#824

I wonder if golf course tenders keep records, and if they’re public. I’d like to see this. What are they spraying that is so toxic? Herbicide doesn’t do this.

We have to keep records. They can come and look at them whenever they want.

3 Likes
#825

Although I believe pesticides play a role in PD:

  • Golf courses are often in wealthy areas where people move to after a certain age: if you retire at age 65 and move to a gated community next to a golf course and are diagnosed with PD 5y later, is the biggest impact the 5y next to the golf course of the 65y before?
  • The thesis is that water is polluted I guess? Did they find pesticides in water next to golf courses?
1 Like
#826

How do you interpret these papers?

So mitochondrial dysfunction might be secondary (vs lysosomal, synaptic, and immune).

1 Like
#827

As I sit here overlooking a golf course, I have to point out what an insightful comment you made!

Eureka, yes, pretty much no one grows up in a golf course community… most of us move here when we are already old!!!

Our water is from our community wells here and they do analyzing it so it’s ’apparently’ safe water, but yeah, there is probably Round Up in it!

#828

The mitochondria are initially ok, but those in the SNc get damaged mainly because they have high OxPhos usage.

#829

Given that the study included age and gender matched controls, I think the first point you’ve made could be partially challenged on those grounds. Of course, there is considerable confounding variables here and the first decades of life undoubtably have an impact.

To an extent the thesis is that the water (and air to some extent) is polluted and the data gathered does suggest this to be the case.

#830

Roundup kills all plants that it touches, so not used on golf courses much if at all. Most likely 2,4-D, Banvel, Triclopyr, MCPP-P, possibly a couple more but that would be it and will kill any bad plants.

Fungicide is possibly harmful and used frequently on golf courses according to Grok. They use the same ones we do, only throughout the season. We just do it once. There are 4 or 5 of these that are common.

Insecticide, the bad ones have been banned for golf courses. They’re stuck with a couple that aren’t great including pyrethroids. I think they were blamed for an airpland of soldiers sprayed and they got ALS (at higher rates). I read about this back when Dad had ALS. So since the others were banned, they may use this more. Still I doubt it.

1 Like
#831

FWIW:

Geographic and Ethnic Variation in Parkinson Disease: A Population-Based Study of US Medicare Beneficiaries

“Prevalence and incidence in urban counties were greater than in rural ones (p < 0.01). Cluster analysis supported a nonrandom distribution of both incident and prevalent Parkinson disease cases (p < 0.001).”

We always heard that there’s a high prevalence of PD in farm workers, especially those handling pesticides and herbicides. But if just living near farmland was a factor, wouldn’t you expect people in rural areas to have higher rates of PD than urban folk? Instead, it’s the opposite. Also, we read about more minority people living near polluted areas, waste dumps and the like. Wouldn’t you expect them to have higher PD rates? Instead, it’s whites. Also, geographic distribution - retired people move to warm climates like Florida and Arizona. So, like with your golf course retirement people moving, you’d expect more PD in those areas - instead, it’s Midwest and Northeast - and retired people famously move away from the urban Northeast (especially to Florida), and especially whites. So you’d expect huge PD numbers to where white urban retirees move to - but it’s not true. There is a lot of farming in the Midwest, but there is more farming in the South than Northeast, besides they already established that there’s more PD in urban than rural. In the end I don’t know how informative is this kind of data - it seems pretty prone to confounding.

1 Like
#832

I’ve heard that California’s ‘salad bowl’ is a hotbed for Parkinson’s.

It’s Salinas, where a lot of our crops are grown

From AI

AI Overview

Research suggests a link between Parkinson’s disease and

pesticide exposure, particularly in agricultural areas like the Salinas Valley (nicknamed the “Salad Bowl of the World”).

Key Findings:

  • Paraquat is a pesticide linked to an increased risk of Parkinson’s disease, and its use is prevalent in agricultural regions of California, including Monterey County.
  • Studies have shown that farmworkers and individuals living or working near fields where paraquat is sprayed are at higher risk of developing Parkinson’s.
  • A potential link exists between the pesticide benomyl and Parkinson’s disease, according to UCLA researchers.
  • Some studies indicate that living in rural or agricultural areas may be associated with increased Parkinson’s risk, possibly due to higher exposure to pesticides and other environmental factors.

And, @Bicep my grandmother had ALS, so I know how awful it is… sorry about your dad. And good point about round up. I do know they use it, but to your point, it’s probably limited to certain weeds. On that note, I nag my husband and brother about their love for that stuff, and just this week my brother had an ‘incident’ where the sprayer broke and he was drenched… he said all I could think about was you on my shoulder :slight_smile:

1 Like
#833

They note in limitations:

Nevertheless, even if golf courses were present decades earlier, our results were based on home address information 2 to 3 years prior to symptom onset, which does not capture the complete prodromal period. Our population was relatively stable, living at their address for approximately 6 to over 43 years. We were unable to extend address history back further than 3 years because residency data becomes more incomplete the further we move back in time due to lack of interaction with the health care system.

They did not look at all at the water content. That’s their central hypothesis but they did not even bother to just analyze the water. That’s a massive weakness in the paper.

A better study would be to look at a random group of people and ask them: “Have you ever lived next to a golf course?” and see the PD prevalence.

#834

This explains why SNc mitochondria are among the first casualties, but that does not make idiopathic PD a purely mitochondrial disease?

#835

I suppose it depends upon what the meaning of “mitochondrial disease” is. As you know I think one of the reasons in both PD and ALS/MND for mitochondrial decay is a shortage of CSF supplied melatonin which otherwise is used to maintain the dynamic equilibrium.

However, if you have other problems with the dynamic equilibrium such as flaws in the PINK1/Parkin system of selective mitophagy then the same vicious circle can happen.

I have been looking at the genes that make ALS more likely recently and lots link to this.

I am using the term “mitochondrial disease” to mean one where the phenotype is caused primarily by mitochondrial decay.

#836

https://www.nature.com/articles/s41531-025-00967-4

Insights into ancestral diversity in Parkinson’s disease risk: a comparative assessment of polygenic risk scores

Abstract
Risk prediction models play a crucial role in advancing healthcare by enabling early detection and supporting personalized medicine. Nonetheless, polygenic risk scores (PRS) for Parkinson’s disease (PD) have not been extensively studied across diverse populations, contributing to health disparities. In this study, we constructed 105 PRS using individual-level data from seven ancestries and compared two different models. Model 1 was based on the cumulative effect of 90 known European PD risk variants, weighted by summary statistics from four independent ancestries (European, East Asian, Latino/Admixed American, and African/Admixed). Model 2 leveraged multi-ancestry summary statistics using a p-value thresholding approach to improve prediction across diverse populations. Our findings provide a comprehensive assessment of PRS performance across ancestries and highlight the limitations of a “one-size-fits-all” approach to genetic risk prediction. We observed variability in predictive performance between models, underscoring the need for larger sample sizes and ancestry-specific approaches to enhance accuracy. These results establish a foundation for future research aimed at improving generalizability in genetic risk prediction for PD.

O3 summary

Paper at a glance
Insights into ancestral diversity in Parkinson’s disease risk: a comparative assessment of polygenic risk scores was published on 3 July 2025 in npj Parkinson’s Disease by Saffie-Awad et al. (nature.com)

1 • Research question & rationale

Polygenic risk scores (PRS) for Parkinson’s disease (PD) have been built almost exclusively from Europeans, raising the danger that “precision” tools will perform poorly—and unfairly—in other populations. The authors ask: How well do different PRS construction strategies work across seven global ancestries, and what drives any performance gaps? (nature.com)

2 • Data & methods

  • Cohort: 45 799 individuals (29 097 PD cases, 16 702 controls) from the Global Parkinson’s Genetics Program (GP2), spanning African, African-admixed, Ashkenazi Jewish, Central Asian, East Asian, European and Latino/admixed American ancestries.
  • Model 1 (“90-SNP”): the 90 established European PD loci weighted by four ancestry-specific GWAS summary statistics, with 56 distinct PRS after covariate choices.
  • Model 2 (“thresholding”): best-fit p-value thresholding on multi-ancestry GWAS meta-analysis results (Kim et al. 2023), yielding 49 PRS.
  • Evaluation: area under the ROC curve (AUC), balanced accuracy, odds ratios, and DeLong tests for pairwise model comparisons. (nature.com)

3 • Key findings

Finding Evidence Take-home
Pronounced heterogeneity in risk‐allele frequencies and effect sizes across ancestries Upset/forest plots show many of the “European 90” are rare or directionally opposite elsewhere A single European-derived panel cannot capture global PD genetics (nature.com)
Performance ceiling remains modest Best AUC ≈ 0.67 (Ashkenazi Jewish, Model 2); worst ≈ 0.54 (African, Model 1) Today’s PRS are far from clinical-grade across the board (nature.com)
Model choice matters and interacts with ancestry Model 1 with European weights beats Model 2 in Europeans and East Asians; Model 2 excels in Africans Statistical power of the base GWAS often overrides ancestry matching (nature.com)
Certain loci dominate prediction SNCA (rs356182) is top in many groups; LRRK2 G2019S and GBA1 variants inflate scores in Europeans/Ashkenazi Jews Highlights need for ancestry-specific fine-mapping and LD correction (nature.com)

4 • Authors’ conclusions

A “one-size-fits-all” PRS strategy entrenches disparities. Larger, well-powered ancestry-specific or genuinely multi-ancestry GWAS—including replication datasets—are required before PRS can support equitable PD risk prediction. (nature.com)

Critical appraisal

Aspect Strengths Limitations / Points to challenge
Scope & novelty First head-to-head comparison of >100 PD-PRS across seven ancestries; highlights equity gap that policymakers and funders increasingly prioritise. Still heavily European (83 % of cases in the meta-analysis feeding Model 2); African, Central-Asian and Latino sub-samples remain under-powered.
Data & design Uses individual-level GP2 data—currently the largest multi-ancestry PD resource; transparent workflow, figures, and supplementary tables. No external replication cohort per ancestry → risk of overfitting to GP2 quirks; self-reported 23andMe and proxy UK Biobank cases may dilute phenotype accuracy.
Analysis choices Sensitivity analyses (PC vs. admixture adjustment, DeLong tests) are thoughtful; comparison of weighting schemes clarifies why “power beats proximity.” Predictive metrics stop at ROC; no calibration curves, decision-curve analysis, or net reclassification improvement—key for clinical usefulness.
Interpretation & fairness Authors candidly discuss limitations and ethical stakes; suggest local-ancestry PRS‐CSx and multi-modal ML as next steps. Discussion could better quantify how ancestry mis-classification (23andMe vs. GP2 pipelines) affects score transferability; environmental and clinical covariates were omitted though easily available for many GP2 cohorts.
Clinical relevance Paper rightly tempers hype: AUC ≤ 0.67 is insufficient for standalone screening, but informs research stratification. The manuscript does not simulate real-world deployment—e.g., whether combining PRS with smell testing or REM-sleep-behaviour-disorder data lifts performance into actionable territory.

Additional considerations

  • Ethics: Even moderate mis-calibration can exacerbate health disparities if PRS inform trial inclusion or insurance decisions. External validation in truly community-based cohorts is critical.
  • Statistical nuance: The 90-SNP panel may underperform in part because effect-size shrinkage differs by ancestry; Bayesian or LD-pred–style methods that model ancestry-specific LD could help.
  • Biological insight vs. prediction: While prediction is the focus, variant-level heterogeneity (e.g., SNCA vs. LRRK2 dominance) also offers biological leads; integrating eQTL or epigenomic data could enhance both interpretability and accuracy.

Bottom line

This study is an important wake-up call: current PD polygenic scores, though scientifically intriguing, remain under-powered, Euro-centric, and not yet clinic-ready—but the paper provides a roadmap for how larger, more diverse GWAS and smarter modelling can get us there. (nature.com)