Frequency of testing for COVID 19 infection and the presence of higher number of available beds per country predict outcomes with the infection, not the GDP of the country - A descriptive statistical analysis

IntroductionThe novel coronavirus epidemic which originated in late 2019 from China has wreaked havoc on millions across the world with illness, death and socioeconomic recession. As of now no valid treatment or preventative strategy has evolved worldwide and governments across the world have been forced to take the draconian step of social isolation in communities by enforcing "lockdowns". Aim of this StudyThis study aims to correlate the rates of infection with the novel coronavirus and total deaths as the primary output variable. In addition the strength of association between infection rates and total death in comparison to GDP share of the respective countries, physicians, hospital beds and rates of testing for COVID 19 infection per thousand patients, is being assessed, in a bid to develop a model which would help to develop tools to reduce the impact of this disease. Material & MethodsData relating to number of cases, severity, cases recovered and deaths worldwide and specifically for the top six countries affected was collected from the WHO COVID-19 situation report which is being updated on a daily basis till 22nd March 2020, the date of analysis. Additional data related to GDP, physician and hospital bed per 1000 patients were procured from the World Bank database. All data were collected in a file in CSV format. Analysis was conducted in Jupyter notebook with Python 3.8.2 software and also with XL-Stat statistical software for excel. The analytical strategy was descriptive with no inferential overtones. ResultsCOVID 19 infection strongly correlates with total deaths (r : 0.89), with a predicted death rate of 25 patients per 1000 affected. There was no correlation between the GDP growth of the country and number of treating physicians/1000 patient population with any COVID 19 related outcome. However there was a negative correlation between COVID 19-related deaths and the number of beds available per 1000 population [r=-0.34]. Importantly there is an inverse correlation between the number of tests conducted per million population with the rates of active infections [r=-0.12], new cases [r=-0.38] and new deaths [r=-0.28] in COVID 19. ConclusionThis is the first study to assess parameters other than age and sex and sets out a robust dataset which indicates an increased risk of worsening outcomes with lesser number of beds and testing, suggesting that the need of the hour is to increase available bed numbers and to increase rates of testing.