Head injury raises Parkinson's risk

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Raquel Gardner, Amy Byers, Deborah Barnes, Yixia Li, John Boscardin and Kristine Yaffe (University of California, San Francisco), in the Chronic Effects of Neurotrauma Consortium. Europe PMC lists support from the US National Institute on Aging (K24 AG031155) and National Institute of Neurological Disorders and Stroke (K23 NS095755).

What happened

The team used three nationwide Veterans Health Administration databases (October 2002 to September 2014). They matched 1:1 by age patients with and without a diagnosis of traumatic brain injury (TBI), all free of Parkinson's and dementia at the start: 325,870 people, mean age 47.9, mean follow-up 4.6 years. During follow-up 1,462 were diagnosed with Parkinson's. Incidence was 0.31% without TBI, 0.58% with TBI of any severity, 0.47% with mild TBI and 0.75% with moderate-to-severe TBI. Adjusted hazard ratios (95% CI): any TBI 1.71 (1.53–1.92), mild 1.56 (1.35–1.80), moderate-to-severe 1.83 (1.61–2.07). Those with prior TBI were diagnosed about two years earlier than controls. Neurology 90(20):e1771–e1779, 2018.

Why it moved us toward a cure

It singled out a preventable exposure and pointed at mechanisms (speculative here) by which injury could feed sporadic disease. It strengthens the "genes plus environment" picture (landmark 25) with a non-chemical cause. It extends earlier work: a 2013 meta-analysis of 22 studies found a pooled odds ratio of 1.57 (1.35–1.83) for head trauma causing concussion (Jafari et al., Mov Disord), and Gardner's own 2015 California hospital study found 44% higher risk after TBI than after fractures (Ann Neurol).

Outcome

Open. This is observational: the TBI group had more medical and psychiatric conditions, follow-up was short and ICD-coded diagnoses may miss cases. Reverse causation is a real concern: a Medicare study of 89,790 people with incident Parkinson's found their risk of TBI rose steadily in the five years before diagnosis (hazard ratio 1.64 at five years to 3.93 in the last year), mostly from falls (Camacho-Soto et al., Ann Neurol 2017). Veterans are not the general population. How a single injury could lead to Parkinson's years later is not established, and it is not a disease-modifying therapy.

Sources

• Gardner et al., "Mild TBI and risk of Parkinson disease: A Chronic Effects of Neurotrauma Consortium Study", Neurology 2018;90:e1771–e1779, DOI 10.1212/WNL.0000000000005522, PMID 29669907 (abstract via Europe PMC)
• Jafari et al., Mov Disord 2013;28:1222–1229, DOI 10.1002/mds.25458 · Gardner et al., Ann Neurol 2015;77:987–995, DOI 10.1002/ana.24396 · Camacho-Soto et al., Ann Neurol 2017;82:744–754, DOI 10.1002/ana.25074 (abstracts via Europe PMC)
• Hospitalist summary of the study
• Not yet verified: the biological mechanism; whether repeated injury changes risk; anything after 2018 (no newer large cohort or meta-analysis was searched). No correction or retraction found.