A year in numbers: Cambodia’s hopeful Covid-19 statistics

Phnom Penh-based data specialist David Benaim has spent recent months trawling through previously scattered information on Covid-19 in Cambodia. A year since the Kingdom reported its first case, his research provides a surprisingly positive public health outlook

David Benaim
January 29, 2021
A year in numbers: Cambodia’s hopeful Covid-19 statistics
A person has their temperature checked at a shop in Phnom Penh on 22 December. EPA-EFE/Kith Serey

Ranking lowest in the world in deaths per million and third lowest in cases per population, Cambodia’s impressive Covid-19 statistics, questioned by some, become even more striking if we apply sensible filters to the raw data.

From research spanning recent months Cambodia-based data specialist David Benaim walks us through his key findings and methodology having analysed data in ways not reported before in the Kingdom.

But before diving into the data, some definition of terms is necessary to separate the different categories of infections reflected in the statistics.

Community transmission is typically an in-country transmission where the source of the first carrier is not known – this differs from local transmission, when the source is known. An imported case is infected abroad, but tested positive in Cambodia – these make up the large majority of those seen in the Kingdom.

Community transmission is vital, as it represents local infections that are not as easily accounted for, have less clear points of origin and are thus more likely to spread unmonitored by health officials. Community transmission also represents the deeper intrusion of the virus into Cambodia from its borders of ports of entry.

This week one year ago saw Cambodia import its first Covid-19 case, with only three threats of in-country transmissions since then. This interactive dashboard allows you to click through the phases. Click here for mobile view.

From March to mid-April there were a reported 22 local infections before a long pause. Two in-country outbreaks in November were thwarted by swift government action thanks to the commendable rapid assessment and rapid containment approach by the Ministry of Health supported by Institut Pasteur du Cambodge and the World Health Organisation. Closure of schools and entertainment venues were imposed three times but full lockdown was never imposed as shops, markets, and eateries were kept open.

Taking the 463 unfiltered reported Covid-19 cases at face value doesn’t offer the full picture, as 86% of reported cases have been imported, which present a near zero threat thanks to Cambodia’s watertight quarantine entry restrictions. I believe that excluding imported cases gives a more accurate assessment of the risk over the pandemic. This interactive dashboard shows the breakdown by source. Click here for mobile view.

Country comparison

When comparing across countries the Worldometer site is an incredibly valuable resource, but ranking from low to high is cluttered. I found, once again, that filtering and reshaping the data before analysis, mostly using Excel’s Power Query feature, can give us a more accurate view of the most resilient nations. 

To paint a more realistic picture, rankings mentioned in this article make certain logical exclusions of non-country territories, as well as the 12 countries with 20 or fewer reported cases (North Korea, Turkmenistan and ten Pacific Island nations). Two other exclusions are also in place, countries with populations under 100,000 and those with no reported testing data according to Worldometer – Tanzania, for example. reports very low cases per 1M, but announced on May 8 that Covid-19 cases would no longer be reported 

After applying these filters, of the three countries with zero deaths (all of which are in Southeast Asia), Cambodia ranks the best in all three categories – deaths per population, deaths per case and deaths per test. Were Cambodia to go from zero to three deaths, it would still have the lowest deaths per capita in the world, as well as the lowest deaths per case.

Population of largest country’s with zero Covid-19 deaths (excludes countries with less than 20 reported cases).

At 463, Cambodia ranks sixth lowest globally for total Covid-19 cases to date, with Laos being number one. Laos is also lowest in cases per million people, with its total of 44 equating to only six cases/1M. Vietnam, often commended as a world leader in its efforts, outperforms Cambodia to rank second lowest at 16 cases/1M. Thailand ranks 14th in cases/1M, although its cases are rising fast. Staggeringly, these numbers suggest that most affected countries, such as the UK and USA, have been 2-3,000 times more affected by Covid than these Mekong countries.

Australian think-tank the Lowy institute published a report on January 26 ranking Covid-19 performance across 98 countries, measuring performance across six categories. Had Cambodia not been excluded, it would have ranked highly in five out of six categories, but significantly less well on tests per 1M – albeit well above Vietnam, with 60% more tests per capita than its much-celebrated neighbour. 

How Cambodia would have ranked in the recently published Lowy Institute report.

Vietnam and Thailand were both placed in the top four of the 98 countries ranked, so would Cambodia not have been excluded it would have been highly ranked as well (note that the stated methodology for this ranking excludes more countries than my data does – e.g. Laos would have been omitted).

It was encouraging to see the fifth metric recognised by the think tank, positive cases per test, as it is neglected in almost every study I have seen. In Cambodia, only 1.2 tests per thousand has been positive, the fourth lowest globally. The US, on the other hand, is around 70 times higher at 86, implying for every undetected case in the Kingdom, other countries may have many, many more. 

Every country has undetected Covid-19 cases, but Cambodia’s count seems to be small. Of the two outbreaks where local transmissions were detected in the last nine months, the first had four local transmissions detected out of 1,400 contacts tested, whilst the second had more than 6,500 people tested (over 4,800 outside Phnom Penh), with 41 positive cases. The actual number of tests was much higher as Cambodia tests each contact up to four times, spread over the two week incubation period. The first or second test are the more likely ones to be positive, but there are exceptions. The last in-country transmission detected was an infant positive on their third test, even though they likely contracted the virus over a week prior, a positive case that would have likely gone undetected in other countries who are less vigilant.

Most of Cambodia’s tests are done in conjunction with the internationally-accredited Pasteur Institut du Cambodge (IPC). The Ministry of Health (MoH) closely follows guidelines from the World Health Organisation when it comes to clear, rapid and unambiguous communication across all mediums, and has gained praise for its efforts. IPC and the MoH have established testing centres for sampling around the Kingdom, in addition to the checkpoint screenings, increasing accessibility to testing and somewhat reducing wait times for results – though samples are currently processed centrally in a few labs in Phnom Penh.

Open Development Cambodia breaks down cases by demographic, allowing me to create this interactive dashboard, most cases are in their 20s and 30s but this is inline with the country’s low median age, males made up the majority of case counts until one month ago when the migrant workers from Thailand (mostly female) started returning following a Covid-19 outbreak there. You can click on any aspect to filter other charts by that aspect (e.g. click “female” and all other charts will apply that filter). Click for mobile view.

The Kingdom’s status as a tropical country with low urbanisation, generally low population density even in urban areas, and a low median age in the mid-20s may contribute to few (mostly mild) cases. While it’s impossible to fully attribute these factors as the reason why Cambodia has been so fortunate, this interview with epidemiologist Dr Michael KInzer offers some ideas and further testing data.

Many worried that Cambodia would succumb to the horrors of exponential case growth, but with only 67 in-country infections detected across the year, with December 15 the most recent local infection, the theat for now seems very much contained.

Data specialist David was given the prestigious Microsoft MVP award in early 2020 which places him in the top 100 or so Excel people in the world. David has taught Excel to over 2000 people, has a YouTube channel about tech tutorials. He runs Xlconsulting, a Cambodia based technology consulting firm with more global Covid-19 data interactive dashboards on their website.

Read more articles