Article Text

Download PDFPDF

Original article
Associations of hip osteoarthritis with history of recurrent exposure to manual handling of loads over 20 kg and work participation: a population-based study of men and women
  1. Leena Kaila-Kangas1,
  2. Jari Arokoski2,
  3. Olli Impivaara3,
  4. Eira Viikari-Juntura1,
  5. Päivi Leino-Arjas1,
  6. Ritva Luukkonen1,
  7. Markku Heliövaara3
  1. 1Finnish Institute of Occupational Health, Helsinki, Finland
  2. 2Department of Physical and Rehabilitation Medicine, Kuopio University Hospital, Kuopio, Finland and the Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland
  3. 3National Institute for Health and Welfare, Helsinki, Finland
  1. Correspondence to Leena Kaila-Kangas, Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A, FI-00250 Helsinki, Finland; leena.kaila-kangas{at}


Objectives We reviewed work histories of manual handling of loads >20 kg in relation to hip osteoarthritis by age, exposure and work participation.

Methods A nationally representative sample of 3110 Finnish men and 3446 women aged 30–97 was recruited. Diagnosis of hip osteoarthritis was based on standardised clinical examination by trained physicians. Previous exposure to physically loading work was evaluated through interviews. Logistic regression was used to estimate associations between work factors and hip osteoarthritis.

Results 1.9% of men and 2.1% of women had hip osteoarthritis. Almost half the men and a quarter of the women had recurrently handled heavy loads at work. Subjects who had manually handled loads >20 kg had a 1.8-fold increased risk of hip osteoarthritis compared to non-exposed references, when age, body mass index, traumatic fractures and smoking were accounted for. Results were similar for men (OR 2.0; 95% CI 1.0 to 4.0) and women (1.8; 1.1 to 2.8). In a sub-analysis of subjects with hip replacement, the OR was 1.7 (1.0 to 2.9). Risk increased first after 12 years' exposure: among men it was 2.2 (0.8 to 5.9) for 13–24 years' exposure, and 2.3 (1.2 to 4.3) for >24 years' exposure. Among women it was 3.8 (1.7 to 8.1) for 13–24 years' exposure. Work participation among men aged <60 years with hip osteoarthritis was 20% lower compared with subjects without osteoarthritis.

Conclusions A work history of manual handling of loads >20 kg showed a strong association with hip osteoarthritis in all age groups except the youngest.

  • Epidemiology
  • public health
  • musculoskeletal
  • gender
  • physical work
  • epidemiology
  • public health
  • musculoskeletal
  • gender
  • physical work

Statistics from

What this paper adds

  • Physically loading work factors may play a considerable role in the development of hip osteoarthritis.

  • Almost half the men and a quarter of the women had recurrently handled loads over 20 kg at some point during their working life.

  • The age distribution of hip osteoarthritis differed between genders: 44% of men but only 19% of women with hip osteoarthritis were under 60 years of age.

  • Hip osteoarthritis can be a significant cause of premature health-related selection out of work, especially among men.

  • Employers should protect workers from manual handling of excessively heavy loads at work.


Osteoarthritis of the hip is a major cause of musculoskeletal morbidity and disability in Western societies, a problem which will only increase with the ageing of the population. As well as having considerable economic costs, it causes a great deal of trouble at the individual level1: besides pain and discomfort, it results in greater dependency regarding walking, stair climbing and other lower extremity tasks than any other disease, especially among the elderly.2 It is estimated that hip osteoarthritis affects approximately 7%–25% of white people over the age of 55 and is more common among men than women.3

The natural history of osteoarthritis is poorly understood and there is no treatment to prevent or cure it once it has started to develop.4 Primary prevention of osteoarthritis is therefore crucial. As a chronic disease with a multi-factorial aetiology which includes modifiable risk factors, hip osteoarthritis could, in theory, be prevented.2 A person may have an inherited predisposition to the disease5 but is only likely to develop it when a biomechanical insult occurs.3 6 Thus, it is most important that the risk factors for hip osteoarthritis and the mechanisms behind it are identified. Moreover, the possible effect of modification by gender should also be taken into account.

Some factors seem to increase the probability of hip osteoarthritis. Twin studies have found a strong genetic basis7–9 for it, and ageing also seems to contribute to its development.10–12 Moderate evidence exists for associations between obesity,13 sporting activities,6 previous physical load at work14 15 or during leisure time,16 and hip osteoarthritis. Several studies also suggest that injury is a major risk factor.10 12 17–19 In a recent prospective study representative of the Finnish population, heavy manual labour or permanent damage as a consequence of any musculoskeletal injury predicted future hip osteoarthritis.19 It has even been proposed that hip osteoarthritis is a common occupational disease among farmers.20

Several jobs, including farming, involve a great deal of physically loading work tasks and awkward postures. Based on previous studies, however, manual handling of heavy loads seems to be the most probable risk factor for hip osteoarthritis, although the definitions of heavy loads have varied greatly. We analysed the associations between a history of recurrent manual handling of loads over 20 kg and hip osteoarthritis. In addition, we observed the distribution of hip osteoarthritis cases by age, exposure and work participation in representative data of Finnish men and women.

Materials and methods

The Health 2000 Survey was conducted in Finland between September 2000 and June 2001.21 Its main purpose was to achieve an overall view of the population's health. The survey was carried out in several phases and included a number of questionnaires, an extensive face-to-face home interview, laboratory and functional capacity tests, and a clinical examination. The nationally representative population sample of Finnish adults was formed through the use of a two-stage cluster sampling method whereby Finland was stratified into 20 sections consisting of the 15 largest cities and five university hospital districts. The 15 cities together with 65 of the 234 municipalities or groups of municipalities with joint primary care (within the five university hospital districts) selected by systematic sampling, formed 80 clusters. Methods and processes are described in detail elsewhere.22 All participants provided written informed consent, and the Ethics Committee for Epidemiology and Public Health of the hospital district of Helsinki and Uusimaa in Finland approved the study.

A total of 8028 subjects aged between 30 and 99 were sampled from the clusters. However, as 49 had died before the data were gathered, the final sample included 7979 subjects, of whom 7087 (89%) were interviewed. Our study ultimately consisted of 6556 (82%) subjects who participated in both the clinical examination and the home interview.

Assessment of hip osteoarthritis

The clinical diagnosis of hip osteoarthritis was made on the basis of disease history, symptoms and clinical findings according to the standard criteria (box 1) evaluated by specially trained physicians. The examining physicians categorised the diagnosis as either probable or definite. Only definite cases were included in the present study.

Box 1

Diagnostic criteria for clinically diagnosed definite hip osteoarthritis in the Health 2000 Survey

  • Documented history of previously diagnosed hip osteoarthritis or hip arthroplasty due to osteoarthritis, based on convincing findings


  • At least moderate restrictions in extension (limitation over 5°) in inner rotation (maximal range <20°), or in outer rotation (maximal range <30°), especially if tenderness accompanies movement


  • Slight restrictions in extension (limitation <5°) or in inner rotation (maximal range 20–30°) or in outer rotation (maximal range 30–40°) or at least moderately restricted abduction–adduction (maximal range <50°) AND either of the following:

    • Documented history of previously diagnosed hip osteoarthritis but no grounds for diagnosis given.

    • Typical symptoms of hip osteoarthritis (stiffness, pain when moving after inactivity, pain during prolonged strain).

Hip and knee radiographs were taken to validate the clinical diagnosis of osteoarthritis among a sub-sample of 136 participants in the Health 2000 Survey.23 The agreement between definite clinical and radiological hip osteoarthritis diagnosis (Kellgren and Lawrence grading scale 2–4)24 proved to be moderate. The κ value was 0.66 (95% CI 0.29 to 1.00).

Workload history

The history of work-related physical loading was assessed through interviews. The interviewers asked the respondents whether they had been exposed daily to different work-related factors either in their current job (yes/no) or in their five longest lasting previous jobs. They also enquired about job duration (in years). Few had held more than five jobs, so the cumulative index covered over 99% of the subjects' complete occupational histories. Information on the manual handling of heavy loads was elicited by asking: “Did your work involve the manual handling of heavy objects, such as lifting, carrying or pushing loads over 20 kg on average of at least 10 times per working day?”.


We selected the possible confounders on the basis of previous literature. Body weight and height were measured in the clinical examination and body mass index (BMI, kg/m2) was calculated and classified into four categories (<25.0, 25.0–29.9, 30.0–35.0 and >35.0). As regards smoking, we categorised the subjects as never smokers, former/occasional smokers and current smokers. Participants were asked (yes/no) whether they had ever had a traumatic fracture in their hip, pelvis or a lower extremity.

Statistical analyses

The dependent variable was clinically evaluated definite hip osteoarthritis (yes/no). We fitted logistic models for the associations between the manual handling of heavy loads and the outcome, adjusted for covariates. We calculated the ORs for risk factors, so that the reference category was subjects who had not been exposed or whose length of exposure had lasted for less than 1 year. We also calculated the prevalence of hip osteoarthritis by age and gender/exposure to manual handling of heavy loads, and presented the results in bar charts.

We fitted logistic regression models to the data using the LOGISTIC procedure of the SAS software package (v 9.1; SAS Institute). In addition, possible dependencies between the individuals in the same cluster were taken into account through multilevel analyses using the RLOGIST procedure of SUDAAN (v 9.1; RTI International). We used population weighting in the analyses to adjust for the age, sex, living district and language distributions of the sample so that they corresponded with those of the total Finnish population.


Table 1 presents the characteristics of the participants. Mean age was 51 years (SD 14) for the men and 53 years (SD 14) for the women. The crude prevalence of definite hip osteoarthritis was 2.0% for the whole population aged 30–97, and 1.9% for the men and 2.1% for the women. The prevalence increased with age (figure 1). The disease was bilateral among 33% of the affected men and 32% of the affected women. Hip replacement had been undergone by 42% (N=53) of cases, 59% of whom were women.

Table 1

Characteristics of participants

Figure 1

Prevalence of hip osteoarthritis by age and gender.

Hip osteoarthritis was strongly related to age, with 70% of all diagnosed cases being aged 60 or over. The age distribution of hip osteoarthritis differed between genders, however. While 44% of the men with hip osteoarthritis were under 60 years of age, only 19% of women with hip osteoarthritis were under 60 (table 1, figure 1). Of the cases under the age of 60, 67% of the men and 78% of the women were still working, whereas the corresponding figures were 87% and 84% among subjects without hip osteoarthritis. Overall, 47% of the men and 24% of the women had worked in a job involving handling loads >20 kg; the mean duration of such exposure was 21 years (SD 15) for the men and 20 years (SD 13) for the women. The prevalence of hip osteoarthritis among subjects who had been handling loads >20 kg was higher in all age groups except the youngest (30–39 years) (figure 2).

Figure 2

Prevalence of hip osteoarthritis by age and manual handling of loads >20 kg.

Among all subjects, an exposure history of manual handling of loads heavier than 20 kg showed a 1.8-fold increased risk of hip osteoarthritis (table 2). The OR was 2.0 (95% CI 1.0 to 4.0) for the men and 1.8 (95% CI 1.1 to 2.8) for the women. The risk increased first after 12 years of exposure in both genders. Among men the OR was 2.2 (0.8 to 5.9) for 13–24 years of exposure and 2.3 (1.2 to 4.3) for >24 years of exposure. Among the women this was 3.8 (1.7 to 8.1) for 13–24 years of exposure but decreased thereafter. In an additional analysis (not shown) among subjects whose hip osteoarthritis had required surgery, the results were in line with the results of the general analysis. The OR for the association between overall heavy materials handling and hip replacement was 1.7 (95% CI 1.0 to 2.9) when all covariates were adjusted for.

Table 2

Associations of manual handling of loads >20 kg according to years of exposure, with clinically verified hip osteoarthritis among Finns aged 30–97


The present study showed that recurrent work-related manual handling of workloads over 20 kg was associated with hip osteoarthritis in both genders between the ages of 30 and 97. The risk was visible after 12 years' exposure and in all age groups except the youngest. These results suggest that hip osteoarthritis takes a long time to develop, and that young people with hip osteoarthritis do not choose to work in physically loading jobs. We also showed that the age-related distribution of hip osteoarthritis was different among genders, with hip osteoarthritis being much more common among working-age men than women. In men with hip osteoarthritis, the proportion of those participating in work was significantly decreased.

Our results are in line with previous studies which have suggested that manual handling of heavy objects is a significant risk factor for clinical and radiographic hip osteoarthritis.14 15 According to Felson,25 susceptibility to osteoarthritis may be the result of an interaction between systemic, intrinsic and extrinsic factors. Manual handling of heavy loads is an extrinsic risk factor, while ageing and genetic susceptibility are systemic factors. Osteoarthritis may occur when activity is excessive or when lower loading acts on a vulnerable joint.25 In our study, the workload of some subjects had probably been excessive in relation to their individual capabilities, and the disease developed over a long exposure history. It is also possible that the handling of very heavy loads indicates a higher injury risk.

Hip osteoarthritis was more common among working-aged men than women, among whom in turn occurrence was strongly related to age. Women usually have a much higher prevalence of osteoarthritis in older age, presumably due to hormonal factors.25 Despite these differences, the age-adjusted risk estimates that we observed were of a similar magnitude in both genders. In other words, among the exposed, the risk of hip osteoarthritis seemed to be uniform, although the development of hip osteoarthritis related to work exposures may be a more prolonged process in women. However, no estimates of the severity of the disease were available.

Exposure to manual handling of heavy loads was higher among men than women. The facts that hip osteoarthritis was much more prevalent among working-age men than women, and that there was less work participation among men with hip osteoarthritis, may be related to work activities. Men more often work in occupations involving very high physical workloads and men with low education have fewer opportunities than women to transfer from strenuous to lighter work. A previous study found that early onset of osteoarthritis was seen in heavy labour jobs with almost 40% of patients reporting their first symptoms before the age of 50.26 The prevalence of hip osteoarthritis among the women was slightly higher than among the men probably because the women's mean age was higher. The life expectancy of women in Finland, as in many other Western countries, is longer than that of men.

The strengths of our study were that the sample was representative of the Finnish population and the participation rate was high with 82% of subjects taking part. Most items in the questionnaires, face-to-face interviews and health examination protocols were selected on the basis of standardised, generally accepted recommendations, or nationally established practice.22 The clinical diagnosis of hip osteoarthritis was made by specially trained physicians who had a specific protocol for the health examination. It is possible, however, that there was some inconsistency between physicians in the assessment of diagnoses. Therefore, the repeatability of the diagnoses was tested and shown to be acceptable, as is also shown in the previous national health examination survey using a similar protocol and the same diagnostic criteria.10 In the current survey, the agreement between definite clinical and radiological diagnoses of hip osteoarthritis proved to be moderate. In addition, however, we also conducted a separate analysis of subjects who had received a hip replacement.

Due to the cross-sectional design of the study, the precise temporal relationship of workload factors with the development of hip osteoarthritis cannot be demonstrated. The study results showed, however, that more than 12 years of exposure was associated with an increased risk of hip osteoarthritis. Another limitation of our study was the retrospective assessment of work factors. Although the dependability of self-reported exposure information has been debated, a recent review showed that self-reports on work-related physical factors are quite reliable.27 Presumably people more easily recall issues such as the number of years they have held a particular job or whether manual handling of loads over 20 kg has been part of their work, than the more detailed quantitative information of their work characteristics. We also found that 20% of men with hip osteoarthritis had stopped working before retirement age. This may have diluted our estimates of the associations between manual handling of workloads over 20 kg with hip osteoarthritis, at least in the category of subjects who had been exposed for over 24 years.

According to our study results, manual handling of loads over 20 kg was strongly associated with hip osteoarthritis. A substantial proportion of working subjects, especially men, had a history of such exposure. The proportion of men under the age 60 with hip osteoarthritis participating in work was 20% lower than among those without the disease. These results suggest that work factors may contribute substantially to the development of hip osteoarthritis, and that this disease may be a significant cause of premature health-related selection out of the workforce, especially among men. Many cases of hip osteoarthritis could probably be avoided by protecting people from manual handling of excessively heavy loads.


The English language was revised by Alice Lehtinen, BA Hons.



  • Funding This work was supported by the Finnish Work Environment Fund (no. 106293) and the Farmers' Social Insurance Institution (no. 0527).

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Ethics Committee for Epidemiology and Public Health of the hospital district of Helsinki and Uusimaa in Finland.

  • 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.