Retired Soccer Players Reported More Depression and Anxiety While Their Cognitive Tests Came Back Normal (new study)

A study presented at the Alzheimer’s Association International Conference in London in July compared 142 former professional soccer players against 56 people who had never played a contact sport. Most of the coverage reported that 31% of former players scored in the range indicating clinically significant depression, compared with 9% of the comparison group.

Playing soccer is good for the brain. Brain imaging in people with regular soccer training found higher gray matter volume than in non-players, in the cerebellum, the thalamus and the visual cortex, with larger differences the longer someone had been training. Physical activity is among the modifiable factors the Lancet Commission lists for dementia prevention and a sport that combines aerobic exercise with fast decision-making and regular social contact covers several of those factors at once. The study described below concerns repeated head impact and what happens when a playing career ends, which is different from the brain benefits of the sport itself.

The players reported more mood symptoms but their cognitive tests were normal. Does depression damage the brain over time, or does it show up early because the brain is already changing? That question applies to everyone, not only to people who played contact sports.

What the Imperial College team measured

The design. The study comes from the Advanced BRAIN Health Clinic Research Programme at the Institute of Sport, Exercise and Health, with Caleigh Grace Lynch as lead author and Dr. Thomas Parker as senior author, both at Imperial College London and the UK Dementia Research Institute Centre for Care Research and Technology. The players were aged 30 to 60, comprising 126 men who had held a full-time professional contract in England for at least three years and 16 women who had competed in the top two tiers of the UK women’s game. The comparison group of 56 people, 43 of them men, had not played contact sports, had not served in the military and had no history of repeated head impacts or neurological conditions.

The symptom findings. Former players reported more symptoms of depression and anxiety, along with more difficulty planning, concentrating and managing everyday tasks. Alongside the depression figures, 42% of players scored in the range indicating clinically significant anxiety against 25% of the comparison group.

The imaging findings. Brain scans in 124 former players showed lower gray matter volume in frontal, cingulate and thalamic regions. Gray matter is the tissue containing the cell bodies of neurons and these particular regions support memory, attention, decision-making and emotional regulation. A neuroradiologist reviewing the scans clinically found atrophy suggestive of neurodegeneration in about 2% of them, which in a sample of 124 is two or three people and is within the range of incidental findings in any middle-aged group.

What the researchers said about it. Ms. Lynch noted that although there were not clear differences between former players and non-players on standard cognitive tests, there were differences in reported symptoms and in brain imaging. The team described the combination as possibly suggesting trauma-related neurodegeneration while stating that more research is needed to establish that.

Six explanations for the mood finding

The study is an interim analysis of a comparison in which everyone was measured at a single point in time rather than followed over years, which means it cannot distinguish among the following.

Head impact acting on emotional circuitry. The frontal, cingulate and thalamic regions where volume was lower are part of the circuitry that supports mood regulation and impulse control. The deep folds of the frontal cortex and the cingulate are also early sites where tau, a protein that builds up abnormally inside neurons, accumulates in chronic traumatic encephalopathy. Neurobehavioral dysregulation is a described feature of the CTE clinical syndrome that typically appears before measurable cognitive impairment. Under this reading, the mood symptoms are the first clinical sign of injury and the cognitive tests are not yet sensitive enough to register anything.

Retirement from elite sport. Professional soccer players stop playing in their early to mid thirties and lose income, structure, physical identity, social network and daily purpose at once, often involuntarily because of injury. Nobody in the comparison group experienced anything similar. This is a described phenomenon in sports psychiatry and it would produce elevated depression and anxiety in this exact age window without any head impact at all. Every former player in the study was also a retired athlete, so the two exposures cannot be separated.

How the two groups were assembled. The programme these players came through is a clinical service for retired athletes. If players entered because they were already concerned about their brains, then people with symptoms are overrepresented among them by design. A comparison group recruited as healthy volunteers selects in the opposite direction. Selection operating in both directions at once can produce a threefold difference in reported symptoms with no underlying biological difference and the published materials do not describe how either group was recruited.

Expectation. Soccer players in the UK have spent more than a decade reading about dementia in former players. Anticipating brain disease can raise both reported mood symptoms and subjective complaints about thinking. This explanation predicts exactly the pattern observed, which is elevated self-report alongside normal performance on objective testing.

Other exposures that come with a soccer career. Orthopedic injury and chronic pain, extended use of pain medication, disrupted sleep and higher alcohol consumption are all more common after a professional playing career. Chronic pain on its own is among the more consistent contributors to depression in middle-aged men.

The direction of the imaging finding. Depression is itself associated with lower volume in prefrontal cortex, anterior cingulate and related regions. The cingulate result could therefore reflect head trauma, or depression, or both. Presenting the symptom finding and the volume finding as confirming one another assumes a direction that has not been established.

What would separate these. A dose-response pattern would, meaning one where more exposure goes with more symptoms. If depression scores rise with heading exposure or career length within the player group, the head impact explanation gains support, because retirement, expectation and self-selection do not obviously track heading volume. Companion abstracts at the same meeting reported that heading exposure in amateur play was linked to short-term rises in blood markers of neural injury, and that longer soccer careers were associated with CTE. The player study itself did not report a dose-response analysis of the mood measures.

Depression and dementia in people who never played a contact sport

The consensus position. The 2024 Lancet standing Commission report, led by Dr. Gill Livingston, keeps depression among 14 modifiable risk factors that together account for an estimated 45% of dementia cases worldwide. Treating depression effectively is one of the recommendations that follows from it.

The pooled numbers. An umbrella review and meta-analysis published in eClinicalMedicine in 2025 is a recent synthesis. Depression present in late life roughly doubled the rate of dementia across 18 studies covering 901,762 participants and 7,595 incident cases. Carried through to lifetime terms, that moves a 42% baseline to somewhere around 65%. Depression assessed in midlife raised the rate by about half across seven studies covering more than 2.5 million participants, which corresponds to roughly 57% instead of 42%.

The same direction problem in the general population

The eClinicalMedicine authors note that shorter intervals between depression and dementia diagnosis may reflect depression appearing as an early symptom of a disease process already underway rather than as a cause of it. Depression earlier in life is less likely to be explained that way, which is part of why the midlife and late-life estimates differ.

A 2025 scoping review concluded that the relationship runs in both directions, with clinical, imaging and longitudinal evidence supporting depression as a risk factor, as an early sign, and as a consequence of neurodegeneration. An earlier meta-analysis in the Journal of Neuropsychiatry and Clinical Neurosciences reached a similar impasse, with the authors noting that their results allow a possible causal relationship to be proposed without settling whether depression is cause or consequence.

The soccer players are an unusual case in this literature because the exposure came first and is not in question. Nobody develops a professional soccer career as an early symptom of neurodegeneration. That makes the cohort a cleaner test of direction than the general epidemiology, provided the recruitment question can be answered.

Beyond depression

Other psychiatric conditions show associations of varying size and varying credibility with dementia risk.

Stress-related disorders. A Korean national cohort matching 8,906 patients with stress-related disorders to 26,718 controls found a 15% higher adjusted rate of dementia overall, which against a 42% lifetime baseline works out to about 47%. Post-traumatic stress disorder was highest, at 78% higher, or roughly 62% lifetime. Work in US veterans has reported a further doubling with PTSD, in the range of 70%.

Severe mental illness. A Taiwanese database study of 84,824 people aged 45 to 69 reported rates of Alzheimer’s disease around ten times higher with bipolar disorder, nine times higher with major depression and four and a half times higher with schizophrenia. Converting those to lifetime terms produces figures above 95%, which shows they cannot be taken at face value. Effect sizes of that magnitude in administrative claims data usually reflect a combination of detection bias, since people already in psychiatric care are more likely to be assessed for cognitive problems, and confounding by the metabolic and vascular conditions that accompany long-term psychiatric illness and its treatment. The ordering is plausible, since severe and persistent mental illness would be expected to carry more risk than a stress-related diagnosis and both studies place it higher. The claims data cannot pin down how much higher.

What might connect mood and brain aging

A framework from gene activity (RNA levels). Every cell carries the same genes but switches them on and off at different levels depending on the tissue, the person and the age. Measuring all of those levels at once, across the whole genome, is called transcriptomics, and it gives a readout of what a tissue is actually doing rather than what it is capable of doing. One account of the connection between mood and brain aging is that depression and normal brain aging move gene activity in the same direction, so that having depression places someone further along a trajectory the brain follows anyway. Work I did during my PhD in Dr. Etienne Sibille’s lab at the University of Pittsburgh developed a way to assign a molecular age to a brain sample from its pattern of gene activity. Comparing four brain areas across cohorts, the pathways involved in neurological disease overlapped substantially with the pathways that change during normal aging and people carrying a low-expressing variant of the SIRT5 gene had brains that were molecularly older than their chronological age. A companion review laid out the genetic and molecular pathways involved, including the pattern where synaptic and calcium signaling genes decrease with age while inflammation-related genes in glial cells increase.

Applying the same measure to depression, a study from the same lab examined tissue from the amygdala, a region involved in processing emotion, in 21 women with major depressive disorder and 21 age-matched controls and found that the gene activity changes associated with depression were the changes that normally occur later in aging, appearing earlier. The authors concluded that an older molecular brain age may be an early biological event in depression or a marker of risk for later symptoms.

Other proposed pathways. Other candidate explanations include chronic inflammation, dysregulation of the hypothalamic-pituitary-adrenal axis that governs the cortisol stress response, vascular changes, altered levels of neurotrophic factors that support neuronal survival, and shared genetic and lifestyle contributions. None of these has been established as the operative pathway in humans.

Several of them overlap with pathways covered here before. Chronic inflammation and vascular change are among the processes reflected in neurofilament light chain, the blood marker of neuronal injury. Autonomic regulation, which is what heart rate variability measures, is disturbed in both depression and cognitive decline. Sleep disruption is a feature of depression and is independently associated with cognitive outcomes. The overlap is part of why isolating depression as a distinct causal factor has been difficult.

Where this leaves things

No randomized trial has treated depression in adults without dementia and then followed them long enough to measure whether their rate of dementia changed, so the study that would settle the question has not been run. The long-term follow-ups of the largest late-life depression treatment trial, IMPACT, tracked diabetes incidence and healthcare costs rather than cognitive outcomes. The randomized evidence that does exist on antidepressants and the brain comes from people who already carry a dementia diagnosis, where a Cochrane review found the evidence variable in quality and not strongly supportive of efficacy beyond twelve weeks. That is a different question from whether treatment in cognitively healthy adults changes what happens later.

Mood symptoms in midlife are associated with brain outcomes decades later across several populations studied in different ways, the association is stronger when depression is measured closer to diagnosis, and at least part of the late-life association reflects depression appearing as an early feature of neurodegeneration rather than driving it. Multi-modal measurement of the kind discussed in what biological age means is one approach to the underlying problem, which is that a single measure taken at a single time point cannot distinguish a cause from an early symptom.

The soccer players may end up providing part of the answer. Dr. Parker’s group plans to follow the same people over time using diffusion imaging, which tracks the movement of water along nerve fibers to assess the health of the brain’s wiring, along with blood markers of neuronal injury in an expanded sample. Mood symptoms appearing before those markers change would point one way and the markers changing first would point the other. Either result would say something about the larger group of people with depression unrelated to playing professional soccer.

Mental health belongs in the same conversation as sleep, exercise and the other things people already think of as brain health measures, regardless of how that resolves. Whether depression drives brain aging, reflects it, or does both, the association is consistent enough across populations that treating mood as separate from cognitive aging does not match the evidence. That is why taking care of your mental health is one of the nine pillars of healthy brain aging that we use at NeuroAge.