Parkinson’s disease is a puzzle. I know because I have had it for more than a decade. Some of its symptoms, such as tremors, are easy to understand, but others are weird. For example, turning my body is difficult, and it’s even more difficult if I try to turn it clockwise rather than counterclockwise. I have lost my ability to swim. And what happens when I suddenly freeze, as if my feet were glued to the floor? My brain has sent a message to my feet to step forward. Did the message not arrive, or did my feet simply ignore it? It’s impossible to know. (This article has a shared byline, but the “I” refers to Torrey.)
Parkinson’s disease is not just a puzzle; it’s an expensive one. A recent detailed study, based on 2017 data, reported that just over 1 million individuals in the United States were living with Parkinson’s. The disease costs our health care system $51.9 billion annually—and that price is expected to balloon to $79.1 billion by 2037, or roughly $1.36 billion a year. Since 90 percent are 65 or older, these patients place a particularly heavy burden on Medicare.
Even more alarming, the researchers estimated that by 2037 an additional 600,000 people will be diagnosed with Parkinson’s. Such projections are consistent with other studies showing that Parkinson’s is the fastest-growing neurological disease globally, increasing even faster than Alzheimer’s disease. Indeed, an editorial in The Lancet Neurology reported that “the prevalence, burden of disability, and number of deaths associated with Parkinson’s disease all more than doubled between 1990 and 2016.” Some observers call this a “Parkinson’s pandemic.”
Part of this increase is attributable to people living longer and to the large, aging Baby Boomer population. But that is only a partial explanation. Another part might be due to factors related to the causes of this disease. Studies have shown that you are more likely to get Parkinson’s if you have red hair or melanoma, or if you still have your appendix. Other studies have reported that having numerous dental amalgam fillings or living downwind from a golf course are risk factors. Perhaps strangest is data showing that drinking large amounts of milk or never having smoked tobacco increases your chances of getting Parkinson’s disease. Examining some of the leading theories of causation—genetics, infection and inflammation, toxic metals, and pesticides—can illuminate this puzzling potpourri of claims and determine whether research dollars are being effectively deployed in halting the rise of the disease.
Genetic research on Parkinson’s has identified more than 20 genes that can potentially cause the disease. However, most of them rarely do. Some genes were only identified in single families, and several only in individuals with early disease onset. The relative lack of importance of genes as a primary cause of Parkinson’s disease has been confirmed by studies in which identical twins, who share 100 percent of their genes, are compared to fraternal twins, who share only 50 percent. If a disease has a genetic cause, one expects both identical twins to be affected—this is called the “concordance rate”—significantly more often than both fraternal twins. Large studies of twins and Parkinson’s disease, however, have reported that the concordance rates for identical and fraternal twins are not significantly different. Twin studies thus confirm the relative lack of importance of genes as a primary cause of Parkinson’s disease and support researchers’ estimates that genes cause only 5 to 15 percent of cases.
In addition to genes that directly cause diseases, there are dozens, and perhaps hundreds, of so-called risk genes for each disease. Risk genes do not cause the illness but, rather, predispose individuals to it or protect them from it. For example, scientists know that a particular kind of Mycobacterium can cause tuberculosis. Once M. tuberculis enters a body, risk genes determine whether it will cause clinical symptoms and, if so, how severe they will be.
Scientists have identified more than 90 risk genes for Parkinson’s disease. One of these also controls the quantity and distribution of melanin, which determines hair color. Redheads have almost double the chance of developing Parkinson’s compared to people with black hair; blondes and brunettes have intermediate risks. This genetic association also explains why someone with Parkinson’s has an almost fourfold increased probability of developing melanoma and why someone with melanoma has a fourfold increased likelihood of developing Parkinson’s.
Infection and Inflammation
Infectious agents are also potential causes of Parkinson’s disease. Many viruses that affect the brain can cause symptoms of Parkinson’s disease, such as tremor and stiffness. Research has shown that the influenza virus, for example, caused the epidemic of encephalitis lethargica—a neurological syndrome—in the 1920s, which followed the 1918 influenza pandemic and resulted in thousands of cases of parkinsonism. (When individuals have symptoms of Parkinson’s disease, but it is unclear whether they have the full disease, it is referred to as “parkinsonism.”) The residual cases from that epidemic became the subject of Oliver Sacks’s Awakenings, subsequently made into a movie starring Robin Williams as the doctor giving his patients a new medication that produces a dramatic but temporary improvement in their symptoms. The possible role of the influenza virus in causing Parkinson’s disease is a subject of ongoing debate among researchers. Other viruses that have been shown to cause parkinsonism include Coxsackie, Japanese encephalitis, western equine encephalitis, herpes simplex, hepatitis C, and the virus that causes acquired immune deficiency disorder. How often such viruses actually do so is not yet known.
Other researchers take a broader approach to the relationship between infections and Parkinson’s disease. A large Swedish study identified individuals who had been hospitalized for any infection of the central nervous system. They found that individuals with Parkinson’s were 50 percent more likely than controls to have previously been hospitalized for a CNS infection. The researchers reasoned that it was not the specific infectious agent that was the problem but that all the infections produced inflammation in the brain.
In support of this theory, many studies have reported increased levels of inflammatory markers in the blood of individuals with Parkinson’s disease. This has led to speculation that the disease is not a brain disease but a disease of the immune system.
Some studies show a relationship between inflammatory bowel disease and Parkinson’s disease. Additional evidence linking the gastrointestinal tract to Parkinson’s is that the wall of the intestine contains alpha-synuclein, a protein also found in the brain of individuals with Parkinson’s. Alpha-synuclein is especially prominent in the wall of the appendix. An international consortium of researchers extensively studied the appendix in individuals with Parkinson’s disease. Looking at the records of more than 1.6 million individuals over 52 years, they found that patients who had had their appendix removed had a modest but statistically significant 19 percent reduction in their chances of developing Parkinson’s. Furthermore, those who had had their appendix removed at least 30 years previously but still developed the disease did so almost four years later than those who still had their appendix.
Researchers have also noted that exposure to high levels of certain metals—among them aluminum, bismuth, copper, iron, lead, manganese, mercury, and zinc—causes parkinsonism. (Workers at a manganese ore–crushing facility, for example, reported parkinsonian symptoms.) Other studies show geographic associations, such as a higher prevalence of Parkinson’s disease in urban counties that also report a higher industrial release of copper or manganese.
Among the most promising research involving metals and Parkinson’s is in studies about mercury. Most human exposure to mercury comes from amalgam dental fillings and eating fish. Amalgam fillings, which consist of 50 percent mercury and a 50 percent mix of silver, copper, zinc, and other metals, were introduced almost 200 years ago. Individuals with amalgam fillings have between two and 12 times more mercury in their bodies than those without amalgam fillings. An autopsy study of 34 individuals reported a statistically significant correlation between the number of amalgam fillings and the mercury level in their brain’s occipital lobe. Studies have also shown that amalgam fillings slowly leak mercury vapor; when inhaled, it can easily pass through the blood-brain barrier. (Based on toxicity studies, amalgam fillings have been banned in Denmark, Norway, and Sweden. They are still used in the United States, where 58 percent of adults have them.)
Studies of mercury that were specifically focused on Parkinson’s patients have confirmed this link. A study of 54 Parkinson’s patients and 95 controls reported a significant association between blood mercury levels and the diagnosis. A large study from Taiwan found that amalgam fillings significantly increased the risk of a subsequent diagnosis of Parkinson’s disease. One study used data from Denmark’s Faroe Island, where there is a high prevalence of Parkinson’s, to examine the dietary history of 79 individuals with the disease and 154 matched controls. A statistically significant association was found between individuals who ate more whale, which is high in mercury, and those with Parkinson’s. Most recently, an Australian study compared the distribution of mercury in the autopsied brains of two individuals who had died with Parkinson’s disease and 12 who had not. Some mercury was found in all the brains, but only in the Parkinson’s-infected brains was it found in neurons in the substantia nigra, striatum, and thalamus, areas associated with this disease. The researchers often found Lewy bodies—abnormal deposits of alpha-synuclein and one of the hallmarks of Parkinson’s disease—along with the mercury.
In the 1980s, an unusual outbreak of Parkinson’s disease occurred among young adults in California. It turned out that all had used a designer street drug, MPTP, that was chemically similar to a widely used pesticide called paraquat. This caused researchers to wonder whether pesticides or other chemicals might be causes of Parkinson’s disease.
Four decades later, research suggests that the answer is yes. A 2017 analysis of 23 such studies concluded that pesticide exposure increases the risk of an individual developing Parkinson’s by 50 percent or greater. This was true for pesticides in general and also for each class of pesticides—insecticides, herbicides, and fungicides—examined individually. Some pesticides seem worse than others; for example, a meta-analysis of 13 case-control studies of paraquat alone demonstrates its association with Parkinson’s disease. Paraquat has already been banned in more than 30 countries, but it is still widely available in the United States, and according to data from the U.S. Geological Survey, its use more than doubled between 2008 and 2018.
The assessment of pesticide exposure differs widely in these studies. For example, a study in Iowa and North Carolina determined the incidence of Parkinson’s disease in individuals who worked in agriculture as professional pesticide applicators. By contrast, a study in Nebraska reported a geographic association between the incidence of Parkinson’s disease and the use of pesticides by counties. Several studies have examined whether rural residents who live next to fields on which pesticides have been used have a higher incidence of Parkinson’s. The results have been mixed. In 2012, researchers in Raleigh, North Carolina, published a letter in a neurology journal asking, “Is Living Downwind of a Golf Course a Risk Factor for Parkinsonism?” They observed that among 26 cases of parkinsonism, 19 individuals lived on or within two miles of a golf course. Furthermore, 16 of the 19 lived downwind from the course, and two others were said to have had additional golf course exposure. The researchers invited readers to confirm their findings with a larger sample size, but we could not ascertain whether anyone had done so.
Many Americans associate Parkinson’s disease with head trauma. The boxer Muhammad Ali was diagnosed with Parkinson’s in 1984, at the age of 42. Studies were subsequently done asking people with Parkinson’s about their history of head trauma. Such studies were subject to recall bias because subjects might have been more likely than controls to remember such incidents. A meta-analysis of 22 such studies, all done since 1984, reported a significant association with Parkinson’s disease, but only for head trauma that resulted in a loss of consciousness or a concussion. Recent studies have also emphasized that sports-related head injuries are more likely to result in chronic traumatic encephalopathy, with symptoms such as depression and cognitive deficits; some subjects will also have tremors, but other symptoms of parkinsonism are not prominent in such cases.
Clues to the causes of Parkinson’s disease have also been sought in prospective, longitudinal health studies in which data on dietary habits and smoking is collected on large groups who are followed for years. A meta-analysis of five longitudinal studies unexpectedly identified milk—but not cheese, yogurt, or butter—as a risk factor for Parkinson’s disease. The more milk people drank, the greater the risk. In a study of men in Hawaii, those who consumed the most milk doubled their chances of developing Parkinson’s disease later in life. The risk does not appear to be related to milk’s calcium; milk causes a decrease in blood chemicals, which are thought to be a protective factor for Parkinson’s.
Even stranger than the milk story is the nicotine story. Everyone knows smoking is bad for your health and is associated with several cancers, chronic obstructive pulmonary disease (COPD), heart disease, and stroke. However, several longitudinal studies have identified nicotine, including in cigarettes, cigars, pipes, and chewing tobacco, as one of the strongest protective factors for Parkinson’s disease, reducing the risk of developing it by more than half. Current smokers have the lowest risk, followed by past smokers and then by people who have never smoked. The danger is inversely related to how long and heavily the person smoked. The decrease in smoking among men in recent decades has been cited as one possible reason Parkinson’s disease is increasing in prevalence. It is possible that nicotine has neuroprotective effects on the brain.
These are all important clues to ultimately discovering the causes and better treatments for Parkinson’s. Given the disease’s increasing prevalence, and the rising cost of caring for Parkinson’s patients, more research needs to be done.
How much federally funded research on Parkinson’s disease is currently being carried out? In 2021, the National Institutes of Health supported 526 research projects on Parkinson’s, totaling $254 million. Based on the titles of the 526 projects, it appears that 58 focus on genetic causes; 17 on infectious and inflammatory causes; 14 on pesticides (including paraquat); and only seven on toxic metals, none of which include mercury. Given the limited number of cases of Parkinson’s disease known to be caused by genes, genetic research is probably being adequately covered. However, the other possible causes all appear to be disappointingly under-researched.
How much should the NIH be spending? One way to assess this is to compare it to research expenditures for Alzheimer’s disease. In 2021, the NIH spent $254 million on Parkinson’s research and roughly $3.1 billion on Alzheimer’s research. There are 6.5 million people with Alzheimer’s, compared to the approximately 1 million with Parkinson’s, meaning that the NIH spent roughly $254 per Parkinson’s patient and more than $470 per Alzheimer’s patient. If we use this comparison, Parkinson’s is being underfunded by the NIH by about $216 million a year. Perhaps it is time to review the NIH research portfolio. This may lead to a better understanding of the causes of Parkinson’s disease, leading to better treatments and control of Medicare costs.