New insights for measuring community transmission of SARS-CoV-2 in Dhaka

Understanding community transmission of SARS-CoV-2 remains key to developing effective strategies to combat its spread. icddr,b scientists investigated community transmission of the virus in Dhaka, Bangladesh between June and September of 2020, and published their results in the journal Tropical Medicine and Infectious Disease.

The authors identified laboratory-confirmed cases of COVID-19 in six high-density slums and seven low-density wards. The study enrolled individuals who had come into contact with COVID-19 confirmed cases within the first two days before the symptoms appeared or fourteen days after the onset of symptoms. To evaluate community transmission, they measured the secondary attack rate — the probability of infection among an exposed group — by dividing the number of positive SARS-CoV-2 cases among the enrolled contacts on days 7, 14, 21, and 28 by the number of contacts enrolled. While this measure can capture multiple generations of spread, it cannot rule out transmission by sources beyond the enrolled contacts. To estimate how many individuals were infected by each initial case on average, the authors calculated the basic reproduction number (R0) by dividing the number of SARS-CoV-2-positive contacts during 14 days of follow-up by the number of primary infectious cases. This is less likely to capture multiple generations of transmission and infection by external sources.

In contrast to a previous study conducted in the USA, the authors found that the secondary attack rate was higher in low-density areas (20% or 50/497) compared to high-density areas (10% or 37/187). This is surprising as an airborne virus like SARS-CoV-2 would be expected to have higher transmission rates in high-density areas where people come into greater proximity to their neighbors. However, the basic reproduction number was higher in high-density areas (2.7) than in low-density areas (1.0), indicating that each confirmed case in a high-density area was likely to infect more individuals than those in low-density areas.

To address this apparent contradiction, the authors offer a possible explanation that many individuals living in slums lost employment and income as a result of lockdown measures at the time of the study and migrated back to their home villages, which affected the actual density of the slum populations. It is possible that there was greater transmission from sources beyond the enrolled contacts in the low-density areas compared to the high-density slums, biasing the estimation of the secondary attack rate.

“The higher secondary attack rate in low-density areas is a bit unexpected,” confirms Dr Satter, Assistant Scientist, and Deputy Project Coordinator at icddr,b and the first author of the paper. “But our estimates of the measures of community transmission of SARS-CoV-2 are within the range of what others have reported worldwide.”

The results of the study — the first of its kind in Bangladesh and wider Southeast Asia — point to the need for more extensive future studies on community transmission of SARS-CoV-2 and highlight the importance of considering local socioeconomic factors when interpreting measures of viral spread. In addition, it provides a baseline for assessing how vaccination efforts and past surges in infection have affected community transmission as the pandemic has progressed.


Ornob Alam