## Monday, May 04, 2020

### Covid-19 in the tropics: How will high heat and humidity shape the outcomes?

by Suja Thomas.

There are many factors thought to influence the spread of the COVID-19 disease: the mode of presymptomatic transmission (fomite/aerosol), asymptomatic incubation period in a host (up to 14 days), environmental conditions (temperature, humidity - relative/absolute, insolation, pollution levels, etc.), local and national public health capacity, population density, and population immunity levels.

In this article, we focus on two environmental conditions potentially relevant in a country like India in the summer:

1. What do we know about how the environment - temperature and humidity - influence the transmission of COVID-19?
2. How will the epidemic be impacted by these conditions in the coming weeks in India?

The National Academies Press report (2020) summarises two distinct approaches we can use to ultimately answer these questions:

1. Experimental studies of laboratory-propagated virus under controlled conditions, and
2. Natural history studies of disease transmission in different locations and seasons.

The controlled conditions in (a) may insufficiently mimic real-life settings, and in (b), we have almost no control of natural conditions and sometimes of other confounding factors.

As the pandemic is just 4 months in, we are currently limited to data available by geography and time periods. In this article, we pull together an admittedly narrow picture from the currently existing literature. In addition, we augment our knowledge using more established knowledge about similar pathogens, most notably SARS-CoV.

### Evidence from similar pathogens

There are 6 other coronaviruses known to infect humans (CDC, 2020): four are less pathogenic and cause common colds; the two that are less common but highly pathogenic are MERS-CoV and SARS-CoV.

Strong evidence of seasonality is reported for less-pathogenic coronaviruses in the fall-winter months in the United States and United Kingdom (Gaunt et al., 2010; Price et al., 2019) and Hong Kong (Chiu et al., 2005; Lau et al., 2006; Martinez,2018). Such seasonality was not found for MERS-CoV or SARS-CoV. van Doremalen et al. (2013) reported increased incidence of MERS-CoV between April-August in the Middle East: high temperature, high ultraviolet index, low wind speed, and low relative humidity were associated with an increase in MERS-CoV cases (median average monthly temperature and humidity of 27 °C and 22%, respectively).

In a laboratory environment, MERS-CoV was most stable at low temperature/humidity conditions and could still be recovered after 48 hours (van Doremalen et al., 2013). In comparison, SARS-CoV-2 was undetectable after 30 minutes at 56 °C (Chin et al., 2020). SARS-CoV, which is genetically most similar to SARS-CoV-2, was inactivated by UV light at 254 nm and heat treatment of 65 °C or greater (Darnell et al., 2004). It was found to be viable between 22-25 °C and at a relative humidity of 40-50% over 5 days, but was rapidly lost at higher temperatures (>= 38°C and >95% relative humidity) (Chan et al., 2011). SARS-CoV-2 is an enveloped virus: certain non-enveloped viruses were found to be present throughout the year and were most active at a relative humidity of around 30% and an average temperature of 9.2 °C (observed range: 4-16 °C) (Price et al., 2019).

Based on what we know about pathogens similar to SARS-CoV-2, high temperatures (>= 56 °C ) and relative humidity values (>= 95%) could be leveraged to inactivate it, but we do not have sufficient information on associations of seasonality with COVID-19 disease spread.

### The temperature and humidity in existing studies

In China, where the pandemic started, the average recorded temperature in cities with COVID-19 spread was 6-8 °C (range: -17 to 21 °C), with 46-100% relative humidity (Wang et al., 2020). Up to March 22, 2020, globally, the pandemic spread over a temperature range of 3-17 °C and 13-39% relative humidity (Bukhari and Jameel, 2020). The table below compares the temperature and humidity recorded for Wuhan, Lombardy and New York city during the outbreak with those expected for cities in the tropics such as Singapore, Trivandrum and Delhi in June 2020, which are much higher: we do not have data yet on spread of the virus under conditions similar to those for India in June.

LocationTemperatureRelative humidity
Wuhan, China 3 - 11 C 59 - 97%
Lombardy, Italy 4 - 9 C 47 - 97%
New York, USA 0 - 10 C 58.5%

LocationTemperatureRelative humidity
Singapore 26 - 32 C 64 - 96%
Trivandrum 24 - 30 C 85%
New Delhi 28 - 39 C 26%

The current pandemic has been most prolific in temperate areas, with temperatures and relative humidity values far below those expected in India in June-August. Hence, extrapolation from natural history studies is of limited value.

### Natural history studies

Several studies report correlations of 0.4-0.6 between R0 (the reproduction number of a disease) and temperature and humidity. While Wang et al. (2020) focus solely on data from 100 cities in China, Bukhari and Jameel (2020), Chen at al. (2020) and Oliveiros et al. (2020) incorporate data from all countries with cases that could be included based on the study design (and time of publication).

Ficetola and Rubolini (2020) studied data on 189 countries/regions and found that growth rates peaked in regions with an average temperature of 5 °C and specific humidity of 4-6 g/m3. Notari (2020) found a 40-50% increase in doubling time of virus transmission from 5 to 25 °C, implying a slowing of disease spread at warmer temperatures. Sajadi et al. (2020) compared cities with significant community spread to those without and found greater disease rates in areas with lower average temperature and humidity (30-50° N). Araújo and Naimi (2020) also report on models showing that SARS-CoV-2 spread is favored in temperate climates (similar to SARS-CoV), but also report high model uncertainties across much of sub-Saharan Africa, Latin America and South-East Asia.

Even if we do not have data for the temperature and humidity values we are interested in, there is some evidence indicating that increase in temperature and relative humidity could be associated with decrease in disease spread.

### Implications

There are currently no strong infereable associations between temperature and relative humidity, but we are beginning to see evidence that high temperatures and high relative humidity in the laboratory can be used to inactivate the virus. The bulk of the outbreak occurred in temperate regions (30-60 °N and S). There are significant limitations in extrapolating what we currently know to forecast any potential effect of environmental conditions for countries such as India (8°4' - 37°6' N) in the upcoming summer months:

1. We have to weigh the quality of data (how cases are tested, counted, and reported and how many), how estimates of disease spread (e.g., R0, community spread, growth rates, etc.) are calculated, time periods for which we have data (Dec - April), and location constraints.

2. The relative importance of different factors may be more relevant for disease spread. Locations with high temperatures and relative humidity, but with higher population densities and longer exposure to public transport, might have much higher R0 than cities such as New York or Milan. For example, perhaps MERS spread so quickly in the Middle East in the summer months because people congregated in indoor air-conditioned environments to avoid the heat. A comparable problem may arise in the Indian summer, in the class of households which face greater densities in air conditioned rooms in summer.

This literature guides us towards the following unique elements of health strategy for India in the summer:

Sunlight (and, thereby, UVA and UVB radiation)
Bäcker (2020) reports that high irradiance reduces the transmission of COVID-19. UVC radiation is currently being used to disinfect hospital floors and transport vehicles (Gorvett, 2020). Sunlight is also a source of vitamin D, which is being recommended to elderly populations who are being asked to quarantine indoors (Gorey, 2020).
Temperature
The US DHS (DHS, 2020) reported that at 35 °C and 80% humidity, the half-life of SARS-CoV-2 was only one hour. Hospitals can leverage this heat treatment for disinfection; similarly, packages brought in from outside could be disinfected with exposure to sunlight for at least 2 hours on days with high temperature (>= 35 °C).
Internal and external environmental ventilation
To avoid the summer heat, most people will seek air-conditioned environments, with high potential for recirculation of contaminated air. Qian et al. (2020), based on data from China, reported that significant outbreaks occurred in indoor environments. While a ventilation rate of 8 to 10 L/s per person is required for good indoor air quality, they point out that the observed ventilation rate was only 3·9 L/s per person in shopping malls and 2·8 L/s per person in public buses. Malls, retail stores and theatres in India, popular destinations in the summer months, should assess how to maintain sufficient physical distance between patrons in addition to ventilation capacity. In private and public air-conditioned environments, we should think carefully about how we can better manage airflow stagnation (open windows, fans, etc.)

As new evidence continues to build and as the scientific community continues to analyze and monitor the situation, we should supplement contact-mitigation strategies with adequate control of temperature and humidity in our environments, along with safe exposure to sunlight.

We should reiterate that numerous factors will shape how the pandemic will work out in the tropics. These include public health capabilities, demographic and genetic characteristics of individuals, the extent of crowding on streets and in public transport, etc. There is no one monocausal explanation. In this article, we have tried to peer into the crystal ball, pulling together the existing research literature, and anticipating what future researchers will discover about the impact of temperature, humidity and insolation upon COVID-19 in the tropics.

### Bibliography

DHS study on temperature and humidity (White House press briefing, Apr 23, 2020).

Emergencies, Preparedness, Response, WHO (2020).

Global Covid-19 Case Fatality Rates.

Human Coronavirus Types, CDC (2020).

Johns Hopkins University of Medicine Coronavirus Resource Center.

Rapid Expert Consultation on SARS-CoV-2 Survival in Relation to Temperature and Humidity and Potential for Seasonality for the COVID-19 Pandemic, April 7, 2020, The National Academies Press (2020).

Altamimi et al., Climate factors and incidence of Middle East respiratory syndrome coronavirus, Journal of Infection and Public Health (2020).

Araújo and Naimi, Spread of SARS-CoV-2 Coronavirus likely constrained by climate, medRxiv (2020).

Bäcker, Follow the Sun: Slower COVID-19 Morbidity and Mortality Growth at Higher Irradiances, SSRN (2020).

Bannister-Tyrrell et al., Preliminary evidence that higher temperatures are associated with lower incidence of COVID-19, for cases reported globally up to 29th February 2020, medRxiv (2020).

Bukhari and Jameel, Will Coronavirus Pandemic Diminish by Summer? SSRN (2020).

Casanova et al., Effects of Air Temperature and Relative Humidity on Coronavirus Survival on Surfaces, Applied and Environmental Microbiology (2010).

Chan et al., The Effects of Temperature and Relative Humidity on the Viability of the SARS Coronavirus, Advances in Virology (2011).

Chen et al., Roles of meteorological conditions in COVID-19 transmission on a worldwide scale, medRxiv (2020).

Chin et al., Stability of SARS-CoV-2 in different environmental conditions, The Lancet Microbe (2020).

Chiu et al., Human coronavirus NL63 infection and other coronavirus infections in children hospitalized with acute respiratory disease in Hong Kong, China, Clinical Infectious Diseases (2005).

Darnell et al., Inactivation of the coronavirus that induces severe acute respiratory syndrome, Journal of Virological Methods (2004).

Ficetola and Rubolini, Climate affects global patterns of COVID-19 early outbreak dynamics, medRxiv (2020).

Gaunt et al., Epidemiology and Clinical Presentations of the Four Human Coronaviruses 229E, HKU1, NL63, and OC43 Detected over 3 Years Using a Novel Multiplex Real-Time PCR Method, Journal of Clinical Microbiology (2010).

Gorey, Vitamin D supplements could be crucial for elderly `cocooning' from coronavirus, Silicon republic (2020).

Gorvett, Can you kill coronavirus with UV light? BBC Future (2020).

Lau et al., Coronavirus HKU1 and Other Coronavirus Infections in Hong Kong, Journal of Clinical Microbiology (2006).

Martinez, The calendar of epidemics: Seasonal cycles of infectious diseases, PLOS Pathogens (2018).

Miller et al., Correlation between universal BCG vaccination policy and reduced morbidity and mortality for COVID-19: an epidemiological study, medRxiv (2020).

Notari, Temperature dependence of COVID-19 transmission, medRxiv (2020).

Oliveiros et al., Role of temperature and humidity in the modulation of the doubling time of COVID-19 cases, medRxiv (2020).

Price et al., Association between viral seasonality and meteorological factors, Nature Scientific Reports (2019).

Qian et al., Indoor transmission of SARS-CoV-2, medRxiv (2020).

Sajadi et al., Temperature, Humidity and Latitude Analysis to Predict Potential Spread and Seasonality for COVID-19, SSRN (2020).

Wang et al., High Temperature and High Humidity Reduce the Transmission of COVID-19, arxiv.org (2020).

van Doremalen et al., Stability of Middle East respiratory syndrome coronavirus (MERS-CoV) under different environmental conditions, Eurosurveillance (2013).

Suja Thomas is a Senior Data Scientist at Abt Associates. She thanks Laura Lee, Jessica Levy, Padmaja Patnaik, and Maria Razzoli for valuable discussions.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.