The difference between a doubling of risk for "some" and the less strong 60% increase, for all pesticides and herbicides is due to the fact that the organic chlorine and phosphorus insecticides were the most harmful of the compounds examined.
The newspaper story is based on an US study that looked at people who developed Parkinson's and their family members, and compared their use of pesticides, drinking water from wells, and if they worked as farmers or in other similar occupations. Age is known to be the greatest known risk factor for this degenerative condition, and men and those with affected family members are known to be at slightly increased risk. The use of pesticides and herbicides, and living in a farming area or having a farming job has been proposed as potential risk factors for the disease and this study supports this theory. However, the study does not attempt to put a value on the size of the risk from use of pesticides or determine the level or type of exposure that constitutes a risk. Research is likely to continue in this field and will aim to identify the extent of this risk.
Where did the story come from?
Dana B Hancock and colleagues of Duke University Medical Center and the University of Miami Miller School of Medicine, US. The study was funded by the National Institutes of Health and the National Institute on Neurological Disorders and Stroke. It was published in the peer-reviewed BMC Neurology.
What kind of scientific study was this?
This was a case- control study in which the researchers looked at those with Parkinson's disease and family members without, and compared their use of pesticides, drinking of water from wells, and farming or other occupations.
The researchers recruited 319 people who had Parkinson's disease via university medical centres, doctor's referrals and self-referrals. The patients were then asked to contact family members, both with and without the condition, and this could include partners, siblings, parents or other relatives. The 'controls' were selected from the relatives (296 in total, 237 of whom were siblings) so that they were similar to the cases in terms of genetic and demographic factors such as living environment.
At the start of the study, everyone completed a medical health questionnaire, a family history report and an environmental risk factor questionnaire (including lifestyle factors, occupation, pesticide and other chemical exposure, whether they had lived on a farm or drank water from a well) by telephone. They also had a clinical examination, mental state examination, and had blood samples taken. A neurologist certified the presence of Parkinson's disease by the presence of clinical signs of the disease. The "cases" were also asked to report the age at which they developed the first signs of the condition. Controls were also examined to confirm the absence of Parkinson's disease.
Families with only one case were classified as negative history families and those with more than one affected family member were classified as positive history families. The researchers say they were not able to establish study participation rates among case and controls, and it is unclear how many were family members were considered for each individual case.
Pesticide exposure was determined by asking, "Have you ever applied pesticides to kill weeds, insects or fungus at work, in your home, in your garden or on your lawn?" to which the participants answered yes or no. If they responded yes, they were asked to list the names of chemicals that they had used, to estimate frequency of use, the method of use (e.g. spraying by hand or by tractor), and whether they used any protective clothing, masks, etc. Exposure to pesticide was classified as "those who reported a direct application of any pesticide prior to the [age of onset of Parkinson's]" and a cumulative exposure measure was estimated from their response to the questions. The researchers used statistical models to see how the level of pesticide exposure related to risk of Parkinson's, adjusting for other factors that may confound results such as age, smoking and coffee drinking.
What were the results of the study?
The researchers found that the cases with Parkinson's disease were 60% more likely to report ever having been exposed to pesticides than the controls. There was a trend towards increased dose and exposure to pesticides having greater association with Parkinson's, with only the highest levels of exposure giving significantly increased risk (e.g. over 10 days per year, over 26 years).
However, the link between Parkinson's disease and pesticide use was only significant amongst those without a family history of the disease. The researchers found that the use of organochlorine and organophosphorus insecticide compounds in particular, were significantly associated with Parkinson's disease. They did not find any significant links between working or living on a farm, or drinking water from a well.
What interpretations did the researchers draw from these results?
The authors conclude that their data agrees with positive associations that have been suspected between Parkinson's disease and pesticide exposure, particularly for sporadic cases. They say that further research will be required with a narrower definition of pesticide exposure following the links that they have found with specific classes of the chemical.
What does the NHS Knowledge Service make of this study?
Use of pesticides and herbicides and living in a farming area or having farming occupation have often been indicated as potential risk factors for Parkinson's disease and this study supports that theory. Putting a value on the size of the risk from use of pesticides and determining the level or type of exposure that constitutes a risky level is difficult. People who occasionally use pesticides in their garden should not be overly concerned by these findings.
- Age is the greatest known risk factor for the degenerative condition of Parkinson's, while those with affected family members, and men, are known to be at slightly increased risk. By having the patient's family member as a control, the researchers hoped to balance genetic risk factors. As this study only found significant links between pesticide use and Parkinson's disease in those who did not have a positive family history, it would indicate that family history remains a significant risk factor.
- Determining the quantity of chemical and pesticide used over a lifetime's occupation was a challenge. Pesticide use and other environmental risk factors were given by self-report, which may not be entirely reliable as it relies upon recall of exposure over a long period. In particular, cases with Parkinson's disease who may have been suffering from dementia may not be able to report this with accuracy. Final classification of exposure as "those who reported a direct application of any pesticide" prior to getting Parkinson's disease doesn't give a clear indication of level of exposure, and it cannot be confirmed that all exposures did in fact occur before the disease began as this measure was self-reported. The authors also report that the environmental questionnaire used had not been, "formally evaluated for reliability over time".
- Pesticides cover a large group of chemicals and, as the authors state, misclassification is likely to occur. Therefore, further research that considers a narrower range of chemicals may give valuable results.
- The reporting of risk by the newspaper was slightly misinterpreted: the study found that those with Parkinson's were 60% more likely to report exposure to pesticides, not that using pesticides increased the risk of developing Parkinson's by 60%.
- This study sample of case-controls, several of whom were self-referred, may not be an entirely representative sample of the population, particularly as many were from farming areas of the US. Levels of exposure to environmental chemicals in this group is likely to be much higher than that found in urban areas and other countries, and cannot be assumed to be similar to domestic pesticide use in the garden in the UK. The family controls for the cases were also self-selected, which may introduce some bias. Those included may be closer to cases and have slightly different exposure characteristics to other family members, who did not share a close home environment to the patient, or if other unrelated controls had been selected who were matched to the patients on other factors such as age or sex.
Research is likely to continue in this field. It seems sensible for those people working in occupations where organochlorine and organophosphorus compounds are routinely used as insecticides to be particularly aware of the safety precautions necessary.
Links to the headlines
Pesticides increase risk of Parkinson's disease Daily Telegraph, March 28 2008
Study links Parkinson's disease to long-term pesticide exposure. The Guardian, March 28 2008
Pesticide Parkinson's link strong BBC News, March 28 2008
Links to the science
Hancock DB, Martin ER, Mayhew GM, et al.
BMC Neurology 2008, 8:6
This news comes from NHS Choices