Our World In Data Drops World Health Organization COVID-19 Data Because of Inaccuracies

Interesting post by Hannah Ritchie at Our World In Data about why the site decided to stop using World Health Organization data on the COVID-19 pandemic in favor of data from the European Centre for Disease Prevention and Control.

Unfortunately, in the publication of WHO data on 18th March – Situation Report 58 – they shifted the reporting cutoff time from 0900 CET to 0000 CET. This means that comparability is compromised because there is an overlap between the last two WHO data publications (Situation Reports 57 and 58).

. . .

In published WHO Situation Reports were several inconsistencies in the number of total confirmed cases, and new confirmed cases that we noticed between the WHO Situation Reports and the WHO Dashboard, which also presents these statistics. These discrepancies are detailed below.

We have informed the WHO about these inconsistencies and are working closely with the World Health Organization (WHO) in an effort to resolve these issues. We continue to be in close contact with the WHO data team to ensure the latest statistics are presented accurately.

The inconsistencies are small and do not affect the overall perspective on the development of COVID-19 in a major way, but errors were too frequent and this was the second reason why we stopped to rely on the WHO data.

Privacy-Preserve Contact Tracing?

Contact tracing is a method of stopping the spread of diseases by quickly finding and treating people who have come into contact with an infected person. According to the World Health Organization,

This monitoring process is called contact tracing, which can be broken down into 3 basic steps:

1. Contact identification: Once someone is confirmed as infected with a virus, contacts are identified by asking about the person’s activities and the activities and roles of the people around them since onset of illness. Contacts can be anyone who has been in contact with an infected person: family members, work colleagues, friends, or health care providers.

2. Contact listing: All persons considered to have contact with the infected person should be listed as contacts. Efforts should be made to identify every listed contact and to inform them of their contact status, what it means, the actions that will follow, and the importance of receiving early care if they develop symptoms. Contacts should also be provided with information about prevention of the disease. In some cases, quarantine or isolation is required for high risk contacts, either at home, or in hospital.

3. Contact follow-up: Regular follow-up should be conducted with all contacts to monitor for symptoms and test for signs of infection.

The Zcash Foundation is looking to develop a decentralized, privacy-preserving contact tracing tool that would springboard off of the Singapore government’s TraceTogether application.

One incredibly exciting technological development is TraceTogether, a mobile application that assists with contact tracing produced by the Government of Singapore and the Singapore Ministry of Health (MoH). The app creates a temporary ID by encrypting a user ID to a MoH-owned public key, and then broadcasts the temporary ID over Bluetooth. This temporary ID is refreshed at regular intervals, so that it cannot be used as a long-term identifier for third-party tracking. Nearby mobile devices running the app log all observed broadcasts. If a user later develops symptoms and tests positive for COVID-19, they can upload their log of contacts to the MoH, who functions as a trusted third party that can decrypt the log entries and notify all of that user’s contacts of potential COVID-19 exposure. The MoH promises to use the log data only for the purposes of contact notification.

While this application is not perfectly privacy-preserving, it is far superior to location-tracking, and reveals personal information only upon infection, rather than using the threat of COVID-19 as a justification to build permanent surveillance infrastructure, or exposing patient data to the public. Public health requires public trust, and the developers should be congratulated for building privacy protections into the system.

World Health Organization Declares Second Strain of Polio Eradicated Worldwide

On October 24, 2019, the World Health Organization announced that wild poliovirus type 3 has been eradicated worldwide other than sample specimens held in laboratory containment.

In an historic announcement on World Polio Day, an independent commission of experts concluded that wild poliovirus type 3 (WPV3) has been eradicated worldwide. Following the eradication of smallpox and wild poliovirus type 2, this news represents a historic achievement for humanity.

. . .

There are three individual and immunologically-distinct wild poliovirus strains: wild poliovirus type 1 (WPV1), wild poliovirus type 2 (WPV2) and wild poliovirus type 3 (WPV3). Symptomatically, all three strains are identical, in that they cause irreversible paralysis or even death. But there are genetic and virologic differences which make these three strains three separate viruses that must each be eradicated individually.

WPV3 is the second strain of the poliovirus to be wiped out, following the certification of the eradication of WPV2 in 2015. The last case of WPV3 was detected in northern Nigeria in 2012. Since then, the strength and reach of the eradication programme’s global surveillance system has been critical to verify that this strain is truly gone. Investments in skilled workers, innovative tools and a global network of laboratories have helped determine that no WPV3 exists anywhere in the world, apart from specimens locked in secure containment.

Death By Snake Bite

The Guardian has an interesting story about snake bite deaths in the Democratic Republic of the Congo. The story documents about how the effects of snakebites are compounded by the DRC’s lack of infrastructure and poverty.

One of the fascinating statistics in the story is how many people die worldwide from snakebites,

Globally, about 5m snake bites occur worldwide each year, according to the World Health Organization, resulting in between 81,000 and 138,000 deaths. A bite from a viper, cobra or mamba can kill in a matter of hours or leave a victim suffering life-changing injury.

I would not have thought the total deaths would be so high, so decided to track down the WHO statistics that The Guardian cites,

The World Health Organization (WHO) estimates that about 5 million snakebites occur each year, resulting in up to 2.7 million envenomings. Published reports suggest that between 81,000 and 138,000 deaths occur each year. Snakebite envenoming causes as many as 400,000 amputations and other permanent disabilities. Many snakebites go unreported, often because victims seek treatment from non-medical sources or do not have access to health care. As a result it is believed that many cases of snakebite go unreported.

That underreporting means the actual total of snake bites and deaths may be significantly higher,

One of the consequences of inadequate efforts to control snakebite envenoming in the past is that the available epidemiological data are fragmented and lack both resolution and completeness. Accuracy is further reduced by the fact that many victims do not attend health centres or hospitals, and instead rely on traditional treatments. As a result, in some countries the degree of under-reporting is greater than 70% especially in rural areas with poor infrastructure.

When Is a Dead Baby Not a Dead Baby? (When It’s Born Outside the United States)

Save The Children has released a report on infant mortality that, among other things, claims the United States has the second highest infant mortality rate among industrialized countries. The problem with this claim, however, is that the United States and other countries can’t quite agree on what counts as a dead baby. As such, infant mortality rates aren’t directly comparable between the United States and other countries.

The problem lies in a fact that, as the Save The Children report notes, most infants who die in the industrialized world die because they are either a) born too early or b) have a very low birth weight (and, of course, the earlier the delivery, the lower the birth weight tends to be).

In the United States, an infant born prematurely and weighing less than 400 grams will receive intensive medical interventions to try to keep it alive. If, as is likely, such expensive interventions fail, the event will be recorded as a) a live birth and b) a death.

In much of the rest of the world — including the industrialized world — such extreme medical interventions would never be attempted. Moreover, this would not be recorded as a live birth or as a subsequent death.

The Save The Children report simply relies on World Health Organization statistics, and the WHO itself recommends that births of less than 1,000 grams not be registered as live births in official records. Most countries follow this definition, whereas the United States doesn’t.

How big of a difference does this make? According to the National Center for Health Statistics, in 2003 infants weighing less than 1,000 grams accounted for 48.7 percent of infant deaths in the United States.

Assuming that rate holds for the 2004 U.S. statistics the Save the Children report covers, that results in a >1,000g infant mortality rate of 3.5 per 1,000 births for the United States. That rate puts the United States barely behind countries like Japan, Finland, and Sweden which clock in at 3 deaths per 1,000 births. Not bad, especially given the largely mono-racial nature of those societies as compared to the United States.

Of course, why should we expect advocacy groups or the media to bother with such arcane statistics when U.S. has second worst newborn death rate in modern world, report says makes such a good headline?

Plague Outbreak in the Democratic Republic of Congo

At least 61 people died in February during an outbreak of the pneumonic plauge in the eastern part of the Democratic Republic of Congo.

About 350 people who worked in a mine in the northern Oriental province were infected with the disesae earlier this year, with at least 61 of them ultimately succumbing to the disease.

The pneumonic plague is the rarest and most deadly of the three types of plague. Unlike bubonic and septicimic plague, the pneumonic form of the disease can be passed from person to person through infected droplets transmitted by coughing or sneezing.

According to the World Health Organization, it is almost always fatal if not treated, but responds well to antibiotics. Unfortunately, the Democratic Republic of Congo is still a relatively chaotic place after the end of its four-year civil war in 2002, and more than 2,000 people who worked at the mine quickly left and dispersed after the outbreak of the disease became widely known.

Plague, of course, used to be a major worldwide killer, famously wiping out a significant proportion of the European population in the late medieval period. The World Health Organization reports that in 2003 there were only about 2,000 cases of the disease worldwide, but almost all of those occurred in Africa.


Plague outbreak kills 60 in Congo. The BBC, February 18, 2005.

DR Congo plague outbreak spreads. The BBC, February 23, 2005.

Plague Outbreak in Eastern Congo. Cynthia Kirk, Voice of America, March 2, 2005.

Deadly Plague Outbreak Feared in Congo . Craig Timberg, Washington Post, February 18, 2005.

Locust invasions on West Africa. IRIN News, December 2004.