High prevalence of underlying health conditions explains increased COVID-19 mortality seen in people with HIV in US study

People living with HIV do not have an increased risk of dying due to COVID-19 once underlying health conditions are taken into account, investigators from the United States report in the online edition of AIDS. However, the research did find that, even after taking into account co-morbidities such as obesity, lung and kidney disease, people with HIV were more likely than closely matched HIV-negative individuals to be hospitalised because of COVID-19.

Dr Yousaf Hadi and his co-researchers at the University of West Virginia conducted the research. With over 400 people with HIV and COVID-19 co-infection, it is one of the largest studies on the topic to date and had a robust methodology, directly comparing mortality and hospitality rates between individuals with HIV and closely matched HIV-negative controls.

“Our analysis finds that outcomes of COVID in HIV populations are no different than HIV-negative population, even in patients with a history of treatment with antiretrovirals," write Dr Hadi and his colleagues.

But he found that people with HIV becoming sick with COVID-19 had a high prevalence of serious underlying illnesses compared to HIV-negative individuals – indeed because of this, their first analysis showed higher mortality rates among people with HIV than HIV-negative people with COVID-19.

The implications of COVID-19 infection for people with HIV are unclear. On the one hand, some studies suggest that people living with HIV have a slightly elevated risk of COVID-19-related death, an excess mortality risk that may be explained by the presence of underlying health conditions or social factors. But it has also been argued that commonly used anti-HIV drugs might offer some protection against COVID-19 and its consequences.

In the present research, Dr Hadi and his colleagues examined the characteristics and outcomes of over 50,000 individuals with confirmed COVID-19 who received care at healthcare organisations in cities across the US. A total of 404 of these people were HIV positive.

The two study outcomes were death within 30 days of COVID-19 diagnosis and hospitalisation because of the infection.

Data were also gathered on possible risk factors for poorer COVID-19 outcomes, such as age, gender, chronic kidney and/or lung disease and cigarette smoking. These were all taken into account when the researchers conduced their comparisons.

The HIV-positive individuals had a mean age of 48 years, 71% were male and 50% were African American. Just under three-quarters (70%) had ever taken antiretrovirals but only 46% had been prescribed HIV therapy within six months of their COVID-19 diagnosis.

Comparison with HIV-negative COVID-19 patients showed that those with HIV were more likely to be male (71% vs 51%, p < 0.001), African American (50% vs 25%, p < 0.001) and to have underlying health conditions including obesity (26% vs 21%, p = 0.03), hypertension (46% vs. 28%, p < 0.001), chronic lung disease (25% vs 16%, p < 0.001), chronic kidney disease (17% vs 7%, p < 0.001), diabetes (22% vs 15%, p < 0.001), heart disease (14% vs 8%, p < 0.001) and be cigarette smokers (14% vs 7%, p < 0.001).

"High mortality is driven by higher burden of risk factors for severe COVID in the HIV patients."

The investigators’ first analysis found that mortality rates within 30 days of COVID-19 diagnosis were higher among people with HIV compared to HIV-negative individuals (5.0% vs 3.2%), a statistically significant 55% difference (risk ratio (RR) = 1.55; 95% CI, 1.01-2.39).

People with HIV were also more likely to be hospitalised because of COVID-19 (19% vs 11%; RR = 1.83; 95% CI, 1.50-2.24).

The investigators then undertook a second analysis, matching each of the HIV-positive individuals with an HIV-negative person with similar characteristics, including underlying health conditions.

In this analysis, the difference in mortality risk between the HIV-positive and HIV-negative groups ceased to be statistically significant (5.0% vs 3.7%; RR = 1.33; 95% CI, 0.69-2.57).

This finding was unaffected when the risk was compared according to the use of antiretrovirals. Matching people taking HIV therapy with HIV-negative people showed that the mortality rate was 3.9% compared to 3.2% for their matched controls, a non-significant difference.

However, it was still the case that people with HIV were more likely be hospitalised after a COVID-19 diagnosis (19% vs 11%, RR = 1.70; 95% CI, 1.21-2.38).

“COVID crude mortality is higher in HIV-positive patients when compared to non-HIV patients, however propensity matched analyses revealed no difference in outcomes, showing that high mortality is driven by higher burden of risk factors for severe COVID in the HIV patients,” conclude Dr Hadi and his colleagues. “Patients living with HIV represent a cohort of patients with many risk factors for severe disease that needs special consideration in public health efforts. Early diagnosis and intensive surveillance may be needed to prevent a ‘Syndemic’ of diseases in this vulnerable cohort.”