Edit: An error was found in the graphic and this post will be updated soon.
New data on maternal death rates for the US was recently reported on by Dr. McCullough. Dr. McCullough lamented about how the CDC was completely quiet about the possibility that the maternal deaths are linked to the COVID jabs.
But year-by-year statistical analysis, using the CDC’s own methods (inverse gamma function), shows that while some rise in maternal death rates had been seen prior to 2021, it was the year of 2021 which was the real blow-out:
[click to enlarge]
I ran simultaneous confidence intervals around the maternal death rates each year, using 3 different confidence levels (99%, 95%, 85.6%). If the upper bound of the rate of a prior year (green) is below the lower bound of the rate the next year (blue), then a significant increase was found.
Orange-highlighted cells show when there was a signficant increase in Total, non-Hispanic Black (NH Black), non-Hispanic White (NH White), or Hispanic (H) maternal death rates.
I can tell from my results that CDC utilized an alpha significance level of 5% — so that results would not be significant unless they are so extreme that they only happen randomly 5% of the time or less.
That bottom analysis leads to the same mosaic of significance that CDC reported, where “some” significant increase was found in “some” group each year after 2018.
Edit: An error was found in the graphic and this post will be updated soon.
Levels of Significance
But to really know that you don’t have a false alarm due to the play of chance alone, restricting alpha even further below “a = .05” helps. If you adjust the “magnification” on the looking glass (restrict alpha further), you find out that only 2021 had extremely high significance (alpha < .001).
Cells L12 through L15 show unprecendented increases in maternal death rates for the year of 2021 only. At that level of “magnification” only 2021 was signficant. This shows that 2021 was “special” — that it was especially harmful — which makes the CDC report all the more suspicious by not targeting COVID jabs as the likely cause.
For the year of 2021, even 99% confidence intervals around the death rates from 2020 and 2021 did not overlap. And that isn’t true for any of the other comparisons.
Reference
[probabilities of overlapping Confidence Intervals] — Knol MJ, Pestman WR, Grobbee DE. The (mis)use of overlap of confidence intervals to assess effect modification. Eur J Epidemiol. 2011 Apr;26(4):253-4. doi: 10.1007/s10654-011-9563-8. Epub 2011 Mar 19. PMID: 21424218; PMCID: PMC3088813. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3088813/
[original CDC report] — Hoyert DL. Maternal mortality rates in the United States, 2021. NCHS Health E-Stats. 2023.
DOI: https://dx.doi.org/10.15620/cdc:124678.