Traditional pneumonia under the COVID-19 epidemic
Coronavirus, also known as COVID-19, was first discovered in Wuhan, China in 2019. It was an unknown virus with strong transmissibility and lethality at that time. Different from traditional pneumonia, COVID-19 is caused by a new coronavirus infection, with fever, dry cough, and fatigue as the main manifestations. Severe patients often develop dyspnea or hypoxemia one week after the onset, and severe cases can quickly progress to acute respiratory distress syndrome and sepsis Shock, difficult-to-correct metabolic acidosis, coagulation dysfunction, and multiple organ failure, etc., and a very small number of patients may also have symptoms such as central nervous system involvement and extremity ischemic necrosis. In the subsequent mutation process, because its symptoms are highly similar to traditional pneumonia, it is difficult for people to distinguish between traditional pneumonia and covid-19 without testing. So I’m wondering if CDC could still get accurate data after the padamic. And after I looked at the CDC’s death statistics after the pandemic in 2020, I found something interesting.
It is unscientific to still believe in the accuracy of the data when the data sample is too large and there are very obvious interference factors. In order to confirm my guess, I set up 3 groups of control groups and in order to make my data more credible, I chose Colorado and California as my samples because one has a large population base and the other has a smaller population base.
— — The first group is Colorado and California The number of deaths from new coronavirus infection from 2020 to 2022.
— — The second group is the number of deaths from traditional influenza pneumonia in the above two states from 2020 to 2022.
— — The third group is from 2014 to 2016 before the epidemic Number of deaths from traditional pneumonia.
When I had an idea about how to pick the sample datas, I felt that the most suitable way to analyze the data was to make them into tables, so I used python’s Pandas to make some tables. And because I want to test data with a time span of up to 3 years, what I need is not the exact number of deaths in the two states in each year but to observe the changing trend of the entire table, so directly generating the table and comparing it will be better solution.
After I compared the tables of the first group and the second group, I found that 2021 is the most severe year of the epidemic. In this year, 47,034 people died of COVID-19 in California. The number of deaths from pneumonia has dropped significantly, from 6,192 in 2020 to 4,643 in 2021. This change is more pronounced in states with fewer people. When I looked at the data in Colorado, I was surprised to find that it actually dropped from 264 people in 2020 to 22 people in 2021. In order to verify the rationality of my results, I obtained the traditional pneumonia deaths in Colorado and California from 2014 to 2018 from the table of the third group, and compared it with the table of the second group, I It was found that the number of deaths from traditional pneumonia has dropped significantly since the start of the epidemic. When the epidemic was at its worst in California in 2021, the number of deaths was once reduced to 4,643, while the average number in previous years was about 6,000. These data further confirmed my initial thoughts: it is difficult for people to distinguish between traditional pneumonia and COVID-19 during the epidemic.
In this data analysis, the two control groups I used were the disease group (COVID-19, pneumonia); the time group (before the epidemic, during the epidemic). After comparing these data separately, I think it is reasonable to draw a conclusion: CDC’s death data on COVID-19 and pneumonia is not accurate, and a considerable part of the data may have been miscounted. I think the factors that lead to this situation are more because the similarity of the symptoms is too high, and the large number of new deaths in the epidemic has also brought troubles to the statistics. I think we need a new data collection method that can effectively help avoid such errors in future data statistics.
