SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

To investigate the risk of radiographic knee osteoarthritis (OA) and lumbar spondylosis associated with occupational activity in elderly Japanese subjects using the large-scale population-based cohort of the Research on Osteoarthritis Against Disability (ROAD) study.

Methods

From the baseline survey of the ROAD study, 1,471 participants age ≥50 years (531 men and 940 women) living in mountainous and seacoast communities were analyzed. Information collected included a lifetime occupational history and details of specific work place physical activities. Radiographic severity at the knee and lumbar spine was determined by the Kellgren/Lawrence (K/L) grading system.

Results

The prevalence of K/L grade ≥2 knee OA and lumbar spondylosis among agricultural, forestry, and fishery workers was significantly higher than among clerical workers and technical experts in the overall population. For occupational activities, sitting on a chair had a significant inverse association with K/L grade ≥2 knee OA and lumbar spondylosis. Standing, walking, climbing, and heavy lifting were associated with K/L grade ≥2 knee OA, but were not associated with K/L grade ≥2 lumbar spondylosis. Kneeling and squatting were associated with K/L grade ≥3 knee OA.

Conclusion

This cross-sectional study using a population-based cohort suggests that sitting on a chair is a significant protective factor against both radiographic knee OA and lumbar spondylosis in Japanese subjects. An occupational activity that includes heavy lifting appears to have a greater effect on knee OA than on lumbar spondylosis.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Osteoarthritis (OA) and spondylosis, which cause cartilage and disc degeneration and osteophyte formation at joints in the extremities and spine, are major public health issues causing chronic disability in the elderly in developed countries (1–6). Despite the urgent need for strategies to prevent and treat these conditions, epidemiologic data on OA and spondylosis are sparse. Established risk factors for knee OA in whites include older age, female sex, evidence of OA in other joints, obesity, and previous injury or surgery of the knee (7–12). Evidence is accumulating in whites that the disease is more common in people who have performed heavy physical work (13–18), particularly in those whose jobs have involved kneeling or squatting (19–24). However, published work has tended to concentrate on the knee, and few studies have focused on risk factors for lumbar spondylosis associated with occupational activity (25–28). In addition, there have been no large-scale population-based epidemiologic studies that have simultaneously evaluated the risk of both knee OA and lumbar spondylosis associated with occupational activity in the same population. Furthermore, most epidemiologic studies of OA and spondylosis associated with occupation are limited in terms of the quality of the information collected about occupational exposure. Occupational histories are not always complete, and exposure has often only been inferred from the subject's job title (13–18). To provide accurate data on the relationship of occupational activities with knee OA and lumbar spondylosis, collected information has to include a lifetime occupational history and details of specific work place physical activities.

With the goal of establishing epidemiologic indexes to evaluate clinical evidence for the development of disease-modifying treatment, we set up a large-scale nationwide OA cohort study called the Research on Osteoarthritis Against Disability (ROAD) study in 2005. In the present study, we used the data of participants living in mountainous and seacoast communities to investigate the association of job title and occupational activity with radiographic knee OA and lumbar spondylosis.

PARTICIPANTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Participants.

The ROAD study is a nationwide prospective study for bone and joint diseases consisting of population-based cohorts established in several communities in Japan. Because the Miyama cohort has been profiled in detail elsewhere (29), the characteristics of the participants are briefly summarized here. To date, we have created a baseline database including clinical and genetic information on 3,040 inhabitants (1,061 men and 1,979 women) ages 23–95 years (mean 70.6 years) who were recruited from listings of resident registrations in 3 communities. All participants provided written informed consent, and the study was conducted with the approval of ethical committees of the University of Tokyo and the Tokyo Metropolitan Institute of Gerontology. Information collected about job title and occupational activity included a lifetime occupational history with details of 7 types of specific work place physical activities, including sitting on a chair, kneeling, squatting, standing, walking, climbing, and heavy lifting. Participants were asked whether they engaged in the following activities: sitting on a chair for ≥2 hours/day, kneeling for ≥1 hour/day, squatting for ≥1 hour/day, standing for ≥2 hours/day, walking ≥3 km/day, climbing up slopes or steps for ≥1 hour/day, and lifting loads weighing ≥10 kg at least once a week. Information on these activities was obtained for the principal job, defined as the job at which the participant had worked the longest. Anthropometric measurements included height, weight, bilateral grip strength, and body mass index (BMI; weight [kg]/height [m2]). All participants were interviewed regarding knee pain and low back pain by asking them, “In the past 1 month, have you had knee pain on most days lasting?” and “In the past 1 month, have you had low back pain on most days lasting?” Participants who answered yes were defined as having knee pain or low back pain, respectively. From the baseline data of all participants, the present study analyzed 1,471 participants (531 men and 940 women) age ≥50 years living in mountainous and seacoast cohorts.

Radiographic assessment.

All participants had a radiographic examination of both knees using anteroposterior and lateral views with weight-bearing and foot map positioning, and an examination of the lumbar spine, including intervertebral levels from L1–L2 to L5–S1 with anteroposterior and lateral views. Knee and lumbar spine radiographs were read without knowledge of participant clinical status by a single well-experienced orthopedist (SM) using the Kellgren/Lawrence (K/L) radiographic atlas, and the severity was determined by K/L grading (30). We defined knee OA and lumbar spondylosis as a K/L grade ≥2 in at least one knee and in one intervertebral level, respectively.

To evaluate the intraobserver variability of K/L grading, 100 randomly selected radiographs of the knee and the lumbar spine were scored by the same observer more than 1 month after the first reading. One hundred other radiographs were also scored by 2 experienced orthopedic surgeons (SM, HO) using the same atlas for interobserver variability. The evaluated intra- and intervariability were confirmed by the kappa analysis to be sufficient for assessment (0.86 and 0.80 for knee OA, 0.84 and 0.76 for lumbar spondylosis, respectively).

Statistical analysis.

The differences of age and BMI between men and women were examined by the unpaired t-test. To compare the prevalence of radiographic knee OA and lumbar spondylosis between men and women, we performed a logistic regression analysis after adjustment for age and BMI. The percentage of each occupational activity was compared between men and women by a chi-square test. To determine risk factors for knee OA and lumbar spondylosis with K/L grades ≥2 as well as K/L grades ≥3, logistic regression analyses were used to estimate the odds ratio (OR) and the associated 95% confidence interval (95% CI) for variables such as job title and occupational activities after adjustment for age and BMI compared with K/L = 0 or 1 (for K/L grades ≥2) and K/L = 0, 1, or 2 (for K/L grades ≥3). Furthermore, the overall population was classified into 4 subpopulation groups based on the presence or absence of knee OA and lumbar spondylosis, and a multinomial logistic regression analysis was performed to determine factors associated with knee OA, lumbar spondylosis, and their combination after adjustment for age, sex, and BMI. The subpopulation with neither knee OA nor lumbar spondylosis was used as a reference group. Data analyses were performed using SAS, version 9.0 (SAS Institute, Cary, NC).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Characteristics of the 1,471 participants age ≥50 years in the 2 cohorts of the ROAD study are shown in Table 1. The prevalence of K/L grade ≥2 and K/L grade ≥3 knee OA was significantly higher in women than in men, whereas that of K/L grade ≥2 lumbar spondylosis was significantly lower in women than in men. The prevalence of K/L grade ≥3 lumbar spondylosis was comparable between sexes.

Table 1. Characteristics of participants*
 OverallMenWomen
  • *

    Values are the mean ± SD unless otherwise indicated. BMI = body mass index; K/L = Kellgren/Lawrence grading system; OA = osteoarthritis.

  • P < 0.05 versus men by unpaired t-test.

  • P < 0.05 versus men by logistic regression analysis after adjustment for age and BMI.

  • §

    P < 0.05 versus men by chi-square test.

No. of subjects1,471531940
Age, years68.4 ± 9.269.1 ± 9.168.0 ± 9.2
Height, cm154.3 ± 9.3162.3 ± 7.1149.8 ± 7.2
Weight, kg55.2 ± 10.561.0 ± 10.351.8 ± 9.1
BMI, kg/m223.1 ± 3.323.1 ± 3.123.1 ± 3.5
Grip strength, kg26.7 ± 9.334.7 ± 8.422.1 ± 6.1
K/L ≥2 knee OA, %55.645.661.2
K/L ≥3 knee OA, %23.016.826.5
K/L ≥2 lumbar spondylosis, %65.379.157.6
K/L ≥3 lumbar spondylosis, %38.738.838.7
Current smoker, no. (%)169 (11.5)140 (26.4)29 (3.1)§
Current alcohol drinking, no. (%)562 (38.2)343 (64.6)219 (23.3)§

There was great diversity in the job titles of the study participants (Table 2). Although a substantial proportion includes clerical workers and technical experts, there were many agricultural, forestry, and fishery workers. Among various occupational activities, agricultural, forestry, and fishery workers had the highest rates of kneeling, squatting, standing, walking, climbing, and lifting weights and the lowest rates of sitting on a chair, whereas clerical workers and technical experts had the lowest rates of kneeling, squatting, standing, walking, climbing, and lifting weights and the highest rates of sitting on a chair (Figure 1).

Table 2. Participants with job title and occupational activity reported as the principal job
 OverallMenWomen
  • *

    P < 0.05 versus men by chi-square test.

Job titles, no. (%)   
 Clerical workers/technical experts363 (24.7)170 (32.0)193 (20.5)
 Agricultural/forestry/fishery workers318 (21.6)164 (30.9)154 (16.4)
 Factory/construction workers153 (10.4)68 (12.8)85 (9.0)
 Shop assistants/managers132 (9.0)25 (4.7)107 (11.4)
 Housekeepers126 (8.6)0 (0.0)126 (13.4)
 Teachers82 (5.6)42 (7.9)40 (4.3)
 Dressmakers51 (3.5)1 (0.2)50 (5.3)
 Clinical workers41 (2.8)1 (0.2)40 (4.3)
 Hairdressers17 (1.2)6 (1.3)11 (1.2)
 Others (cooks, taxi drivers, etc.)72 (4.9)22 (4.1)50 (5.3)
 No answer116 (7.9)32 (6.0)84 (8.9)
Occupational activities, no. (%)   
 Sitting on a chair ≥2 hours/day657 (44.7)254 (47.8)403 (42.8)
 Kneeling ≥1 hour/day292 (19.9)96 (18.1)196 (20.9)
 Squatting ≥1 hour/day386 (26.2)131 (24.7)255 (27.1)
 Standing ≥2 hours/day1,235 (84.0)456 (85.9)779 (82.9)
 Walking ≥3 km/day673 (45.8)268 (50.5)405 (43.1)
 Climbing ≥1 hour/day346 (23.5)185 (34.8)161 (17.1)*
 Lifting weights ≥10 kg at least once a week788 (53.6)347 (65.3)441 (46.9)*
thumbnail image

Figure 1. Percentages of participants engaged in each occupational activity: sitting on a chair ≥2 hours/day, kneeling ≥1 hour/day, squatting ≥1 hour/day, standing ≥2 hours/day, walking ≥3 km/day, climbing ≥1 hour/day, or lifting weights ≥10 kg at least once a week among agricultural, forestry, and fishery workers; factory and construction workers; clerical workers and technical experts; and others.

Download figure to PowerPoint

To determine factors associated with K/L grade ≥2 knee OA and lumbar spondylosis, we performed a logistic regression analysis to estimate ORs and 95% CIs (Tables 3 and 4). Analysis of job titles revealed that agricultural, forestry, and fishery workers had a significantly higher risk of knee OA and lumbar spondylosis compared with clerical workers and technical experts in the overall population. We then examined the association of occupational activities with knee OA and lumbar spondylosis (Tables 3 and 4). Sitting on a chair for ≥2 hours/day was a significant protective factor for knee OA and lumbar spondylosis in the overall population and in men. Neither kneeling for ≥1 hour/day nor squatting for ≥1 hour/day was associated with knee OA in the overall population. Standing for ≥2 hours/day, walking ≥3 km/day, climbing for ≥1 hour/day, and lifting weights ≥10 kg at least once a week were significantly associated with knee OA in the overall population and in both sexes (Table 3). A multiple logistic regression analysis after adjustment for age, BMI, sex, and the above 4 occupational activities showed that climbing and lifting weights were significantly associated with knee OA overall (OR 1.65, 95% CI 1.18–2.32 and OR 1.51, 95% CI 1.16–1.95, respectively) and in men (OR 1.75, 95% CI 1.10–2.80 and OR 1.76, 95% CI 1.14–2.73, respectively), suggesting that among the 4 activities that required a standing position, climbing and lifting weights had an independent association with knee OA. In contrast, these occupational activities had no significant association with lumbar spondylosis except for lifting weights in women (Table 4).

Table 3. Association of K/L grade ≥2 knee OA with job title and occupational activity*
 Overall, OR (95% CI)Men, OR (95% CI)Women, OR (95% CI)
  • *

    ORs were calculated by a logistic regression analysis after adjustment for age, sex, and BMI in the overall population, and for age and BMI in both sexes. K/L = Kellgren/Lawrence grading system; OA = osteoarthritis; OR = odds ratio; 95% CI = 95% confidence interval; BMI = body mass index.

  • Includes all participants except for agricultural/forestry/fishery workers, factory/construction workers, and clerical workers/technical experts.

Job titles (vs. clerical workers/technical experts)   
 Agricultural/forestry/fishery workers1.69 (1.19–2.41)1.58 (0.98–2.56)1.90 (1.14–3.20)
 Factory/construction workers1.52 (0.99–2.36)1.33 (0.72–2.47)1.64 (0.90–3.06)
 Other1.18 (0.88–1.60)1.21 (0.73–2.00)1.20 (0.82–1.76)
Occupational activities   
 Sitting on a chair ≥2 hours/day0.73 (0.57–0.92)0.63 (0.44–0.92)0.80 (0.60–1.09)
 Kneeling ≥1 hour/day1.11 (0.83–1.48)0.79 (0.49–1.26)1.36 (0.93–1.97)
 Squatting ≥1 hour/day1.23 (0.94–1.61)0.89 (0.58–1.35)1.50 (1.06–2.13)
 Standing ≥2 hours/day1.97 (1.43–2.72)2.31 (1.32–4.17)1.78 (1.21–2.63)
 Walking ≥3 km/day1.80 (1.42–2.29)2.17 (1.49–3.16)1.59 (1.17–2.16)
 Climbing ≥1 hour/day2.24 (1.65–3.04)2.43 (1.64–3.60)1.85 (1.19–2.96)
 Lifting weights ≥10 kg at least once a week1.90 (1.50–2.42)2.26 (1.52–3.40)1.68 (1.24–2.26)
Table 4. Association of K/L grade ≥2 lumber spondylosis with job title and occupational activity*
 Overall, OR (95% CI)Men, OR (95% CI)Women, OR (95% CI)
  • *

    ORs were calculated by a logistic regression analysis after adjustment for age, sex, and BMI in the overall population, and for age and BMI in both sexes. See Table 3 for definitions.

  • Includes all participants except for agricultural/forestry/fishery workers, factory/construction workers, and clerical workers/technical experts.

Job titles (vs. clerical workers/technical experts)   
 Agricultural/forestry/fishery workers1.46 (1.02–2.11)1.49 (0.83–2.68)1.42 (0.89–2.28)
 Factory/construction workers1.05 (0.68–1.55)1.52 (0.76–3.22)0.84 (0.49–1.44)
 Other1.22 (0.91–1.64)1.53 (0.87–2.76)1.11 (0.78–1.58)
Occupational activities   
 Sitting on a chair ≥2 hours/day0.78 (0.62–0.99)0.48 (0.30–0.76)0.93 (0.71–1.23)
 Kneeling ≥1 hour/day0.96 (0.72–1.28)0.95 (0.55–1.70)0.97 (0.70–1.35)
 Squatting ≥1 hour/day1.05 (0.81–1.38)0.95 (0.58–1.61)1.09 (0.80–1.48)
 Standing ≥2 hours/day1.11 (0.81–1.50)1.14 (0.61–2.04)1.10 (0.77–1.57)
 Walking ≥3 km/day1.00 (0.79–1.26)0.89 (0.57–1.40)1.04 (0.79–1.37)
 Climbing ≥1 hour/day1.02 (0.76–1.38)1.09 (0.68–1.78)0.98 (0.67–1.44)
 Lifting weights ≥10 kg at least once a week1.15 (0.91–1.45)1.09 (0.69–1.72)1.23 (1.01–1.55)

We next performed a multinomial logistic regression analysis to determine factors associated with K/L grade ≥2 knee OA, lumbar spondylosis, and their combination after adjustment for age, sex, and BMI. Sitting on a chair was confirmed to be a significant protective factor for the presence of both knee OA and lumbar spondylosis (OR 0.62, 95% CI 0.45–0.86). Although neither kneeling nor squatting was associated with the presence of knee OA or lumbar spondylosis, standing (OR 2.03, 95% CI 1.32–3.12), walking (OR 1.56, 95% CI 1.12–2.17), climbing (OR 2.14, 95% CI 1.38–3.40), and lifting weights (OR 2.05, 95% CI 1.48–2.86) were associated with the presence of both knee OA and lumbar spondylosis. For the subpopulation group with knee OA and without lumbar spondylosis, standing (OR 1.69, 95% CI 1.04–2.79), climbing (OR 2.34, 95% CI 1.39–3.97), and lifting weights (OR 1.92, 95% CI 1.31–2.81) were also significantly associated, although there were no significant associations of the subpopulation group with lumbar spondylosis and without knee OA compared with the subpopulation group without knee OA or lumbar spondylosis.

We further analyzed the association of K/L grade ≥2 knee OA and lumbar spondylosis with job titles and occupational activities according to the presence of knee pain and low back pain at the baseline examination (Supplementary Tables A and B, available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home). Although some of the job titles and occupational activities showed higher ORs in the subpopulation with knee pain, the direction of association was similar regardless of the presence of pain, and the results did not differ between the overall population and the subpopulation without knee pain or low back pain.

We next determined factors associated with K/L grade ≥3 knee OA and lumbar spondylosis using logistic regression analysis after adjustment for age and BMI. Analysis of occupational activities revealed that sitting on a chair was a significant protective factor for lumbar spondylosis in men (OR 0.58, 95% CI 0.40–0.84). In the overall population and in women, kneeling (OR 1.40, 95% CI 1.01–1.93 and OR 1.69, 95% CI 1.16–2.47, respectively), squatting (OR 1.34, 95% CI 1.00–1.80 and OR 1.51, 95% CI 1.06–2.15, respectively), and lifting weights (OR 1.60, 95% CI 1.21–3.12 and OR 1.73, 95% CI 1.25–2.43, respectively) were associated with knee OA. A multinomial logistic regression analysis also showed that sitting on a chair was a protective factor for the presence of both K/L grade ≥3 knee OA and lumbar spondylosis, as well as for the presence of lumbar spondylosis and the absence of knee OA in men (OR 0.46, 95% CI 0.23–0.87 and OR 0.63, 95% CI 0.42–0.94, respectively). Lifting weights (OR 1.57, 95% CI 1.10–2.23) was associated with the presence of both knee OA and lumbar spondylosis. For the subpopulation group with knee OA and without lumbar spondylosis, kneeling (OR 1.76, 95% CI 1.13–2.72), squatting (OR 1.85, 95% CI 1.23–2.77), and lifting weights (OR 1.77, 95% CI 1.19–2.65) were significantly associated, although there were no significant associations of the subpopulation group with lumbar spondylosis and without knee OA compared with the subpopulation group without knee OA or lumbar spondylosis.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Using baseline data from the ROAD study, the present investigation evaluated the risk of occupational activity for radiographic knee OA and lumbar spondylosis, and revealed that sitting on a chair was a significant protective factor for both radiographic knee OA and lumbar spondylosis in Japanese subjects. For other occupational activities, kneeling, squatting, standing, walking, climbing, and heavy lifting were significantly associated with radiographic knee OA, whereas there was no significant occupational activity for radiographic lumbar spondylosis in the overall population. To our knowledge, this is the first epidemiologic study using a large-scale population-based cohort to determine the risk of occupational activity for both knee OA and lumbar spondylosis simultaneously in the same population. Information on occupational activities was collected by direct inquiry rather than being inferred from the job title.

In the present study, agricultural, forestry, and fishery workers had a significantly higher prevalence of both radiographic knee OA and lumbar spondylosis compared with clerical workers and technical experts in the overall population. These jobs have historically been among the first to be identified in relation to knee OA in whites (31, 32), which is also compatible with our data in this Japanese population. As other authors have hypothesized, the combination of intense exposure to heavy labor of varied nature and repeated local stresses, especially at a young age, could contribute to some systemic mechanism in the development of OA (33). This argument would support the implementation of preventive measures as a priority to reduce the intensity of physical labor in this sector, particularly for young male and female farm workers.

For occupational activities, standing, walking, climbing, and heavy lifting were associated with K/L grade ≥2 knee OA in the overall population, whereas kneeling and squatting were not, which was similar to previous studies in Japan and China (34, 35). Comparison of characteristics and ORs for knee OA associated with occupational activity among epidemiologic studies is shown in Tables 5 and 6. The present study showed different results from other previously published studies (Table 6). Because each study defined knee OA and cases somewhat differently (in some studies, a case was defined as a subject with K/L grade ≥3 OA with knee pain, while it was defined as a subject with K/L grade ≥2 or K/L grade ≥3 OA in the present study), our results are not directly comparable with those of other studies. Even so, studies of whites have suggested that occupational activities of kneeling and squatting and job titles that required kneeling and squatting were associated with knee OA (19–24), whereas these activities were not associated with K/L grade ≥2 OA in this study. The discrepancies between white and Japanese subjects may be partly explained by the Japanese traditional lifestyle, which includes sitting on the heels on a mat and using the Japanese-style lavatory, where subjects have to take a deep squatting position. These positions may cause mechanical stress to the knee joint and possibly lead to the acceleration of knee OA. Among elderly Japanese subjects, kneeling and squatting are common postures in daily life, which could obscure the association between knee OA and occupational activities of kneeling and squatting.

Table 5. Comparison of characteristics of epidemiologic studies
Author, ref.Ethnicity/countryAge, yearsTotal no.Men:women
Muraki et al, current studyJapan≥501,471531:940
Yoshimura et al,34Japan≥452020:202
Lau et al,35Chinese 1,316332:984
Anderson and Felson,19Blacks and whites/US55–641,250606:644
Felson et al,20Whites/US≥631,376569:807
Cooper et al,21UK≥5532790:237
Coggon et al,22UK≥471,036410:626
Sandmark et al,23Sweden≥551,173589:584
Manninen et al,24Finland≥55805195:610
Table 6. Comparison of odds ratios for knee osteoarthritis associated with occupational activity among epidemiologic studies*
 Muraki et al (current study)Yoshimura et al (34), K/L ≥3 with knee painLau et al (35), K/L ≥3Anderson and Felson (19), K/L ≥2Felson et al (20)Cooper et al (21), K/L ≥3 with knee painCoggon et al (22), listed for knee surgerySandmark et al (23), TKAManninen et al (24), TKA
K/L ≥2K/L ≥3K/L ≥2K/L ≥3
  • *

    K/L = Kellgren/Lawrence grading system; TKA = total knee arthroplasty.

  • P < 0.05.

  • P < 0.05. Kneeling or squatting.

Sitting on a chair0.70.8    1.2  
 Men0.60.8     0.7 
 Women0.80.80.4     0.9 
Kneeling1.11.4   3.41.81.7
 Men0.80.91.4   1.72.11.7
 Women1.41.71.00.9   2.01.51.8
Squatting1.21.36.92.31.7
 Men0.91.01.22.52.22.02.22.91.7
 Women1.51.51.11.13.50.40.72.81.11.8
Standing2.01.4    0.8 0.6
 Men2.31.1     1.70.4
 Women1.81.51.2     1.60.7
Walking1.81.1   0.91.9 1.1
 Men2.20.92.2   1.7 1.5
 Women1.61.10.91.4   2.1 1.1
Climbing2.21.3   2.71.51.6
 Men2.41.04.1   2.31.22.8
 Women1.91.50.96.1   0.71.41.5
Lifting weights1.91.6   1.41.71.0
 Men2.31.31.7   1.93.00.9
 Women1.71.71.01.5   1.51.71.1

The direction of the association of kneeling and squatting with knee OA was also different between sexes in the present study, although these differences were not significant, except for squatting in women. Because men are known to have greater muscle strength than women of all ages and muscle strength has a protective effect on knee OA (36–38), it might be that the greater muscle strength obscures the harmful effects of kneeling and squatting on knee OA in men, resulting in lower ORs for knee OA than in women.

For K/L grade ≥2 lumbar spondylosis, there were no occupational activities associated with the increased prevalence except for heavy lifting in women. Few studies have focused on risk factors for lumbar spondylosis associated with occupational activity (25–28), and no increased risk of lumbar osteophytes due to physical activities has been reported (25, 39, 40).

In the present study, the occupational activity of sitting on a chair was inversely associated with both K/L grade ≥2 knee OA and lumbar spondylosis. For knee OA, our previous small-scale study showed that prolonged sitting on a chair at work was associated with a reduced prevalence of knee OA (34) (Table 5). Regarding the relationship between sedentary work and OA, the results of studies investigating the influence of sedentary work on knee OA are controversial (21, 22). Although sitting on a chair clearly involves reduced load on many joints compared with other working activities, no other studies have reported a relationship between sedentary activity and knee OA. Sitting on a chair as a physical activity in the work place appears to represent a characteristic protective factor for OA in Japan.

Contrary to K/L grade ≥2 knee OA, occupational activities of kneeling and squatting were significantly associated with K/L grade ≥3 knee OA, whereas those of standing, walking, and climbing were not. Considering the definition of the K/L grade, this may suggest distinct risk factors between osteophytosis and joint space narrowing. In this population-based cohort study, the prevalence of K/L grade ≥2 knee OA was 45.6% in men and 61.2% in women, which was higher than that in whites, whereas that of K/L grade ≥3 was 16.8% and 26.5%, which is comparable with that in whites (41), suggesting that the Japanese lifestyle may be associated with osteophytosis rather than joint space narrowing. Therefore, regarding K/L grade ≥2 knee OA, the Japanese lifestyle could obscure the association between knee OA and occupational activities of kneeling and squatting as mentioned above. Furthermore, the discrepancy between risk factors for K/L grade ≥2 and K/L grade ≥3 knee OA may also be due to differences between the mechanism of osteophytosis and joint space narrowing. There is accumulating evidence that osteophytosis and joint space narrowing have distinct etiologic mechanisms (25, 42–47). A previous prospective study using a large-scale OA cohort reported that there was no association between the 2 representative features of knee OA (44). A recent cross-sectional study also showed that osteophytosis was unrelated not only to joint space narrowing on plain radiographs, but also to cartilage loss measured by quantitative magnetic resonance imaging (45). Furthermore, our study on an experimental mouse model for OA has identified a cartilage-specific molecule, carminerin, which regulates osteophytosis without affecting joint cartilage destruction during OA progression (46, 47). Further clinical and basic research will disclose the distinct backgrounds of these 2 features of OA.

There are several limitations in the present study. First, this is a cross-sectional study on factors associated with knee OA and lumbar spondylosis, so a causal association with occupational activity could not be determined. However, information collected included a lifetime occupational history and details of specific work place physical activities; therefore, ample evidence on the background of knee OA and lumbar spondylosis could be obtained. Second, information regarding past occupational exposures was obtained by self-report and there is a possibility that both self-selection bias and recall bias may have occurred. People with painful conditions may choose work that allows them to avoid aggravation of their conditions, so the impact of job titles and occupational activities on knee OA and lumbar spondylosis may be underestimated in the present study. Conversely, people with painful knee and lumbar conditions are likely to look for and assign a cause when asked about past work exposures. To determine the impact of working conditions on knee OA and lumbar spondylosis independently of the presence of pain at the examination, we analyzed the association of knee OA and lumbar spondylosis with job titles and occupational activities according to the presence of knee pain and low back pain at the baseline examination. The direction of association was similar regardless of the presence of pain, and the results between the overall population and the subpopulation without knee pain or low back pain were not different, suggesting that pain at the examination may not affect the results of the overall population very much in this study.

In conclusion, the present cross-sectional study using a large-scale population from the ROAD study revealed distinct risk factors of occupational activities for radiographic knee OA and lumbar spondylosis in Japanese subjects. Sitting on a chair was a significant protective factor for both radiographic knee OA and lumbar spondylosis. Other occupational activities of kneeling, squatting, standing, walking, climbing, and heavy lifting were risk factors for radiographic knee OA, but not for radiographic lumbar spondylosis. Further studies, along with longitudinal data in the ROAD study, will elucidate the environmental backgrounds of OA and spondylosis and clarify clinical evidence for the development of disease-modifying treatments.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Muraki had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Muraki, Akune, Oka, Mabuchi, En-yo, Yoshida, Saika, Nakamura, Kawaguchi, Yoshimura.

Acquisition of data. Muraki, Akune, Oka, Mabuchi, En-yo, Yoshida, Saika, Nakamura, Kawaguchi, Yoshimura.

Analysis and interpretation of data. Muraki, Akune, Oka, Mabuchi, En-yo, Yoshida, Saika, Nakamura, Kawaguchi, Yoshimura.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PARTICIPANTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES