Chief Epidemiologist Samaraweera misclassifies COVID-19 spread in SL

Fact Check

In his statement, the Chief Epidemiologist Sudath Samaraweera claims that the spread of COVID-19 in Sri Lanka has not reached community transmission level. He also states that community transmission can only be established where confirmed cases cannot be traced to an “origin”.

To check the claim, FactCheck consulted the data on daily COVID-19 cases published by the Epidemiology Unit of the Ministry of Health (MOH) and the definitions for COVID-19 adopted by the World Health Organisation (WHO).

The WHO defines four transmission scenarios for COVID-19 spread: no cases, sporadic cases, clusters of cases and community transmission. Exhibit 1 presents the definitions for each of these scenarios. COVID-19 spread is described as limited to clusters when the discovered cases are limited to well-defined parameters, denoted by time, geographic location, and common exposures. This definition means that for a case to be classified as being in a particular cluster, it should be linked to that cluster by a well-defined: (i) time period; (ii) geographic location; and (iii) common exposure.

Exhibit 2 shows the change in the number of cases for the period 1 July – 9 November 2020 (when the statement was made). It can be observed that since the emergence of the Minuwangoda/Peliyagoda (M/P) cluster on 5 October until the time at which the claim was made, 97% of all new reported cases (or 10,446 of 10,533 cases) were attributed to that cluster. Hence, the time-period of the spread of this cluster has exceeded 37 days (i.e., nearly three times the maximum incubation period). Therefore, the basis of defining M/P as a cluster cannot be on the criteria of a well-defined time period.

Evaluating on the basis of the other two criteria, we find that cases that were being attributed to the M/P cluster were discovered across 15 districts in the country. Therefore, the basis of defining M/P as a cluster cannot be on the criteria of a well-defined geographic location either.

Due to the 37-day spread in time and 15-district spread in location of the discovered cases, the basis of defining M/P as a cluster also cannot be on the criteria of a well-defined “common exposure”.

Therefore, the basis of attributing much of the discovered cases to the M/P cluster fails on all three of the criteria set out by the WHO for defining a cluster.

Hence, we classify the chief epidemiologist’s statement as FALSE.

Note: The error in the official classification arises possibly by deriving a backward chain speculatively connecting each new case to a previous case, and doing so recursively until a potential connection is made to a case that originated in the geographic location of M/P. This flawed approach would allow for a situation where 100% of the Sri Lankan population is COVID-19 positive and still defined as being limited to the M/P cluster – and render meaningless the concept of a cluster.

*FactCheck’s verdict is based on the most recent information that is publicly accessible. As with every fact check, if new information becomes available, FactCheck will revisit the assessment.

Additional Notes:

To understand how the government’s interpretation of a cluster deviates from the WHO definition, it is useful to look at the available definitions of “contact” of a COVID-19 case, and how they are reported in Sri Lanka.

The WHO defines a contact as anyone with the following exposures to a COVID-19 case, from 2 days before to 14 days after the case’s onset of illness:

• being within 1 metre of a COVID-19 case for >15 minutes;
• direct physical contact with a COVID-19 case;
• providing direct care for patients with COVID-19 disease without using proper personal protective equipment (PPE);
• other definitions, as indicated by local risk assessments [these include definitions for different settings such as household closed settings, healthcare settings, public/share transport and other well-defined settings and gatherings such as places of worship, workplaces, schools, private social events].

The MOH defines a “close contact” as “a person staying in an enclosed environment for more than 15 mins (same household/workplace/social gathering/travelling in same vehicle) or had direct physical contact.” The MOH also defines “second level contacts” as contacts “identified from environments that have higher risk of transmission such as patients from highly overcrowded areas/patients who had very high mobility with large number of contacts/people living in congregate settings like hostels/camps/institutional care facilities”].

The MOH reports both close contacts and second level contacts as part of a cluster. By doing so, the MOH widens the definition of a cluster beyond that set out by the WHO, thus allowing it to maintain the claim that the recent spread is contained within the M/P cluster.

Notably, the WHO’s definition of cluster spread implies “a low risk to the general population if exposure to the cluster is avoided. However, the MOH’s own categorisation of several areas of the country as “high risk” suggests that the characteristics of the spread and the government’s response to it are more aligned with the WHO’s definition of community transmission.

Disclaimer: This factcheck based its assessment of the claim on the definitions for the four transmission scenarios provided by the WHO. FactCheck was not able to identify alternative MOH definitions for the transmission scenarios in Sri Lanka that the Chief Epidemiologist may have used in making his claim. 

Exhibit 1: COVID-19 Transmission Scenarios as defined by WHO

The Lancet COVID-19 Commissioners also provides the following guidance for transmission in a population: 

Low transmission: 10 or fewer new cases per million population per day

Moderate transmission: 10–50 new cases per million per day.

High transmission: 50–100 new cases per million per day, 

Very high transmission: 100 or more new cases per million population per day.

Exhibit 2: Number of COVID-19 Cases in Sri Lanka (1 July – 9 November 2020)