Baseline psychiatric and neurological disorders predict Long Covid while the latter brings internalizing morbidity, female-skewed cerebrovascular risk, and more emergency visits
Tabi, Younes Adam; Meyer, Eva Christina
Abstract
Long COVID is linked to substantial neuropsychiatric morbidity, but bidirectional risks, sex differences, and implications for routine care remain underexplored.
We conducted three complementary retrospective cohort analyses in the TriNetX Global Network (157 healthcare organizations), using sex-stratified 1:1 propensity-score matching for age and a fixed 1,095-day follow-up.
Pre-existing psychiatric disorders (anxiety, depression, bipolar, stress/adjustment, obsessive–compulsive, eating, personality) and several neurological conditions (migraine, peripheral neuropathy, multiple sclerosis) were associated with a higher probability of ICD-10 U09 post-COVID coding, with consistently larger excesses in women; schizophrenia and Alzheimer’s disease showed no increase.
Conversely, among patients with COVID-19 infection, U09 coding was followed predominantly by internalizing-spectrum diagnoses (anxiety, depression, stress/adjustment), smaller increases in eating, obsessive–compulsive, personality, and substance-related disorders, and rises in migraine, tension-type headache, peripheral neuropathy, epilepsy, transient ischemic attack, and stroke; no elevation was detected for schizophrenia or Alzheimer’s disease. Absolute psychiatric event rates after U09 were higher in women.
In patients with baseline disorders, U09 did not uniformly worsen survival. Exceptions included higher three-year mortality in men with prior transient ischemic attack, anxiety, or depression; modestly higher post-U09 stroke/TIA in women with anxiety, depression, stress-related, or headache disorders (and in anxious/stressed men); and increased emergency-department use after U09 in stroke survivors and in primary headache and affective/anxiety diagnoses.
Together, these findings suggest that in healthcare routine, the U09 label preferentially captures a phenotype dominated by internalizing psychopathology and pain/dysautonomia, with sex-specific risk architectures. Integrated neurology–psychiatry follow-up and targeted vascular surveillance for identified high-risk subgroups may help mitigate the long-term burden.
Web | DOI | PDF | Brain Disorders | Open Access
Tabi, Younes Adam; Meyer, Eva Christina
Abstract
Long COVID is linked to substantial neuropsychiatric morbidity, but bidirectional risks, sex differences, and implications for routine care remain underexplored.
We conducted three complementary retrospective cohort analyses in the TriNetX Global Network (157 healthcare organizations), using sex-stratified 1:1 propensity-score matching for age and a fixed 1,095-day follow-up.
Pre-existing psychiatric disorders (anxiety, depression, bipolar, stress/adjustment, obsessive–compulsive, eating, personality) and several neurological conditions (migraine, peripheral neuropathy, multiple sclerosis) were associated with a higher probability of ICD-10 U09 post-COVID coding, with consistently larger excesses in women; schizophrenia and Alzheimer’s disease showed no increase.
Conversely, among patients with COVID-19 infection, U09 coding was followed predominantly by internalizing-spectrum diagnoses (anxiety, depression, stress/adjustment), smaller increases in eating, obsessive–compulsive, personality, and substance-related disorders, and rises in migraine, tension-type headache, peripheral neuropathy, epilepsy, transient ischemic attack, and stroke; no elevation was detected for schizophrenia or Alzheimer’s disease. Absolute psychiatric event rates after U09 were higher in women.
In patients with baseline disorders, U09 did not uniformly worsen survival. Exceptions included higher three-year mortality in men with prior transient ischemic attack, anxiety, or depression; modestly higher post-U09 stroke/TIA in women with anxiety, depression, stress-related, or headache disorders (and in anxious/stressed men); and increased emergency-department use after U09 in stroke survivors and in primary headache and affective/anxiety diagnoses.
Together, these findings suggest that in healthcare routine, the U09 label preferentially captures a phenotype dominated by internalizing psychopathology and pain/dysautonomia, with sex-specific risk architectures. Integrated neurology–psychiatry follow-up and targeted vascular surveillance for identified high-risk subgroups may help mitigate the long-term burden.
Web | DOI | PDF | Brain Disorders | Open Access