There Is Nothing More Deceptive Than An Obvious Fact. Arthur Conan Doyle

Dr. Why
Infectious diseases, Hydroxychloroquine, Coronavirus, Azithromycin, Mortality, Whypinion

May 11 2020

JAMA (Journal of American Medical Association) published original investigation titled “Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in New York State”.

The authors retrospectively (e.g. after the fact) analyzed 1438 cases of patients hospitalized in NYC comparing mortality amongst 4 groups of parents:

  1. Treated with Hydroxychloroquine alone (HC);

  2. Treated with Azithromycin alone (AZ);

  3. Treated with both Hydroxychloroquine & Azithromycin (HC+AZ);

  4. Treated with the best supportive therapy (neither HC no AZ)

Authors came to the conclusion that among patients hospitalized with COVID-19, treatment with hydroxychloroquine, azithromycin, or both WAS NOT associated with significantly LOWER in-hospital mortality. Let’s call it study #2.

The study design differed slightly from the one published on April 11 of 2020, which looked at 368 patients treated with:

  1. Hydroxychloroquine alone - 97 patients (HC);

  2. Hydroxychloroquine with azithromycin - 113 patients (HC+AZ); or

  3. No Hydroxychloroquine - 158 patients (no HC).

The study concluded that patients treated with HC alone were 2.61 times more likely to die. When AZ was used alone or with HC - there was not significant difference in mortality. See blog from April 21 2020 One More Nail . Let’s call it study #1.

It sounds like both studies came to the same conclusion, or did they? The “devil is in the details”!

First, study #1 clearly states that hydroxychloroquine is associated with INCREASED mortality. Study #2 concludes that AZ, HC, or their combination DOES NOT LOWER the mortality;

Second, study #1 concludes that when both drugs are used together, this excess mortality is alleviated, e.g. AZ possibly compensates for the bad effects of HC;

Third, study #2 indicates that hazard ratio (HR) with HC alone is 1.08, e.g. 8% higher than without any drugs (HR = 1.0), which is not statistically significant;

Forth, study #2 concludes that HR for HC and AZ together is 1.35, e.g. 35% higher but, again, not statistically significant.

Fifths, study #2 concludes that HR for AZ alone is 0.56, e.g. 44% lower than without either drug, again, not statistically significant;

Six, study #2 FAILS to compare AZ alone with HR 0.56 vs. AZ+HC together with HR 1.35! This clearly indicates that AZ alone is by at least a factor of 2 better than AZ+HC.

Dr. Why’s interpretation:

  1. AZ lowers mortality and possibly compensates for ill effects of HC;

  2. HC increases mortality both alone and in combination with AZ;

  3. it appears to me that study #2, or, to be accurate, the way results are presented is designed to deceive, to make us believe that even although HC does not lower mortality, it also does not increase mortality. This reminds me half-full vs. half-empty comparison which are, technically, the same but they are not.

The bottom line: stay away from HC, use AZ when indicated, and DO NOT USE HC+AZ! One cannot come to this conclusion without reading the “small print”.

Stroke On Strike

Viva Belarus!