Humidity as a non-pharmaceutical intervention for influenza A

Abstract

Influenza is a global problem infecting 5-10% of adults and 20-30% of children annually. Non-pharmaceutical interventions (NPIs) are attractive approaches to complement vaccination in the prevention and reduction of influenza. Strong cyclical reduction of absolute humidity has been associated with influenza outbreaks in temperate climates. This study tested the hypothesis that raising absolute humidity above seasonal lows would impact influenza virus survival and transmission in a key source of influenza virus distribution, a community school. Air samples and objects handled by students (e.g. blocks and markers) were collected from preschool classrooms. All samples were processed and PCR used to determine the presence of influenza virus and its amount. Additionally samples were tested for their ability to infect cells in cultures. We observed a significant reduction (p < 0.05) in the total number of influenza A virus positive samples (air and fomite) and viral genome copies upon humidification as compared to control rooms. This suggests the future potential of artificial humidification as a possible strategy to control influenza outbreaks in temperate climates. There were 2.3 times as many ILI cases in the control rooms compared to the humidified rooms, and whether there is a causal relationship, and its direction between the number of cases and levels of influenza virus in the rooms is not known. Additional research is required, but this is the first prospective study suggesting that exogenous humidification could serve as a scalable NPI for influenza or other viral outbreaks.

Fig 1. Absolute humidity and influenza hospitalized cases.

(A) Outdoor absolute humidity (AH) values (n = 65; one measurement per day) from Rochester, MN and influenza hospitalized cases in MN (n = 1070, week ending in January 16^th- March 19^th). Applying the national trend model described by Shaman et al.[34] to the local humidity and illnesses, onset of influenza followed the predicted delay of 10–16 days (grey box) after an absolute humidity trough (blue box). Peak cases follow (pink box), as there is an incubation period of 1–4 days with viral shedding up to 7 days after symptoms resolve. (B) AH in 4 preschool classrooms (average of two sensors from 10 minute intervals over 150 minutes (n = 16 per sensor, room D) or (n = 17 per sensor, rooms A,B, C) per class period per sensor). Center values are mean of both sensors during class time and error bars are s.d. and corresponding outdoor AH (n = 15, 1 per day) on the 15 days of sample collection. Humidifiers were running in humidified rooms through sample collection on February 23.

Fig 2. Influenza A virus genomic copies of positive samples.

Horizontal bars indicate mean copy number and error bars are 95% CI. Fomites control, n = 31; Fomites humidified, n = 27; Air (total) control, n = 33; Air (total) humidified, n = 21. * Indicates air samples calculated mean per cubic meter of air based on air sample volume. * P<0.001.

Fig 3. Air particle concentrations by size in classrooms.

Bars indicate mean particle counts per cubic cm of air by size of air particles and error bars are 95% CI. Control, n = 23; Humidified, n = 30. * P <0.05, ** P <0.01.