Open Access Highly Accessed Research article

Predictors of indoor absolute humidity and estimated effects on influenza virus survival in grade schools

Tyler H Koep1, Felicity T Enders2, Chris Pierret3, Stephen C Ekker3, Dale Krageschmidt4, Kevin L Neff3, Marc Lipsitch5, Jeffrey Shaman6 and W Charles Huskins7*

Author Affiliations

1 Clinical and Translational Sciences, Mayo Graduate School, Mayo Clinic, Rochester, MN, 55905, USA

2 Department of Health Sciences Research, Mayo Clinic, Rochester, MN, 55905, USA

3 Department of Biochemistry and Molecular Biology, Mayo Clinic, Rochester, MN, 55905, USA

4 Department of Facilities and Support Services, Mayo Clinic, Rochester, MN, 55905, USA

5 Center for Communicable Disease Dynamics, Department of Epidemiology and Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, MA, 02115, USA

6 Department of Environmental Health Sciences, Mailman School of Public Health, Columbia University, New York, NY, 10032, USA

7 Division of Pediatric Infectious Diseases, Department of Pediatric and Adolescent Medicine, Mayo Clinic Children’s Center, Rochester, MN, 55905, USA

For all author emails, please log on.

BMC Infectious Diseases 2013, 13:71  doi:10.1186/1471-2334-13-71

Published: 5 February 2013



Low absolute humidity (AH) has been associated with increased influenza virus survival and transmissibility and the onset of seasonal influenza outbreaks. Humidification of indoor environments may mitigate viral transmission and may be an important control strategy, particularly in schools where viral transmission is common and contributes to the spread of influenza in communities. However, the variability and predictors of AH in the indoor school environment and the feasibility of classroom humidification to levels that could decrease viral survival have not been studied.


Automated sensors were used to measure temperature, humidity and CO2 levels in two Minnesota grade schools without central humidification during two successive winters. Outdoor AH measurements were derived from the North American Land Data Assimilation System. Variability in indoor AH within classrooms, between classrooms in the same school, and between schools was assessed using concordance correlation coefficients (CCC). Predictors of indoor AH were examined using time-series Auto-Regressive Conditional Heteroskedasticity models. Classroom humidifiers were used when school was not in session to assess the feasibility of increasing indoor AH to levels associated with decreased influenza virus survival, as projected from previously published animal experiments.


AH varied little within classrooms (CCC >0.90) but was more variable between classrooms in the same school (CCC 0.81 for School 1, 0.88 for School 2) and between schools (CCC 0.81). Indoor AH varied widely during the winter (range 2.60 to 10.34 millibars [mb]) and was strongly associated with changes in outdoor AH (p < 0.001). Changes in indoor AH on school weekdays were strongly associated with CO2 levels (p < 0.001). Over 4 hours, classroom humidifiers increased indoor AH by 4 mb, an increase sufficient to decrease projected 1-hour virus survival by an absolute value of 30% during winter months.


During winter, indoor AH in non-humidified grade schools varies substantially and often to levels that are very low. Indoor results are predicted by outdoor AH over a season and CO2 levels (which likely reflects human activity) during individual school days. Classroom humidification may be a feasible approach to increase indoor AH to levels that may decrease influenza virus survival and transmission.

Influenza; Humidity; Schools; Climate