“Declining prevalence of antibody positivity to SARS-COV-2” in UK seroprevalence study: MedRxiv and Imperial College London.

A new study, published on MedRxiv on October 27, 2020, shows a decrease in antibody positivity to SARS-COV-2 among randomly selected UK (United Kingdom) patients over a period of three months. The specimens were collected at three times between late June and September 2020 and showed positivity rates of 6.0%, 4.8%, and 4.4%– a drop of 26%. Evaluations of subgroups, however, showed only 22% loss of antibodies among those who had definitely or probably been sick with COVID-19, as is explained below.

The study collected finger-prick blood specimens at home and performed surveys over the phone or online. The three rounds of testing used roughly 100,000, 106,000, and 159,000 patients. The tests were conducted about 12, 18, and 24 weeks after the peak of the pandemic in England in early April.

The results showed the highest prevalence of antibodies among people who had confirmed COVID-19 with RT-PCR tests at the first round: 96% had antibodies. On the second and third rounds, these rates fell to 90% and 75%. Those who did not have RT-PCR tests, but were suspected by a doctor to have COVID-19, showed 35% positive antibody tests. Those who thought they might have had COVID-19 were 19% positive, whereas those who didn’t think they had it had a 0.9% prevalence.

People who had no symptoms had a 1% prevalence of positivity; those who had atypical symptoms were 10.5% positive, and those who had typical symptoms were 25.1% positive.

Among demographics, those in the youngest group (18-24) had the highest prevalence of antibodies: 7.9%. Those who lived in London had a 13% prevalence of antibodies. Those who were Black had a 17.3% prevalence, versus a 5% prevalence among White people.

Those in the lowest socioeconomic group had a 7.3% prevalence, versus those in the highest, who had a 5% prevalence. Those in the smallest household group, of one person, had a 4% prevalence, versus 9.8% in the largest households of 7 or more people. Those in the lowest population density areas had a 3.2% prevalence versus 7.4% in the highest density areas.

Those who worked in healthcare with patients had 12.9% positivity at the first test, increasing to 13.4% at the third test. Those who worked in nursing homes with patients had 19.6% positive rates, decreasing to 11.1% at the third test. Those who worked in jobs outside healthcare had 6.5% positive rates at the first test, decreasing to 4.3% at the third test, and those who did not work had 4.2% positive at first, decreasing to 3.1% at the end.

The greatest declines in positivity were among those with the lowest rates to begin with: those who had no symptoms and no positive tests had antibody positivity rates fall by 64% (from 1% to 0.4%) and 59% (from 0.9% to 0.3%) respectively. Those who had confirmed COVID-19 by RT-PCR only had positivity rates fall by 22%. Residents of nursing homes actually had rates increase by 82%, but the total numbers of people in that group was small, only 150-348 patients.

These numbers indicate that the rate of antibodies among people in England fell over the three-month period after the sudden onslaught of cases that peaked in the second week of April. Numbers of new cases fell precipitously three weeks after the country was locked down around March 23. The antibody positivity rates fell the most among people who probably had mild or inapparent infections, while those who had confirmed infections retained most of their antibody prevalence.

Those who worked in health care, whether they worked with patients or not, had increasing rates of antibody positivity– more than double the rates for the entire group. Essential workers had decreasing positivity rates, overall similar to the average rates for the whole population.

Most important, those who had confirmed or suspected COVID-19 lost their antibody prevalences at a less than 10% rate, while those who were not thought to have been sick lost antibody prevalence at a 45% rate. This suggests that most people who clearly had COVID-19 retained antibodies after three months. The overall decrease in antibody prevalence of 19% includes a lot of people who were not thought to have been sick; many of these people must have had mild, inapparent infections.

So far, we do not have any such population statistics for the United States. These statistics would probably be very different from those in the United Kingdom.