Risk Factors for Parkinson Disease

February 22, 2016

Researchers gather environmental data to better understand possible mechanisms of Parkinson disease development.

A study by Robert Abbott and colleagues, recently published in Neurology, examines the relationship between midlife milk consumption and the risk of Parkinson disease (PD).1 The study also discusses the role of putative risk factors for PD – pesticide exposure, serum uric acid level, and smoking – in the degeneration of neurons in the substantia nigra. A recent umbrella meta-analysis by Bellou and colleagues lends additional support to these findings.

Abbott and colleagues analyzed histories and decedent brains of participants of the Honolulu-Asia Aging Study. (The study was initiated in 1991 to examine the causes of dementia in aged Japanese-American men.) The authors found that neuronal density was the lowest in men who consumed over 16 ounces of milk a day and never smoked. These men had 41.5% (95% confidence interval, 22.7%-55.7%; P < 0.001) lower neuronal density than men who drank less that 16 ounces of milk a day; men with clinical PD or dementia with Lewy bodies were excluded from this analysis. In current or past smokers, neuronal density and milk consumption were not associated.

The association of milk and nigral degeneration may be mediated by the presence of heptachlor epoxide in the brains of these men; it was detected in the brains of 9/10 men who drank over 16 ounces of milk a day vs 26/41 of men who did not drink milk (P = 0.017). Heptachlor epoxide is an organochlorine pesticide used widely until the 1980s. It was used in pineapple fields in Hawaii and accumulated in milk (possibly because contaminated pineapple plants were used as cattle feed). The authors stated that further studies were needed to confirm the link between pesticide contamination of milk and nigral degeneration. If pesticides are not the cause, can milk alone cause nigral degeneration? Indeed, consumption of pesticide-free dairy has been linked to PD via the mechanism discussed below. 

In the editorial that accompanies the study, Honglei Chen and Karen Marder summarize its key findings and limitations. They praise the authors for using epidemiologic data to understand the mechanism of PD and mention the importance of collecting environmental data during the ongoing large clinical studies.2 While additional data will indeed help better understand causes and risk factors of PD, existing data also have value. Vanesa Bellou and colleagues recently conducted an umbrella meta-analysis of environmental risk factors for PD by synthesizing data from 75 unique, already published meta-analyses.3 Their findings were published in Parkinsonism and Related Disorders in December 2015.

Bellou and colleagues evaluated smoking, pesticide exposure, farming, physical activity, constipation, etc. as risk factors for PD. They found class I evidence supporting the association of constipation and physical activity with PD. They also identified other putative risk factors. Head injury, anxiety, depression, and the use of beta-blockers increased the risk of PD, while serum uric acid and smoking decreased it.

Curiously, both smoking and serum uric acid likely play a role in the study by Abbott and colleagues. First, the association of milk consumption with PD disappeared in smokers. Second, dairy consumption may decrease the serum level of uric acid and hence increase the risk of PD regardless of the presence of pesticides.

The authors and editor of the meta-analysis caution readers about the many limitations of this approach, which “render these associations essentially unproven,” according to Zbigniew Wszolek, editor in chief of Parkinsonism and Related Disorders.




1. Abbott RD, et al. Midlife milk consumption and substantia nigra neuron density at death. Neurology. 2016 Feb 9;86(6):512-519.

2. Chen H, Marder K. Milk consumption and the risk of nigral degeneration. Neurology. 2016;86:496-497.

3. Bellou V, et al. Environmental risk factors and Parkinson's disease: an umbrella review of meta-analyses. Parkinsonism Relat Disord. 2016;23:1-9.