Lower than 10% of US adults have SARS-CoV-2 antibodies

A new study by researchers at Stanford University and the Ascend Clinical Laboratory published in The Lancet in September 2020 reports the prevalence of antibodies to the acute respiratory syndrome severe coronavirus 2 (SARS-CoV-2) virus in a dialysis patient population in the US.

Advantages of the sample

The researchers selected this population because these patients will be tested every month to monitor their therapy and identify complications. This makes it easy to test the seropositivity of this virus without the need for another venipuncture as the dialysis port is present. Additionally, these individuals typically have the risk factors that are believed to increase the risk of infection and severe COVID-19, including older age, non-white race, poverty, and diabetes, at a far higher frequency than the rest of the USA population. Therefore, testing this population for seroprevalence using residual plasma has been viewed as a population representative monitoring strategy that can be maintained lengthways.

Study characteristics

Researchers tested a randomly selected group of dialysis patients to get an estimate of national exposure to SARS-CoV-2 from the start of the pandemic to July 2020 by age, sex, region and race.

The study enrolled over 28,500 people, with the majority tested in the first two weeks. The study also uses drop test data (polymerase chain reaction nasal swab test, PCR) that can help assess the correlation between seroprevalence and other measures of the pandemic such as number of cases, deaths, percentage of positive tests, and community-based risk factors for seropositivity.

Less than ten percent seropositivity

A total of 8% of the tests were positive for antibodies. The highest seroprevalence was in the northeast with ~ 27% and the lowest in the south with ~ 4%. There was no difference between men and women, but there was a slight decrease in patients over 80 compared to patients aged 45 to 64 years. The former was 20% less seropositive.

When classifying by race, non-Hispanic black patients were 2-3 times more likely to be seropositive, and non-Hispanic white patients were the least likely.

Based on June 2020 data, the prevalence of PCR positive cases is estimated to be 826 / 100,000 US adults. In contrast, the current estimate of seroprevalence is close to 9,000 / 100,000 residents, indicating that fewer than one in ten seropositive people will be admitted during the infection period – despite the fact that the highest number of confirmed cases in the US currently has.

On the other hand, a seroprevalence of 9% indicates that most American adults are still free of the virus. This includes dialysis patients who have one of the highest risk levels for COVID-19-related mortality in the world.

Bad correlation with other measures

The largest difference between the estimates was in New York at ~ 34%, but there were no differences in seven states. The best correlation was with the number of cumulative deaths per 100,000 population. People from impoverished areas had twice the risk, while people from overcrowded areas had ten times the risk. The measures implemented in March this year to restrict mobility reduced seroconversion rates by 60%.

Possible restrictions

The study presents robust data as the samples were collected routinely, which, unlike most other studies, eliminated numerous sources of bias. On the other hand, the seroprevalence in the dialysis group can overestimate the prevalence in the community for several reasons.

First, this group contains a higher proportion of poorer people and people from ethnic minorities. Black Americans are known to have four times the risk of end-stage kidney disease compared to whites. Second, patients could use shared transport to get to the hemodialysis center and spend 10 to 12 hours in the center each time they visit, which could encourage higher transmission.

On the other hand, dialysis patients work less often and are more likely to be confined to the home, which reduces the chances of transmission. Finally, seroconversion in these patients may be weaker due to a weaker immune response, as shown by studies of hepatitis B immunization in this group, in which only up to 75% of those vaccinated were seroconverted, compared with 95% in the general population.

Another confusing factor is the possible absence of the most seriously ill SARS-CoV-2 patients in the dialysis group due to death or hospitalization with other indications, which ensures their exclusion from the dialysis group.


Estimated seroprevalence is consistent with previous studies in both the US and Geneva, Switzerland, showing that by July 2020, less than a tenth of Americans had been exposed to the virus. As such, herd immunity is still a distant dream despite the major outbreaks in the first half of the year, as well as other large-scale seroprevalence surveys in the UK and Spain.

The study also highlights the differences in health care between ethnic, racial and socioeconomic groups that have been further exaggerated with the current pandemic. The tenfold risk of seroprevalence in people living in high density areas such as crowded cities, in confined spaces, and in large gatherings such as carnivals shows the essential nature of the virus spread.

The slightly lower likelihood of seropositivity in older people may be due to their careful distance from other potentially infectious people. However, it cannot be ruled out that more elderly people will be hospitalized or have died from the infection. This excludes them from the sample group in this study.

Given these factors, the study provides a model for monitoring the pandemic in the community and highlights the feasibility and cost-effectiveness of using repetitive, routine-collected samples from a group of people who are already being monitored to fully understand the actual incidence of SARS-CoV-2 to measure infection and complete faster and more limited surveys. The accuracy of this estimate is shown to some extent in the lack of correlation between this and other prevalence measures currently in use, with the exception of cumulative deaths.

Such surveys can also help assess the appropriateness of the tests, and longitudinal follow-up can predict hospital stays and intensive care admissions, since there is only an average of 10 days between exposure and seroconversion. This helps with precautionary measures. They can also help assess how well preventive and therapeutic interventions are working. All of these benefits come with little added cost as the remaining plasma is used, eliminating the need for venipuncture with human, equipment, and infrastructure requirements and including groups that have traditionally been left out or underrepresented.

The study includes: “Serial sampling of residual dialysis plasma should be used to determine trends in disease prevalence and the impact of various strategies being implemented in the US to reduce the exposure of the general population to COVID-19. “

Comments are closed.