Abstract

Mechanical stress producing head injury is associated with Parkinson's disease, suggesting that relations with other physical hazards such as whole-body vibration (WBV) should be tested. In this study, the authors evaluated the relation between occupational exposure to WBV and Parkinson's disease. A population-based case-control study with 403 cases and 405 controls was conducted in British Columbia, Canada, between 2001 and 2008. From detailed occupational histories and published measurements, metrics of occupational WBV exposure were constructed and tested for associations with Parkinson's disease using logistic regression and adjusting for age and sex first, and then also for smoking and history of head injury. While ever being occupationally exposed to WBV was inversely associated with Parkinson's disease (odds ratio = 0.67, 95% confidence interval: 0.48, 0.94), higher intensities had consistently elevated odds ratios, with a statistically significant effect being noted for intermediate intensities when exposures were restricted to the 10 years or more prior to diagnosis. Possible mechanisms of an inverse relation between low levels of WBV exposure and Parkinson's disease could include direct protective effects or correlation with other protective effects such as exercise. Higher intensities of WBV could result in micro-injury, leading to vascular or inflammatory pathology in susceptible neurons.

Parkinson's disease (PD) is a neurodegenerative disorder that involves loss of dopamine-producing neurons, resulting in tremors, rigidity, and impaired mobility (1), together with various nonmotor symptoms. The causes are not known (2), although environmental exposures are logical targets because of the low heritability of PD (3, 4). Previous epidemiologic research on occupational risk factors has largely focused on potential chemical hazards. However, physical hazards should also be considered because head injury is associated with increased risk of PD (5, 6), possibly attributable to neuroinflammatory effects that result when injuries are incurred early in life (7). This relation has led to the hypothesis that exposure to whole-body vibration (WBV) could also be a risk factor for PD.

WBV exposure is repetitive physical displacement in any of 3 dimensions. The motion can be sinusoidal or regular (e.g., from engine vibration) and can include intermittent shocks (e.g., from travelling over uneven surfaces). Exposure most often occurs in the operation of heavy equipment and vehicles (8, 9), when the body rests on a supporting surface that itself is vibrating (10). While exposure is commonplace, variation in exposure is attributable to differences in occupation (11), suggesting that epidemiologic studies should characterize occupational exposure. A case-series analysis carried out by Iivanainen in 1975 found a correlation between occupational vibration exposure and diffuse cerebral atrophy (12). However, to our knowledge, only 1 subsequent study examined a potential neurologic risk relation—a small case-control study of Alzheimer's disease (13) that found no association with an expert-assessed dichotomous variable for WBV exposure. We conducted a population-based case-control study, and we have described elsewhere the combination of self-report and literature-derived measurements of occupational WBV used to construct quantitative exposure metrics (14).

MATERIALS AND METHODS

The source population included the 2.1 million residents of Greater Vancouver, British Columbia, Canada, and 400,000 residents of Vancouver Island (excluding the city of Victoria). Living persons between the ages of 40 and 69 years with sufficient English-language skill, health, and stamina to complete the interview were eligible. The upper age limit was applied to reduce challenges to self-reporting, including cognitive impairment and length of recall. The study used a 2-stage contact and consent procedure: Potential participants were first contacted by the British Columbia Ministry of Health to obtain their permission for contact information to be released to our study team. Those who agreed were then contacted by the team, screened for eligibility, and invited to participate in an occupational history interview. Study methods were approved by the University of British Columbia Behavioural Research Ethics Board. Interviews were conducted between 2001 and 2008.

Cases: users of antiparkinsonian medications

During the study period, the British Columbia government offered a program to reimburse residents for annual medication costs over $800. We identified cases as anyone who, between 1995 and 2002, was reimbursed for at least 1 prescription for the antiparkinsonian medications levodopa, bromocriptine mesylate, pergolide mesylate, levodopa/benserazide hydrochloride, levodopa/carbidopa, or selegiline hydrochloride. Patients who also filled prescriptions for antipsychotic drugs (who might have had parkinsonism secondary to those drugs) or were living in assisted-care facilities were excluded. The data extraction process included a random sample of other persons in the database who did not meet these criteria. This 20% “camouflage” sample was intended to disguise the disease status of list members. Consenting potential cases were screened by our team to determine eligibility. Persons who reported taking the specified drugs for a purpose other than PD were excluded. Those who reported having PD were further screened during an interview in which a detailed physical assessment checklist was used to record information on: the first symptoms noticed by the patient; the patient's family history; the use of antiparkinsonian medications and when the last dose had been taken; and the presence of symptoms, including resting tremor, stiffness, slowness of movements, loss of dexterity, changes in writing (a writing sample was obtained from each patient), loss of balance, and reduced facial expression. Interviewers observed whether resting tremor and masked face were present and were trained by a neurologist (J. K. C. T.) to administer simple tests for stiffness (wrist flexibility test) and bradykinesia (finger-tapping test). Completed forms were individually reviewed by the neurologist to confirm PD status.

Controls: registrants of the provincial health insurer

Controls were identified from the list of persons insured by the British Columbia Medical Services Plan, a public insurer that covers approximately 97% of the province's population. We applied the same age and geographic restrictions as for cases, and frequency-matched potential controls with the extracted potential case sample. Each control was randomly assigned a case's diagnosis date to censor exposures.

Exposure assessment

A more detailed description of the exposure assessment process is available elsewhere (14). Overall, our task was to derive commonly used metrics of vibration exposure (10, 15) from questionnaire data collected during in-depth interviews conducted by trained interviewers. While blinding was precluded because of the evident symptoms of PD, interviewers were not informed of the study hypotheses, to minimize the potential for bias. Where physical difficulties impaired speech, a family member occasionally helped to interpret. Participants reported their complete job histories for all jobs held longer than 6 months and were prompted on potential sources of WBV exposure by reference to a list of equipment and vehicles. For each reported exposure, participants were asked to report operations (from a list provided) conducted during use of the equipment, as well as numbers of weeks per year and hours per week in which the equipment was used.

Occupational hygienists reviewed each participant's job history (blinded to case status) to ensure that all relevant exposures were reported (in only 4 instances did the hygienists identify possible unreported vibration exposures). We then applied restrictions to ensure that reported exposures were likely to be above background levels. We excluded reports of exposure to vehicles which were indirect only (e.g., working near equipment but having no direct contact). We also restricted all reports to exposures incurred for more than 30 minutes/week (10 hours/week for cars and 5 hours/week for vans and light trucks).

To construct metrics of exposure, we obtained measurements of vibration intensity for each equipment type from relevant peer-reviewed literature on vibration exposure assessment (the literature search strategy and the 13 articles used are described in detail elsewhere (14)). We calculated 3 metrics to capture different elements of lifetime equipment exposure for each individual: All metrics were initially calculated to include all exposures incurred up to the year of diagnosis, and then recalculated to exclude any exposures occurring less than 10 years and less than 20 years before diagnosis.

  1. A duration metric that summed the total working years of exposure across all sources of WBV exposure.

  2. A greatest-intensity metric that reflected the piece of equipment with the greatest estimated vibration intensity (derived from peer-reviewed measurement articles) of all equipment exposures reported. Vibration intensity is measured as acceleration (in m/s2), and it increases with the amplitude (displacement) of vibration oscillations. The measurement articles we referenced used customary summary measures of vibration intensity, such as the root mean square vector sum acceleration (10).

  3. A cumulative-dose metric that multiplied vibration intensity (raised to the fourth power) and duration of use, summed across all of the reported equipment exposures. This metric was similar to that in other vibration exposure articles used to summarize dose, in that it multiplies intensity estimates by duration of exposure, although sources differ as to the power to which intensity estimates are raised (10, 15). We used a fourth-power exponent on intensity estimates, similar to that used by Griffin (10). This allows the cumulative metric to better reflect both intensity and duration; without this, duration is highly correlated with the cumulative metric.

Statistical analyses

We constructed unconditional logistic regression models to compare odds of PD in persons occupationally exposed to vibration with the odds for those with no occupational exposure, and to examine dose-response relations using categorized metrics of exposure. Categories for each metric were based on quartiles among the exposed. We used the lowest nonzero exposure category as the reference group, because we suspected that persons who were exposed could represent a distinct subpopulation. The first models were adjusted for age (year of birth in 5-year categories) and sex. In a second set of models, we also adjusted for smoking (in pack-years), because occupation can be related to smoking behavior (16) and smoking is associated with an inverse risk of PD (17). Previous head injury (ever/never) was also included as a covariate, because it is associated with PD (6) and might be related to employment in industries involving vibration exposure. In a previous analysis in the data set, Rugbjerg et al. (18) had examined the relation between PD and pesticides; we used the hygienist-reviewed pesticide exposure variable from these analyses to adjust the vibration results in a third set of models. Because of the insidious onset of PD, we were interested in the effect of including only those exposures that occurred well before diagnosis. Therefore, we constructed regression models to test relations for exposures and covariates censored at 10 and 20 years prior to diagnosis. Analyses were performed using SAS, version 9.2 (SAS Institute Inc., Cary, North Carolina).

RESULTS

We recruited and interviewed 808 participants (403 cases and 405 controls). Figure 1 shows the classification of all potential participants. Eligibility was unknown for the potential participants who refused further contact or were never contacted. However, if we assume that the proportion of contacted subjects who were eligible was the same as that in the samples initially extracted by the Ministry of Health, the participation rate was 51% for cases and 32% for controls (see the article by Rugbjerg et al. (18) for calculation details).

Figure 1.

Distribution of 3,783 potential participants for a case-control study of Parkinson's disease (PD), British Columbia (BC), Canada, 2001–2008. Potential cases were extracted from records of the provincial Pharmacare medication reimbursement program, representing users of antiparkinsonian medications who were reimbursed for prescription costs between 1995 and 2002. A 20% camouflage sample of nonusers was added to the extraction to disguise commonalities. Controls were extracted from the client registry of the provincial health insurer. (UBC, University of British Columbia).

Figure 1.

Distribution of 3,783 potential participants for a case-control study of Parkinson's disease (PD), British Columbia (BC), Canada, 2001–2008. Potential cases were extracted from records of the provincial Pharmacare medication reimbursement program, representing users of antiparkinsonian medications who were reimbursed for prescription costs between 1995 and 2002. A 20% camouflage sample of nonusers was added to the extraction to disguise commonalities. Controls were extracted from the client registry of the provincial health insurer. (UBC, University of British Columbia).

The average age of cases at the time of interview was 65.0 years (standard deviation, 6.6 years), and for controls it was 62.2 years (standard deviation, 9.0 years). The average age at PD diagnosis among cases was 56.0 years. Although the initially extracted potential case and control samples were frequency-matched on age and sex, Table 1 shows that the final samples were not. Many of the potential cases were ineligible because they used antiparkinsonian drugs for reasons other than PD (see Figure 1). This ineligibility was related to age and sex: The non-Parkinson's users of antiparkinsonian drugs were more likely to be younger and female; therefore, adjustment for these variables was particularly important. While most participants were not occupationally exposed to WBV, the exposure was not uncommon (Table 2).

Table 1.

Parkinson's Disease-related Characteristics of Cases and Controls in a Study of Occupational Whole-Body Vibration Exposure and Parkinson's Disease Risk, British Columbia, Canada, 2001–2008

Characteristic Cases
 
Controls
 
No. % No. % 
Sex 
 Male 266 66.0 204 50.4 
 Female 137 34.0 201 49.6 
Year of birth 
 1929–1933 109 27.0 81 20.0 
 1934–1938 136 33.8 94 23.2 
 1939–1943 75 18.6 68 16.8 
 1944–1948 56 13.9 61 15.1 
 1949–1953 20 5.0 51 12.6 
 1953–1958 1.7 50 12.4 
History of head injury 
 Never 311 77.2 354 87.4 
 Ever 92 22.8 51 12.6 
History of pesticide exposure 
 Never 366 90.8 382 94.3 
 Ever 37 9.2 23 5.7 
Mean cumulative pack-years of smoking 11.4 (20.4)a  15.4 (22.4)  
Characteristic Cases
 
Controls
 
No. % No. % 
Sex 
 Male 266 66.0 204 50.4 
 Female 137 34.0 201 49.6 
Year of birth 
 1929–1933 109 27.0 81 20.0 
 1934–1938 136 33.8 94 23.2 
 1939–1943 75 18.6 68 16.8 
 1944–1948 56 13.9 61 15.1 
 1949–1953 20 5.0 51 12.6 
 1953–1958 1.7 50 12.4 
History of head injury 
 Never 311 77.2 354 87.4 
 Ever 92 22.8 51 12.6 
History of pesticide exposure 
 Never 366 90.8 382 94.3 
 Ever 37 9.2 23 5.7 
Mean cumulative pack-years of smoking 11.4 (20.4)a  15.4 (22.4)  

a Numbers in parentheses, standard deviation.

Table 2.

Vibration-related Characteristics of Cases and Controls in a Study of Occupational Whole-Body Vibration Exposure and Parkinson's Disease Risk, British Columbia, Canada, 2001–2008

Characteristic Cases
 
Controls
 
No. % No. % 
History of occupational exposure to whole-body vibration 
 Ever 145 36.0 147 36.3 
 Never 258 64.0 258 63.7 
Greatest intensity, m/s2 
 No occupational exposure 258 64.0 258 63.7 
  > 0–0.68 44 10.9 61 15.1 
  > 0.68–0.88 42 10.4 44 10.9 
  > 0.88–1.19 23 5.7 17 4.2 
  > 1.19 36 8.9 25 6.2 
Duration, WYEa 
 No occupational exposure 258 64.0 258 63.7 
  > 0–2.80 33 8.2 40 9.9 
  > 2.80–8.48 40 9.9 33 8.2 
  > 8.48–21.69 39 9.7 34 8.4 
  > 21.69 33 8.2 40 9.9 
Cumulative dose, m4/s8 . WYE 
 No occupational exposure 258 64.0 258 63.7 
  > 0–0.61 32 7.9 41 10.1 
  > 0.61–2.48 39 9.7 34 8.4 
  > 2.48–7.23 33 8.2 40 9.9 
  > 7.23 41 10.2 32 7.9 
Characteristic Cases
 
Controls
 
No. % No. % 
History of occupational exposure to whole-body vibration 
 Ever 145 36.0 147 36.3 
 Never 258 64.0 258 63.7 
Greatest intensity, m/s2 
 No occupational exposure 258 64.0 258 63.7 
  > 0–0.68 44 10.9 61 15.1 
  > 0.68–0.88 42 10.4 44 10.9 
  > 0.88–1.19 23 5.7 17 4.2 
  > 1.19 36 8.9 25 6.2 
Duration, WYEa 
 No occupational exposure 258 64.0 258 63.7 
  > 0–2.80 33 8.2 40 9.9 
  > 2.80–8.48 40 9.9 33 8.2 
  > 8.48–21.69 39 9.7 34 8.4 
  > 21.69 33 8.2 40 9.9 
Cumulative dose, m4/s8 . WYE 
 No occupational exposure 258 64.0 258 63.7 
  > 0–0.61 32 7.9 41 10.1 
  > 0.61–2.48 39 9.7 34 8.4 
  > 2.48–7.23 33 8.2 40 9.9 
  > 7.23 41 10.2 32 7.9 

Abbreviation: WYE, working-year equivalents.

a 1 working year was defined as equivalent to 2,000 hours.

Occupational exposure to WBV (versus none) was inversely related to PD (Table 3). This relation was apparent only after adjustment for sex, because WBV exposure was much more common in men (14). A similar effect was observed with categories of most intense vibrating exposure: Persons with no occupational exposure had greater odds of PD than those in the lowest categories of exposure (Table 4).

Table 3.

Associations Between Parkinson's Disease and Occupational Exposure to Whole-Body Vibration Among 403 Cases and 405 Controls, British Columbia, Canada, 2001–2008

Exposure Period and WBV Exposure Status Model 1a
 
Model 2b
 
Model 3c
 
OR 95% CI OR 95% CI OR 95%CI 
Before diagnosis 
 Never 1.00  1.00  1.00  
 Ever 0.67* 0.48, 0.94 0.71 0.50, 1.00 0.67* 0.47, 0.96 
≥10 years before diagnosis 
 Never 1.00  1.00  1.00  
 Ever 0.67* 0.47, 0.94 0.69* 0.49, 0.99 0.67* 0.47, 0.96 
≥20 years before diagnosis 
 Never 1.00  1.00  1.00  
 Ever 0.64* 0.45, 0.91 0.66* 0.46, 0.95 0.63* 0.44, 0.92 
Exposure Period and WBV Exposure Status Model 1a
 
Model 2b
 
Model 3c
 
OR 95% CI OR 95% CI OR 95%CI 
Before diagnosis 
 Never 1.00  1.00  1.00  
 Ever 0.67* 0.48, 0.94 0.71 0.50, 1.00 0.67* 0.47, 0.96 
≥10 years before diagnosis 
 Never 1.00  1.00  1.00  
 Ever 0.67* 0.47, 0.94 0.69* 0.49, 0.99 0.67* 0.47, 0.96 
≥20 years before diagnosis 
 Never 1.00  1.00  1.00  
 Ever 0.64* 0.45, 0.91 0.66* 0.46, 0.95 0.63* 0.44, 0.92 

Abbreviations: CI, confidence interval; OR, odds ratio; WBV, whole-body vibration.

* P < 0.05.

a Adjusted for age (in 5-year intervals) and sex.

b Adjusted for age (in 5-year intervals), sex, smoking (in pack-years), and head injury (ever).

cAdjusted for age (in 5-year intervals), sex, smoking (in pack-years), head injury (ever), and hygienist-reviewed pesticide exposure (ever).

Table 4.

Associations Between Parkinson's Disease and Intensity of Occupational Exposure to Whole-Body Vibration Among 403 Cases and 405 Controls, British Columbia, Canada, 2001–2008

Exposure Period and Greatest Intensity, m/s2 Model 1a
 
Model 2b
 
Model 3c
 
OR 95% CI OR 95% CI OR 95% CI 
Before diagnosis 
 No occupational exposure 1.97* 1.02, 2.60 1.81* 1.13, 2.91 1.80* 1.12, 2.89 
  > 0–0.68 1.00  1.00  1.00  
  > 0.68–0.88 1.35 0.74, 2.47 1.23 0.67, 2.28 1.18 0.63, 2.19 
  > 0.88–1.19 1.79 0.82, 3.89 1.98 0.89, 4.41 1.78 0.79, 4.02 
  > 1.19 1.74 0.90, 3.37 1.65 0.84, 3.24 1.51 0.76, 3.00 
≥10 years before diagnosis 
 No occupational exposure 2.14* 1.33, 3.45 1.97* 1.21, 3.21 1.97* 1.21, 3.20 
  > 0–0.68 1.00  1.00    
  > 0.68–0.88 1.51 0.81, 2.80 1.37 0.73, 2.56 1.33 0.71, 2.50 
  > 0.88–1.19 2.12 0.95, 2.36 2.35* 1.04, 5.33 2.22 0.97, 5.10 
  > 1.19 1.96 0.97, 3.97 1.76 0.88, 3.44 1.67 0.81, 3.46 
≥20 years before diagnosis 
 No occupational exposure 2.06* 1.15, 3.11 1.98* 1.19, 3.29 1.98* 1.19, 3.29 
  > 0–0.68 1.00  1.00    
  > 0.68–0.88 1.43 0.73, 2.81 1.38 0.70, 2.75 1.35 0.68, 2.69 
  > 0.88–1.19 1.70 0.72, 4.02 1.98 0.82, 4.80 1.85 0.75, 4.57 
  > 1.19 1.63 0.77, 3.45 1.51 0.70, 3.24 1.43 0.66, 3.13 
Exposure Period and Greatest Intensity, m/s2 Model 1a
 
Model 2b
 
Model 3c
 
OR 95% CI OR 95% CI OR 95% CI 
Before diagnosis 
 No occupational exposure 1.97* 1.02, 2.60 1.81* 1.13, 2.91 1.80* 1.12, 2.89 
  > 0–0.68 1.00  1.00  1.00  
  > 0.68–0.88 1.35 0.74, 2.47 1.23 0.67, 2.28 1.18 0.63, 2.19 
  > 0.88–1.19 1.79 0.82, 3.89 1.98 0.89, 4.41 1.78 0.79, 4.02 
  > 1.19 1.74 0.90, 3.37 1.65 0.84, 3.24 1.51 0.76, 3.00 
≥10 years before diagnosis 
 No occupational exposure 2.14* 1.33, 3.45 1.97* 1.21, 3.21 1.97* 1.21, 3.20 
  > 0–0.68 1.00  1.00    
  > 0.68–0.88 1.51 0.81, 2.80 1.37 0.73, 2.56 1.33 0.71, 2.50 
  > 0.88–1.19 2.12 0.95, 2.36 2.35* 1.04, 5.33 2.22 0.97, 5.10 
  > 1.19 1.96 0.97, 3.97 1.76 0.88, 3.44 1.67 0.81, 3.46 
≥20 years before diagnosis 
 No occupational exposure 2.06* 1.15, 3.11 1.98* 1.19, 3.29 1.98* 1.19, 3.29 
  > 0–0.68 1.00  1.00    
  > 0.68–0.88 1.43 0.73, 2.81 1.38 0.70, 2.75 1.35 0.68, 2.69 
  > 0.88–1.19 1.70 0.72, 4.02 1.98 0.82, 4.80 1.85 0.75, 4.57 
  > 1.19 1.63 0.77, 3.45 1.51 0.70, 3.24 1.43 0.66, 3.13 

Abbreviations: CI, confidence interval; OR, odds ratio.

* P < 0.05.

a Adjusted for age (in 5-year intervals) and sex.

b Adjusted for age (in 5-year intervals), sex, smoking (in pack-years), and head injury (ever).

cAdjusted for age (in 5-year intervals), sex, smoking (in pack-years), head injury (ever), and hygienist-reviewed pesticide exposure (ever).

In contrast, the higher intensities of exposure were associated with increased odds of PD when compared with the lowest nonzero intensities of exposure (Table 4). These associations were not substantially attenuated when results were adjusted for smoking and previous head injury; indeed, some of the point estimates were greater following adjustment. The effect of intensity was modestly attenuated after adjustment for reported exposure to pesticides (as reviewed by an occupational hygienist).

The relations between PD and both duration and cumulative dose of WBV were less consistent. Several categories of duration of exposure showed null or inverse associations, while others were elevated (Table 5). Cumulative dose of WBV showed a pattern similar to that observed for intensity, although with weaker effects (Table 6).

Table 5.

Associations Between Parkinson's Disease and Duration of Occupational Exposure to Whole-Body Vibration Among 403 Cases and 405 Controls, British Columbia, Canada, 2001–2008

Exposure Period and Duration, WYEa Model 1b
 
Model 2c
 
Model 3d
 
OR 95% CI OR 95% CI OR 95% CI 
Before diagnosis 
 No occupational exposure 1.43 0.84, 2.45 1.49 0.87, 2.57 1.57 0.91, 2.71 
  > 0–2.80 1.00  1.00  1.00  
  > 2.80–8.48 1.27 0.64, 2.53 1.48 0.73, 2.99 1.47 0.72, 3.00 
  > 8.48–21.69 0.98 0.49, 1.96 1.05 0.52, 2.10 1.05 0.52, 2.11 
  > 21.69 0.65 0.32, 1.29 0.78 0.38, 1.57 0.77 0.38, 1.56 
≥10 years before diagnosis 
 No occupational exposure 1.00 0.60, 1.69 1.05 0.62, 1.77 1.09 0.64, 1.85 
  > 0–2.80 1.00  1.00  1.00  
  > 2.80–8.48 0.57 0.29, 1.12 0.64 0.32, 1.27 0.64 0.32, 1.28 
  > 8.48–21.69 0.62 0.32, 1.20 0.66 0.34, 1.29 0.66 0.34, 1.29 
  > 21.69 0.44* 0.21, 0.95 0.55 0.25, 1.20 0.54 0.25, 1.19 
≥20 years before diagnosis 
 No occupational exposure 1.34 0.81, 2.22 1.42 0.85, 2.36 1.47 0.88, 2.46 
  > 0–2.80 1.00  1.00  1.00  
  > 2.80–8.48 0.76 0.39, 1.45 0.83 0.42, 1.65 0.84 0.42, 1.66 
  > 8.48–21.69 0.76 0.38, 1.51 0.85 0.42, 1.71 0.84 0.42, 1.70 
  > 21.69 1.04 0.31, 3.52 1.88 0.51, 6.90 1.89 0.52, 6.96 
Exposure Period and Duration, WYEa Model 1b
 
Model 2c
 
Model 3d
 
OR 95% CI OR 95% CI OR 95% CI 
Before diagnosis 
 No occupational exposure 1.43 0.84, 2.45 1.49 0.87, 2.57 1.57 0.91, 2.71 
  > 0–2.80 1.00  1.00  1.00  
  > 2.80–8.48 1.27 0.64, 2.53 1.48 0.73, 2.99 1.47 0.72, 3.00 
  > 8.48–21.69 0.98 0.49, 1.96 1.05 0.52, 2.10 1.05 0.52, 2.11 
  > 21.69 0.65 0.32, 1.29 0.78 0.38, 1.57 0.77 0.38, 1.56 
≥10 years before diagnosis 
 No occupational exposure 1.00 0.60, 1.69 1.05 0.62, 1.77 1.09 0.64, 1.85 
  > 0–2.80 1.00  1.00  1.00  
  > 2.80–8.48 0.57 0.29, 1.12 0.64 0.32, 1.27 0.64 0.32, 1.28 
  > 8.48–21.69 0.62 0.32, 1.20 0.66 0.34, 1.29 0.66 0.34, 1.29 
  > 21.69 0.44* 0.21, 0.95 0.55 0.25, 1.20 0.54 0.25, 1.19 
≥20 years before diagnosis 
 No occupational exposure 1.34 0.81, 2.22 1.42 0.85, 2.36 1.47 0.88, 2.46 
  > 0–2.80 1.00  1.00  1.00  
  > 2.80–8.48 0.76 0.39, 1.45 0.83 0.42, 1.65 0.84 0.42, 1.66 
  > 8.48–21.69 0.76 0.38, 1.51 0.85 0.42, 1.71 0.84 0.42, 1.70 
  > 21.69 1.04 0.31, 3.52 1.88 0.51, 6.90 1.89 0.52, 6.96 

Abbreviations: CI, confidence interval; OR, odds ratio; WYE, working-year equivalents.

* P < 0.05.

a 1 working year was defined as equivalent to 2,000 hours.

b Adjusted for age (in 5-year intervals) and sex.

cAdjusted for age (in 5-year intervals), sex, smoking (in pack-years), and head injury (ever).

dAdjusted for age (in 5-year intervals), sex, smoking (in pack-years), head injury (ever), and hygienist-reviewed pesticide exposure (ever).

Table 6.

Associations Between Parkinson's Disease and Cumulative Dose of Occupational Exposure to Whole-Body Vibration Among 403 Cases and 405 Controls, British Columbia, Canada, 2001–2008

Exposure Period and Cumulative Dose, m4/s8. WYEa Model 1b
 
Model 2c
 
Model 3d
 
OR 95% CI OR 95% CI OR 95% CI 
Before diagnosis 
 No occupational exposure 1.59 0.94, 2.71 1.56 0.91, 2.68 1.60 0.93, 2.76 
  > 0–0.61 1.00  1.00  1.00  
  > 0.61–2.48 1.31 0.66, 2.62 1.32 0.66, 2.66 1.33 0.66, 2.68 
  > 2.48–7.23 0.87 0.44, 1.72 0.90 0.45, 1.79 0.87 0.43, 1.75 
  > 7.23 1.13 0.57, 2.23 1.28 0.63, 2.58 1.16 0.57, 2.37 
≥10 years before diagnosis 
 No occupational exposure 1.38 0.82, 2.34 1.39 0.82, 2.37 1.42 0.83, 2.43 
  > 0–0.61 1.00  1.00  1.00  
  > 0.61–2.48 0.90 0.46, 1.77 0.92 0.47, 1.82 0.92 0.47, 1.83 
  > 2.48–7.23 0.74 0.37, 1.48 0.79 0.39, 1.60 0.77 0.38, 1.57 
  > 7.23 1.06 0.52, 2.15 1.26 0.61, 2.62 1.18 0.56, 2.48 
≥20 years before diagnosis 
 No occupational exposure 1.42 0.84, 2.41 1.47 0.86, 2.51 1.49 0.87, 2.55 
  > 0–0.61 1.00  1.00  1.00  
  > 0.61–2.48 0.75 0.39, 1.45 0.81 0.41, 1.58 0.80 0.41, 1.57 
  > 2.48–7.23 0.96 0.40, 2.34 1.01 0.41, 2.50 0.94 0.38, 2.35 
  > 7.23 1.05 0.48, 2.29 1.29 0.57, 2.93 1.23 0.54, 2.81 
Exposure Period and Cumulative Dose, m4/s8. WYEa Model 1b
 
Model 2c
 
Model 3d
 
OR 95% CI OR 95% CI OR 95% CI 
Before diagnosis 
 No occupational exposure 1.59 0.94, 2.71 1.56 0.91, 2.68 1.60 0.93, 2.76 
  > 0–0.61 1.00  1.00  1.00  
  > 0.61–2.48 1.31 0.66, 2.62 1.32 0.66, 2.66 1.33 0.66, 2.68 
  > 2.48–7.23 0.87 0.44, 1.72 0.90 0.45, 1.79 0.87 0.43, 1.75 
  > 7.23 1.13 0.57, 2.23 1.28 0.63, 2.58 1.16 0.57, 2.37 
≥10 years before diagnosis 
 No occupational exposure 1.38 0.82, 2.34 1.39 0.82, 2.37 1.42 0.83, 2.43 
  > 0–0.61 1.00  1.00  1.00  
  > 0.61–2.48 0.90 0.46, 1.77 0.92 0.47, 1.82 0.92 0.47, 1.83 
  > 2.48–7.23 0.74 0.37, 1.48 0.79 0.39, 1.60 0.77 0.38, 1.57 
  > 7.23 1.06 0.52, 2.15 1.26 0.61, 2.62 1.18 0.56, 2.48 
≥20 years before diagnosis 
 No occupational exposure 1.42 0.84, 2.41 1.47 0.86, 2.51 1.49 0.87, 2.55 
  > 0–0.61 1.00  1.00  1.00  
  > 0.61–2.48 0.75 0.39, 1.45 0.81 0.41, 1.58 0.80 0.41, 1.57 
  > 2.48–7.23 0.96 0.40, 2.34 1.01 0.41, 2.50 0.94 0.38, 2.35 
  > 7.23 1.05 0.48, 2.29 1.29 0.57, 2.93 1.23 0.54, 2.81 

Abbreviations: CI, confidence interval; OR, odds ratio; WYE, working-year equivalents.

a 1 working year was defined as equivalent to 2,000 hours.

b Adjusted for age (in 5-year intervals) and sex.

cAdjusted for age (in 5-year intervals), sex, smoking (in pack-years), and head injury (ever).

dAdjusted for age (in 5-year intervals), sex, smoking (in pack-years), head injury (ever), and hygienist-reviewed pesticide exposure (ever).

DISCUSSION

Our results suggest a possible nonlinear relation between occupational WBV exposure and PD, in which persons with low levels of WBV exposure may be at reduced risk compared with the unexposed, while those with higher intensities of exposure may be at increased risk. We are cautious in our interpretation of these newly observed associations, but we consider potential causal and noncausal explanations for our observations below.

The inverse association with low levels of occupational WBV exposure could be due to a causally protective effect against PD. Very-low-intensity and -frequency WBV has been examined as a possible treatment for PD, with the hypothesized effect of improvements in proprioception and ease of movement (19, 20), although a trial that included a placebo found no benefit (21). Furthermore, even clear treatment benefits might not indicate a protective effect for premorbid exposure.

A second possibility is that persons who work in industries involving vibration exposure may share an unmeasured protective exposure (e.g., physical activity). Physical activity has been hypothesized as a protective factor for PD because forced exercise in parkinsonian animal models was found to preserve dopamine production (22). In epidemiologic cohort studies, greater levels of physical activity long before diagnosis were associated with lower risk of PD (23, 24). Some vibration-exposed work (such as agricultural and construction work) could be hypothesized to entail physical activity above the population background level. However, the correlation is imperfect, because other vibration-exposed work would be considered sedentary (driving of buses or semitrailer trucks). Furthermore, there is evidence that those who work in exposed industries may be less likely to participate in leisure (non-work-related) physical activity (25), so it is not clear that WBV exposure would be a valid proxy for total physical activity.

A third possible explanation for the inverse relation between ever having occupational WBV exposure and PD is that persons who are susceptible to the effects of vibration may be more sensitive to it and subsequently avoid exposure; that is, they leave the exposed workforce earlier or do not enter the exposed workforce. An analogous effect has been observed in respiratory epidemiology, where persons who are most sensitive to exacerbating exposures modify their occupations to avoid exposure (26).

Higher-intensity equipment exposure was positively associated with PD. While most of these results were not significant at the P < 0.05 level, many approached statistical significance, and the point estimates were consistently greater than 1 for the higher categories of intensity (contrasting with the unstable associations between PD and duration of exposure). The opposing effects of background exposure and high intensities of exposure highlight the importance of examining dose-response relations and may be one reason why a previous study of Alzheimer's disease and WBV using a dichotomous construction of exposure did not detect an effect (13). The associations between PD and higher categories of most-intense-vibration exposure were slightly attenuated by adjustment for pesticide exposure; however, the point estimate for pesticide exposure was also weakened by inclusion in models with vibration intensity, and indeed was weaker than the point estimates for the top 2 categories of vibration exposure intensity. In our previous analysis of pesticide exposure, we did not find strong associations with PD and found some evidence that associations between PD and pesticide exposure may be subject to recall bias (18). To assess the potential for recall bias, our interview included an open-ended question: “What do you think causes PD?” Only 1 participant made reference to vibration in the response, whereas 154 participants reported “chemicals” as a suspected cause (18).

If higher intensities of WBV could be causally related to PD, what could the mechanism of this effect be? Previous studies of WBV have found that the accelerative forces associated with WBV are indeed transmitted to the head (27–29). Transmissibility is affected by the axis in which vibration occurs (27–29), body posture (27–29), and the presence of head rests (29) and back rests (30). The effect of this transmission on the brain itself is not as well-studied. However, increases in cerebral blood flow and oxygenation have been observed in men exposed to WBV at frequencies comparable to those of vehicle exposures (31). Although this increased blood flow may not be a hazardous effect, it demonstrates a response of brain tissue to WBV exposure. Vascular impairment is a known risk of vibration exposure, particularly in the extremities (32). Curry et al. (33) found that high-intensity vibration exposure damaged rat arterial cells after only 9 hours of exposure.

Other animal studies have suggested a possible immunoreactive response to vibration stress (34, 35), which may relate to an inflammatory response (36). Neuroinflammation has been suggested as an important factor in the development of PD (7). Liu et al. (7) discussed neuroinflammation as a mechanism by which both infection and injury may increase risk of PD. Hachiya et al. (37) hypothesized that mechanical stress may stimulate the production of protein aggregates which are associated with PD. This mechanical stress was discussed in the context of head injuries, but it is possible that the large and repetitive shocks associated with high-intensity vibration exposure could exert effects comparable to those of single head-injury events. Further study is required to test possible associations between occupational WBV exposure and PD and to elucidate the mechanism of action. It would be useful to study the acute effects of WBV exposure of differing intensities on previously observed biomarkers of neuroinflammation (38).

Because this study indicates intensity of vibration as more relevant than duration of exposure, investigators attempting to replicate our findings must carefully construct metrics of exposure that distinguish among intensities of exposure. Job titles or expert assessments that depend on job or industry to make judgments about exposure probably will not suffice. For example, persons in the highest category of vibration intensity were exposed to equipment such as snowmobiles, high-speed marine craft, tanks, and motorized dirt bikes (14) while working in a diverse array of jobs. Where exposure might be specific to job title or industry (e.g., tank use was military), job title or industry might not be specific to exposure, since many workers with that job title or within that industry may not be exposed. Job title and industry (or expert assessment that relies on these) might not be able to identify exposure to equipment items such as snowmobiles and marine craft, which were used during work in recreation, law enforcement, conservation, and even the arts (film). No single industry dominated among persons exposed to higher-intensity equipment. While this highlights the limitations of job title as an exposure variable, it also suggests that it is unlikely that another exposure shared by the diverse group of workers represented could straightforwardly confound and explain the findings presented here.

Our study included prevalent cases of PD, but survival bias is less likely in PD because of the long survival of PD patients (39). Relying on self-reports of equipment exposure may have introduced error into our exposure assessment, although we attempted to minimize this by using recognizable prompts. A test of the validity of self-reported WBV exposure suggested that self-reports are fairly accurate (40), but the periods of recall tested were much shorter than those in the current study, which was a study of complete occupational history. Although there are no studies of the reliability and accuracy of self-reported WBV exposure in PD patients, self-reported environmental exposures and PD show high reliability, with no differences in reliability between cases and controls (41). To address the nondifferential errors associated with retrospective and self-reported exposures, a future cohort study could incorporate contemporary measurements of occupational WBV exposure, perhaps for a subset of exposed participants.

To our knowledge, this study was the first to examine a possible relation between occupational WBV exposure and PD. Our results suggest that this relation warrants further scrutiny in the continuing effort to explain and prevent the occurrence of PD.

ACKNOWLEDGMENTS

Author affiliations: School of Population and Public Health, Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada (M. Anne Harris, Stephen A. Marion, John J. Spinelli, Kay Teschke); Pacific Parkinson's Research Centre, University of British Columbia, Vancouver, British Columbia, Canada (Joseph K. C. Tsui); Cancer Control Research, British Columbia Cancer Agency, Vancouver, British Columbia, Canada (John J. Spinelli); and Occupational Cancer Research Centre, Cancer Care Ontario, Vancouver, British Columbia, Canada (M. Anne Harris).

This study was funded by the Medical Research Council of Canada (now the Canadian Institutes of Health Research), WorkSafe BC, and the Pacific Parkinson's Research Centre. M. A. H. also acknowledges salary support from the Canadian Institutes of Health Research, the Public Health Agency of Canada, the Michael Smith Foundation for Health Research, and the University of British Columbia Bridge Program.

The authors thank Dr. Hui Shen for valuable analytic assistance and support. They thank the staffs of both the University of British Columbia and the British Columbia Ministry of Health for their contributions. The authors also appreciate the guidance on exposure assessment received from Dr. Peter Cripton.

Conflict of interest: none declared.

REFERENCES

1
Weiner
WJ
A differential diagnosis of Parkinsonism
Rev Neurol Dis
 , 
2005
, vol. 
2
 
3
(pg. 
124
-
131
)
2
Obeso
JA
Rodriguez-Oroz
MC
Goetz
CG
, et al.  . 
Missing pieces in the Parkinson's disease puzzle
Nat Med
 , 
2010
, vol. 
16
 
6
(pg. 
653
-
661
)
3
Tanner
CM
Ottman
R
Goldman
SM
, et al.  . 
Parkinson disease in twins: an etiologic study
JAMA
 , 
1999
, vol. 
281
 
4
(pg. 
341
-
346
)
4
Wirdefeldt
K
Gatz
M
Schalling
M
, et al.  . 
No evidence for heritability of Parkinson disease in Swedish twins
Neurology
 , 
2004
, vol. 
63
 
2
(pg. 
305
-
311
)
5
Dick
FD
De Palma
G
Ahmadi
A
, et al.  . 
Environmental risk factors for Parkinson's disease and parkinsonism: the Geoparkinson Study
Occup Environ Med
 , 
2007
, vol. 
64
 
10
(pg. 
666
-
672
)
6
Goldman
SM
Tanner
CM
Oakes
D
, et al.  . 
Head injury and Parkinson's disease risk in twins
Ann Neurol
 , 
2006
, vol. 
60
 
1
(pg. 
65
-
72
)
7
Liu
B
Gao
HM
Hong
JS
Parkinson's disease and exposure to infectious agents and pesticides and the occurrence of brain injuries: role of neuroinflammation
Environ Health Perspect
 , 
2003
, vol. 
111
 
8
(pg. 
1065
-
1073
)
8
Cann
AP
Salmoni
AW
Vi
P
, et al.  . 
An exploratory study of whole-body vibration exposure and dose while operating heavy equipment in the construction industry
Appl Occup Environ Hyg
 , 
2003
, vol. 
18
 
12
(pg. 
999
-
1005
)
9
Futatsuka
M
Maeda
S
Inaoka
T
, et al.  . 
Whole-body vibration and health effects in the agricultural machinery drivers
Ind Health
 , 
1998
, vol. 
36
 
2
(pg. 
127
-
132
)
10
Griffin
MJ
Handbook of Human Vibration
 , 
1990
London, United Kingdom
Elsevier Academic Press
11
Palmer
KT
Griffin
MJ
Bendall
H
, et al.  . 
Prevalence and pattern of occupational exposure to whole body vibration in Great Britain: findings from a national survey
Occup Environ Med
 , 
2000
, vol. 
57
 
4
(pg. 
229
-
236
)
12
Iivanainen
M
Statistical correlations of diffuse cerebral atrophy, with special reference to diagnostic and aetiological clues
Acta Neurol Scand
 , 
1975
, vol. 
51
 
5
(pg. 
365
-
379
)
13
Gun
RT
Korten
AE
Jorm
AF
, et al.  . 
Occupational risk factors for Alzheimer disease: a case-control study
Alzheimer Dis Assoc Disord
 , 
1997
, vol. 
11
 
1
(pg. 
21
-
27
)
14
Harris
MA
Cripton
PA
Teschke
K
Retrospective assessment of occupational exposure to whole body vibration for a case-control study
J Occup Environ Hyg
 , 
2012
, vol. 
9
 
6
(pg. 
371
-
380
)
15
Boshuizen
HC
Bongers
PM
Hulshof
CT
Self-reported back pain in tractor drivers exposed to whole-body vibration
Int Arch Occup Environ Health
 , 
1990
, vol. 
62
 
2
(pg. 
109
-
115
)
16
Bang
KM
Kim
JH
Prevalence of cigarette smoking by occupation and industry in the United States
Am J Ind Med
 , 
2001
, vol. 
40
 
3
(pg. 
233
-
239
)
17
Hernán
MA
Takkouche
B
Caamaño-Isorna
F
, et al.  . 
A meta-analysis of coffee drinking, cigarette smoking, and the risk of Parkinson's disease
Ann Neurol
 , 
2002
, vol. 
52
 
3
(pg. 
276
-
284
)
18
Rugbjerg
K
Harris
MA
Shen
H
, et al.  . 
Pesticide exposure and risk of Parkinson's disease—a population-based case-control study evaluating the potential for recall bias
Scand J Work Environ Health
 , 
2011
, vol. 
37
 
5
(pg. 
427
-
436
)
19
King
LK
Almeida
QJ
Ahonen
H
Short-term effects of vibration therapy on motor impairments in Parkinson's disease
NeuroRehabilitation
 , 
2009
, vol. 
25
 
4
(pg. 
297
-
306
)
20
Haas
CT
Buhlmann
A
Turbanski
S
, et al.  . 
Proprioceptive and sensorimotor performance in Parkinson's disease
Res Sports Med
 , 
2006
, vol. 
14
 
4
(pg. 
273
-
287
)
21
Arias
P
Chouza
M
Vivas
J
, et al.  . 
Effect of whole body vibration in Parkinson's disease: a controlled study
Mov Disord
 , 
2009
, vol. 
24
 
6
(pg. 
891
-
898
)
22
Tillerson
JL
Cohen
AD
Philhower
J
, et al.  . 
Forced limb-use effects on the behavioral and neurochemical effects of 6-hydroxydopamine
J Neurosci
 , 
2001
, vol. 
21
 
12
(pg. 
4427
-
4435
)
23
Chen
H
Zhang
SM
Schwarzschild
MA
, et al.  . 
Physical activity and the risk of Parkinson disease
Neurology
 , 
2005
, vol. 
64
 
4
(pg. 
664
-
669
)
24
Xu
Q
Park
Y
Huang
X
, et al.  . 
Physical activities and future risk of Parkinson disease
Neurology
 , 
2010
, vol. 
75
 
4
(pg. 
341
-
348
)
25
Burton
NW
Turrell
G
Occupation, hours worked, and leisure-time physical activity
Prev Med
 , 
2000
, vol. 
31
 
6
(pg. 
673
-
681
)
26
Le Moual
N
Kauffmann
F
Eisen
EA
, et al.  . 
The healthy worker effect in asthma: work may cause asthma, but asthma may also influence work
Am J Respir Crit Care Med
 , 
2008
, vol. 
177
 
1
(pg. 
4
-
10
)
27
Paddan
GS
Griffin
MJ
The transmission of translational seat vibration to the head—I. Vertical seat vibration
J Biomech
 , 
1988
, vol. 
21
 
3
(pg. 
191
-
197
)
28
Paddan
GS
Griffin
MJ
The transmission of translational seat vibration to the head—II. Horizontal seat vibration
J Biomech
 , 
1988
, vol. 
21
 
3
(pg. 
199
-
206
)
29
Paddan
GS
Griffin
MJ
Transmission of yaw seat vibration to the head
J Sound Vib
 , 
2000
, vol. 
229
 
5
(pg. 
1077
-
1095
)
30
Wang
W
Rakheja
S
Boileau
PE
Effect of back support condition on seat to head transmissibilities of seated occupants under vertical vibration
J Low Freq Noise Vib Active Control
 , 
2006
, vol. 
25
 
4
(pg. 
239
-
259
)
31
Maikala
RV
King
S
Bhambhani
YN
Cerebral oxygenation and blood volume responses to seated whole-body vibration
Eur J Appl Physiol
 , 
2005
, vol. 
95
 
5-6
(pg. 
447
-
453
)
32
Cherniack
M
Clive
J
Seidner
A
Vibration exposure, smoking, and vascular dysfunction
Occup Environ Med
 , 
2000
, vol. 
57
 
5
(pg. 
341
-
347
)
33
Curry
BD
Bain
JL
Yan
JG
, et al.  . 
Vibration injury damages arterial endothelial cells
Muscle Nerve
 , 
2002
, vol. 
25
 
4
(pg. 
527
-
534
)
34
Nakamura
H
Moroji
T
Nagase
H
, et al.  . 
Changes of cerebral vasoactive intestinal polypeptide- and somatostatin-like immunoreactivity induced by noise and whole-body vibration in the rat
Eur J Appl Physiol Occup Physiol
 , 
1994
, vol. 
68
 
1
(pg. 
62
-
67
)
35
Nakamura
H
Moroji
T
Nohara
S
, et al.  . 
Effects of whole-body vibration stress on substance P- and neurotensin-like immunoreactivity in the rat brain
Environ Res
 , 
1990
, vol. 
52
 
2
(pg. 
155
-
163
)
36
Chorny
A
Gonzalez-Rey
E
Varela
N
, et al.  . 
Signaling mechanisms of vasoactive intestinal peptide in inflammatory conditions
Regul Pept
 , 
2006
, vol. 
137
 
1–2
(pg. 
67
-
74
)
37
Hachiya
NS
Kozuka
Y
Kaneko
K
Mechanical stress and formation of protein aggregates in neurodegenerative disorders
Med Hypotheses
 , 
2008
, vol. 
70
 
5
(pg. 
1034
-
1037
)
38
Piskunov
AK
Neuroinflammation biomarkers
Neurochem J
 , 
2010
, vol. 
4
 
1
(pg. 
55
-
63
)
39
Silva
MT
Watts
PM
Jenner
P
Effect of adding selegiline to levodopa in early, mild Parkinson's disease. Parkinson's disease is rarely a primary cause of death [letter]
BMJ
 , 
1996
, vol. 
312
 
7032
pg. 
703
 
40
Palmer
KT
Haward
B
Griffin
MJ
, et al.  . 
Validity of self reported occupational exposures to hand transmitted and whole body vibration
Occup Environ Med
 , 
2000
, vol. 
57
 
4
(pg. 
237
-
241
)
41
Gartner
CE
Battistutta
D
Dunne
MP
, et al.  . 
Test-retest repeatability of self-reported environmental exposures in Parkinson's disease cases and healthy controls
Parkinsonism Relat Disord
 , 
2005
, vol. 
11
 
5
(pg. 
287
-
295
)

Author notes

Abbreviations: PD, Parkinson's disease; WBV, whole-body vibration.