A new way of looking at Covid
CLINICAL MATTERS
Some media outfits and doomsayers continue to bandy the rising Covid-19 numbers and positivity rates in a way that tends to scare people into a panic. Alarmist headlines and overblown pronouncements of alleged impending lockdowns do not help. Covid-19 is no longer the killer it was two years ago. When the Philippines first locked down, we had no immunity, no proven treatments, and no idea how fast the virus was spreading. Nowadays, we have effective vaccines and boosters that substantially decrease the risk of severe Covid-19. We have effective medications that prevent progression to severe disease. Soon, we expect updated vaccines with increased efficacy against the new Omicron sublineages, which will further temper the impact and spread of these new variants.
To understand why and how the initial community quarantines and alert levels were implemented in 2020, we need to look at the metrics that the government decided to use early in the pandemic and compare these to what are relevant now.
The most intuitive and rational parameter for gauging the impact of Covid-19 on society is the healthcare utilization rate (HCUR). This measures the number of hospital beds occupied by Covid-19 cases and how many are still available. Unfortunately, HCUR tended to be a late indicator in the early days of the pandemic when a significant proportion of cases were severe. By the time a significant number of beds are occupied, any intervention may not be timely enough to take effect to prevent an overshoot of patients needing confinement. With the Covid-19 incubation period being up to 14 days, infection would have already occurred in a substantial number of people and some of these would need admission whatever measures are taken. HCUR also doesn’t include patients in the emergency room, who are awaiting confinement, which can suddenly use up existing bed capacity.
With the relatively high proportion of severe cases prior to the widespread availability of vaccines, HCUR was insufficient to protect the healthcare system and needed to be supplemented with more sensitive indicators.
Since HCUR is a lagging indicator, the Department of Health came up with other metrics that served as an earlier signal for an impending spike in cases. These metrics were the two-week growth rate (TWGR) and the average daily attack rate (ADAR).
The two-week growth rate compares the average number of daily new cases from the previous two weeks to the current two weeks. TWGR captures the speed at which cases are climbing and can anticipate the impact on HCUR faster than just passively observing HCUR. The basic problem with TWGR is that it can be misleading when cases are low. For instance, a TWGR from 10 cases a day from the previous two weeks to 30 cases a day for the current two weeks would be considered a 200 percent increase. This has the same weight and percent value if the cases went from 10,000 a day for two weeks to 20,000 a day for two weeks. There can be very wide fluctuations even when the actual increase in cases are few in number. Because of this problem, the TWGR was recently dropped as a risk metric in the last IATF-EID meeting under the Duterte administration.
The average daily attack rate considers the population of a place and measures the number of cases per 100,000 people. This is particularly useful in gauging the severity of an outbreak in a locality. It is a proportional metric and does not penalize cities and towns with large populations for having more overall cases than less populated areas just because there are more people living in an area. The cutoff numbers for ADAR have evolved throughout the pandemic. Earlier when it was essential to trace and contain outbreaks quickly and there were no vaccines or treatment available, an ADAR > 1 was already considered moderate risk. This was increased to an ADAR of six once there was significant population protection from vaccination. It also takes into account the increased transmissibility of Omicron. If the current metrics are retained, ADAR cutoffs may need to be adjusted upward further as much fewer cases now result in severe infection and these have a lesser impact on the healthcare system.
The new administration has not yet decided whether it will keep the current alert level system and metrics beyond July 15, 2022. The latest matrix that was approved by the previous IATF now uses HCUR as the most important metric such that alert levels will not escalate from level 1 to level 2 unless HCUR goes beyond 50 percent. Even then, it isn’t automatic if ADAR remains below 6. The latest iteration of alert level metrics was conceived with a transition to endemicity in mind.
An early metric, which WHO used to define adequate testing prior to the availability of vaccines and effective treatments, was a sample positivity rate of five percent or less over a two-week period. There is no point in using positivity rate as a gauge for adequate testing nowadays. Testing policies and objectives are no longer the same as they were at the start of the pandemic. With 70 million Filipinos already vaccinated, testing every single case of suspected Covid-19 with an RT-PCR is pointless and wasteful. The remaining reasons to test with RT-PCR are to confirm the diagnosis in patients with suspected severe and critical Covid-19 who are for hospitalization and treatment; to confirm the diagnosis of moderate Covid-19 for vulnerable at-risk groups so they can receive treatment with antivirals; and for surveillance purposes through systematic sampling and for genome sequencing. We therefore expect Covid-19 testing positivity rate to potentially rise above five percent in the future because we are selecting people with a higher probability of disease for testing and decreasing the total pool of people (hence the denominator) for testing for RT-PCR. This would have little bearing on any decisions to increase alert levels.
Case fatality rates for Covid-19 worldwide have been dropping from a high of seven percent near the start of the pandemic to just above one percent currently. Covid-19 is starting to look more like the flu and less like SARS, Covid-19, however, is still currently more deadly and transmissible than influenza. SARS-CoV-2, particularly the Omicron variant, is much more transmissible with an R0 of 8 to 14 (each case can infect eight to fourteen people) compared to an R0 of 1 to 2 for influenza. At the start of the pandemic, Covid-19 was more than 10 times deadlier than the flu but this has gone down substantially to about twice as deadly, thanks to widespread and effective treatments. Since Covid-19 remains very contagious, it is better to keep masks on for now, at least until variant-specific vaccines arrive later this year.
The graph below shows the case fatality rate for confirmed Covid-19 cases going down from about seven percent at the start of the pandemic to just above one percent at present. For comparison, pandemic flu (A H1N1pdm09) had a case fatality rate of 0.5 percent among symptomatic confirmed cases.
The best metrics for transitioning to endemicity are healthcare utilization and the number of new severe/critical cases. Elderly and other high-risk groups remain vulnerable to long Covid-19 and post-Covid-19 complications like stroke and heart attacks and so we need to keep mask mandates in place for now. Many countries are thinking about putting back their mask mandates as cases continue to rise as a result of widespread BA.5 transmission. Removing mask mandates can be revisited when the pandemic is declared over by WHO and to give time for updated vaccines with better infection prevention as boosters to arrive. In the meantime, increasing booster uptake will keep severe and critical cases low and ensure that the healthcare system remains open even if cases continue to increase.