With current media reports about the flu epidemic, people are being encouraged to wash and sanitize their hands often. We explore hand sanitizer usage in various settings (retail, medical, gyms, kitchens) with and without a sign inviting the public to use the hand sanitizer. In our experiment, we use a randomized treatment/control schedule to study the effects of the sign, and weigh the hand sanitizer bottles daily. We also consider covariate effects from Google Trend data on flu and superbug searches. While we found a small naive effect in a basic OLS when the sign is introduced, the use of more appropriate time-series models fails to reject the null hypothesis that the sign has no effect on usage.
The flu is a major health hazard that poses a significant risk to our health and our lives. In any given year as much as 20% of the US population is affected by the virus, with tens of thousands of hospitalizations and thousands of deaths.
Aside from the influenza vaccine, proper hand hygiene using soap and water as well as alcohol based rubs has been found to be effective in reducing the presence of some types of influenza virus on human hands. Other studies also find a reduction in the transmission of the flu and other illnesses and absenteeism at work in response to hand hygiene programs incorporating alcohol based hand sanitizers,,.
This paper discusses an experiment to test the effectiveness of placing an indicator sign, pointing to the location of a hand sanitizer dispenser, on increasing the usage in multiple locations where people congregate. Considering the sobering statistics and the simplicity of the proposed treatment, even a small treatment effect would have significant practical public health implications.
For more details please see the file An_Insignificant_Amount_of_Purell.pdf in this repository.