Comparison of Hand Eczema Search Terms in Iraq Before and During SARS-CoV-2 Pandemic Using Frequentist Statistics and Polynomial Models

Abstract

Introduction.SARS-CoV-2 pandemic spread around the world exponentially. People use disinfectants excessively as a form of protection from the novel coronavirus, which may result in contact eczema. This, in turn, may be monitored by the local health authorities. Our study explores the internet in order to detect significant changes in online information search behaviors associated with eczema in Iraq during the pandemic. Material and Methods. We searched the internet, via Google Trends, using five search terms; "اكزيما", "الاكزيما", "اكزيما اليد", "كحول", and "مطهر"; these are the Arabic translation for "eczema", "the eczema", "hand eczema", "alcohol", and "disinfectant". We explored the temporal mapping covering two years, before and during the pandemic, using frequentist statistics, polynomial models, and neural networks to evaluate the time series which reflects web users' information-seeking behavior with regard to these terms. Results. Spatial mapping conveyed data from six Iraq governorates, including Ninawa, Babil, Al-Najaf, Baghdad, Basrah, and Erbil. Basrah governorate had the highest score (interest) for the search term "the eczema" (الاكزيما), while Al-Najaf had the highest score regarding the search term "disinfectant" (مطهر). Temporal mapping exhibited high variability, the highest of which was for the "the eczema" (الاكزيما) and "alcohol" (كحول). Exploring the time series using polynomial models demonstrated a weak power over the two years. However, in the course of the pandemic year, all models possessed moderate power. Conclusions. Changes in the human behavior during pandemic events are of prime importance for the pharmacovigilance experts. Pandemics may affect medical conditions, including hand eczema, as a manifestation of disinfectants overuse. Combining statistics and artificial intelligence facilitates screening, detecting, and collecting pharmacovigilance safety data.

Keywords

• Digital epidemiology
• online systems
• pharmacovigilance
• spatio-temporal analysis
• web search