In a rigorous medRxiv * preprint paper, a group of researchers shows shifts in the age structure and clinical characteristics of people with coronavirus disease (COVID-19) after social and economic reopening in three US states, showing how older people are less in were able to reduce these contacts compared to the younger ones.
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of the ongoing COVID-19 pandemic, has led to significant morbidity and mortality – especially in the US.
Accordingly, reopenings at the state level in spring 2020 offered a perfect opportunity for the resurgence of SARS-CoV-2 transmission. During this time, a pertinent question was whether human contact and mixing patterns could gradually increase without virus transmission increasing.
The reason for the latter was that mixed patterns are likely associated with better distancing, masking, and hygiene practices. Additionally, there was a question of whether the clinical features of the outbreak would improve after the onset of the cases.
A group of researchers led by Dr. Nathan Wikle of Pennsylvania State University analyzed the age-structured time series for cases, hospitalizations and deaths in the three US states (Rhode Island, Massachusetts and Pennsylvania) that had no major epidemic setbacks in the summer of 2020 compared to March / April .
Transferable mixing βt (in gray and blue) and mobility changes (yellow) from March 1st to August 31st. The average population mix for March 5-15 is set at 1.0 as the pre-epidemic level of communicable mix. All other values are relative to this. Gray lines show 1000 posterior beta trajectories sampled, with blue lines showing the median and 95% believable intervals. Note that there are significant uncertainties in these estimates in the first few weeks of March, as the number of cases was low and reporting may not have captured a large proportion of the actual cases at this point. Yellow lines show the proportion of Facebook and SafeGraph users who left their homes at least once a day. The correlation between population movement (yellow) and transferable population movement (gray + blue) begins to disappear in all three states in early May.
Assessment of epidemiological patterns and clinical data streams
These researchers evaluated eleven clinical data streams that were described by the respective state health departments in a Bayesian inference frame. Basically, this is based on an age-structured ordinary differential equation (ODE) epidemic model that includes compartments (i.e. clinical conditions) for hospital stays, critical care, and mechanical ventilation.
In particular, they derived parameters related to clinical features, surveillance and transmission patterns of the first wave of epidemics in Rhode Island, Massachusetts and Pennsylvania.
They then described the patterns of consistently low transmission in the three aforementioned states through August 31, then compared them to changes in human mobility metrics, and assessed changes in age structure and clinical outcomes.
Finally, they measured the impact of the spring epidemic on the elderly in these three states and compared the infection mortality rates with the estimates available from other parts of the United States.
Shifts in the age structure and clinical outcomes
"We show that population-average blending rates decreased by more than 50% during the March / April embargo period and that the correlation between total population and transferable mobility was broken in May when these states were partially reopened," say the authors of the study in this medRxiv paper.
In their analysis, infection death rates were estimated to be higher compared to the recently published summary studies, with the differences particularly notable in the 60- to 79-year-old age group, where rates 1.5 to 2.5 times that high.
Likewise, the researchers showed that older people were much less able to decrease contacts than younger ones during the lockdown, which resulted in the outbreak – despite the lockdown – being focused on older communities.
In addition, the study also indicated that people infected during the spring wave and summer interval were more likely to develop symptomatic illnesses compared to the average person in the population.
This is in line with the observation that children were least exposed in the spring and summer months. Therefore, the exposed population was more prone to obvious symptoms and more likely to develop severe clinical outcomes.
Posterior distributions of reporting rate (panel A) and clinical parameters (panels B to E) for Rhode Island (purple, left column), Massachusetts (orange, middle column), and Pennsylvania (green, right column). (A) report parameter ρ, i.e. H. The proportion of symptomatic SARS-CoV-2 cases reported to the healthcare system, plotted as a function of time. In Rhode Island, it was known that no tests were available and cases could not be confirmed in March. therefore a spline function was suitable for ρ. The same feature provided a better fit for Pennsylvania data, but not for Massachusetts data. (B) The median length of hospital stay on the medical floor was 7.5 days in RI, 11.9 days in MA, and 15.7 days in PA. In MA, this parameter was limited to 11.8 to 12.8 days, since without this restriction there were identifiability problems due to the lack of the data stream “cumulative hospital stays”. (C) Probabilities of Death at Home for Age Groups 60-69, 70-79, and 80+; The age group 60-69 was only included for RI as the data for PA and MA were insufficient. These largely reflect the epidemics that occur in nursing home populations where individuals are not considered hospitalized if they remain in care in their community care facility in a severe or advanced clinical condition. These probabilities are important when predicting hospital bed capacity. (D) Age-adjusted probability of occurrence in the intensive care unit during the blocking period in spring 2020 (lighter color) and after the blocking period (darker color). (E) Probability of hospital stay (median and 95% CIs) for symptomatic SARS-CoV-2 infections by age group; MA estimates are excluded because these priorities were established based on estimates in RI.
How can you protect the weakest?
In short, the statistical inferences described in this article on attack rates, underreporting, and shifting age profiles can actually provide improved guidance for real-time decision making and appropriate public health messaging.
"As recent policy discussions have been surrendered by the surrender and laziness of an epidemic management approach that would encourage younger / healthier populations to become infected, we should reiterate that our state-level analysis suggests that older people are not fully isolating during lockdown can periods ", emphasize the authors of the study.
Basically, this means that the strategy of protecting vulnerable people is impractical because vulnerable people continue to need the necessary care and contact. Hence, any policy aimed at protecting only the most vulnerable while allowing the rest of the population to move about freely would almost certainly prevent the introduction of viruses.
Since people in the oldest age groups are more or less unaffected by the lockdown, the best solution to protecting these (but also other) vulnerable populations is to limit its spread to the general population.
* Important NOTE
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be considered conclusive, guide clinical practice / health-related behavior, or treated as established information.