Cohort Study; the HCS Research Staff; and Vanessa Cox who managed the data. They would like to thank Dr Clare Harris for her work coding the occupational exposures and Dr Richard Dodds for his support with the literature review for this manuscript.


View Abstract


  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval East and North Hertfordshire NHS Research Ethics Committees.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles