|
I have been fully vindicated by the world's most objective epidemiologist. The COVID vaccines are not safe.
I should have done this sooner. I used gold-standard record-level Medicare data and had it analyzed by an objective expert. The result: 10% higher all-cause mortality risk by brand.
|
| |||||||||||||||
 |
Executive summary
I lied to ChatGPT. I told it I was giving it cancer drug data and wanted to know which cancer drug was safer based on 1 year mortality from the time the drug was given.
In reality, I fed it the Medicare record-level data from Connecticut.
The CDC has this data, but they will never let the public see it. But I will.
This is Gold standard data. Anyone can replicate what I did below in about 1 minute of effort. If they wanted to know the truth.
In any sane society, we should be done.
But in today’s society, not a single public health official wants to know the truth. So they won’t look. Ever.
The data I fed to ChatGPT
Here’s the 66MB file. It’s record-level COVID vaccine pulled from Medicare from Connecticut. I got it. The CDC could easily get it too if they wanted to look.
I just changed the headings and the vaccine brands. Purple is Pfizer. Mixed is Moderna. Cherry is Janssen.
The reason is because I didn’t want to bias ChatGPT.
ChatGPT analysis summary
10% higher all-cause 1-year age-standardized mortality of Pfizer over Moderna.
Highly statistically significant. p=2.39e-176.
This means the vaccines are not safe and that the health authorities are not looking at the data.
The conversation with ChatGPT
You said:
Here is a spreadsheet of everyone in the US who got cancer treatment with drug Purple vs. Mixed brands. Which brand was more effective in saving lives, or are they about the same? What is the percentage all-cause mortality difference between the brands based on the mortality 1 year from the first shot?
ChatGPT said:
ChatGPT
Based on the data, here are the 1-year all-cause mortality rates for the different brands:
Purple: 4.18%
Mixed: 2.59%
Analysis:
The 1-year mortality rate for those treated with Purple is higher (4.18%) compared to the Mixed brands (2.59%).
This suggests that Purple has a higher all-cause mortality rate within 1 year compared to Mixed brands.
Percentage difference:
To calculate the percentage difference in mortality rates between Purple and Mixed:
Percentage increase=4.18%−2.59%2.59%×100\text{Percentage increase} = \frac{4.18\% - 2.59\%}{2.59\%} \times 100Percentage increase=2.59%4.18%−2.59%×100
I'll compute that for you next.
The 1-year all-cause mortality rate for those treated with Purple is approximately 61.46% higher than for those treated with Mixed brands.
This suggests that Mixed brands may be safer, given the lower mortality rate within 1 year from the first dose. However, it's essential to consider other factors such as patient demographics, health conditions, and treatment protocols before making definitive conclusions about efficacy and safety.
4o
You said:
How confident are you in that assessment? What are the 95% confidence bands?
ChatGPT said:
ChatGPT
Here are the 95% confidence intervals for the 1-year all-cause mortality rates:
Purple:
Mortality rate: 4.18%
95% confidence interval: 4.11% to 4.25%
Mixed brands:
Mortality rate: 2.59%
95% confidence interval: 2.51% to 2.66%
Interpretation:
Both mortality rates have fairly narrow confidence intervals due to the relatively large sample sizes, which suggests that we can be reasonably confident in these estimates.
Since the confidence intervals do not overlap, this strongly indicates that the higher mortality rate for Purple compared to Mixed brands is statistically significant.
Thus, I am confident in the assessment that Mixed brands have a lower 1-year all-cause mortality rate compared to Purple based on this data.
4o
You said:
did you do an age standardized mortality rate calculation to compare the brands? the age distribution of the brands was not the same so you have to take that into account. did you?
ChatGPT said:
ChatGPT
You're absolutely correct. The simple comparison I made earlier did not account for potential differences in the age distribution between the groups treated with Purple and Mixed brands. This could significantly affect the mortality rates, as older populations typically have higher mortality rates.
To make a more accurate comparison, we need to calculate an age-standardized mortality rate (ASMR), which adjusts for differences in age distribution between the two groups. This will allow us to make a fair comparison of the effectiveness and safety of the two treatments, independent of age differences.
Let me calculate the age-standardized mortality rate for each brand now.
The age distribution summary for the two main brands, Purple and Mixed, shows a difference in the average age at which patients received the treatment:
Purple:
Mean age: 74.12 years
Median age: 74.0 years
Mixed brands:
Mean age: 72.53 years
Median age: 72.0 years
This indicates that the Purple group is, on average, slightly older than the Mixed group. Because age is a significant factor in mortality, we need to account for these differences using age-standardized mortality rates.
I'll proceed with calculating the age-standardized mortality rates for both brands.
To calculate the age-standardized mortality rate (ASMR), we typically use a standard population (such as the World Health Organization's standard population) to adjust the mortality rates based on age groups.
Here’s the process:
Group the data into age bands (e.g., 20-29, 30-39, etc.).
Calculate mortality rates within each age band for each brand.
Apply weights based on the standard population’s age distribution.
Calculate the standardized rate for each brand.
I'll prepare the data into age groups and calculate the ASMR now.
Here are the 1-year mortality rates for each age group and brand:
Purple:
20-29: 0.41%
30-39: 0.61%
40-49: 1.08%
50-59: 2.39%
60-69: 1.49%
70-79: 2.23%
80-89: 6.17%
90-99: 17.61%
100+: 32.09%
Mixed:
20-29: 0.45%
30-39: 0.64%
40-49: 0.97%
50-59: 2.53%
60-69: 1.19%
70-79: 1.68%
80-89: 4.43%
90-99: 12.18%
100+: 24.18%
Next Step:
Now that we have the mortality rates for each age group, I will calculate the age-standardized mortality rate (ASMR) by applying weights based on a standard population distribution. This will allow us to adjust for differences in age structure between the groups.
Let me proceed with that calculation.
To compute the age-standardized mortality rates (ASMR), we need to apply a standard population distribution to the age-specific mortality rates for each brand. Typically, a standard population (such as the WHO's or a national standard) provides the proportion of the population in each age group, which we can use as weights.
Since we don't have the standard population immediately available, I can apply an equal weighting for a simplified calculation or use a known standard population distribution if you have one in mind.
Would you prefer to proceed with equal weighting, or do you have a specific standard population distribution you would like me to use for the calculation?
4o
You said:
use a known one such as the US distribution please or any common standard one.
ChatGPT said:
ChatGPT
I'll use the US 2000 standard population distribution, which is commonly used for age-standardization. Here is the age group breakdown for that population:
Age GroupProportion of Population (%)20-2913.5%30-3914.6%40-4915.0%50-5911.5%60-697.7%70-794.7%80-891.9%90-990.2%100+0.0%
I'll now calculate the age-standardized mortality rates for both brands using this distribution.
The age-standardized mortality rates (ASMR) for the two brands are as follows:
Purple: 0.95%
Mixed brands: 0.87%
Interpretation:
After adjusting for differences in age distribution, the Purple brand still has a slightly higher mortality rate than the Mixed brands.
The difference is smaller than the unadjusted rates, but Purple remains associated with a higher 1-year all-cause mortality rate, suggesting that it may still be slightly less safe than the Mixed brands.
Would you like to explore any further details or conduct more analysis?
4o
You said:
so purple looks about 10% higher mortality ? is that statistically significant?
ChatGPT said:
ChatGPT
Yes, the Purple brand has an approximately 10% higher mortality rate compared to the Mixed brands based on the age-standardized rates:
Purple: 0.95%
Mixed brands: 0.87%
To determine if this difference is statistically significant, we can conduct a hypothesis test, such as a z-test for proportions, comparing the mortality rates between the two groups.
I will perform the test to assess whether the 10% difference is statistically significant.
The z-statistic is 28.31, and the p-value is extremely small (2.39e-176), which is far below any typical significance level (e.g., 0.05).
Conclusion:
The 10% higher mortality rate for the Purple brand is statistically significant.
This means that the difference in mortality between Purple and Mixed brands is unlikely to be due to random chance, and there is strong evidence to suggest that Purple has a higher 1-year all-cause mortality rate than the Mixed brands.
Let me know if you'd like further analysis or clarification!
Summary
If ChatGPT made an error, please let me know.
If not, the “debate” is over.
You're currently a free subscriber to Steve Kirsch's newsletter. For the full experience, upgrade your subscription.
| ||||||||||
No comments:
Post a Comment