Dementia Patients Have Twice the Risk of COVID-19

Dementia patients had a significantly increased risk for COVID-19 and that risk was even higher for African Americans with dementia, a retrospective analysis of U.S. electronic health record (EHR) data showed.

After adjusting for age, sex, race, and COVID-19 risk factors including comorbidities and nursing home stay, people with dementia had twice the risk of developing COVID-19 as other adults (adjusted OR 2.00, 95% CI 1.94-2.06, P<0.001), according to Rong Xu, PhD, of Case Western Reserve University School of Medicine in Cleveland, and co-authors.

Among people with dementia, African-American patients were nearly three times as likely to be infected with SARS-CoV-2 as white patients (adjusted OR 2.86, 95% CI 2.67-3.06, P<0.001), they reported in Alzheimer’s & Dementia.

The 6-month mortality risk for patients with dementia and COVID-19 was 20.99%. Hospitalization risk was 59.26%.

“We have identified patients with dementia as a group that is especially vulnerable to becoming infected with SARS-CoV2,” said co-author Pamela Davis, MD, PhD, also of Case Western Reserve University. “Once infected, these patients are especially vulnerable to severe disease and death from COVID.”

“Both of these vulnerabilities occur over and above concomitant known risk factors for COVID, such as nursing-home living, advanced age, hypertension, and diabetes,” Davis told MedPage Today. “Physicians should be attuned to this increased risk and take whatever additional measures they can to protect this vulnerable population, such as vaccination, masking, and careful attention to masking and social distancing of caregivers.”

The findings illustrate well-known health disparities that must be addressed, said Carl Hill, PhD, MPH, chief diversity, equity, and inclusion officer of the Alzheimer’s Association in Chicago, which wasn’t involved with the study.

“We know African Americans are disproportionately impacted by both Alzheimer’s disease and COVID-19, and this paper suggests African Americans with dementia are three times as likely to contract COVID-19 as their white counterparts,” Hill told MedPage Today. “This may be explained by the African American’s higher risk for comorbidities associated with dementia, like diabetes and hypertension, that may make them more vulnerable to symptomatic and severe cases of COVID-19.”

In their analysis, Xu and co-authors used EHR data from 360 hospitals and 317,000 providers across the U.S., incorporating records from 61.9 million patients from the start of the pandemic in February until August 21, 2020. In total, 1,064,960 people had dementia, including 351,590 people with Alzheimer’s disease, 172,630 with senile dementia, 126,450 with post-traumatic dementia, 117,860 with vascular dementia, and 31,960 with presenile dementia.

Of 15,770 people diagnosed with COVID-19, 810 had dementia. The highest risk was for patients with vascular dementia (adjusted OR 3.17) followed by presenile dementia (adjusted OR 2.62), senile dementia (adjusted OR 1.99), Alzheimer’s disease (adjusted OR 1.86), and post-traumatic dementia (adjusted OR 1.67; all P<0.001).

Among people with both COVID-19 and dementia, hospitalization risk during the 6-month study period was 73.08% among African-American patients and 53.85% among white patients (P<0.01). Mortality risk for people with COVID-19 and dementia was 23.08% for African Americans and 19.23% for white patients.

Overall, the 6-month mortality risk for people with dementia and COVID-19 (20.99%) was higher than it was for people with COVID-19 but not dementia (4.81%, P<0.001), or for people with dementia but not COVID-19 (7.64%, P<0.001).

More work is needed to understand what drives elevated COVID-19 risk in dementia patients, Davis said. The findings have several limitations, the researchers noted: EHR data may be subject to underdiagnosis, overdiagnosis, or misdiagnosis. The study had limited information about socioeconomic and lifestyle determinants, and looked only at people who had encounters with healthcare systems, Xu and colleagues added. Findings are not causal, and unknown confounders may have influenced results.