Golf Courses and Parkinson's Disease: What Researchers Have Found and What Drinking Water Has to Do With It

A study published in JAMA Network Open linked living near golf courses to a higher risk of Parkinson’s disease (PD). The odds of developing the disease were highest for people living within 1–3 miles (≈1.6–4.8 km) of a course; the risk decreased with distance. The association was especially strong where the water supply served the area containing the course and where groundwater was vulnerable. The authors speculate that pesticides used to care for lawns may be responsible, as they can leach into the air and drinking water.

Background

• Why look at golf courses at all. To maintain perfect turf, courses use a variety of herbicides, fungicides, and insecticides; some of these substances may leach into groundwater or be dispersed by aerosols. Field and survey work has shown the presence of pesticides in aquifers and water bodies on/near golf courses (e.g., USGS and current regional risk assessments). This provides a plausible exposure pathway for the surrounding population.
• What is already known about pesticides and PD. Over the decades, many studies have accumulated in which pesticide exposures (occupational, household, agricultural) have been associated with an increased risk of PD, although the strength and consistency of the effects vary. Current reviews confirm the general signal of association of some classes of pesticides with PD, but highlight the limitations of observational data.
• The role of drinking water is a controversial but important hypothesis. Several studies have found associations between well water use and BP (as a proxy for exposure to soluble pesticides), but a more recent meta-analysis showed no consistent association, likely due to coarse exposure proxies and regional heterogeneity. This calls for more accurate geohydrological models and consideration of aquifer vulnerability.
• Regulatory differences across countries. For example, in the EU, substances such as paraquat are banned; this means lower exposure to certain high-risk pesticides and limits the direct transferability of risk assessments between the US and Europe.
• What the current work adds. The study used data from the Rochester Epidemiology Project (USA, 1991–2015): 419 PD cases and 5,113 controls. Exposure was assessed by distance to a golf course, household affiliation with a municipal water supply that included a golf course, and groundwater vulnerability. A gradient dependence of risk was found, with a decrease with distance and the strongest effects in water supply zones with golf courses and high groundwater vulnerability.
• Key limitations to remember: The design remains observational; individual pesticide doses and specific substances were not measured, and residual confounding is possible. Therefore, the results show association, not causation, and require confirmation in studies with direct biomarkers of exposure and detailed water/air monitoring.
• Summary context: Against the background of an accumulated but heterogeneous body of data on pesticides and BP, this work is valuable for its more nuanced geospatial and hydrogeological consideration of exposure pathways (distance + water supply + aquifer vulnerability). It does not close the question of causality, but it does provide clear directions for further research and for sanitary planning in vulnerable water systems.

How the study is structured

• Type: Population-based case-control study within the Rochester Epidemiology Project.
• Period: 1991–2015.
• Participants: 419 newly diagnosed PD cases (median age 73 years) and 5113 age- and sex-matched controls.
• Exposure: distance from home address to the nearest golf course (according to exact course boundaries, 139 objects). The address was taken 2–3 years before the onset of symptoms.
• Additionally: belonging to a water supply zone (from groundwater, surface water or private wells), vulnerability of groundwater (sandy soils, karst, shallow rock occurrence), presence of small municipal wells.
• Model: logistic regression adjusting for age, sex, race/ethnicity, year, area median income, "urbanization," and intensity of contact with the health care system; associations with distance were assessed both categorically and as cubic splines.

What happened?

Connection with distance to golf course

Compared to living further than 6 miles:

• <1 mile: adjusted odds of PD ↑ 2.26 times (95% CI 1.09–4.70).
• 1–2 miles: ↑ 2.98 times (1.46–6.06).
• 2-3 miles: ↑ 2.21 times (1.06-4.59).
• 3–6 miles: trend towards ↑ (1.92; 0.91–4.04).

The spline model showed that up to ~3 miles the association was “flat,” and beyond 3 miles the risk decreased linearly by 13% for each additional mile (aOR 0.87 per mile; 0.77–0.98).

Water supply and groundwater vulnerability

• Residents of groundwater supply areas with a golf course had nearly double the risk compared with areas without golf courses (aOR 1.96; 1.20–3.23) and 49% higher than private well users (aOR 1.49; 1.05–2.13).
• If such a zone was also on vulnerable groundwater, the risk was 82% higher compared to non-vulnerable ones (aOR 1.82; 1.09–3.03).

Important: No association was found for the presence of shallow municipal wells (<100 feet) or wells directly on the golf course.

Why is this plausible?

Golf courses are regularly treated with herbicides, fungicides, and insecticides. Some known substances (e.g., paraquat, rotenone, some organophosphates, and organochlorines) have been linked in experimental and epidemiological studies to mechanisms similar to the pathogenesis of PD: oxidative stress, mitochondrial dysfunction, and death of dopaminergic neurons. Possible routes of action:

1. Pollution of groundwater with subsequent entry into the drinking water of entire areas (water supply zone = total water resource).
2. Airborne drift - aerosols and dust; the connection was more noticeable in urban locations.

What this does not prove (limitations)

• This is observational work: one can talk about connection, not causality.
• The address was taken 2–3 years before the symptoms, whereas PD has a long prodrome (exposures could have been decades earlier).
• There is no data on professions, head injuries, genetics - their influence cannot be ruled out.
• Region is predominantly white; portability to other states/countries requires verification.

Practical conclusions (reasonable, no panic)

For residents near golf courses and municipalities:

• Transparency: Publish treatment schedules, active ingredients, and measures to protect aquifers.
• Water monitoring: regular analyses in vulnerable groundwater areas; reports to the public.
• Integrated lawn protection: minimizing the volume and frequency of treatments, choosing less persistent products, buffer zones and treatment times taking into account the wind.
• Household level: monitor water utility reports; if desired, carbon filters/reverse osmosis as barriers for some pesticides (this is a general safety precaution, not a specific recommendation for the article).

Remember that the absolute risk of BP remains low; the goal is to reduce potential impacts where it is cheap and feasible to do so.

What to explore next

• Measure actual pesticide levels in water/air and biomarkers in residents, not just distance.
• Take into account the length of residence and history of migration (total dose over decades).
• To analyze specific active ingredients and modes of application, seasonality, weather conditions.
• Assess genetic susceptibility and gene-environment interactions.

Conclusion

In this study, living closer to a golf course means having a higher chance of being diagnosed with Parkinson's disease, especially if the area sits on vulnerable groundwater and gets its water from a common groundwater resource. The data are consistent with the pesticide hypothesis and suggest simple directions for public health: monitoring water and air, reducing pesticide loads, and improving public awareness.

Source: Krzyzanowski B. et al. Proximity to Golf Courses and Risk of Parkinson Disease. JAMA Network Open. 2025;8(5):e259198. Open access (PMC).