Category: Information / Topics: History • Information • Statistics • Trends
by Stu Johnson
Posted: August 217, 2021
While only a blip at the end of July, the delta variant and high number of unvaccinated are putting a damper on the optimism felt a month ago…
Putting the COVID-19 pandemic in perspective (Number 13)
This series was spawned by my reaction to reporting early in the COVID-19 pandemic that focused on raw numbers. Big numbers are impressive, even frightening, but hard to comprehend. This series turned into a monthly summary setting the U.S. numbers in global perspective. This analysis is based on data from worldometers.info, which monitors 215 countries. From those, I focus on details for 26 countries since I started more detailed tracking in September 2020. (Indonesia, South Africa, Ukraine, Poland, and Netherlands were added to the list this month because they have appeared in the weekly report of most new activity and are now in the top-20 of cases and deaths).
A note on reliability of data: It should be noted that the statistics reported by worldometers and other sources are only as good as the integrity of the reporting system in each country. China's statistics have increased so little since I started tracking numbers that it is doubtful its numbers are accurate, so they are not included in charts that focus on countries. Other countries have also had gaps in reporting, or made adjustments. There is still enough information to make trends evident. That is why I tend to round some numbers and watch for changes over two or three months rather than focusing on a single month, as significant as that may turn out to be.
These reports will continue as long as the pandemic persists around the world.
Report Sections:
• July at-a-glance
• The Continental View • USA Compared Other Countries
• COVID Deaths Compared to the Leading Causes of Death in the U.S.
• U.S. COVID Cases versus Vaccinations
• Profile of Monitored Continents & Countries • Scope of This Report
This is where I have to suggest how important it is to get information from reliable and knowledgeable sources that know how to interpret the numbers. In an atmosphere wary of misinformation, news-by-anecdote from otherwise trusted sources can itself be a form of misinformation. As I go through the statistics each month, I am reminded often that the numbers do not always line up with the impressions from the news. With that caveat, let's dig into the numbers for July 2021.
While COVID-19 has been classified as a global pandemic, it is not distributed evenly around the world, though we began to see some shifts in patterns starting in April 2021.
Asia accounts for 59.3% of the world's population (Figure 2), but had only 31.2% of COVID cases at the end July (Figure 3A)—affecting a mere 1.3% of its population (COVID cases have been reported for 2.5% of world population). After shrinking in proportion of cases since October, Asia continues to increase, with each month since March surpassing the previous month. The trade-off is that Europe and North America have been going down in share each of those months. Meanwhile South America (with 8.4% of world population) bottomed out at around 15% of COVID cases back in January and has crept up to a share of 18% the last two months.
Where Asia and Africa combined represent about three-quarters (76.4%) of the world's 7.8-billion people, Europe, South America and North America still account for two-thirds of cases (65.3% - Figure 3A) and three quarters of COVID deaths (74.8% - Figure 4A). The shares for Europe and the Americas combined are slowly coming down from their highs (74.7% for cases and 80.8% for deaths in February).
Oceana's population is too small to be statistically significant, so it is not included in my analysis.
While far from the dramatic surges of past months, all continents except South America showed very slight increases in the rate of growth in cases in July, apparently related to the highly-transmissible delta variant. If that is the case, we should see an even greater impact in August.
Asia has the highest number of reported COVID cases, having surpassed both Europe and North America as it experienced a dramatic upturn in April and May followed by slower but continued growth in June and July. South America remains in fourth place and slowed down in rate of growth slightly in July. North America began to flatten its curve back in January (when vaccinations began), dropping away from Europe in number of cases, but showed a slight upward bend in its curve in July. Europe (slower to get vaccinations started) showed a slight increase in pace in March and April before decelerating in May and June, but it too is showing an apparent delta-related increase in rate in July. Africa, second by population, remains a loner, with an almost flat curve compared to the other four large continents, but it also bent upward slightly in July.
The raw numbers of Figure 3B can be deceptive. Figure 3C gives a more realistic picture of impact by translating raw case numbers to percentage of population.
After a sharp increase in January, North America has slowed down in monthly growth, with a very slight downturn in June followed by a similar up-tick in July. That still leaves North America far above the other continents in proportion of infections, with one in ten (11.5%) reported as having contracted COVID—the proportion is increasing about half a percent each month. Even if it goes flat, with virtually no further increases in cases, the likelihood of any other continent matching it in proportion of cases is extremely low.
Europe has slowed slightly below the North America and global pace. Like North America, Europe went down slightly in June, then back up in July. South America remains above the global level, increasing its distance from the global level by a small amount each month, thus bending its curve upward.
Asia and Africa remain below Global levels, though Asia showed an increase in its pace for April and May before slowing slightly in June and July. Despite the reports of COVID surges in India and other parts of Asia similar to what Italy experienced at the beginning of the pandemic, the sheer size of the continent means it can absorb a lot of growth in cases before it causes a dramatic bend in the curve (as we did see in North America in January). However, individual countries in Asia dominate my watch list (see July-at-a-Glance above),
Reports of surges in Africa barely are reflected with just a minor upward movement in its curve in July. Africa reported 6.7-millon cases by the end of July, 0.5% of its population of 1.3-billion—up a million cases, but only increasing one-tenth of a percent in share of its population.
The proportion of deaths between continents is even more distorted than that of cases. Figure 4A shows the steady increase in the proportion of deaths in Europe and South America through April. North America peaked in February. July ended with Asia and Africa (76.4% of world population) accounting for only 25.2% of COVID deaths.
The growing crisis in India showed up in May deaths, with Asia hitting 19.2% and a slower increase to 21.2% in July, pushing it to a new high since my observation of continental data began in November.
Africa was down slightly in May then up to 4.0% by July, still barely below its most recent high of 4.1% in January and February . . . and far below its 17.1% share of world population.
Europe went down in proportion of deaths for the third consecutive month in July, North America for the fifth consecutive month.
South America slowed in proportion of deaths through February, then began increasing each month, gaining 7.5 points by the end of July. Peru made corrections to its death count in June that represent a major part of the jump in South American deaths in June, with a much smaller increase in July.
The overall distribution for Deaths through July shows that while the trajectory lags behind cases and has progressed at a steadier rate, it does reflect the overall changes in Cases by continent.
By numbers alone, we can say that COVID deaths reported for Africa continue at a very slow and steady pace. North America continues to slow down, remaining under the 1-millon mark and falling to third place last month. Europe showed a slight downward bend of its curve in June that continued into July, but remains in first place. Asia and South America began climbing in March. Asia slowed slightly in June and July, keeping it in fourth place, but on a trajectory to pass North America in another month or two. South America climbed at an even higher rate, surpassing North America in June to move into second place by number of deaths, then crossing the 1-million mark in July, poised to overtake Europe if the trend line continues.
Raw numbers are virtually meaningless without relating them to the size of a given country. So, we start with a look at the top-5 countries by proportion of population with reported COVID cases. Argentina moves into #1 this month, passing now #2 USA by a very small margin. Belgium stays at #3, Spain and Columbia, at #4 and #5 respectively, are virtually tied.
The top-5 by proportion of population with COVID cases, have progressed at a rate higher than the Global level, but have stayed within a fairly narrow band that has been tightening since March.
USA reached #1 by December, continuing to climb at a very fast pace until starting to slow in January as vaccine roll-outs began, bringing it closer to the others by the end of June and landing in #2 in July. The others meandered a bit, began to appear as if they were spreading out by March, but drew closer together since then. While USA had a roller-coaster rise in the early months, Argentina's rise to #1 accelerated more slowly but steadily in the past four months. All are significantly higher than the global level of 2.5% cases as proportion of a country's own population.
Another way to look at population proportion is the measure "1 in." The global figure of 2.5% means that 1 in 39 people in the world have been reported with COVID (and that only by official record keeping, not including any unreported and likely asymptomatic cases). For USA it is 1 in
9. For Spain and Columbia it is 1 in 11.
Netherlands tied Argentina at 10.9% of its population with COVID, but since it was added to the monitored list this month, it will not show up in charts for another month or two when there is a trend to illustrate.
Bolivia and Canada traded placed this month. The other three countries remain in the same order from June.
At the low end (Fig. 5B), the five lowest countries for cases by proportion of population are much closer to the global figure of 2.5% of population with confirmed cases of COVID-19 at the end of July, though Bolivia and Canada have been breaking away from the others since May.
Canada was moving upward more than the others until June when it was barely ahead of Bolivia (3.8% to 3.7%). By the end of July, Canada stayed at 3.8%, putting it below Bolivia, which increased to 4.0%. Ecuador has stayed in the middle of the five, always above the Global level. India, at the Global level to start, stayed below it until it stated a more rapid ascent in April, passing Mexico in May, and slowing enough to stay below the Global level in July. Mexico followed close to the Global level through March, then moved below it, ending at the lowest position the past three months.
These countries represent a considerable spread in size, from India, the second largest country, to Bolivia, ranked number 50 of the 215 countries tracked by worldometers. For Canada, its 3.8% level represents 1 in 27 people, for India it is 1 in 44 and for Mexico it is 1 in 46.
Because the size of countries makes the use of raw case numbers illusory, another measure I find helpful is the rate of change from month to month (Figure 5C).
For this chart, countries are selected based on the change over two-months (end of May to end of July for this report). Except for USA (which will appear every month), the mix has changed since the last report, but the tilt toward South America last month, with four countries, is moderated somewhat with the return of UK as it faced surging numbers even as it reopened. Iran also returns to this group. Not returning from last month are India and Chile.
The overall trend (red line)
is flattening as it drops. Interrupting the downward Global trend were a four-point increase in April and a 2-point bump in July. (A polynomial trend line flexes as adjacent data points go up and down, so the leading edge of newest dates can change the shape of the curve as new months are added). Note that some months appear "flatter" than others, with less variations between countries and the global (world) number.
Columbia has had monthly increases in excess of 10% for all but two months since November, the first month it was tracked. From a low of 7% change over the previous month in February and March, it climbed to a 25% change in June before settling down some to a 14% change in July.
UK is notable for its 22% increase of July over June—after five months below 10%—just as the delta variant wrecked havoc on what promised to be a grand reopening. UK also suffered a huge spike of 128% back in October during the fall 2020 surges that swept across Europe and North America.
Argentina and Bolivia both hit recent highs for monthly change in May and have moved downward since then (11% and 8% respectively for July over June), but still had high enough change over two-months to land on this chart.
USA started above the Global level from November through January, then has fallen further below it each month since, although July showed a bump from the low of 1% for June over May back up to 3% for July over June.
Because deaths as a percentage of population is such a small number, the "Deaths-per-Million" metric shown in Figure 6A provides a comparable measure.
Peru and Brazil stay at #1 and #2 from June. Columbia moves up two to #3. Argentina replaces Italy at #4 and Belgium moves down two to #5. That puts four countries in South America in the top-five for Deaths-per-Million in July.
As Figure 6A shows, Peru still soars over the others following a correction to its death data in June. July remains more than double #2 Brazil and ten times the Global level of 543.
The other South American countries have all been moving upward. In fact, Columbia and Argentina are so close, you can barely tell them apart on the chart. Belgium has been slowing down in monthly deaths-per-million, but remains in the top-five of the 26 countries I monitor.
Since this analysis focuses on 26 countries that have been in the top-20 of cases and deaths, there are 7 other countries with Deaths-per-Million between Peru and Brazil. The second place country at the end of July (as it was in June) was Hungary (population 9.6-million) with a Deaths-per-Million of 3,117. Ranked #4 was Czechia (population 10.7-million) at 2,830.
The other countries on the chart, including USA are all above the Global level, and (except for Peru) fairly close to each other—though in July they started to spread out again from their closest proximity in June.
As observed last month, the overall trend shown in Figure 6A is that USA, with aggressive testing and vaccination, has seen slowing COVID deaths since January, while Europe (represented by Belgium) was also beginning to slow deaths in recent months. Progress in Europe and North America, however, is being stymied by the delta variant and plateauing vaccination rates. Even without Peru's astonishing correction in June, South America (also represented by Brazil, Colombia and Argentina) remains a trouble spot for COVD deaths, more so statistically than Asia or Africa.
Among the lowest five for Deaths-per-Million, Russia and Iran traded places for #1 and #2 from June. The remaining three countries appear in the same order.
Though starting to escalate in March, India remained well below the global rate of 543 Deaths-per-Million at the end of July.
The other four countries are all above the Global level. Russia has been moving steadily upward and a further upward bend of its curve in July put it at the top of the lowest-five, just ahead of Iran, whose slowing rate helped put it in second place. Canada, similar to USA, has been slowing since January. Turkey, following an upward bump in May is moving back toward the Global level.
As with the comparable chart for Rate of Change for Cases (Figure 5C), countries for Rate of Change for Deaths (Figure 6C) are selected based on the change over two-months (end of May to end of July for this report) in reported COVID deaths. Except for USA (which will appear every month), the mix has changed since the last report, with Ecuador replacing India. Peru and Russia remaining at #1 and #5, while Columbia and Argentina switched the #3 and #4 positions.
The trend
in the rate of change in deaths (red line in Figure 6C) shows the delayed connection with cases. Monthly Global changes for deaths trended down through May, bumped up in April, then went back down in July. While the Peru increase of 177% in June was exceptional, it was not enough to bend the global trend significantly. (As mentioned with Rate of Change for Cases, the trend line is a polynomial that can change shape as new values are added at the most-recent end.)
An important comparison is individual countries each month against the global (World) level. As shown in Figure 6C, the Global level peaked at 24% in November, with Russia at or above the Global level from October through February. As the Global level dropped, so did the variance of change in the top-5 countries—until June that is, when Peru came in with the 177% increase and Columbia at 20% was nearly double the Global level. In July, even with the Global level of change from the previous month down to 7%, four of the top-five were still higher (Peru had dropped to 2%, but Ecuador was 48%, while Columbia, Argentina and Russia were all above 10%.
USA was higher than the Global level in December, January and February, then has been significantly lower since then, falling to a 1% change in July (over the number of deaths in June).
Mortality Rates (percentage of deaths against reported cases) have generally been slowly declining. This is not surprising as several factors came into play: the ratio of community spread versus outbreaks (nursing homes, similar confined settings) increased, knowledge about treatment increased, along with testing (which would expand the spread between reported cases and deaths), and vaccinations since January (though that should impact both cases and deaths).
The Global mortality rate has dropped from 2.6% in October to 2.1% in July. The median for the 26 countries monitored for this report was 2.9% at the end of July.
The top-5 in mortality rate (among the 26 countries monitored) all return from June, in the same order. Three of the five, not surprisingly, are in South America, along with nearby Mexico (all Central American countries are considered North America by continent). .
Peru was in fourth place and declining until June when its corrected death numbers drove Mortality to 9.4%, just ahead of Mexico at 9.3%. At the end of July, Peru eased back slightly, down to 9.3% and Mexico made a significant drop to 8.5%. Like Peru, Ecuador had been declining faster than the Global level, but then shot up nearly two full points to 6.5% at the end of July as deaths increased 47%.
Bolivia's Mortality Rate has been declining, slowing down since January, but still significantly above the Global level, at 3.8% for July. Italy has been slowly declining since a big drop early on and while above the Global level, the 2.9% at the end of July is close to the median of 2.7% for all 26 monitored countries. South Africa also came in at 2.9%, but it was just added to the monitored list this month, so it will take a month or more of additional data before it appears in trend charts.
Since these represent the best mortality rates, where low is good, the "rank" order is actually in reverse. Spain enters the list at #1, as France moves to #2 and Canada drops off the list. USA, India and Turkey return, in the same order as June.
All five countries have been below the Global Mortality Rate since April. Having come closer together over time, Spain, France, and USA all ended July with Morality Rates of 1.8%. USA has been below the Global level the entire time, Spain and France started higher than dropped below the Global level in April. India started at 1.5% then moved down to 1.1% in April before moving up again due to surging cases in some areas, ending July at 1.3%. Ironically, while television reports showed the human scale of suffering in India, the country is so large that the statistics make it look like not much was happening. Turkey, which was added to the monitored countries list in December, tracked in a pattern similar to but below India, ending July at 0.9%.
How real is the threat of death from COVID? That's where successful mitigation comes in. Worldwide, by the end of July, 1 in 39 people have been reported as having contracted COVID and 1 in 1,848 people have died. In USA, while the mortality rate is low, because the number of cases is so high, 1 in 530 have died through July 2021.
With low mortality, USA. should have been able to keep deaths much lower, but the extraordinarily high number of cases means more deaths. Without a better-than-global mortality rate, the USA death rate would be far higher. Compared to the 1918 pandemic, it could be ten times worse. The response of the health care system and availability of vaccines are part of keeping mortality down, but it's far too early to detail the cause for that positive piece of the COVID picture for USA. Even at the Global mortality rate of 2.2%, USA would have had 785-thousand deaths (for 35.7-million cases) by the end of July, instead of 629-thousand.
The same five countries remain on top in COVID testings, having been in the same order since October 2020.
USA remains ahead of other countries in reported COVID tests administered, at 530-million, 14% ahead of India, but that has narrowed from 56% in April. UK continues at the pace it set with an upturn in February, having moved past Russia in March. Russia and France continue on a steady upward path, roughly 60-million tests apart since January.
These are raw numbers, so it is important to recognize the size of the country. It is also the case that COVID tests can be administered multiple times to the same person, so it cannot be assumed that USA has tested almost all of its population of some 330-million. Some schools and organizations with in-person gatherings are testing as frequently as once a week or more for those who are not yet fully vaccinated. That's a lot of testing!
The same five countries return from June, in the same order.
There is tremendous disparity between countries in terms of testing. Except for Belgium, the other four countries (among the 26 monitored) with the lowest reported number of tests are in South and North America.
Belgium and Peru are climbing at a rate comparable to USA and India. Bolivia and Ecuador are not making much progress, and Mexico falls between the two groups.
As questions arise about equity of testing between countries, check the number of tests for countries of similar size:
Peru (15.5-million tests for its 32.9-million population) is well below Canada (39.5-million tests for its 37.7-million population). Ironically, even though they both fall in the bottom-5 by number of tests among the 26 monitored countries,, Bolivia, with a population of 11.7-million has tallied 2.1-million tests, while Belgium, with 11.6-million people, has reported seven times that number, with 17.2-mllion tests.
Tests per million adds another perspective. Fig. 7C shows the five countries with the highest tests per million. All five return from June, but France moves ahead of USA.
The UK, already the most aggressive in testing, increased its numeric lead each month since February, with a reported 3.58-million tests per million population in July, just over 3.5 tests per person. USA continues on a straight line trajectory, with a very slight dip the last two months, reaching 1.590-million tests per million, more than 1.5 tests per person, allowing France to edge ahead with 1.599. Belgium and Italy track close to each other, both increasing their climb above 1-million tests per million in June, with Belgium at nearly 1.5-million tests per million at the end of July and Italy approaching 1.3-million tests per million. .
Anything over 1,000 (or "x-million tests-per-million") represents more tests than people (1,000 on the chart actually means 1,000,000), but as mentioned above, that does not mean that everyone had been tested. Some people have been tested more than once, and some are being tested regularly or with increased frequency.
The same five countries have appeared as the lowest five in tests per million in the same order since May.
While still at the bottom of the 26 countries monitored for this report, Iran and Brazil have made the most substantial progress, Iran in a steady and slowly accelerating curve reaching 304-thousand tests-per-million, Brazil, in more halting fashion that may indicate inconsistencies in reporting, is not far behind at 257-thousand. Bolivia began to accelerate in March, so it is now nearly double Ecuador which has moved more slowly, but steadily, along with Mexico.
While improvement is evident in all five, the equivalent proportion of tests to population remains very low, from roughly 6% to 30%. This illustrates the arguments over inequity in resources among countries.
Early in the reporting on COVID, as the death rate climbed in the U.S., a great deal of attention was given to benchmarks, most notably as it approached 58,000, matching the number of American military deaths in the Vietnam War. At that time, I wrote the first article in this series, "About Those Numbers," in which I looked at ways of viewing the data, which at the time of that writing in May 2020 was still focused on worst-case models and familiar benchmarks, like Vietnam.
Figure 8 shows the number of US COVID cases and deaths against the top-10 causes of death reported by CDC for 2018 (the latest data available), along with several other mileposts from wars and the 1918 pandemic. Notice that for nearly nine months, the curve for deaths was increasing at a faster rate than cases. Then, starting in October 2020 the curve for cases took a decided turn upward, while deaths increased at a more moderate pace.
Most media sources reported that the U.S. would pass 500,000 deaths on February 22, 2021. The CDC reported 497,415 deaths on February 21, while worldometer.info was already reporting 512,068. As mentioned elsewhere in this report, worldometer tends to be about 3% ahead of CDC and Johns Hopkins, a matter of a few days difference in data collection. At any rate, it is clear that February 2020 saw the U.S. death toll from COVID pass the 500,000 mark, headed toward 600,000.
The February slowdown in cases was met by a barely perceptible slowdown in deaths, because as pointed out earlier, deaths trail changes in cases. That trend continued in March and bent downward even more in April. In fact, if the trajectory from November to February had continued (the upper edge of the blue cone), we would have seen close to 1-million deaths by the end of July. The total of 610-thousand at the end of May did pass the number who died from Cancer in 2018 (599,000).
The next benchmark is the #1 cause of death, heart disease, at 655-thousand. We're close, but the curve has been bending down, so how long will it take to get there? The delta variant could make the answer to that question more troublesome than the trend through July makes it appear.
The latest "Ensemble Forecast" from CDC suggests this for the immediate future (even though worldometers is already ahead of this prediction, it is within the 3% difference between the two sources):
...the number of newly reported COVID-19 deaths will likely increase over the next 4 weeks, with 2,300 to 9,100 new deaths likely reported in the week ending August 28, 2021. The national ensemble predicts that a total of 624,000 to 642,000 COVID-19 deaths will be reported by this date...
Note: As I've referenced in the notes for several charts, the data from worldometers.info tends to be ahead of CDC and Johns Hopkins by about 3%, because of reporting methodology and timing. I use it as a primary source because its main table is very easy to sort and provides the relevant data for these reports.
The 1918-19 Spanish Flu pandemic is estimated to have struck 500 million people, 26.3% of the world population of 1.9-billion at that time. By contrast, we're now at 2.5% of the global population. Deaths a century ago have been widely estimated at between 50- and 100-million worldwide, putting the global mortality rate somewhere between 10 and 20-percent. It has been estimated that 675,000 died in the U.S.
IF COVID-19 hit at the same rate as 1918, we would see about 2-billion cases worldwide by the time COVID-19 is over, with the global population now at 7.8-billion—four times what it was in 1918. There would be 200- to 400-million deaths. The U.S. is estimated to have had 27-million cases (one-quarter of the population of 108-million) and 675,000 deaths. Today, with a population of 330-million (a three-fold increase from 1918) this would mean more than 80-million cases, and 2- to 4-million deaths.
However, at the present rate of confirmed cases and mortality while the total number of global cases could approach 500 million or more—comparable to 1918 in number, that would be one-quarter of 1918 when taking population growth into account . .. and assuming the pandemic persists as long as the Spanish Flu, which went on in three waves over a two year period. At the present rate of increase (approximately 12-million cases per month) it would take another 26 months to reach 500-million, well into 2024.
If the total number of cases globally did approach 500-million, using the global mortality rate of 2.1% in July, there would be roughly 10.5-million deaths worldwide, tragic but far below the number reported for 1918 (50-million) with an even wider gap (200 million) when taking population growth into account.
In my last report, I indicated that with vaccination in progress and expected to be completed in the U.S. by the end of summer, the end of COVID-19 could come sooner. Like 1918, however, there are now complicating factors, such as the combination of the delta variant with a high number of unvaccianted (some by choice, as in USA and Europe, and many more in underdeveloped countries by inequitable access to vaccine). While a month ago we may have thought the end of the pandemic was in sight, it is still too early to make predictions on the duration and severity of the COVID-19 pandemic globally.
Despite the darkening forecast in the past month, the vast difference in scale between the Spanish Flu pandemic a century ago and COVID-19 even a year-and-a-half in, cannot be denied.. Key differences are the mitigation efforts, treatments available today (though still leaving the health care system overwhelmed in some areas during surges), and the availability of vaccines. In addition, in 1918 much of the world was focused on a brutal war among nations (World War I) rather than waging a war against the pandemic, which ran its course and was undoubtedly made much worse by the war, with trans-national troop movements, the close quarters of trench warfare, and large public gatherings supporting or protesting the war.
With remarkable speed (it usually takes years to develop vaccines), two COVID vaccines were granted emergency approval for use in the U.S. starting in January—the one by Pfizer requires super-cold storage, which limits its deployment. The other, by Moderna, requires cold storage similar to other vaccines. Both of these require two doses, which means that vaccine dosages available must be divided in two to determine the number of people covered. By my March report, Johnson & Johnson had been granted approval for a single-dose vaccine. The numbers in Figure 9 represent the status of all three vaccines as of June 30.
A person is considered "fully vaccinated" two weeks after the final (or only) vaccine dose; roughly five to six weeks total for Pfizer and Moderna and two weeks for Johnson & Johnson.
Early on (still somewhat evident in March 2021) there was much anxiety about vaccine availability and scheduling appointments. That is behind us and vaccinations are readily available to most of the population (there will always be some disparities and trouble spots, but in general most people who want the vaccine can get it). Now, the biggest concern is convincing the dubious, the deniers, and the resisters to get vaccinated and help the country reach herd immunity (about 70% of the population fully vaccinated). Some of the latest strategies to encourage vaccination is the provision of incentives, from free beer to lotteries for the fully vaccinated. The seriousness of the delta variant has scared others into taking action.
The CDC (Centers for Disease Control and Prevention) recently revised its mask guidelines. Some resistors claim this proves the science cannot be trust—but that position misunderstands the nature and role of science, which is to observe and constantly assess assumptions and recommendations based on a growing (and developing) body of data.
Some school districts that were prepared to make masking optional have announced that the school year will begin with masks in place, regardless of vaccination status. The district in which we live has had strong arguments on both sides (option versus mandate), but yesterday Illinois Governor Priztker announced a mask mandate for all private and public schools pre-school through grade 12.
Masking policy is actually evidence that the real battle now is vaccination. Will the rise of infections and the call for more masking be enough to spur more vaccinations? For some, perhaps, but whether low levels of vaccinations prolonged the pandemic (around the world) will be the subject of discussion and debate for years to come.
Vaccinating over 300 million people in the United States (much less a majority of the billions around the world) is a daunting task. It is a huge logistical challenge, from manufacture to distribution to administration. Yet, it remains amazing that any of this is possible this soon after the identification of the virus just over a year and a half ago.
There is a delicate balance between maintaining hope with the reality that this is a huge and complicated logistical operation that will take time, though that time is now measured in months instead of years. In addition, getting shots into the arms of the unvaccinated is proving to be a far bigger challenge that most officials assumed a few months ago.
As the richer countries with access to more resources make progress, the global situation is raising issues of equity and fairness within and between countries. Even as the U.S. and other countries launch large scale vaccine distribution to a needy world community, the immensity of the need is so great that a common refrain heard now is whether this aid is too little, too late. As COVID fades into a bad memory in countries able to provide help, will the sense of urgency remain high enough to produce the results needed to end this global pandemic?
Maintaining Perspective
The following section has appeared several months and I repeat it here because there is an even greater need now to evaluate what is happening: to understand the science and its dynamic nature, to put politics in perspective and to show empathy, to truly demonstrate that we are in this together and want to get out of it as soon as possible.
In the tendency to turn everything into a binary right-wrong or agree-disagree with science or government, we ignore the need to recognize the nature of science and the fact that we are dealing with very complicated issues. So, in addition to recommending excellent sources like the Centers for Disease Control and Prevention (CDC), it is also wise to consider multiple qualified sources.
While there has been much focus placed in trusting "the science," it is important to recognize that science itself changes over time based on research and available data. In the highly volatile political atmosphere we find ourselves in (not just in the U.S., but around the world), there is a danger of not allowing the experts to change their views as their own understanding expands, or of trying to silence voices of experts whose views are out of sync with "the science" as reported by the majority of media outlets.
In an earlier report, I mentioned the Greater Barrington Declaration, currently signed by nearly 58-thousand medical & public health scientists and medical practitioners, which states "As infectious disease epidemiologists and public health scientists we have grave concerns about the damaging physical and mental health impacts of the prevailing COVID-19 policies, and recommend an approach we call Focused Protection."
For a personal perspective from a scholar and practitioner who espouses an approach similar to the Focused Protection of the Greater Barrington Declaration, see comments by Scott W. Atlas, Robert Wesson Senior Fellow at the Hoover Institution at Stanford University, in an article "Science, Politics, and COVID: Will Truth Prevail?"
Several months ago on SeniorLifestyle I posted an article by Mallory Pickett of The New Yorker, "Sweden's Pandemic Experiment," which provides a fair evaluation of the very loose protocols adopted by Sweden, essentially a variation of the "Focused Protection" approach. The "jury is still out" on this one, so judge for yourself whether Sweden hit the mark any better than the area in which you live.
UPDATE ON SWEDEN: At the end of July Sweden reported 1.1-million cases of COVID, or 10.8% of its population of 10.2-million. There have been 14,620 deaths, for a mortality rate of 1.3%. Ranked 89 by population, Sweden was number 29 in cases and number 39 in deaths. That would put Sweden in a virtual tie with Argentina and USA at the top of my cases as a proportion of population, but the mortality rate would be tied with India, near the bottom among the 26 countries monitored for this report and well below the global rate of 2.1%.
How we evaluate the many approaches used to deal with COVID will determine how we prepare for and approach the next global event.
What is clear in July 2021 compared to July 2020 is that while debates rage about masking and other policies, the response to surges in countries with vaccine readily available is much more like the "focused protection" approach discussed by the sources cited here than the full lockdowns of a year ago.
My purpose in mentioning these sources is to recognize that there are multiple, sometimes dissenting, voices that should be part of the conversation. The purpose of these monthly reports remains first and foremost to present the numbers about COVID-19 in a manner that helps you understand how the pandemic is progressing and how the U.S. compares to the world—and how to gain more perspective than might be gathered from the news alone.
(Data from worldometers.info).
Rank | Country | Population | Share of World Population |
Density People per square km |
Urban Population |
Median Age |
WORLD | 7.82B | 100% | -- | -- | -- | |
Top 10 Countries by Population, plus Five Major Continents See lists of countries by continent |
||||||
- | ASIA | 4.64B | 59.3% | 150 | 51 countries | 32 |
1 | China | 1.44B | 18.4% | 153 | 61% | 38 |
2 | India | 1.38B | 17.7% | 454 | 35% | 28 |
- | AFRICA | 1.34BM | 17.1% | 45 | 59 countries | 20 |
- | EUROPE | 747.7M | 9.6% | 34 | 44 countries | 43 |
- | S AMERICA | 653.8M | 8.4% | 32 | 50 countries | 31 |
- | N AMERICA | 368.9M | 4.7% | 29 | 5 countries | 39 |
3 | USA | 331.5M | 4.3% | 36 | 83% | 38 |
4 | Indonesia** | 274.5M | 3.5% | 151 | 56% | 30 |
5 | Pakistan* | 220.9M | 2.8% | 287 | 35% | 23 |
6 | Brazil | 212.9M | 2.7% | 25 | 88% | 33 |
7 | Nigeria* | 206.1M | 2.6% | 226 | 52% | 18 |
8 | Bangladesh* | 165.2M | 2.1% | 1,265 | 39% | 28 |
9 | Russia | 145.9M | 1.9% | 9 | 74% | 40 |
10 | Mexico | 129.3M | 1.7% | 66 | 84% | 29 |
*these countries do not appear in the details because they have not yet reached a high enough threshold to be included **Indonesia was added to the monitored list in July 2021 Other Countries included in Analysis most have been in top 20 of cases or deaths |
||||||
Rank | Country | Population | Share of World Population |
Density People per square km |
Urban Population |
Median Age |
17 | Turkey | 84.3M | 1.1% | 110 | 76% | 32 |
18 | Iran | 83.9M | 1.1% | 52 | 76% | 32 |
19 | Germany | 83.8M | 1.1% | 240 | 76% | 46 |
21 | United Kingdom | 67.9M | 0.9% | 281 | 83% | 40 |
22 | France | 65.3M | 0.8% | 119 | 82% | 42 |
23 | Italy | 60.4M | 0.8% | 206 | 69% | 47 |
25 | South Africa* | 59.3M | 0.8% | 94 | 67% | 28 |
29 | Colombia | 50.9M | 0.7% | 46 | 80% | 31 |
30 | Spain | 46.8M | 0.6% | 94 | 80% | 45 |
32 | Argentina | 45.2M | 0.6% | 17 | 93% | 32 |
35 | Ukraine* | 43.7M | 0.6% | 75 | 69% | 41 |
39 | Poland* | 37.8M | 0.5% | 124 | 60% | 42 |
39 | Canada | 37.7M | 0.5% | 4 | 81% | 41 |
43 | Peru | 32.9M | 0.4% | 26 | 79% | 31 |
63 | Chile | 19.1M | 0.2% | 26 | 85% | 35 |
67 | Ecuador | 17.6M | 0.2% | 71 | 63% | 28 |
69 | Netherlands* | 17.1M | 0.2% | 508 | 92% | 43 |
80 | Bolivia | 11.7M | 0.1% | 11 | 69% | 26 |
81 | Belgium | 11.6M | 0.1% | 383 | 98% | 42 |
*Added to the monitored list in July 2021 |
From the worldometers.info website I track the following Categories:
Instead of reporting Cases per Million directly, I try to put raw numbers in the perspective of several key measures. These are a different way of expressing "per Million" statistics, but it seems easier to grasp.
My analysis covers countries that have appeared in the top-10 of the worldometers categories since September 2020. This includes most of the world's largest countries as well as some that are much smaller (see the chart in the previous section).
This article was also posted on SeniorLifestyle, which I edit
Search all articles by Stu Johnson
Stu Johnson is owner of Stuart Johnson & Associates, a communications consultancy in Wheaton, Illinois focused on "making information make sense."
• E-mail the author (moc.setaicossajs@uts*)* For web-based email, you may need to copy and paste the address yourself.
Posted: August 217, 2021 Accessed 3,973 times
Go to the list of most recent InfoMatters Blogs
Search InfoMatters (You can expand the search to the entire site)
Category: Information / Topics: History • Information • Statistics • Trends
by Stu Johnson
Posted: August 217, 2021
While only a blip at the end of July, the delta variant and high number of unvaccinated are putting a damper on the optimism felt a month ago…
Putting the COVID-19 pandemic in perspective (Number 13)
This series was spawned by my reaction to reporting early in the COVID-19 pandemic that focused on raw numbers. Big numbers are impressive, even frightening, but hard to comprehend. This series turned into a monthly summary setting the U.S. numbers in global perspective. This analysis is based on data from worldometers.info, which monitors 215 countries. From those, I focus on details for 26 countries since I started more detailed tracking in September 2020. (Indonesia, South Africa, Ukraine, Poland, and Netherlands were added to the list this month because they have appeared in the weekly report of most new activity and are now in the top-20 of cases and deaths).
A note on reliability of data: It should be noted that the statistics reported by worldometers and other sources are only as good as the integrity of the reporting system in each country. China's statistics have increased so little since I started tracking numbers that it is doubtful its numbers are accurate, so they are not included in charts that focus on countries. Other countries have also had gaps in reporting, or made adjustments. There is still enough information to make trends evident. That is why I tend to round some numbers and watch for changes over two or three months rather than focusing on a single month, as significant as that may turn out to be.
These reports will continue as long as the pandemic persists around the world.
Report Sections:
• July at-a-glance
• The Continental View • USA Compared Other Countries
• COVID Deaths Compared to the Leading Causes of Death in the U.S.
• U.S. COVID Cases versus Vaccinations
• Profile of Monitored Continents & Countries • Scope of This Report
This is where I have to suggest how important it is to get information from reliable and knowledgeable sources that know how to interpret the numbers. In an atmosphere wary of misinformation, news-by-anecdote from otherwise trusted sources can itself be a form of misinformation. As I go through the statistics each month, I am reminded often that the numbers do not always line up with the impressions from the news. With that caveat, let's dig into the numbers for July 2021.
While COVID-19 has been classified as a global pandemic, it is not distributed evenly around the world, though we began to see some shifts in patterns starting in April 2021.
Asia accounts for 59.3% of the world's population (Figure 2), but had only 31.2% of COVID cases at the end July (Figure 3A)—affecting a mere 1.3% of its population (COVID cases have been reported for 2.5% of world population). After shrinking in proportion of cases since October, Asia continues to increase, with each month since March surpassing the previous month. The trade-off is that Europe and North America have been going down in share each of those months. Meanwhile South America (with 8.4% of world population) bottomed out at around 15% of COVID cases back in January and has crept up to a share of 18% the last two months.
Where Asia and Africa combined represent about three-quarters (76.4%) of the world's 7.8-billion people, Europe, South America and North America still account for two-thirds of cases (65.3% - Figure 3A) and three quarters of COVID deaths (74.8% - Figure 4A). The shares for Europe and the Americas combined are slowly coming down from their highs (74.7% for cases and 80.8% for deaths in February).
Oceana's population is too small to be statistically significant, so it is not included in my analysis.
While far from the dramatic surges of past months, all continents except South America showed very slight increases in the rate of growth in cases in July, apparently related to the highly-transmissible delta variant. If that is the case, we should see an even greater impact in August.
Asia has the highest number of reported COVID cases, having surpassed both Europe and North America as it experienced a dramatic upturn in April and May followed by slower but continued growth in June and July. South America remains in fourth place and slowed down in rate of growth slightly in July. North America began to flatten its curve back in January (when vaccinations began), dropping away from Europe in number of cases, but showed a slight upward bend in its curve in July. Europe (slower to get vaccinations started) showed a slight increase in pace in March and April before decelerating in May and June, but it too is showing an apparent delta-related increase in rate in July. Africa, second by population, remains a loner, with an almost flat curve compared to the other four large continents, but it also bent upward slightly in July.
The raw numbers of Figure 3B can be deceptive. Figure 3C gives a more realistic picture of impact by translating raw case numbers to percentage of population.
After a sharp increase in January, North America has slowed down in monthly growth, with a very slight downturn in June followed by a similar up-tick in July. That still leaves North America far above the other continents in proportion of infections, with one in ten (11.5%) reported as having contracted COVID—the proportion is increasing about half a percent each month. Even if it goes flat, with virtually no further increases in cases, the likelihood of any other continent matching it in proportion of cases is extremely low.
Europe has slowed slightly below the North America and global pace. Like North America, Europe went down slightly in June, then back up in July. South America remains above the global level, increasing its distance from the global level by a small amount each month, thus bending its curve upward.
Asia and Africa remain below Global levels, though Asia showed an increase in its pace for April and May before slowing slightly in June and July. Despite the reports of COVID surges in India and other parts of Asia similar to what Italy experienced at the beginning of the pandemic, the sheer size of the continent means it can absorb a lot of growth in cases before it causes a dramatic bend in the curve (as we did see in North America in January). However, individual countries in Asia dominate my watch list (see July-at-a-Glance above),
Reports of surges in Africa barely are reflected with just a minor upward movement in its curve in July. Africa reported 6.7-millon cases by the end of July, 0.5% of its population of 1.3-billion—up a million cases, but only increasing one-tenth of a percent in share of its population.
The proportion of deaths between continents is even more distorted than that of cases. Figure 4A shows the steady increase in the proportion of deaths in Europe and South America through April. North America peaked in February. July ended with Asia and Africa (76.4% of world population) accounting for only 25.2% of COVID deaths.
The growing crisis in India showed up in May deaths, with Asia hitting 19.2% and a slower increase to 21.2% in July, pushing it to a new high since my observation of continental data began in November.
Africa was down slightly in May then up to 4.0% by July, still barely below its most recent high of 4.1% in January and February . . . and far below its 17.1% share of world population.
Europe went down in proportion of deaths for the third consecutive month in July, North America for the fifth consecutive month.
South America slowed in proportion of deaths through February, then began increasing each month, gaining 7.5 points by the end of July. Peru made corrections to its death count in June that represent a major part of the jump in South American deaths in June, with a much smaller increase in July.
The overall distribution for Deaths through July shows that while the trajectory lags behind cases and has progressed at a steadier rate, it does reflect the overall changes in Cases by continent.
By numbers alone, we can say that COVID deaths reported for Africa continue at a very slow and steady pace. North America continues to slow down, remaining under the 1-millon mark and falling to third place last month. Europe showed a slight downward bend of its curve in June that continued into July, but remains in first place. Asia and South America began climbing in March. Asia slowed slightly in June and July, keeping it in fourth place, but on a trajectory to pass North America in another month or two. South America climbed at an even higher rate, surpassing North America in June to move into second place by number of deaths, then crossing the 1-million mark in July, poised to overtake Europe if the trend line continues.
Raw numbers are virtually meaningless without relating them to the size of a given country. So, we start with a look at the top-5 countries by proportion of population with reported COVID cases. Argentina moves into #1 this month, passing now #2 USA by a very small margin. Belgium stays at #3, Spain and Columbia, at #4 and #5 respectively, are virtually tied.
The top-5 by proportion of population with COVID cases, have progressed at a rate higher than the Global level, but have stayed within a fairly narrow band that has been tightening since March.
USA reached #1 by December, continuing to climb at a very fast pace until starting to slow in January as vaccine roll-outs began, bringing it closer to the others by the end of June and landing in #2 in July. The others meandered a bit, began to appear as if they were spreading out by March, but drew closer together since then. While USA had a roller-coaster rise in the early months, Argentina's rise to #1 accelerated more slowly but steadily in the past four months. All are significantly higher than the global level of 2.5% cases as proportion of a country's own population.
Another way to look at population proportion is the measure "1 in." The global figure of 2.5% means that 1 in 39 people in the world have been reported with COVID (and that only by official record keeping, not including any unreported and likely asymptomatic cases). For USA it is 1 in
9. For Spain and Columbia it is 1 in 11.
Netherlands tied Argentina at 10.9% of its population with COVID, but since it was added to the monitored list this month, it will not show up in charts for another month or two when there is a trend to illustrate.
Bolivia and Canada traded placed this month. The other three countries remain in the same order from June.
At the low end (Fig. 5B), the five lowest countries for cases by proportion of population are much closer to the global figure of 2.5% of population with confirmed cases of COVID-19 at the end of July, though Bolivia and Canada have been breaking away from the others since May.
Canada was moving upward more than the others until June when it was barely ahead of Bolivia (3.8% to 3.7%). By the end of July, Canada stayed at 3.8%, putting it below Bolivia, which increased to 4.0%. Ecuador has stayed in the middle of the five, always above the Global level. India, at the Global level to start, stayed below it until it stated a more rapid ascent in April, passing Mexico in May, and slowing enough to stay below the Global level in July. Mexico followed close to the Global level through March, then moved below it, ending at the lowest position the past three months.
These countries represent a considerable spread in size, from India, the second largest country, to Bolivia, ranked number 50 of the 215 countries tracked by worldometers. For Canada, its 3.8% level represents 1 in 27 people, for India it is 1 in 44 and for Mexico it is 1 in 46.
Because the size of countries makes the use of raw case numbers illusory, another measure I find helpful is the rate of change from month to month (Figure 5C).
For this chart, countries are selected based on the change over two-months (end of May to end of July for this report). Except for USA (which will appear every month), the mix has changed since the last report, but the tilt toward South America last month, with four countries, is moderated somewhat with the return of UK as it faced surging numbers even as it reopened. Iran also returns to this group. Not returning from last month are India and Chile.
The overall trend (red line)
is flattening as it drops. Interrupting the downward Global trend were a four-point increase in April and a 2-point bump in July. (A polynomial trend line flexes as adjacent data points go up and down, so the leading edge of newest dates can change the shape of the curve as new months are added). Note that some months appear "flatter" than others, with less variations between countries and the global (world) number.
Columbia has had monthly increases in excess of 10% for all but two months since November, the first month it was tracked. From a low of 7% change over the previous month in February and March, it climbed to a 25% change in June before settling down some to a 14% change in July.
UK is notable for its 22% increase of July over June—after five months below 10%—just as the delta variant wrecked havoc on what promised to be a grand reopening. UK also suffered a huge spike of 128% back in October during the fall 2020 surges that swept across Europe and North America.
Argentina and Bolivia both hit recent highs for monthly change in May and have moved downward since then (11% and 8% respectively for July over June), but still had high enough change over two-months to land on this chart.
USA started above the Global level from November through January, then has fallen further below it each month since, although July showed a bump from the low of 1% for June over May back up to 3% for July over June.
Because deaths as a percentage of population is such a small number, the "Deaths-per-Million" metric shown in Figure 6A provides a comparable measure.
Peru and Brazil stay at #1 and #2 from June. Columbia moves up two to #3. Argentina replaces Italy at #4 and Belgium moves down two to #5. That puts four countries in South America in the top-five for Deaths-per-Million in July.
As Figure 6A shows, Peru still soars over the others following a correction to its death data in June. July remains more than double #2 Brazil and ten times the Global level of 543.
The other South American countries have all been moving upward. In fact, Columbia and Argentina are so close, you can barely tell them apart on the chart. Belgium has been slowing down in monthly deaths-per-million, but remains in the top-five of the 26 countries I monitor.
Since this analysis focuses on 26 countries that have been in the top-20 of cases and deaths, there are 7 other countries with Deaths-per-Million between Peru and Brazil. The second place country at the end of July (as it was in June) was Hungary (population 9.6-million) with a Deaths-per-Million of 3,117. Ranked #4 was Czechia (population 10.7-million) at 2,830.
The other countries on the chart, including USA are all above the Global level, and (except for Peru) fairly close to each other—though in July they started to spread out again from their closest proximity in June.
As observed last month, the overall trend shown in Figure 6A is that USA, with aggressive testing and vaccination, has seen slowing COVID deaths since January, while Europe (represented by Belgium) was also beginning to slow deaths in recent months. Progress in Europe and North America, however, is being stymied by the delta variant and plateauing vaccination rates. Even without Peru's astonishing correction in June, South America (also represented by Brazil, Colombia and Argentina) remains a trouble spot for COVD deaths, more so statistically than Asia or Africa.
Among the lowest five for Deaths-per-Million, Russia and Iran traded places for #1 and #2 from June. The remaining three countries appear in the same order.
Though starting to escalate in March, India remained well below the global rate of 543 Deaths-per-Million at the end of July.
The other four countries are all above the Global level. Russia has been moving steadily upward and a further upward bend of its curve in July put it at the top of the lowest-five, just ahead of Iran, whose slowing rate helped put it in second place. Canada, similar to USA, has been slowing since January. Turkey, following an upward bump in May is moving back toward the Global level.
As with the comparable chart for Rate of Change for Cases (Figure 5C), countries for Rate of Change for Deaths (Figure 6C) are selected based on the change over two-months (end of May to end of July for this report) in reported COVID deaths. Except for USA (which will appear every month), the mix has changed since the last report, with Ecuador replacing India. Peru and Russia remaining at #1 and #5, while Columbia and Argentina switched the #3 and #4 positions.
The trend
in the rate of change in deaths (red line in Figure 6C) shows the delayed connection with cases. Monthly Global changes for deaths trended down through May, bumped up in April, then went back down in July. While the Peru increase of 177% in June was exceptional, it was not enough to bend the global trend significantly. (As mentioned with Rate of Change for Cases, the trend line is a polynomial that can change shape as new values are added at the most-recent end.)
An important comparison is individual countries each month against the global (World) level. As shown in Figure 6C, the Global level peaked at 24% in November, with Russia at or above the Global level from October through February. As the Global level dropped, so did the variance of change in the top-5 countries—until June that is, when Peru came in with the 177% increase and Columbia at 20% was nearly double the Global level. In July, even with the Global level of change from the previous month down to 7%, four of the top-five were still higher (Peru had dropped to 2%, but Ecuador was 48%, while Columbia, Argentina and Russia were all above 10%.
USA was higher than the Global level in December, January and February, then has been significantly lower since then, falling to a 1% change in July (over the number of deaths in June).
Mortality Rates (percentage of deaths against reported cases) have generally been slowly declining. This is not surprising as several factors came into play: the ratio of community spread versus outbreaks (nursing homes, similar confined settings) increased, knowledge about treatment increased, along with testing (which would expand the spread between reported cases and deaths), and vaccinations since January (though that should impact both cases and deaths).
The Global mortality rate has dropped from 2.6% in October to 2.1% in July. The median for the 26 countries monitored for this report was 2.9% at the end of July.
The top-5 in mortality rate (among the 26 countries monitored) all return from June, in the same order. Three of the five, not surprisingly, are in South America, along with nearby Mexico (all Central American countries are considered North America by continent). .
Peru was in fourth place and declining until June when its corrected death numbers drove Mortality to 9.4%, just ahead of Mexico at 9.3%. At the end of July, Peru eased back slightly, down to 9.3% and Mexico made a significant drop to 8.5%. Like Peru, Ecuador had been declining faster than the Global level, but then shot up nearly two full points to 6.5% at the end of July as deaths increased 47%.
Bolivia's Mortality Rate has been declining, slowing down since January, but still significantly above the Global level, at 3.8% for July. Italy has been slowly declining since a big drop early on and while above the Global level, the 2.9% at the end of July is close to the median of 2.7% for all 26 monitored countries. South Africa also came in at 2.9%, but it was just added to the monitored list this month, so it will take a month or more of additional data before it appears in trend charts.
Since these represent the best mortality rates, where low is good, the "rank" order is actually in reverse. Spain enters the list at #1, as France moves to #2 and Canada drops off the list. USA, India and Turkey return, in the same order as June.
All five countries have been below the Global Mortality Rate since April. Having come closer together over time, Spain, France, and USA all ended July with Morality Rates of 1.8%. USA has been below the Global level the entire time, Spain and France started higher than dropped below the Global level in April. India started at 1.5% then moved down to 1.1% in April before moving up again due to surging cases in some areas, ending July at 1.3%. Ironically, while television reports showed the human scale of suffering in India, the country is so large that the statistics make it look like not much was happening. Turkey, which was added to the monitored countries list in December, tracked in a pattern similar to but below India, ending July at 0.9%.
How real is the threat of death from COVID? That's where successful mitigation comes in. Worldwide, by the end of July, 1 in 39 people have been reported as having contracted COVID and 1 in 1,848 people have died. In USA, while the mortality rate is low, because the number of cases is so high, 1 in 530 have died through July 2021.
With low mortality, USA. should have been able to keep deaths much lower, but the extraordinarily high number of cases means more deaths. Without a better-than-global mortality rate, the USA death rate would be far higher. Compared to the 1918 pandemic, it could be ten times worse. The response of the health care system and availability of vaccines are part of keeping mortality down, but it's far too early to detail the cause for that positive piece of the COVID picture for USA. Even at the Global mortality rate of 2.2%, USA would have had 785-thousand deaths (for 35.7-million cases) by the end of July, instead of 629-thousand.
The same five countries remain on top in COVID testings, having been in the same order since October 2020.
USA remains ahead of other countries in reported COVID tests administered, at 530-million, 14% ahead of India, but that has narrowed from 56% in April. UK continues at the pace it set with an upturn in February, having moved past Russia in March. Russia and France continue on a steady upward path, roughly 60-million tests apart since January.
These are raw numbers, so it is important to recognize the size of the country. It is also the case that COVID tests can be administered multiple times to the same person, so it cannot be assumed that USA has tested almost all of its population of some 330-million. Some schools and organizations with in-person gatherings are testing as frequently as once a week or more for those who are not yet fully vaccinated. That's a lot of testing!
The same five countries return from June, in the same order.
There is tremendous disparity between countries in terms of testing. Except for Belgium, the other four countries (among the 26 monitored) with the lowest reported number of tests are in South and North America.
Belgium and Peru are climbing at a rate comparable to USA and India. Bolivia and Ecuador are not making much progress, and Mexico falls between the two groups.
As questions arise about equity of testing between countries, check the number of tests for countries of similar size:
Peru (15.5-million tests for its 32.9-million population) is well below Canada (39.5-million tests for its 37.7-million population). Ironically, even though they both fall in the bottom-5 by number of tests among the 26 monitored countries,, Bolivia, with a population of 11.7-million has tallied 2.1-million tests, while Belgium, with 11.6-million people, has reported seven times that number, with 17.2-mllion tests.
Tests per million adds another perspective. Fig. 7C shows the five countries with the highest tests per million. All five return from June, but France moves ahead of USA.
The UK, already the most aggressive in testing, increased its numeric lead each month since February, with a reported 3.58-million tests per million population in July, just over 3.5 tests per person. USA continues on a straight line trajectory, with a very slight dip the last two months, reaching 1.590-million tests per million, more than 1.5 tests per person, allowing France to edge ahead with 1.599. Belgium and Italy track close to each other, both increasing their climb above 1-million tests per million in June, with Belgium at nearly 1.5-million tests per million at the end of July and Italy approaching 1.3-million tests per million. .
Anything over 1,000 (or "x-million tests-per-million") represents more tests than people (1,000 on the chart actually means 1,000,000), but as mentioned above, that does not mean that everyone had been tested. Some people have been tested more than once, and some are being tested regularly or with increased frequency.
The same five countries have appeared as the lowest five in tests per million in the same order since May.
While still at the bottom of the 26 countries monitored for this report, Iran and Brazil have made the most substantial progress, Iran in a steady and slowly accelerating curve reaching 304-thousand tests-per-million, Brazil, in more halting fashion that may indicate inconsistencies in reporting, is not far behind at 257-thousand. Bolivia began to accelerate in March, so it is now nearly double Ecuador which has moved more slowly, but steadily, along with Mexico.
While improvement is evident in all five, the equivalent proportion of tests to population remains very low, from roughly 6% to 30%. This illustrates the arguments over inequity in resources among countries.
Early in the reporting on COVID, as the death rate climbed in the U.S., a great deal of attention was given to benchmarks, most notably as it approached 58,000, matching the number of American military deaths in the Vietnam War. At that time, I wrote the first article in this series, "About Those Numbers," in which I looked at ways of viewing the data, which at the time of that writing in May 2020 was still focused on worst-case models and familiar benchmarks, like Vietnam.
Figure 8 shows the number of US COVID cases and deaths against the top-10 causes of death reported by CDC for 2018 (the latest data available), along with several other mileposts from wars and the 1918 pandemic. Notice that for nearly nine months, the curve for deaths was increasing at a faster rate than cases. Then, starting in October 2020 the curve for cases took a decided turn upward, while deaths increased at a more moderate pace.
Most media sources reported that the U.S. would pass 500,000 deaths on February 22, 2021. The CDC reported 497,415 deaths on February 21, while worldometer.info was already reporting 512,068. As mentioned elsewhere in this report, worldometer tends to be about 3% ahead of CDC and Johns Hopkins, a matter of a few days difference in data collection. At any rate, it is clear that February 2020 saw the U.S. death toll from COVID pass the 500,000 mark, headed toward 600,000.
The February slowdown in cases was met by a barely perceptible slowdown in deaths, because as pointed out earlier, deaths trail changes in cases. That trend continued in March and bent downward even more in April. In fact, if the trajectory from November to February had continued (the upper edge of the blue cone), we would have seen close to 1-million deaths by the end of July. The total of 610-thousand at the end of May did pass the number who died from Cancer in 2018 (599,000).
The next benchmark is the #1 cause of death, heart disease, at 655-thousand. We're close, but the curve has been bending down, so how long will it take to get there? The delta variant could make the answer to that question more troublesome than the trend through July makes it appear.
The latest "Ensemble Forecast" from CDC suggests this for the immediate future (even though worldometers is already ahead of this prediction, it is within the 3% difference between the two sources):
...the number of newly reported COVID-19 deaths will likely increase over the next 4 weeks, with 2,300 to 9,100 new deaths likely reported in the week ending August 28, 2021. The national ensemble predicts that a total of 624,000 to 642,000 COVID-19 deaths will be reported by this date...
Note: As I've referenced in the notes for several charts, the data from worldometers.info tends to be ahead of CDC and Johns Hopkins by about 3%, because of reporting methodology and timing. I use it as a primary source because its main table is very easy to sort and provides the relevant data for these reports.
The 1918-19 Spanish Flu pandemic is estimated to have struck 500 million people, 26.3% of the world population of 1.9-billion at that time. By contrast, we're now at 2.5% of the global population. Deaths a century ago have been widely estimated at between 50- and 100-million worldwide, putting the global mortality rate somewhere between 10 and 20-percent. It has been estimated that 675,000 died in the U.S.
IF COVID-19 hit at the same rate as 1918, we would see about 2-billion cases worldwide by the time COVID-19 is over, with the global population now at 7.8-billion—four times what it was in 1918. There would be 200- to 400-million deaths. The U.S. is estimated to have had 27-million cases (one-quarter of the population of 108-million) and 675,000 deaths. Today, with a population of 330-million (a three-fold increase from 1918) this would mean more than 80-million cases, and 2- to 4-million deaths.
However, at the present rate of confirmed cases and mortality while the total number of global cases could approach 500 million or more—comparable to 1918 in number, that would be one-quarter of 1918 when taking population growth into account . .. and assuming the pandemic persists as long as the Spanish Flu, which went on in three waves over a two year period. At the present rate of increase (approximately 12-million cases per month) it would take another 26 months to reach 500-million, well into 2024.
If the total number of cases globally did approach 500-million, using the global mortality rate of 2.1% in July, there would be roughly 10.5-million deaths worldwide, tragic but far below the number reported for 1918 (50-million) with an even wider gap (200 million) when taking population growth into account.
In my last report, I indicated that with vaccination in progress and expected to be completed in the U.S. by the end of summer, the end of COVID-19 could come sooner. Like 1918, however, there are now complicating factors, such as the combination of the delta variant with a high number of unvaccianted (some by choice, as in USA and Europe, and many more in underdeveloped countries by inequitable access to vaccine). While a month ago we may have thought the end of the pandemic was in sight, it is still too early to make predictions on the duration and severity of the COVID-19 pandemic globally.
Despite the darkening forecast in the past month, the vast difference in scale between the Spanish Flu pandemic a century ago and COVID-19 even a year-and-a-half in, cannot be denied.. Key differences are the mitigation efforts, treatments available today (though still leaving the health care system overwhelmed in some areas during surges), and the availability of vaccines. In addition, in 1918 much of the world was focused on a brutal war among nations (World War I) rather than waging a war against the pandemic, which ran its course and was undoubtedly made much worse by the war, with trans-national troop movements, the close quarters of trench warfare, and large public gatherings supporting or protesting the war.
With remarkable speed (it usually takes years to develop vaccines), two COVID vaccines were granted emergency approval for use in the U.S. starting in January—the one by Pfizer requires super-cold storage, which limits its deployment. The other, by Moderna, requires cold storage similar to other vaccines. Both of these require two doses, which means that vaccine dosages available must be divided in two to determine the number of people covered. By my March report, Johnson & Johnson had been granted approval for a single-dose vaccine. The numbers in Figure 9 represent the status of all three vaccines as of June 30.
A person is considered "fully vaccinated" two weeks after the final (or only) vaccine dose; roughly five to six weeks total for Pfizer and Moderna and two weeks for Johnson & Johnson.
Early on (still somewhat evident in March 2021) there was much anxiety about vaccine availability and scheduling appointments. That is behind us and vaccinations are readily available to most of the population (there will always be some disparities and trouble spots, but in general most people who want the vaccine can get it). Now, the biggest concern is convincing the dubious, the deniers, and the resisters to get vaccinated and help the country reach herd immunity (about 70% of the population fully vaccinated). Some of the latest strategies to encourage vaccination is the provision of incentives, from free beer to lotteries for the fully vaccinated. The seriousness of the delta variant has scared others into taking action.
The CDC (Centers for Disease Control and Prevention) recently revised its mask guidelines. Some resistors claim this proves the science cannot be trust—but that position misunderstands the nature and role of science, which is to observe and constantly assess assumptions and recommendations based on a growing (and developing) body of data.
Some school districts that were prepared to make masking optional have announced that the school year will begin with masks in place, regardless of vaccination status. The district in which we live has had strong arguments on both sides (option versus mandate), but yesterday Illinois Governor Priztker announced a mask mandate for all private and public schools pre-school through grade 12.
Masking policy is actually evidence that the real battle now is vaccination. Will the rise of infections and the call for more masking be enough to spur more vaccinations? For some, perhaps, but whether low levels of vaccinations prolonged the pandemic (around the world) will be the subject of discussion and debate for years to come.
Vaccinating over 300 million people in the United States (much less a majority of the billions around the world) is a daunting task. It is a huge logistical challenge, from manufacture to distribution to administration. Yet, it remains amazing that any of this is possible this soon after the identification of the virus just over a year and a half ago.
There is a delicate balance between maintaining hope with the reality that this is a huge and complicated logistical operation that will take time, though that time is now measured in months instead of years. In addition, getting shots into the arms of the unvaccinated is proving to be a far bigger challenge that most officials assumed a few months ago.
As the richer countries with access to more resources make progress, the global situation is raising issues of equity and fairness within and between countries. Even as the U.S. and other countries launch large scale vaccine distribution to a needy world community, the immensity of the need is so great that a common refrain heard now is whether this aid is too little, too late. As COVID fades into a bad memory in countries able to provide help, will the sense of urgency remain high enough to produce the results needed to end this global pandemic?
Maintaining Perspective
The following section has appeared several months and I repeat it here because there is an even greater need now to evaluate what is happening: to understand the science and its dynamic nature, to put politics in perspective and to show empathy, to truly demonstrate that we are in this together and want to get out of it as soon as possible.
In the tendency to turn everything into a binary right-wrong or agree-disagree with science or government, we ignore the need to recognize the nature of science and the fact that we are dealing with very complicated issues. So, in addition to recommending excellent sources like the Centers for Disease Control and Prevention (CDC), it is also wise to consider multiple qualified sources.
While there has been much focus placed in trusting "the science," it is important to recognize that science itself changes over time based on research and available data. In the highly volatile political atmosphere we find ourselves in (not just in the U.S., but around the world), there is a danger of not allowing the experts to change their views as their own understanding expands, or of trying to silence voices of experts whose views are out of sync with "the science" as reported by the majority of media outlets.
In an earlier report, I mentioned the Greater Barrington Declaration, currently signed by nearly 58-thousand medical & public health scientists and medical practitioners, which states "As infectious disease epidemiologists and public health scientists we have grave concerns about the damaging physical and mental health impacts of the prevailing COVID-19 policies, and recommend an approach we call Focused Protection."
For a personal perspective from a scholar and practitioner who espouses an approach similar to the Focused Protection of the Greater Barrington Declaration, see comments by Scott W. Atlas, Robert Wesson Senior Fellow at the Hoover Institution at Stanford University, in an article "Science, Politics, and COVID: Will Truth Prevail?"
Several months ago on SeniorLifestyle I posted an article by Mallory Pickett of The New Yorker, "Sweden's Pandemic Experiment," which provides a fair evaluation of the very loose protocols adopted by Sweden, essentially a variation of the "Focused Protection" approach. The "jury is still out" on this one, so judge for yourself whether Sweden hit the mark any better than the area in which you live.
UPDATE ON SWEDEN: At the end of July Sweden reported 1.1-million cases of COVID, or 10.8% of its population of 10.2-million. There have been 14,620 deaths, for a mortality rate of 1.3%. Ranked 89 by population, Sweden was number 29 in cases and number 39 in deaths. That would put Sweden in a virtual tie with Argentina and USA at the top of my cases as a proportion of population, but the mortality rate would be tied with India, near the bottom among the 26 countries monitored for this report and well below the global rate of 2.1%.
How we evaluate the many approaches used to deal with COVID will determine how we prepare for and approach the next global event.
What is clear in July 2021 compared to July 2020 is that while debates rage about masking and other policies, the response to surges in countries with vaccine readily available is much more like the "focused protection" approach discussed by the sources cited here than the full lockdowns of a year ago.
My purpose in mentioning these sources is to recognize that there are multiple, sometimes dissenting, voices that should be part of the conversation. The purpose of these monthly reports remains first and foremost to present the numbers about COVID-19 in a manner that helps you understand how the pandemic is progressing and how the U.S. compares to the world—and how to gain more perspective than might be gathered from the news alone.
(Data from worldometers.info).
Rank | Country | Population | Share of World Population |
Density People per square km |
Urban Population |
Median Age |
WORLD | 7.82B | 100% | -- | -- | -- | |
Top 10 Countries by Population, plus Five Major Continents See lists of countries by continent |
||||||
- | ASIA | 4.64B | 59.3% | 150 | 51 countries | 32 |
1 | China | 1.44B | 18.4% | 153 | 61% | 38 |
2 | India | 1.38B | 17.7% | 454 | 35% | 28 |
- | AFRICA | 1.34BM | 17.1% | 45 | 59 countries | 20 |
- | EUROPE | 747.7M | 9.6% | 34 | 44 countries | 43 |
- | S AMERICA | 653.8M | 8.4% | 32 | 50 countries | 31 |
- | N AMERICA | 368.9M | 4.7% | 29 | 5 countries | 39 |
3 | USA | 331.5M | 4.3% | 36 | 83% | 38 |
4 | Indonesia** | 274.5M | 3.5% | 151 | 56% | 30 |
5 | Pakistan* | 220.9M | 2.8% | 287 | 35% | 23 |
6 | Brazil | 212.9M | 2.7% | 25 | 88% | 33 |
7 | Nigeria* | 206.1M | 2.6% | 226 | 52% | 18 |
8 | Bangladesh* | 165.2M | 2.1% | 1,265 | 39% | 28 |
9 | Russia | 145.9M | 1.9% | 9 | 74% | 40 |
10 | Mexico | 129.3M | 1.7% | 66 | 84% | 29 |
*these countries do not appear in the details because they have not yet reached a high enough threshold to be included **Indonesia was added to the monitored list in July 2021 Other Countries included in Analysis most have been in top 20 of cases or deaths |
||||||
Rank | Country | Population | Share of World Population |
Density People per square km |
Urban Population |
Median Age |
17 | Turkey | 84.3M | 1.1% | 110 | 76% | 32 |
18 | Iran | 83.9M | 1.1% | 52 | 76% | 32 |
19 | Germany | 83.8M | 1.1% | 240 | 76% | 46 |
21 | United Kingdom | 67.9M | 0.9% | 281 | 83% | 40 |
22 | France | 65.3M | 0.8% | 119 | 82% | 42 |
23 | Italy | 60.4M | 0.8% | 206 | 69% | 47 |
25 | South Africa* | 59.3M | 0.8% | 94 | 67% | 28 |
29 | Colombia | 50.9M | 0.7% | 46 | 80% | 31 |
30 | Spain | 46.8M | 0.6% | 94 | 80% | 45 |
32 | Argentina | 45.2M | 0.6% | 17 | 93% | 32 |
35 | Ukraine* | 43.7M | 0.6% | 75 | 69% | 41 |
39 | Poland* | 37.8M | 0.5% | 124 | 60% | 42 |
39 | Canada | 37.7M | 0.5% | 4 | 81% | 41 |
43 | Peru | 32.9M | 0.4% | 26 | 79% | 31 |
63 | Chile | 19.1M | 0.2% | 26 | 85% | 35 |
67 | Ecuador | 17.6M | 0.2% | 71 | 63% | 28 |
69 | Netherlands* | 17.1M | 0.2% | 508 | 92% | 43 |
80 | Bolivia | 11.7M | 0.1% | 11 | 69% | 26 |
81 | Belgium | 11.6M | 0.1% | 383 | 98% | 42 |
*Added to the monitored list in July 2021 |
From the worldometers.info website I track the following Categories:
Instead of reporting Cases per Million directly, I try to put raw numbers in the perspective of several key measures. These are a different way of expressing "per Million" statistics, but it seems easier to grasp.
My analysis covers countries that have appeared in the top-10 of the worldometers categories since September 2020. This includes most of the world's largest countries as well as some that are much smaller (see the chart in the previous section).
This article was also posted on SeniorLifestyle, which I edit
Search all articles by Stu Johnson
Stu Johnson is owner of Stuart Johnson & Associates, a communications consultancy in Wheaton, Illinois focused on "making information make sense."
• E-mail the author (moc.setaicossajs@uts*)* For web-based email, you may need to copy and paste the address yourself.
Posted: August 217, 2021 Accessed 3,974 times
Go to the list of most recent InfoMatters Blogs
Search InfoMatters (You can expand the search to the entire site)
Category: Information / Topics: History • Information • Statistics • Trends
by Stu Johnson
Posted: August 217, 2021
While only a blip at the end of July, the delta variant and high number of unvaccinated are putting a damper on the optimism felt a month ago…
Putting the COVID-19 pandemic in perspective (Number 13)
This series was spawned by my reaction to reporting early in the COVID-19 pandemic that focused on raw numbers. Big numbers are impressive, even frightening, but hard to comprehend. This series turned into a monthly summary setting the U.S. numbers in global perspective. This analysis is based on data from worldometers.info, which monitors 215 countries. From those, I focus on details for 26 countries since I started more detailed tracking in September 2020. (Indonesia, South Africa, Ukraine, Poland, and Netherlands were added to the list this month because they have appeared in the weekly report of most new activity and are now in the top-20 of cases and deaths).
A note on reliability of data: It should be noted that the statistics reported by worldometers and other sources are only as good as the integrity of the reporting system in each country. China's statistics have increased so little since I started tracking numbers that it is doubtful its numbers are accurate, so they are not included in charts that focus on countries. Other countries have also had gaps in reporting, or made adjustments. There is still enough information to make trends evident. That is why I tend to round some numbers and watch for changes over two or three months rather than focusing on a single month, as significant as that may turn out to be.
These reports will continue as long as the pandemic persists around the world.
Report Sections:
• July at-a-glance
• The Continental View • USA Compared Other Countries
• COVID Deaths Compared to the Leading Causes of Death in the U.S.
• U.S. COVID Cases versus Vaccinations
• Profile of Monitored Continents & Countries • Scope of This Report
This is where I have to suggest how important it is to get information from reliable and knowledgeable sources that know how to interpret the numbers. In an atmosphere wary of misinformation, news-by-anecdote from otherwise trusted sources can itself be a form of misinformation. As I go through the statistics each month, I am reminded often that the numbers do not always line up with the impressions from the news. With that caveat, let's dig into the numbers for July 2021.
While COVID-19 has been classified as a global pandemic, it is not distributed evenly around the world, though we began to see some shifts in patterns starting in April 2021.
Asia accounts for 59.3% of the world's population (Figure 2), but had only 31.2% of COVID cases at the end July (Figure 3A)—affecting a mere 1.3% of its population (COVID cases have been reported for 2.5% of world population). After shrinking in proportion of cases since October, Asia continues to increase, with each month since March surpassing the previous month. The trade-off is that Europe and North America have been going down in share each of those months. Meanwhile South America (with 8.4% of world population) bottomed out at around 15% of COVID cases back in January and has crept up to a share of 18% the last two months.
Where Asia and Africa combined represent about three-quarters (76.4%) of the world's 7.8-billion people, Europe, South America and North America still account for two-thirds of cases (65.3% - Figure 3A) and three quarters of COVID deaths (74.8% - Figure 4A). The shares for Europe and the Americas combined are slowly coming down from their highs (74.7% for cases and 80.8% for deaths in February).
Oceana's population is too small to be statistically significant, so it is not included in my analysis.
While far from the dramatic surges of past months, all continents except South America showed very slight increases in the rate of growth in cases in July, apparently related to the highly-transmissible delta variant. If that is the case, we should see an even greater impact in August.
Asia has the highest number of reported COVID cases, having surpassed both Europe and North America as it experienced a dramatic upturn in April and May followed by slower but continued growth in June and July. South America remains in fourth place and slowed down in rate of growth slightly in July. North America began to flatten its curve back in January (when vaccinations began), dropping away from Europe in number of cases, but showed a slight upward bend in its curve in July. Europe (slower to get vaccinations started) showed a slight increase in pace in March and April before decelerating in May and June, but it too is showing an apparent delta-related increase in rate in July. Africa, second by population, remains a loner, with an almost flat curve compared to the other four large continents, but it also bent upward slightly in July.
The raw numbers of Figure 3B can be deceptive. Figure 3C gives a more realistic picture of impact by translating raw case numbers to percentage of population.
After a sharp increase in January, North America has slowed down in monthly growth, with a very slight downturn in June followed by a similar up-tick in July. That still leaves North America far above the other continents in proportion of infections, with one in ten (11.5%) reported as having contracted COVID—the proportion is increasing about half a percent each month. Even if it goes flat, with virtually no further increases in cases, the likelihood of any other continent matching it in proportion of cases is extremely low.
Europe has slowed slightly below the North America and global pace. Like North America, Europe went down slightly in June, then back up in July. South America remains above the global level, increasing its distance from the global level by a small amount each month, thus bending its curve upward.
Asia and Africa remain below Global levels, though Asia showed an increase in its pace for April and May before slowing slightly in June and July. Despite the reports of COVID surges in India and other parts of Asia similar to what Italy experienced at the beginning of the pandemic, the sheer size of the continent means it can absorb a lot of growth in cases before it causes a dramatic bend in the curve (as we did see in North America in January). However, individual countries in Asia dominate my watch list (see July-at-a-Glance above),
Reports of surges in Africa barely are reflected with just a minor upward movement in its curve in July. Africa reported 6.7-millon cases by the end of July, 0.5% of its population of 1.3-billion—up a million cases, but only increasing one-tenth of a percent in share of its population.
The proportion of deaths between continents is even more distorted than that of cases. Figure 4A shows the steady increase in the proportion of deaths in Europe and South America through April. North America peaked in February. July ended with Asia and Africa (76.4% of world population) accounting for only 25.2% of COVID deaths.
The growing crisis in India showed up in May deaths, with Asia hitting 19.2% and a slower increase to 21.2% in July, pushing it to a new high since my observation of continental data began in November.
Africa was down slightly in May then up to 4.0% by July, still barely below its most recent high of 4.1% in January and February . . . and far below its 17.1% share of world population.
Europe went down in proportion of deaths for the third consecutive month in July, North America for the fifth consecutive month.
South America slowed in proportion of deaths through February, then began increasing each month, gaining 7.5 points by the end of July. Peru made corrections to its death count in June that represent a major part of the jump in South American deaths in June, with a much smaller increase in July.
The overall distribution for Deaths through July shows that while the trajectory lags behind cases and has progressed at a steadier rate, it does reflect the overall changes in Cases by continent.
By numbers alone, we can say that COVID deaths reported for Africa continue at a very slow and steady pace. North America continues to slow down, remaining under the 1-millon mark and falling to third place last month. Europe showed a slight downward bend of its curve in June that continued into July, but remains in first place. Asia and South America began climbing in March. Asia slowed slightly in June and July, keeping it in fourth place, but on a trajectory to pass North America in another month or two. South America climbed at an even higher rate, surpassing North America in June to move into second place by number of deaths, then crossing the 1-million mark in July, poised to overtake Europe if the trend line continues.
Raw numbers are virtually meaningless without relating them to the size of a given country. So, we start with a look at the top-5 countries by proportion of population with reported COVID cases. Argentina moves into #1 this month, passing now #2 USA by a very small margin. Belgium stays at #3, Spain and Columbia, at #4 and #5 respectively, are virtually tied.
The top-5 by proportion of population with COVID cases, have progressed at a rate higher than the Global level, but have stayed within a fairly narrow band that has been tightening since March.
USA reached #1 by December, continuing to climb at a very fast pace until starting to slow in January as vaccine roll-outs began, bringing it closer to the others by the end of June and landing in #2 in July. The others meandered a bit, began to appear as if they were spreading out by March, but drew closer together since then. While USA had a roller-coaster rise in the early months, Argentina's rise to #1 accelerated more slowly but steadily in the past four months. All are significantly higher than the global level of 2.5% cases as proportion of a country's own population.
Another way to look at population proportion is the measure "1 in." The global figure of 2.5% means that 1 in 39 people in the world have been reported with COVID (and that only by official record keeping, not including any unreported and likely asymptomatic cases). For USA it is 1 in
9. For Spain and Columbia it is 1 in 11.
Netherlands tied Argentina at 10.9% of its population with COVID, but since it was added to the monitored list this month, it will not show up in charts for another month or two when there is a trend to illustrate.
Bolivia and Canada traded placed this month. The other three countries remain in the same order from June.
At the low end (Fig. 5B), the five lowest countries for cases by proportion of population are much closer to the global figure of 2.5% of population with confirmed cases of COVID-19 at the end of July, though Bolivia and Canada have been breaking away from the others since May.
Canada was moving upward more than the others until June when it was barely ahead of Bolivia (3.8% to 3.7%). By the end of July, Canada stayed at 3.8%, putting it below Bolivia, which increased to 4.0%. Ecuador has stayed in the middle of the five, always above the Global level. India, at the Global level to start, stayed below it until it stated a more rapid ascent in April, passing Mexico in May, and slowing enough to stay below the Global level in July. Mexico followed close to the Global level through March, then moved below it, ending at the lowest position the past three months.
These countries represent a considerable spread in size, from India, the second largest country, to Bolivia, ranked number 50 of the 215 countries tracked by worldometers. For Canada, its 3.8% level represents 1 in 27 people, for India it is 1 in 44 and for Mexico it is 1 in 46.
Because the size of countries makes the use of raw case numbers illusory, another measure I find helpful is the rate of change from month to month (Figure 5C).
For this chart, countries are selected based on the change over two-months (end of May to end of July for this report). Except for USA (which will appear every month), the mix has changed since the last report, but the tilt toward South America last month, with four countries, is moderated somewhat with the return of UK as it faced surging numbers even as it reopened. Iran also returns to this group. Not returning from last month are India and Chile.
The overall trend (red line)
is flattening as it drops. Interrupting the downward Global trend were a four-point increase in April and a 2-point bump in July. (A polynomial trend line flexes as adjacent data points go up and down, so the leading edge of newest dates can change the shape of the curve as new months are added). Note that some months appear "flatter" than others, with less variations between countries and the global (world) number.
Columbia has had monthly increases in excess of 10% for all but two months since November, the first month it was tracked. From a low of 7% change over the previous month in February and March, it climbed to a 25% change in June before settling down some to a 14% change in July.
UK is notable for its 22% increase of July over June—after five months below 10%—just as the delta variant wrecked havoc on what promised to be a grand reopening. UK also suffered a huge spike of 128% back in October during the fall 2020 surges that swept across Europe and North America.
Argentina and Bolivia both hit recent highs for monthly change in May and have moved downward since then (11% and 8% respectively for July over June), but still had high enough change over two-months to land on this chart.
USA started above the Global level from November through January, then has fallen further below it each month since, although July showed a bump from the low of 1% for June over May back up to 3% for July over June.
Because deaths as a percentage of population is such a small number, the "Deaths-per-Million" metric shown in Figure 6A provides a comparable measure.
Peru and Brazil stay at #1 and #2 from June. Columbia moves up two to #3. Argentina replaces Italy at #4 and Belgium moves down two to #5. That puts four countries in South America in the top-five for Deaths-per-Million in July.
As Figure 6A shows, Peru still soars over the others following a correction to its death data in June. July remains more than double #2 Brazil and ten times the Global level of 543.
The other South American countries have all been moving upward. In fact, Columbia and Argentina are so close, you can barely tell them apart on the chart. Belgium has been slowing down in monthly deaths-per-million, but remains in the top-five of the 26 countries I monitor.
Since this analysis focuses on 26 countries that have been in the top-20 of cases and deaths, there are 7 other countries with Deaths-per-Million between Peru and Brazil. The second place country at the end of July (as it was in June) was Hungary (population 9.6-million) with a Deaths-per-Million of 3,117. Ranked #4 was Czechia (population 10.7-million) at 2,830.
The other countries on the chart, including USA are all above the Global level, and (except for Peru) fairly close to each other—though in July they started to spread out again from their closest proximity in June.
As observed last month, the overall trend shown in Figure 6A is that USA, with aggressive testing and vaccination, has seen slowing COVID deaths since January, while Europe (represented by Belgium) was also beginning to slow deaths in recent months. Progress in Europe and North America, however, is being stymied by the delta variant and plateauing vaccination rates. Even without Peru's astonishing correction in June, South America (also represented by Brazil, Colombia and Argentina) remains a trouble spot for COVD deaths, more so statistically than Asia or Africa.
Among the lowest five for Deaths-per-Million, Russia and Iran traded places for #1 and #2 from June. The remaining three countries appear in the same order.
Though starting to escalate in March, India remained well below the global rate of 543 Deaths-per-Million at the end of July.
The other four countries are all above the Global level. Russia has been moving steadily upward and a further upward bend of its curve in July put it at the top of the lowest-five, just ahead of Iran, whose slowing rate helped put it in second place. Canada, similar to USA, has been slowing since January. Turkey, following an upward bump in May is moving back toward the Global level.
As with the comparable chart for Rate of Change for Cases (Figure 5C), countries for Rate of Change for Deaths (Figure 6C) are selected based on the change over two-months (end of May to end of July for this report) in reported COVID deaths. Except for USA (which will appear every month), the mix has changed since the last report, with Ecuador replacing India. Peru and Russia remaining at #1 and #5, while Columbia and Argentina switched the #3 and #4 positions.
The trend
in the rate of change in deaths (red line in Figure 6C) shows the delayed connection with cases. Monthly Global changes for deaths trended down through May, bumped up in April, then went back down in July. While the Peru increase of 177% in June was exceptional, it was not enough to bend the global trend significantly. (As mentioned with Rate of Change for Cases, the trend line is a polynomial that can change shape as new values are added at the most-recent end.)
An important comparison is individual countries each month against the global (World) level. As shown in Figure 6C, the Global level peaked at 24% in November, with Russia at or above the Global level from October through February. As the Global level dropped, so did the variance of change in the top-5 countries—until June that is, when Peru came in with the 177% increase and Columbia at 20% was nearly double the Global level. In July, even with the Global level of change from the previous month down to 7%, four of the top-five were still higher (Peru had dropped to 2%, but Ecuador was 48%, while Columbia, Argentina and Russia were all above 10%.
USA was higher than the Global level in December, January and February, then has been significantly lower since then, falling to a 1% change in July (over the number of deaths in June).
Mortality Rates (percentage of deaths against reported cases) have generally been slowly declining. This is not surprising as several factors came into play: the ratio of community spread versus outbreaks (nursing homes, similar confined settings) increased, knowledge about treatment increased, along with testing (which would expand the spread between reported cases and deaths), and vaccinations since January (though that should impact both cases and deaths).
The Global mortality rate has dropped from 2.6% in October to 2.1% in July. The median for the 26 countries monitored for this report was 2.9% at the end of July.
The top-5 in mortality rate (among the 26 countries monitored) all return from June, in the same order. Three of the five, not surprisingly, are in South America, along with nearby Mexico (all Central American countries are considered North America by continent). .
Peru was in fourth place and declining until June when its corrected death numbers drove Mortality to 9.4%, just ahead of Mexico at 9.3%. At the end of July, Peru eased back slightly, down to 9.3% and Mexico made a significant drop to 8.5%. Like Peru, Ecuador had been declining faster than the Global level, but then shot up nearly two full points to 6.5% at the end of July as deaths increased 47%.
Bolivia's Mortality Rate has been declining, slowing down since January, but still significantly above the Global level, at 3.8% for July. Italy has been slowly declining since a big drop early on and while above the Global level, the 2.9% at the end of July is close to the median of 2.7% for all 26 monitored countries. South Africa also came in at 2.9%, but it was just added to the monitored list this month, so it will take a month or more of additional data before it appears in trend charts.
Since these represent the best mortality rates, where low is good, the "rank" order is actually in reverse. Spain enters the list at #1, as France moves to #2 and Canada drops off the list. USA, India and Turkey return, in the same order as June.
All five countries have been below the Global Mortality Rate since April. Having come closer together over time, Spain, France, and USA all ended July with Morality Rates of 1.8%. USA has been below the Global level the entire time, Spain and France started higher than dropped below the Global level in April. India started at 1.5% then moved down to 1.1% in April before moving up again due to surging cases in some areas, ending July at 1.3%. Ironically, while television reports showed the human scale of suffering in India, the country is so large that the statistics make it look like not much was happening. Turkey, which was added to the monitored countries list in December, tracked in a pattern similar to but below India, ending July at 0.9%.
How real is the threat of death from COVID? That's where successful mitigation comes in. Worldwide, by the end of July, 1 in 39 people have been reported as having contracted COVID and 1 in 1,848 people have died. In USA, while the mortality rate is low, because the number of cases is so high, 1 in 530 have died through July 2021.
With low mortality, USA. should have been able to keep deaths much lower, but the extraordinarily high number of cases means more deaths. Without a better-than-global mortality rate, the USA death rate would be far higher. Compared to the 1918 pandemic, it could be ten times worse. The response of the health care system and availability of vaccines are part of keeping mortality down, but it's far too early to detail the cause for that positive piece of the COVID picture for USA. Even at the Global mortality rate of 2.2%, USA would have had 785-thousand deaths (for 35.7-million cases) by the end of July, instead of 629-thousand.
The same five countries remain on top in COVID testings, having been in the same order since October 2020.
USA remains ahead of other countries in reported COVID tests administered, at 530-million, 14% ahead of India, but that has narrowed from 56% in April. UK continues at the pace it set with an upturn in February, having moved past Russia in March. Russia and France continue on a steady upward path, roughly 60-million tests apart since January.
These are raw numbers, so it is important to recognize the size of the country. It is also the case that COVID tests can be administered multiple times to the same person, so it cannot be assumed that USA has tested almost all of its population of some 330-million. Some schools and organizations with in-person gatherings are testing as frequently as once a week or more for those who are not yet fully vaccinated. That's a lot of testing!
The same five countries return from June, in the same order.
There is tremendous disparity between countries in terms of testing. Except for Belgium, the other four countries (among the 26 monitored) with the lowest reported number of tests are in South and North America.
Belgium and Peru are climbing at a rate comparable to USA and India. Bolivia and Ecuador are not making much progress, and Mexico falls between the two groups.
As questions arise about equity of testing between countries, check the number of tests for countries of similar size:
Peru (15.5-million tests for its 32.9-million population) is well below Canada (39.5-million tests for its 37.7-million population). Ironically, even though they both fall in the bottom-5 by number of tests among the 26 monitored countries,, Bolivia, with a population of 11.7-million has tallied 2.1-million tests, while Belgium, with 11.6-million people, has reported seven times that number, with 17.2-mllion tests.
Tests per million adds another perspective. Fig. 7C shows the five countries with the highest tests per million. All five return from June, but France moves ahead of USA.
The UK, already the most aggressive in testing, increased its numeric lead each month since February, with a reported 3.58-million tests per million population in July, just over 3.5 tests per person. USA continues on a straight line trajectory, with a very slight dip the last two months, reaching 1.590-million tests per million, more than 1.5 tests per person, allowing France to edge ahead with 1.599. Belgium and Italy track close to each other, both increasing their climb above 1-million tests per million in June, with Belgium at nearly 1.5-million tests per million at the end of July and Italy approaching 1.3-million tests per million. .
Anything over 1,000 (or "x-million tests-per-million") represents more tests than people (1,000 on the chart actually means 1,000,000), but as mentioned above, that does not mean that everyone had been tested. Some people have been tested more than once, and some are being tested regularly or with increased frequency.
The same five countries have appeared as the lowest five in tests per million in the same order since May.
While still at the bottom of the 26 countries monitored for this report, Iran and Brazil have made the most substantial progress, Iran in a steady and slowly accelerating curve reaching 304-thousand tests-per-million, Brazil, in more halting fashion that may indicate inconsistencies in reporting, is not far behind at 257-thousand. Bolivia began to accelerate in March, so it is now nearly double Ecuador which has moved more slowly, but steadily, along with Mexico.
While improvement is evident in all five, the equivalent proportion of tests to population remains very low, from roughly 6% to 30%. This illustrates the arguments over inequity in resources among countries.
Early in the reporting on COVID, as the death rate climbed in the U.S., a great deal of attention was given to benchmarks, most notably as it approached 58,000, matching the number of American military deaths in the Vietnam War. At that time, I wrote the first article in this series, "About Those Numbers," in which I looked at ways of viewing the data, which at the time of that writing in May 2020 was still focused on worst-case models and familiar benchmarks, like Vietnam.
Figure 8 shows the number of US COVID cases and deaths against the top-10 causes of death reported by CDC for 2018 (the latest data available), along with several other mileposts from wars and the 1918 pandemic. Notice that for nearly nine months, the curve for deaths was increasing at a faster rate than cases. Then, starting in October 2020 the curve for cases took a decided turn upward, while deaths increased at a more moderate pace.
Most media sources reported that the U.S. would pass 500,000 deaths on February 22, 2021. The CDC reported 497,415 deaths on February 21, while worldometer.info was already reporting 512,068. As mentioned elsewhere in this report, worldometer tends to be about 3% ahead of CDC and Johns Hopkins, a matter of a few days difference in data collection. At any rate, it is clear that February 2020 saw the U.S. death toll from COVID pass the 500,000 mark, headed toward 600,000.
The February slowdown in cases was met by a barely perceptible slowdown in deaths, because as pointed out earlier, deaths trail changes in cases. That trend continued in March and bent downward even more in April. In fact, if the trajectory from November to February had continued (the upper edge of the blue cone), we would have seen close to 1-million deaths by the end of July. The total of 610-thousand at the end of May did pass the number who died from Cancer in 2018 (599,000).
The next benchmark is the #1 cause of death, heart disease, at 655-thousand. We're close, but the curve has been bending down, so how long will it take to get there? The delta variant could make the answer to that question more troublesome than the trend through July makes it appear.
The latest "Ensemble Forecast" from CDC suggests this for the immediate future (even though worldometers is already ahead of this prediction, it is within the 3% difference between the two sources):
...the number of newly reported COVID-19 deaths will likely increase over the next 4 weeks, with 2,300 to 9,100 new deaths likely reported in the week ending August 28, 2021. The national ensemble predicts that a total of 624,000 to 642,000 COVID-19 deaths will be reported by this date...
Note: As I've referenced in the notes for several charts, the data from worldometers.info tends to be ahead of CDC and Johns Hopkins by about 3%, because of reporting methodology and timing. I use it as a primary source because its main table is very easy to sort and provides the relevant data for these reports.
The 1918-19 Spanish Flu pandemic is estimated to have struck 500 million people, 26.3% of the world population of 1.9-billion at that time. By contrast, we're now at 2.5% of the global population. Deaths a century ago have been widely estimated at between 50- and 100-million worldwide, putting the global mortality rate somewhere between 10 and 20-percent. It has been estimated that 675,000 died in the U.S.
IF COVID-19 hit at the same rate as 1918, we would see about 2-billion cases worldwide by the time COVID-19 is over, with the global population now at 7.8-billion—four times what it was in 1918. There would be 200- to 400-million deaths. The U.S. is estimated to have had 27-million cases (one-quarter of the population of 108-million) and 675,000 deaths. Today, with a population of 330-million (a three-fold increase from 1918) this would mean more than 80-million cases, and 2- to 4-million deaths.
However, at the present rate of confirmed cases and mortality while the total number of global cases could approach 500 million or more—comparable to 1918 in number, that would be one-quarter of 1918 when taking population growth into account . .. and assuming the pandemic persists as long as the Spanish Flu, which went on in three waves over a two year period. At the present rate of increase (approximately 12-million cases per month) it would take another 26 months to reach 500-million, well into 2024.
If the total number of cases globally did approach 500-million, using the global mortality rate of 2.1% in July, there would be roughly 10.5-million deaths worldwide, tragic but far below the number reported for 1918 (50-million) with an even wider gap (200 million) when taking population growth into account.
In my last report, I indicated that with vaccination in progress and expected to be completed in the U.S. by the end of summer, the end of COVID-19 could come sooner. Like 1918, however, there are now complicating factors, such as the combination of the delta variant with a high number of unvaccianted (some by choice, as in USA and Europe, and many more in underdeveloped countries by inequitable access to vaccine). While a month ago we may have thought the end of the pandemic was in sight, it is still too early to make predictions on the duration and severity of the COVID-19 pandemic globally.
Despite the darkening forecast in the past month, the vast difference in scale between the Spanish Flu pandemic a century ago and COVID-19 even a year-and-a-half in, cannot be denied.. Key differences are the mitigation efforts, treatments available today (though still leaving the health care system overwhelmed in some areas during surges), and the availability of vaccines. In addition, in 1918 much of the world was focused on a brutal war among nations (World War I) rather than waging a war against the pandemic, which ran its course and was undoubtedly made much worse by the war, with trans-national troop movements, the close quarters of trench warfare, and large public gatherings supporting or protesting the war.
With remarkable speed (it usually takes years to develop vaccines), two COVID vaccines were granted emergency approval for use in the U.S. starting in January—the one by Pfizer requires super-cold storage, which limits its deployment. The other, by Moderna, requires cold storage similar to other vaccines. Both of these require two doses, which means that vaccine dosages available must be divided in two to determine the number of people covered. By my March report, Johnson & Johnson had been granted approval for a single-dose vaccine. The numbers in Figure 9 represent the status of all three vaccines as of June 30.
A person is considered "fully vaccinated" two weeks after the final (or only) vaccine dose; roughly five to six weeks total for Pfizer and Moderna and two weeks for Johnson & Johnson.
Early on (still somewhat evident in March 2021) there was much anxiety about vaccine availability and scheduling appointments. That is behind us and vaccinations are readily available to most of the population (there will always be some disparities and trouble spots, but in general most people who want the vaccine can get it). Now, the biggest concern is convincing the dubious, the deniers, and the resisters to get vaccinated and help the country reach herd immunity (about 70% of the population fully vaccinated). Some of the latest strategies to encourage vaccination is the provision of incentives, from free beer to lotteries for the fully vaccinated. The seriousness of the delta variant has scared others into taking action.
The CDC (Centers for Disease Control and Prevention) recently revised its mask guidelines. Some resistors claim this proves the science cannot be trust—but that position misunderstands the nature and role of science, which is to observe and constantly assess assumptions and recommendations based on a growing (and developing) body of data.
Some school districts that were prepared to make masking optional have announced that the school year will begin with masks in place, regardless of vaccination status. The district in which we live has had strong arguments on both sides (option versus mandate), but yesterday Illinois Governor Priztker announced a mask mandate for all private and public schools pre-school through grade 12.
Masking policy is actually evidence that the real battle now is vaccination. Will the rise of infections and the call for more masking be enough to spur more vaccinations? For some, perhaps, but whether low levels of vaccinations prolonged the pandemic (around the world) will be the subject of discussion and debate for years to come.
Vaccinating over 300 million people in the United States (much less a majority of the billions around the world) is a daunting task. It is a huge logistical challenge, from manufacture to distribution to administration. Yet, it remains amazing that any of this is possible this soon after the identification of the virus just over a year and a half ago.
There is a delicate balance between maintaining hope with the reality that this is a huge and complicated logistical operation that will take time, though that time is now measured in months instead of years. In addition, getting shots into the arms of the unvaccinated is proving to be a far bigger challenge that most officials assumed a few months ago.
As the richer countries with access to more resources make progress, the global situation is raising issues of equity and fairness within and between countries. Even as the U.S. and other countries launch large scale vaccine distribution to a needy world community, the immensity of the need is so great that a common refrain heard now is whether this aid is too little, too late. As COVID fades into a bad memory in countries able to provide help, will the sense of urgency remain high enough to produce the results needed to end this global pandemic?
Maintaining Perspective
The following section has appeared several months and I repeat it here because there is an even greater need now to evaluate what is happening: to understand the science and its dynamic nature, to put politics in perspective and to show empathy, to truly demonstrate that we are in this together and want to get out of it as soon as possible.
In the tendency to turn everything into a binary right-wrong or agree-disagree with science or government, we ignore the need to recognize the nature of science and the fact that we are dealing with very complicated issues. So, in addition to recommending excellent sources like the Centers for Disease Control and Prevention (CDC), it is also wise to consider multiple qualified sources.
While there has been much focus placed in trusting "the science," it is important to recognize that science itself changes over time based on research and available data. In the highly volatile political atmosphere we find ourselves in (not just in the U.S., but around the world), there is a danger of not allowing the experts to change their views as their own understanding expands, or of trying to silence voices of experts whose views are out of sync with "the science" as reported by the majority of media outlets.
In an earlier report, I mentioned the Greater Barrington Declaration, currently signed by nearly 58-thousand medical & public health scientists and medical practitioners, which states "As infectious disease epidemiologists and public health scientists we have grave concerns about the damaging physical and mental health impacts of the prevailing COVID-19 policies, and recommend an approach we call Focused Protection."
For a personal perspective from a scholar and practitioner who espouses an approach similar to the Focused Protection of the Greater Barrington Declaration, see comments by Scott W. Atlas, Robert Wesson Senior Fellow at the Hoover Institution at Stanford University, in an article "Science, Politics, and COVID: Will Truth Prevail?"
Several months ago on SeniorLifestyle I posted an article by Mallory Pickett of The New Yorker, "Sweden's Pandemic Experiment," which provides a fair evaluation of the very loose protocols adopted by Sweden, essentially a variation of the "Focused Protection" approach. The "jury is still out" on this one, so judge for yourself whether Sweden hit the mark any better than the area in which you live.
UPDATE ON SWEDEN: At the end of July Sweden reported 1.1-million cases of COVID, or 10.8% of its population of 10.2-million. There have been 14,620 deaths, for a mortality rate of 1.3%. Ranked 89 by population, Sweden was number 29 in cases and number 39 in deaths. That would put Sweden in a virtual tie with Argentina and USA at the top of my cases as a proportion of population, but the mortality rate would be tied with India, near the bottom among the 26 countries monitored for this report and well below the global rate of 2.1%.
How we evaluate the many approaches used to deal with COVID will determine how we prepare for and approach the next global event.
What is clear in July 2021 compared to July 2020 is that while debates rage about masking and other policies, the response to surges in countries with vaccine readily available is much more like the "focused protection" approach discussed by the sources cited here than the full lockdowns of a year ago.
My purpose in mentioning these sources is to recognize that there are multiple, sometimes dissenting, voices that should be part of the conversation. The purpose of these monthly reports remains first and foremost to present the numbers about COVID-19 in a manner that helps you understand how the pandemic is progressing and how the U.S. compares to the world—and how to gain more perspective than might be gathered from the news alone.
(Data from worldometers.info).
Rank | Country | Population | Share of World Population |
Density People per square km |
Urban Population |
Median Age |
WORLD | 7.82B | 100% | -- | -- | -- | |
Top 10 Countries by Population, plus Five Major Continents See lists of countries by continent |
||||||
- | ASIA | 4.64B | 59.3% | 150 | 51 countries | 32 |
1 | China | 1.44B | 18.4% | 153 | 61% | 38 |
2 | India | 1.38B | 17.7% | 454 | 35% | 28 |
- | AFRICA | 1.34BM | 17.1% | 45 | 59 countries | 20 |
- | EUROPE | 747.7M | 9.6% | 34 | 44 countries | 43 |
- | S AMERICA | 653.8M | 8.4% | 32 | 50 countries | 31 |
- | N AMERICA | 368.9M | 4.7% | 29 | 5 countries | 39 |
3 | USA | 331.5M | 4.3% | 36 | 83% | 38 |
4 | Indonesia** | 274.5M | 3.5% | 151 | 56% | 30 |
5 | Pakistan* | 220.9M | 2.8% | 287 | 35% | 23 |
6 | Brazil | 212.9M | 2.7% | 25 | 88% | 33 |
7 | Nigeria* | 206.1M | 2.6% | 226 | 52% | 18 |
8 | Bangladesh* | 165.2M | 2.1% | 1,265 | 39% | 28 |
9 | Russia | 145.9M | 1.9% | 9 | 74% | 40 |
10 | Mexico | 129.3M | 1.7% | 66 | 84% | 29 |
*these countries do not appear in the details because they have not yet reached a high enough threshold to be included **Indonesia was added to the monitored list in July 2021 Other Countries included in Analysis most have been in top 20 of cases or deaths |
||||||
Rank | Country | Population | Share of World Population |
Density People per square km |
Urban Population |
Median Age |
17 | Turkey | 84.3M | 1.1% | 110 | 76% | 32 |
18 | Iran | 83.9M | 1.1% | 52 | 76% | 32 |
19 | Germany | 83.8M | 1.1% | 240 | 76% | 46 |
21 | United Kingdom | 67.9M | 0.9% | 281 | 83% | 40 |
22 | France | 65.3M | 0.8% | 119 | 82% | 42 |
23 | Italy | 60.4M | 0.8% | 206 | 69% | 47 |
25 | South Africa* | 59.3M | 0.8% | 94 | 67% | 28 |
29 | Colombia | 50.9M | 0.7% | 46 | 80% | 31 |
30 | Spain | 46.8M | 0.6% | 94 | 80% | 45 |
32 | Argentina | 45.2M | 0.6% | 17 | 93% | 32 |
35 | Ukraine* | 43.7M | 0.6% | 75 | 69% | 41 |
39 | Poland* | 37.8M | 0.5% | 124 | 60% | 42 |
39 | Canada | 37.7M | 0.5% | 4 | 81% | 41 |
43 | Peru | 32.9M | 0.4% | 26 | 79% | 31 |
63 | Chile | 19.1M | 0.2% | 26 | 85% | 35 |
67 | Ecuador | 17.6M | 0.2% | 71 | 63% | 28 |
69 | Netherlands* | 17.1M | 0.2% | 508 | 92% | 43 |
80 | Bolivia | 11.7M | 0.1% | 11 | 69% | 26 |
81 | Belgium | 11.6M | 0.1% | 383 | 98% | 42 |
*Added to the monitored list in July 2021 |
From the worldometers.info website I track the following Categories:
Instead of reporting Cases per Million directly, I try to put raw numbers in the perspective of several key measures. These are a different way of expressing "per Million" statistics, but it seems easier to grasp.
My analysis covers countries that have appeared in the top-10 of the worldometers categories since September 2020. This includes most of the world's largest countries as well as some that are much smaller (see the chart in the previous section).
This article was also posted on SeniorLifestyle, which I edit
Search all articles by Stu Johnson
Stu Johnson is owner of Stuart Johnson & Associates, a communications consultancy in Wheaton, Illinois focused on "making information make sense."
• E-mail the author (moc.setaicossajs@uts*)* For web-based email, you may need to copy and paste the address yourself.
Posted: August 217, 2021 Accessed 3,975 times
Go to the list of most recent InfoMatters Blogs
Search InfoMatters (You can expand the search to the entire site)