Study suggests eating ultra-processed foods may accelerate early Parkinson's symptoms

This new study, following a very sound scientific methodology, provides new evidence to suggest that what we eat today will have consequences for our mental health tomorrow. 

This is a follow-up study based on two large cohort studies at Harvard University, the cohort of American nurses and the cohort of health professionals with more than 42,000 participants [in total]. The researchers observed that those who consumed the most ultra-processed foods (on average 11 servings per day), foods characterised by being industrially processed and often containing a long list of ingredients with additives such as sweeteners, colourings, emulsifiers or flavour enhancers, were 2.5 times more likely to develop three or more non-motor prodromal signs of Parkinson's disease than those who on average consumed three ultra-processed foods per day. To a lesser extent, participants who consumed on average 6 ultra-processed foods were 1.5 times more likely to develop 3 or more prodromal signs including smell problems (hyposmia), constipation, probable REM sleep behaviour disorder, colour vision problems, daytime sleepiness, body pain and depressive symptoms. When each non-motor prodromal sign was analysed separately, a positive association was observed with REM sleep behaviour disorder, constipation, body pain and depressive symptoms. 

When looking at the different ultra-processed food groups, sweet snacks and sugary or sweetened beverages increased the risk of developing these prodromal signs the most. 

Looking at non-motor prodromal signs of Parkinson's is an advantage, as these appear ten to twenty years before clinical diagnosis, which provides an opportunity for early interventions to prevent the development of Parkinson's disease. However, it would be desirable to have further studies that also look at a relationship between ultra-processed food consumption and the incidence of Parkinson's disease diagnosis, not only with early non-motor signs of the disease. 

In any case, these results continue to support the need to discourage the consumption of ultra-processed foods, not only because of their increased risk with metabolic diseases, but also with mental illnesses, and the need to continue promoting the Mediterranean diet in the population, based on minimally processed foods with a high consumption of fruit, vegetables, legumes, whole grains, fish and olive oil.

Firstly, the outcome term 'early symptoms of Parkinson’s disease' is a bit misleading as symptoms such as constipation, and body pain here found to be associated with consumption of Ultra Processed Foods (UPF) are quite common in ageing and are not necessarily indicative of Parkinson’s disease.
 
Even the most likely predictor of Parkinson’s disease – probably REM sleep disorder – is seen in 65% of Parkinson patients but also in 10% of controls, with low (65%) sensitivity for Parkinson's disease, even when people already have this disease (Kakazu, 2024). This symptom only shows relations with the highest intake of UPF.
 
Other symptoms like reduced sense of smell, daytime sleepiness, impaired colour vision and depression by themselves seem not related to consumption of UPF.
 
With regards to the UPF outcome, 30% of food consumption assessed by questionnaire was not agreed on and while experts apparently re-assessed these, it is not clear how they agreed on categorisation of foods, so whether they were UPF or not.
 
It seemed strange that non-UPF food included beef, pork, lamb chicken or turkey sandwich (all processed meats); cream; pancakes or waffles; pie, home-baked or readymade; popcorn; potato or
corn chips; soy milk; and tomato sauce, as well as distilled alcohol and dairy coffee.
 
Individual foods such as UPF breads or cereals and indeed microwaveable meals were by themselves not associated with the 'early Parkinson disease symptoms' while sauces, sweets, artificial sweetened drinks and desserts were as well as savoury snacks, animal and dairy products including yogurts.  Such foods are associated with diabetes mellitus and vascular (heart) disease, respectively, which can impact on brain disease because of their sugar and trans fat contents, respectively.
 
However, it is not the first study to show associations of UPF and brain disease.  We early wrote a piece on studies investigating dementia risk and processed meat consumption.  A healthy varied whole food diet is associated with prevention of many diseases including dementia.
 
Lastly, these two cohorts were mainly white health professionals so the results do not necessarily translate to everyone.
 
So this study may be affected by UPF categorisation as a predictor, where also not all UPF foods showed an association; the limited study group associations were assessed in (only mainly white health professionals and nurses) and also by the outcome, as these symptoms are not necessarily predictive of Parkinson’s disease, nor were these symptoms individually all associated with UPF consumption.

The findings in this study are interesting and appear to be based on solid research with conclusions well supported by the data.  However, it is important to highlight that the symptoms examined in this study are possible early signs of Parkinson’s disease, not definitive indicators that someone will go on to develop it.  The study did not track whether participants were diagnosed with Parkinson’s later on. 

Many of the individual symptoms noted, such as sleep disturbances, constipation, and mood changes, are common in the general population.  While the study found that people who ate more ultra-processed foods tended to report more of these non-motor symptoms, it did not find a direct increase in the risk of Parkinson’s disease itself.  That said, having more of these symptoms suggests a higher risk over time.  For example, a person experiencing a combination of REM sleep behaviour disorder, constipation, and depressive symptoms has a higher likelihood of developing Parkinson’s down the line, but the risk is not absolute.  To better understand the long-term implications, we would need a longer follow-up to see how many participants go on to develop Parkinson’s and how this is associated with their diet. 

In short, this is an interesting piece of research addressing important questions.  But the connection to Parkinson’s disease should be viewed with caution until more definitive evidence becomes available.