SEARCH

SEARCH BY CITATION

Abstract

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

Objective

There is little information on falls by sex and age strata in Japan, and few factors associated with falls have been established. However, the association between bone and joint diseases and falls remains unclear. We examined prevalence of falls by sex and age strata, determined its association with radiographic osteoarthritis (OA) of the knee and lumbar spine, and determined knee and lower back pain after single and multiple falls.

Methods

A questionnaire assessed the number of falls during 12 months preceding baseline. Knee and lumbar spine radiographs were read by Kellgren/Lawrence (K/L) grade; radiographic knee OA and lumbar spondylosis were defined as a K/L grade of 3 or 4. Knee and lower back pain were estimated by an interview.

Results

A total of 587 men and 1,088 women (mean ± SD age 65.3 ± 12.0 years) were analyzed. During 1 year, 79 (13.5%) men and 207 (19.0%) women reported at least 1 fall. With increasing age, the prevalence of multiple falls was higher in women, but lower in elderly men age >60 years. In men, few factors were significantly associated with falls. In women, radiographic knee OA and lumbar spondylosis, as well as knee and lower back pain, were significantly associated with multiple falls without adjustment. Lower back pain and knee pain were independently associated with multiple falls in women after adjustment.

Conclusion

Lower back pain and knee pain were significantly associated with multiple falls in women.


INTRODUCTION

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

Falls are one of the main causes of injury, disability, and death among the elderly (1, 2). In Japan, according to the recent National Livelihood Survey of the Ministry of Health, Labour and Welfare, fall and fracture are ranked fifth among diseases that cause disabilities and subsequently require support with activities of daily living (3). However, there have been few population-based studies for prevalence of fall based on sex and age strata. Further, in terms of factors associated with falls, muscle strength, balance, vision, and functional capacities, there are traits that diminish with aging, and these factors have been suggested as predictive risk factors for falls and fractures (4). Cognitive impairment has also been established as a risk factor for falls (5), but the association of bone and joint diseases, especially osteoarthritis (OA), with falls remains unclear.

The representative sites of OA are the knee and lumbar spine. Knee OA and lumbar spondylosis (LS) are major public health issues since they cause chronic pain and disability (6–11). The prevalence of radiographic knee OA and LS is high in Japan (12, 13), with 25,300,000 and 37,900,000 subjects ages ≥40 years estimated to experience radiographic knee OA and LS, respectively (14). The National Livelihood Survey ranked OA fourth among diseases that cause disabilities and subsequently require support with activities of daily living (3), but there have been few studies of the association between falls and OA (15, 16). In previous studies, knee OA was assessed only by interview and not by radiography. The principal clinical symptom of knee OA is pain (17), but its correlation with the radiographic severity of knee OA is not as strong as expected (12, 18–20). In fact, in a study in Japan, ∼20% of the subjects without knee OA had knee pain, and 30% of the subjects with severe knee OA had no knee pain (12). Therefore, knee OA diagnosed by interview could be limited by variable accuracy. In addition, men and women were not examined separately in these previous studies, although sex differences have been found in the prevalence of knee OA (12). Furthermore, knee OA is conventionally defined according to Kellgren/Lawrence (K/L) grade (21), and our previous study showed that the association of a K/L grade of 2 (knee OA with pain) was weak, but that a K/L grade of 3 or 4 (knee OA with pain) was strong (12); therefore, the association of knee OA with falls may be different between a K/L grade of 2 for knee OA and a K/L grade of 3 or 4 for knee OA. However, there are no population-based studies on the association of severity of knee OA with falls. With regard to LS, to the best of our knowledge, there have been no population-based studies regarding its association with falls.

Previous studies have shown that associations between individual risk factors and a single fall are few in number and weak compared to risk factors for multiple falls (16), indicating that single and multiple falls may have different backgrounds. Therefore, to determine factors associated with falls, single and multiple falls should be analyzed separately.

The objectives of this study were to clarify prevalence of single and multiple falls by sex and age strata in Japan using a large-scale, population-based cohort study known as Research on Osteoarthritis/osteoporosis Against Disability (ROAD). Further, we examined the associations of radiographic knee OA and LS, as well as knee and lower back pain, with single and multiple falls in Japanese men and women.

Significance & Innovations

  • During 1 year, 13.5% of men and 19.0% of women reported at least 1 fall.

  • With increasing age, prevalence of multiple falls was higher in women, but lower in elderly men age >60 years.

  • Lower back pain and knee pain were independently associated with multiple falls in women.

PATIENTS AND METHODS

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

Patients.

The ROAD study is a nationwide prospective study designed to establish epidemiologic indexes for evaluation of clinical evidence for the development of a disease-modifying treatment for bone and joint diseases (OA and osteoporosis are the representative bone and joint diseases, respectively). It consists of population-based cohorts in 3 communities in Japan. A detailed profile of the ROAD study has been described elsewhere (12–14, 22); a brief summary is provided here. To date, we have completed the creation of a baseline database that includes clinical and genetic information for 3,040 subjects (1,061 men and 1,979 women) with a mean age of 70.6 years (range 23–95 years), who were recruited from resident registration listings in 3 communities: an urban region in Itabashi, Tokyo; a mountainous region in Hidakagawa, Wakayama; and a coastal region in Taiji, Wakayama.

Residents of these regions were recruited from the resident registration list of the relevant region. The participants in the urban region were recruited from a randomly selected cohort from the Itabashi-ward residents' registration database (22). The participation rate was 75.6%. The participants in mountainous and coastal regions were also recruited from the resident-registration lists, and the participation rates in these 2 areas were 56.7% and 31.7%, respectively. The inclusion criteria, apart from residence in the communities mentioned above, were the ability to 1) walk to the survey site, 2) report data, and 3) understand and sign an informed consent form. The baseline survey of the ROAD study was completed in 2006. All participants provided their written informed consent, and the study was conducted with the approval of the ethics committees of the University of Tokyo and the Tokyo Metropolitan Institute of Gerontology.

Falls assessment.

All subjects were interviewed with regard to falls and fractures by experienced interviewers and were asked the following questions: “Have you experienced falls during the 12 months preceding baseline, and if yes, how many falls did you experience?” and “Have you experienced any fractures when you fell?” According to a previous study on falls (23), a fall is defined as a sudden, unintentional change in position causing an individual to land at a lower level on an object, the floor, or the ground, other than as a consequence of a sudden onset of paralysis, epileptic seizure, or overwhelming external force.

Pain assessment.

All subjects were also interviewed by experienced orthopedists (SM and HO) with regard to knee pain and lower back pain and were asked the following questions: “Have you experienced knee pain on most days in the past year, in addition to now?” and “Have you experienced lower back pain on most days in the past year, in addition to now?” Those who answered yes were defined as having pain.

Radiographic assessment.

All participants underwent radiographic examination of both knees using anteroposterior and lateral views with weight-bearing and foot map positioning; radiographic examination of the anteroposterior and lateral views of the lumbar spine, including intervertebral levels L1/2 to L5/S, was also performed. Knee and lumbar spine radiographs were read without the knowledge of participant clinical status by a single, experienced orthopedist (SM) using the K/L radiographic atlas (21) to determine the severity of K/L grading. Radiographs were scored as grade 0 through 4, with higher grades being associated with more severe OA. We defined knee OA and LS as a K/L grade of ≥3 in at least 1 knee and 1 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 and HO) using the same atlas for interobserver variability. The intra- and interobserver variabilities evaluated were confirmed by kappa analysis to be sufficient for assessment (0.86 and 0.80 for knee OA, and 0.84 and 0.76 for LS, respectively).

Covariates.

Anthropometric measurements included height, weight, and body mass index (BMI; kg/m2). Grip strength was measured on bilateral sides using a TOEI LIGHT handgrip dynamometer, and the best measurement was used to characterize maximum muscle strength. To measure physical performance, the time taken to walk 6 meters at normal walking speed in a hallway was recorded. Subjects were told to walk from a marked starting line to a 6-meter mark as if they were walking down their hallway at home. Time was measured in seconds with a stopwatch and rounded to the nearest hundredth of a second. The average of 2 trials was recorded. These gait-speed trial measurements are considered highly reliable in community-dwelling elderly subjects (24–27).

The time taken for 5 consecutive chair rises without the use of hands was also recorded. Hands were folded in front of the chest with feet flat on the floor, following the protocol described by Guralnik et al (28) and used by other researchers (25, 29, 30). Time was measured in seconds with a stopwatch and rounded to the nearest hundredth of a second. Timing began with the command “go” and ended when the buttocks contacted the chair on the fifth landing. The reliability of this protocol is adequate (25, 28, 29). Cognition was also evaluated for all subjects using a Mini-Mental State Examination, and a cutoff score of <24 was used to select participants with cognitive impairment (31).

Statistical analyses.

The differences in age, anthropometric measurements, and physical performance measurements between men and women were examined by Student's unpaired t-test, and among groups of nonfallers, single fallers, and multiple fallers using one-way analysis of variance (ANOVA). The prevalence of cognitive impairment, radiographic knee OA and LS, and knee and lower back pain was compared between men and women, and among nonfallers, single fallers, and multiple fallers by using the chi-square test. The prevalence of single and multiple falls was also compared between men and women, among subjects with no knee OA (K/L grade 0 or 1), with K/L grade 2 for knee OA and K/L grade 3 or 4 for knee OA, and among subjects with no LS (K/L grade 0 or 1), with K/L grade 2 for LS, and K/L grade 3 or 4 for LS by using the chi-square test. The association of knee pain and lower back pain with physical performance was determined by logistic regression analysis. Multinomial logistic regression analysis was also used to determine the association of anthropometric measurements, physical performance, cognitive impairment, radiographic knee OA and LS defined as K/L grade 3 or 4, and knee and lower back pain, with single and multiple falls compared with nonfalls. Further, to determine the independent association of radiographic knee OA and LS, and knee and lower back pain with single and multiple falls compared with nonfalls, we first used multinomial logistic regression analysis with age, BMI, cognitive impairment, radiographic knee OA and LS, and knee and lower back pain as independent variables. In addition to the above independent variables, we additionally adjusted for grip strength, 6-meter walking time, and chair stand time. Data analyses were performed using SAS software, version 9.0.

RESULTS

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

Of the 1,690 subjects in the mountainous and seaside cohorts at baseline, 15 subjects provided incomplete fall questionnaires, leaving a total of 1,675 subjects (587 men, 1,088 women). Table 1 shows the age, anthropometric measurements, and physical performance of the participants in the present study. Regarding physical performance, grip strength, 6-meter walking time, and chair stand time were significantly better in men than in women. The prevalence of cognitive impairment was not significantly different between men and women. The prevalence of radiographic knee OA and knee pain was significantly higher in women than in men, while that of LS and lower back pain was not different between men and women.

Table 1. Characteristics of participants*
 OverallMenWomen
  • *

    Values are the mean ± SD unless indicated otherwise. BMI = body mass index; OA = osteoarthritis.

  • P < 0.05 vs. men by Student's unpaired t-test.

  • P < 0.05 vs. men by chi-square test.

Subjects, no.1,6755871,088
Age, years65.3 ± 12.066.3 ± 11.764.7 ± 12.1
Height, cm155.1 ± 9.3163.4 ± 7.2150.6 ± 6.9
Weight, kg55.6 ± 10.862.3 ± 10.952.0 ± 8.9
BMI, kg/m223.0 ± 3.423.3 ± 3.222.9 ± 3.5
Grip strength, kg27.4 ± 9.835.7 ± 9.322.9 ± 6.8
6-meter walking time, seconds5.5 ± 2.55.3 ± 2.25.6 ± 2.6
Chair stand time, seconds10.1 ± 4.49.7 ± 3.610.4 ± 4.8
Cognitive impairment, %4.55.24.2
Radiographic knee OA, %20.315.023.0
Radiographic lumbar spondylosis, %37.137.736.9
Knee pain, %24.418.927.4
Lower back pain, %20.121.721.2

During the 12 months preceding the baseline examination, 79 men (13.5%, 95% confidence interval [95% CI] 10.9–16.5%) and 207 women (19.0%, 95% CI 16.8–21.5%) reported at least 1 fall, and 48 men (8.2%, 95% CI 6.2–10.7%) and 80 women (7.4%, 95% CI 5.9–9.1%) reported multiple falls. Chi-square test showed that the prevalence of single and multiple falls were significantly different between men and women (P < 0.0001). Among 286 subjects with at least 1 fall, 6 subjects (2.1%) had a wrist fracture, 2 (0.7%) had a proximal humerus fracture, 1 (0.3%) had a vertebral fracture, and 12 (4.2%) had fractures at other sites. With increasing age, the prevalence of falls was lower in elderly men age >60 years; however, the prevalence of falls was higher in women with increasing age (Table 2). Moreover, with increasing age, the prevalence of multiple falls was also lower in elderly men age >60 years, but it was higher in women with increasing age (Table 2). The prevalence (95% CI) of a single fall (%) was similar among age strata in men and women (for men: 5.3% [1.8–14.4%], 6.8% [3.3–13.4%], 3.2% [1.4–7.3%], 5.5% [3.2–9.4%], and 7.4% [1.0–12.5%] in the age subgroups of <50 years, 50–59 years, 60–69 years, 70–79 years, and ≥80 years, respectively; for women: 11.9% [7.5–18.5%], 11.1% [7.5–16.1%], 12.0% [8.9–16.0%], 11.6% [8.6–15.6%], and 11.4% [6.7–18.9%] in the age subgroups of <50 years, 50–59 years, 60–69 years, 70–79 years, and ≥80 years, respectively).

Table 2. Prevalence of falls by age strata in men and women*
Age, yearsSingle fallMultiple falls
MenWomenMenWomen
  • *

    Values are the percentage (95% confidence interval).

<5015.8 (8.5–27.4)13.4 (8.7–20.2)10.5 (4.9–21.1)1.5 (0.4–5.3)
50–5910.7 (6.1–18.1)17.4 (12.8–23.1)3.9 (1.5–9.6)6.3 (3.7–10.4)
60–6916.7 (11.6–23.3)18.8 (14.9–23.4)13.5 (9.0–19.7)6.8 (4.5–10.1)
70–7912.4 (8.7–17.5)21.1 (16.9–25.9)6.9 (4.2–11.1)9.4 (6.7–13.1)
≥8011.1 (5.2–22.2)23.8 (16.7–32.8)3.7 (1.0–12.5)12.4 (7.4–20.0)

Table 3 shows the age, anthropometric measurements, physical performance, and prevalence of cognitive impairment among nonfallers, single fallers, and multiple fallers. One-way ANOVA showed that there were no significant associations of age, anthropometric measurements, physical performance, and prevalence of cognitive impairment with falls in men, while age and BMI were higher in multiple fallers than in nonfallers in women. With regard to physical performance, grip strength was lower and 6-meter walking time and chair stand time were longer in multiple fallers than in nonfallers and single fallers in women. Further, prevalence of cognitive impairment was also different among nonfallers, single fallers, and multiple fallers in women. Further, to determine the association of anthropometric measurements, physical performance, and cognitive impairment with single and multiple falls, we also used multinomial logistic regression analysis and found that age (odds ratio [OR] 1.04, 95% CI 1.02–1.06), BMI (OR 1.10, 95% CI 1.03–1.17), grip strength (OR 0.92, 95% CI 0.89–0.96), 6-meter walking time (OR 1.10, 95% CI 1.02–1.17), chair stand time (OR 1.06, 95% CI 1.02–1.10), and cognitive impairment (OR 3.86, 95% CI 1.67–3.83) were significantly associated with multiple falls in women.

Table 3. Comparison of characteristics among nonfallers, single fallers, and multiple fallers in men and women*
 MenWomen
NonfallersSingle fallersMultiple fallersPNonfallersSingle fallersMultiple fallersP
  • *

    Values are the mean ± SD unless indicated otherwise. One-way analysis of variance was used to determine the differences in age, height, weight, body mass index (BMI), grip strength, 6-meter walking time, normal step length, and chair stand time among nonfallers, single fallers, and multiple fallers. Chi-square test was used to determine the differences in prevalence of cognitive impairment among nonfallers, single fallers, and multiple fallers.

Subjects, no.5083148 88112780 
Age, years66.4 ± 11.767.6 ± 11.964.6 ± 11.30.5064.4 ± 12.164.3 ± 12.269.1 ± 10.40.004
Height, cm163.5 ± 7.4162.3 ± 6.3162.9 ± 5.90.56150.9 ± 6.8150.7 ± 7.7148.5 ± 7.00.01
Weight, kg62.6 ± 11.160.7 ± 10.460.3 ± 9.00.2751.8 ± 8.853.3 ± 9.252.8 ± 8.90.15
BMI, kg/m223.3 ± 3.223.0 ± 3.122.7 ± 2.80.2722.7 ± 3.423.4 ± 3.623.9 ± 3.70.002
Grip strength, kg35.8 ± 9.334.0 ± 9.635.5 ± 9.10.5723.3 ± 6.822.6 ± 6.519.9 ± 5.3< 0.001
6-meter walking time, seconds5.2 ± 2.25.8 ± 2.55.6 ± 2.30.215.5 ± 2.65.7 ± 2.66.3 ± 2.70.03
Chair stand time, seconds9.6 ± 3.610.3 ± 3.810.2 ± 3.30.3010.2 ± 4.810.5 ± 4.611.9 ± 5.10.01
Cognitive impairment, %4.66.510.60.263.35.611.70.008

To determine the association of the severity of knee OA with falls, we classified subjects as those with no knee OA (K/L grade 0 or 1), with K/L grade 2 for knee OA, and with K/L grade 3 or 4 for knee OA. The prevalence of falls in subjects with no knee OA, K/L grade 2 for knee OA, and K/L grade 3 or 4 for knee OA was 11.8%, 17.1%, and 12.5%, and 17.7%, 17.6%, and 25.6% in men and women, respectively. There were no significant associations between falls and the severity of knee OA in men (chi-square test; P = 0.27), while prevalence of falls was higher in women with K/L grade 3 or 4 for knee OA than those with no knee OA and K/L grade 2 for knee OA (P = 0.01). Similar to knee OA, we classified subjects as those with no LS (K/L grade 0 or 1), those with K/L grade 2 for LS, and those with K/L grade 3 or 4 for LS. The prevalence of falls in subjects with no LS, K/L grade 2 for LS, and K/L grade 3 or 4 for LS was 16.3%, 11.3%, and 14.0%, and 17.0%, 20.5%, and 20.7% in men and women, respectively. There were no significant associations between falls and the severity of LS in men and women (chi-square test, P = 0.38 and 0.32, respectively). We next used the chi-square test to determine the association of single and multiple falls with knee OA and LS defined as K/L grade 3 or 4 (Table 4). A chi-square test showed that no significant factors were associated with falls in men, but radiographic knee OA, knee pain, and lower back pain were significantly associated with falls in women.

Table 4. Comparison of radiographic knee OA and LS, as well as knee and lower back pain, among nonfallers, single fallers, and multiple fallers in men and women*
 MenWomen
NonfallersSingle fallersMultiple fallersPNonfallersSingle fallersMultiple fallersP
  • *

    Values are the number/total number (percentage) unless otherwise indicated. The chi-square test was used to determine the differences in radiographic findings and pain among nonfallers, single fallers, and multiple fallers. Radiographic knee OA and LS were defined as Kellgren/Lawrence grade 3 or 4. OA = osteoarthritis; LS = lumbar spondylosis.

  • Nine subjects with total knee arthroplasty were excluded.

  • One subject with incomplete information regarding knee pain was excluded.

  • §

    One subject with incomplete information regarding lower back pain was excluded.

Subjects, no.5083148 88112780 
Radiographic knee OA77/507 (15.2)4/31 (12.9)7/47 (14.9)0.9417186/875 (21.3)31/127 (24.4)33/79 (41.8)0.0002
Knee pain97/508 (19.1)3/31 (9.7)11/48 (22.9)0.3268224/880 (25.5)37/127 (29.1)37/80 (46.3)0.0003
Radiographic LS190/508 (37.4)12/31 (38.7)19/48 (39.6)0.9490318/881 (36.1)45/127 (35.4)38/80 (47.5)0.1210
Lower back pain§99/508 (19.5)10/31 (32.3)9/48 (18.8)0.2203177/880 (20.1)31/127 (24.4)28/80 (35.0)0.0062

Multinomial logistic regression analysis also showed that radiographic knee OA, LS, and knee and lower back pain were significantly associated with multiple falls in women (Table 5). Because knee pain and lower back pain were also significantly associated with grip strength, 6-meter walking time, and chair stand time in men and women (logistic regression analysis; P <0.05); to examine the independent association between radiographic knee OA, knee pain, radiographic LS, and lower back pain in women, we first used multinomial logistic regression analysis with age, BMI, cognitive impairment, radiographic knee OA, knee pain, radiographic LS, and lower back pain as independent variables (Table 5). In this analysis, only lower back pain was independently associated with multiple falls in women. In addition to the above independent variables, we also adjusted for grip strength, 6-meter walking time, and chair stand time, and found that the significant association of lower back pain with multiple falls disappeared, while knee pain was independently associated with multiple falls in women (Table 5).

Table 5. Association of radiographic knee OA and LS, as well as knee and lower back pain, with single and multiple falls in women*
 Crude OR (95% CI)Adjusted OR1 (95% CI)Adjusted OR2 (95% CI)
Single fallsMultiple fallsSingle fallsMultiple fallsSingle fallsMultiple falls
  • *

    Radiographic knee osteoarthritis (OA) and lumbar spondylosis (LS) were defined as Kellgren/Lawrence grade 3 or 4. Multinomial logistic regression analysis was used to calculate the odds ratio (OR) and 95% confidence interval (95% CI) compared with nonfallers. Eight subjects with total knee arthroplasty or incomplete information regarding pain were excluded.

  • Adjusted OR1 was calculated using multinomial logistic regression analysis with age, body mass index, cognitive impairment, radiographic knee OA, knee pain, radiographic LS, and lower back pain as independent variables.

  • Adjusted OR2 was calculated using multinomial logistic regression analysis with grip strength, 6 -meter walking time, and chair stand time in addition to the above independent variables.

Radiographic knee OA1.20 (0.76–1.83)2.66 (1.64–4.26)1.07 (0.63–1.82)1.43 (0.78–2.61)1.04 (0.60–1.77)1.31 (0.70–2.43)
Knee pain1.20 (0.79–1.81)2.52 (1.58–4.02)1.00 (0.62–1.61)1.61 (0.92–2.79)0.99 (0.60–1.61)1.87 (1.06–3.28)
Radiographic LS0.97 (0.65–1.43)1.60 (1.01–2.54)0.87 (0.57–1.32)1.12 (0.68–1.85)0.88 (0.57–1.33)1.04 (0.61–1.74)
Lower back pain1.28 (0.82–1.96)2.14 (1.30–3.46)1.34 (0.84–2.08)1.72 (1.01–2.88)1.33 (0.84–2.08)1.58 (0.91–2.70)

DISCUSSION

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

The present study is the first large-scale population-based cohort study of the prevalence of single and multiple falls and their association with radiographic knee OA and LS, as well as pain in Japanese men and women. We found that lower back pain and knee pain were independently associated with multiple falls in women.

There were distinct associations between age strata and single and multiple falls. We found that several factors were associated with multiple falls in women, but no factors were associated with a single fall in women. Previous studies have shown that associations between individual risk factors and a single fall are few in number and weak compared with risk factors for multiple falls (16). A single fall in a year could be accidental and occur due to individual as well as environmental factors, which may partly explain why there were no factors significantly associated with a single fall in our study. In contrast, several factors were associated with multiple falls in the present study, indicating that multiple falls may occur primarily due to individual factors.

In women, the prevalence of multiple falls was higher with increasing age, but in men, the prevalence of multiple falls was lower in subjects ages >60 years, although this could be a random error because of small prevalence, particularly in men. This may be partly explained by the fact that elderly men generally retire from their occupations at approximately ages 60–70 years; therefore, their environment may change and men may become more sedentary as they age, leading to lower risks of falls. Women, however, must often continue to do household chores even after age 60 years, and their environment may therefore change to a smaller extent than that of men, but their health or muscle strength continues to decline (32), leading to the higher risk of falls.

Our study is the first population-based study to examine the association between knee OA and LS diagnosed by radiography and falls in Japanese men and women. Radiographic knee OA and LS were significantly associated with multiple falls in women, but not in men, although no significant association of radiographic knee OA or LS with falls may be due to the small number of falls in men. The sex differences identified in the association between radiographic knee OA and falls may be partly explained by the weaker quadriceps muscles and increased postural sway associated with knee OA (33, 34), both of which are known to be independent risk factors for falls (16, 35). In men, muscle strength was higher than that in women in all decades (32), which may obscure the association between radiographic knee OA and falls. LS was also significantly associated with falls in this study, but the OR was lower than that for knee OA. Therefore, falls may be more strongly associated with problems of the lower extremities rather than the trunk.

After adjustment for age, BMI, and cognitive impairment, lower back pain was independently associated with multiple falls, and after adjustment for age, BMI, grip strength, cognitive impairment, 6-meter walking time, and chair stand time, knee pain was independently associated with multiple falls. Given that the significant association of radiographic knee OA and LS with multiple falls disappeared after adjustment, multiple falls may occur due to symptoms such as pain caused by radiographic knee OA or LS rather than radiographic changes in the knee or lumbar spine itself. A previous study also suggested that subjects with knee pain had an increased risk of falls (15). In other words, falls may be preventable when pain is relieved by medical care, even if subjects have radiographic knee OA or LS.

The present study has several limitations. First, this is a large-scale population-based study with a cross-sectional analysis of baseline data. Therefore, causal relationships could not be determined. The ROAD study is a longitudinal survey; therefore, further progress may help elucidate any causal relationships. Second, our subjects lived in the community, and therefore our findings may not apply to elderly persons residing in institutions. Third, we did not include other weight-bearing OA diseases, such as hip OA, in the analysis, although this disorder also affects falls (36). However, the prevalence of K/L grade 3 or 4 for hip OA is 1.4% and 3.5% in Japanese men and women (37), respectively, which is smaller than that of K/L grade 3 or 4 for knee OA in the present study. Therefore, it is possible that hip OA would not strongly affect the results of the present study. Fourth, the prevalence of fall was comparably small, particularly in men. Therefore, our results regarding the prevalence may include random error, but the present study is the first large-scale, population-based cohort study of the prevalence of falls in Japanese men and women.

In conclusion, the present cross-sectional analysis using a large-scale population from the ROAD study revealed the prevalence and factors associated with falls in men and women. In women, lower back pain and knee pain were significantly associated with multiple falls. Further studies, along with continued longitudinal surveys in the ROAD study, will help elucidate the background of knee OA and LS, and their relationship with falls.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. 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 submitted for publication. 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, En-yo, Yoshida, Nakamura, Kawaguchi, Yoshimura.

Acquisition of data. Muraki, Akune, Oka, En-yo, Yoshimura.

Analysis and interpretation of data. Muraki, Akune, Oka, Yoshimura.

Acknowledgements

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

The authors wish to thank Mrs. Tomoko Takijiri and other members of the Public Office in Hidakagawa, as well as Mrs. Tamako Tsutsumi, Mrs. Kanami Maeda, and other members of the Public Office in Taiji for their assistance in the location and scheduling of participants for examinations.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES
  • 1
    Baker S, O'Neill B, Karpf RS. The injury fact book. Lexington (MA): Lexington Books; 1984.
  • 2
    Fife D, Barancik JI, Chatterjee MS. Northeastern Ohio Trauma Study, II: injury rates by age, sex and cause. Am J Public Health 1984; 74: 4738.
  • 3
    Ministry of Health, Labour and Welfare. The outline of the results of National Livelihood Survey 2007. URL: http://www.mhlw.go.jp/toukei/list/20-19-1.html.
  • 4
    Dargent-Molina P, Favier F, Grandjean H, Baudoin C, Schott AM, Hausherr E, et al. Fall-related factors and risk of hip fracture: the EPIDOS prospective study. Lancet 1996; 348: 1459.
  • 5
    Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med 1988; 319: 17017.
  • 6
    Jackson DW, Simon TM, Aberman HM. Symptomatic articular cartilage degeneration: the impact in the new millennium. Clin Orthop Relat Res 2001; 391 Suppl: S1425.
  • 7
    Reginster JY. The prevalence and burden of arthritis. Rheumatology (Oxford) 2002; 41 Suppl: S36.
  • 8
    Buckwalter JA, Saltzman C, Brown T. The impact of osteoarthritis: implications for research. Clin Orthop Relat Res 2004; 427 Suppl: S615.
  • 9
    Sharma L, Kapoor D. Epidemiology of osteoarthritis. In: Moskowitz RW, Altman RD, Hochberg MC, Buckwalter JA, Goldberg VM, editors. Osteoarthritis: diagnosis and medical/surgical management. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2007. p. 326.
  • 10
    Hadjipavlou AG, Simmons JW, Pope MH, Necessary JT, Goel VK. Pathomechanics and clinical relevance of disc degeneration and annular tear: a point-of-view review. Am J Orthop 1999; 28: 56171.
  • 11
    Emery SE, Ringus VM. Osteoarthritis of the spine. In: Moskowitz RW, Altman RD, Hochberg MC, Buckwalter JA, Goldberg VM, editors. Osteoarthritis: diagnosis and medical/surgical management. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2007. p. 42752.
  • 12
    Muraki S, Oka H, Akune T, Mabuchi A, En-yo Y, Yoshida M, et al. Prevalence of radiographic knee osteoarthritis and its association with knee pain in the elderly of Japanese population-based cohorts: the ROAD study. Osteoarthritis Cartilage 2009; 17: 113743.
  • 13
    Muraki S, Oka H, Akune T, Mabuchi A, En-Yo Y, Yoshida M, et al. Prevalence of radiographic lumbar spondylosis and its association with low back pain in the elderly of population-based cohorts: the ROAD study. Ann Rheum Dis 2008; 68: 14016.
  • 14
    Yoshimura N, Muraki S, Oka H, Mabuch A, En-yo Y, Yoshida M, et al. Prevalence of knee osteoarthritis, lumbar spondylosis and osteoporosis in Japanese men and women: the Research on Osteoarthritis/osteoporosis Against Disability (ROAD). J Bone Miner Metab 2009; 27: 6208.
  • 15
    Arden NK, Crozier S, Smith H, Anderson F, Edwards C, Raphael H, et al. Knee pain, knee osteoarthritis, and the risk of fracture. Arthritis Rheum 2006; 55: 6105.
  • 16
    Nevitt MC, Cummings SR, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls: a prospective study. JAMA 1989; 261: 26638.
  • 17
    Linaker CH, Walker-Bone K, Palmer K, Cooper C. Frequency and impact of regional musculoskeletal disorders. Baillieres Clin Rheumatol 1999; 13: 197215.
  • 18
    Summers MN, Haley WE, Reveille JD, Alarcon GS. Radiographic assessment and psychological variables as predictors of pain and functional impairment in osteoarthritis of the knee or hip. Arthritis Rheum 1988; 31: 2049.
  • 19
    Cicuttini FM, Baker J, Hart DJ, Spector TD. Association of pain with radiological changes in different compartments and views of the knee joint. Osteoarthritis Cartilage 1996; 4: 1437.
  • 20
    Wluka AE, Wolfe R, Stuckey S, Cicuttini FM. How does tibial cartilage volume relate to symptoms in subjects with knee osteoarthritis? Ann Rheum Dis 2004; 63: 2648.
  • 21
    Kellgren JH, Lawrence JS, editors. The epidemiology of chronic rheumatism: atlas of standard radiographs of arthritis. Oxford: Blackwell Scientific; 1963.
  • 22
    Yoshimura N, Muraki S, Oka H, Kawaguchi H, Nakamura K, Akune T. Cohort profile: research on osteoarthritis/osteoporosis against disability (ROAD). Study Int J Epidemiol 2010; 39: 98895.
  • 23
    Tinetti M, Baker D, Dutcher J, Vincent J, Rozett R. Reducing the risk of falls among older adults in the community. Berkeley (CA): Peaceable Kingdom Press; 1997.
  • 24
    Shimada H, Lord SR, Yoshida H, Kim H, Suzuki T. Predictors of cessation of regular leisure-time physical activity in community-dwelling elderly people. Gerontology 2007; 53: 2937.
  • 25
    Judge JO, Davis RB III, Ounpuu S. Step length reductions in advanced age: the role of ankle and hip kinetics. J Gerontol A Biol Sci Med Sci 1996; 51: M30312.
  • 26
    Judge JO, Lindsey C, Underwood M, Winsemius D. Balance improvements in older women: effects of exercise training. Phys Ther 1993; 73: 25464.
  • 27
    Steffan TM, Hacker TA, Mollinger L. Age- and gender-related test performance in community-dwelling older people: six-minute walk test, Berg balance scale, timed up and go test, and gait speeds. Phys Ther 2002; 82: 12837.
  • 28
    Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 1994; 49: M8594.
  • 29
    Bohannon RW. Comfortable and maximum walking speed of adults aged 20-79 years: reference values and determinants. Age Ageing 1997; 26: 159.
  • 30
    Bohannon RW. Sit-to-stand test for measuring performance of lower extremity muscles. Percept Motor Skills 1995; 80: 1636.
  • 31
    Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 18998.
  • 32
    Sinaki M, Nwaogwugwu NC, Phillips BE, Mokri MP. Effect of gender, age, and anthropometry on axial and appendicular muscle strength. Am J Phys Med Rehabil 2001; 80: 3308.
  • 33
    Jones G, Nguyen T, Sambrook PN, Lord SR, Kelly PJ, Eisman JA. Osteoarthritis, bone density, postural stability, and osteoporotic fractures: a population based study. J Rheumatol 1995; 22: 9215.
  • 34
    Wegener L, Kisner C, Nichols D. Static and dynamic balance responses in persons with bilateral knee osteoarthritis. J Orthop Sports Phys Ther 1997; 25: 138.
  • 35
    Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a community-based prospective study of people 70 years and older. J Gerontol 1989; 44: Ml127.
  • 36
    Arden NK, Nevitt MC, Lane NE, Gore LR, Hochberg MC, Scott JC, et al, for the Study of Osteoporotic Fractures Research Group. Osteoarthritis and risk of falls, rates of bone loss, and osteoporotic fractures. Arthritis Rheum 1999; 42: 137885.
  • 37
    Inoue K, Wicart P, Kawasaki T, Huang J, Ushiyama T, Hukuda S, et al. Prevalence of hip osteoarthritis and acetabular dysplasia in French and Japanese adults. Rheumatology (Oxford) 2000; 39: 7458.