Comparison by difficult number of people

It is said that the ** second wave ** is coming. Certainly, according to daily reports, for example, the number of positive PCR tests in Tokyo exceeds the time when the ** state of emergency ** was made. However, in the background, the so-called ** night town cluster ** measures were introduced Shinjuku Ward sympathy money system. In particular, it is mentioned that young people engaged in the water business are more likely to undergo PCR tests. And, when the generations under 50, who are actually said to be low risk, are counted, it can be seen that they have already exceeded the previous peak.

Therefore, the actual severity may not be known from a simple comparison of the number of people.

What is important in Covid-19 / coronavirus 0. Active infected persons should not exceed the capacity of the quarantine facility

Severely ill people using ICU, ventilator, etc. should not exceed their capacity
The death toll does not increase

Is. 0 correlates with the number of positives in a certain period, but 1 and 2, which are indicators of seriousness, cannot be understood from the number of people. Therefore, by totaling the already known mortality rate by age group by the number of positive persons by age group, the ** potential estimated number of deaths ** that will change to severe cases in the future is estimated, and the ** risk level ** I tried to aggregate the transition as.

Age-specific mortality data source NHK special site

Below is a summary graph by date

Next is a summary graph by week number

At first glance, it is clear that it is ** increasing **, but it is far lower than the previous peak. It takes a certain amount of time for the current positive to become severely ill. Therefore, the following can be said

** Due to the increasing trend of PCR-positive individuals, the capacity of isolation facilities is likely to be imminent within a few weeks ** ** However, the severity is lower than when the state of emergency was declared from the estimated potential deaths ** ** Similarly, there will be no shortage of facilities for the severely ill **

I understand that. The background may be the increase in the number of PCR tests for young people mentioned at the beginning and the increasing degree of social distance implementation at facilities with high-risk groups such as elderly care facilities.

On the other hand, it is a fact that the number of young infected people is increasing, and it can be said that if this jumps to a high-risk group, the severity may rise at once. Regarding the current situation, I would like you to consider regulations while trying to balance with the economy after accurately analyzing multifaceted figures.

Below is the analysis code that runs on Jupyter etc. The same result can be obtained by downloading Data source Tokyo Metropolitan New Coronavirus Positive Patient Announcement Details to localdir.

#coding: utf-8
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt 
import datetime as dt
localdir = "/Downloads/"
csv_data = localdir +"130001_tokyo_covid19_patients.csv"
age_dic = {'unknown':0.001,'Under 10 years old':0.001,"'-":0.001,'90s':0.111,'100 years and over':0.111,'70s':0.052,'80s':0.111,'60s':0.017,'50s':0.004,'Forties':0.001,'30s':0.001,'20's':0.001,'10's':0.001}
high_risks = ('90s','100 years and over','70s','80s','60s')
labels = ('Published_date','patient_Age')

dataset = pd.read_csv(csv_data,encoding = 'utf-8',header=0,)
dataset = dataset.loc[:,labels]
print (set(dataset.loc[:,'patient_Age']))
dataset.loc[:,'date'] = pd.to_datetime(dataset.loc[:,'Published_date'])
dataset.loc[:,'weekday'] = dataset.date.dt.week

dataset.loc[:,'val'] = dataset.loc[:,'patient_Age']
high_risk_table = dataset[~dataset.val.isin(high_risks)]

dataset.val = dataset.val.replace(age_dic)

datahighrisk = high_risk_table.loc[:,('date','val')]
databyweek = dataset.loc[:,('weekday','val')]
databydate = dataset.loc[:,('date','val')]
#dataset = dataset.fillna(0)
print(dataset)

highriskcount = datahighrisk.groupby('date').count()
deathbyweek = databyweek.groupby('weekday').sum()
deathbyday = databydate.groupby('date').sum()

plt.title("Risk Index by Date")
plt.plot(deathbyweek.index,deathbyweek.val)
plt.show()

plt.title("Risk Index by Week Number")
plt.plot(deathbyday.index,deathbyday.val)
plt.show()

plt.title("Count of Low Risk Group")
plt.plot(highriskcount.index,highriskcount.val)
plt.show()

[PYTHON] Corona Tokyo potential death toll or risk

Comparison by difficult number of people