California Age Management Institute

Premarin and Provera (PremPro)
risks and benefits on cardiovascular disease

Jama July 17, 2002 Vol 288 No. 3 (full pdf)

response and re-analysis

As you might imagine, there is another view from where I sit.

Once again, there is too much illogic in the headlines and not enough light.

I am referring, of course, to the recent “conclusion” that there was a small, but “worrisome” number of women having an increase in side effects from “hormone therapy.” These were (minor) cardiovascular and breast cancer increases.

Let us examine the real facts.  The erroneous conclusions and policy decisions and advice have caused undue discomfort and hardships in millions of women since the original publication date

The setup

First, in the entire universe of standard hormone therapy, this study used Premarin and Provera.  Premarin, conjugated equilinin estrogens, or as you might remember is a contraction of Pregnant Mares Urine.  Doesn’t sound bio-identical to me.  Technically, Premarin is a zeno-estrogen.  That is, an estrogen foreign to the body.

In fact, there has been a profusion of better plant-based estrogens on the market for years.  Climara, Estrace, Ogen, Vivelle, to name just a few estradiol patches.  These are all estrogens (yes, that’s plural because estrogen is a class of Estrone, Estradiol and Estriol).  We use BiEst which is a compounded form of Estradiol and Estriol.  Bio-identical, natural hormones.

Second, Provera, synthetic medroxy progesterone acetate, which was chosen in the study, has nasty side effects.  These are well known.  None of this is a mystery.  As above, Progestins simulate the effects of Progesterone but is NOT equal to progesterone.  Natural Progesterone, which is what you, both men and women, have been internally producing for decades, could have been used. Natural Progesterone is readily available.  It has been studied in the past and is commercially available as Prometrium.  We use a compounded form of Natural Progesterone.

illogical (misleading) conclusions

Now here is the illogic.  When one drug (and the worst example) becomes a problem you don’t go condemning the entire class of drugs.  When the drug Rezulin was found to have some problems in reducing insulin resistance, they did not condemn the entire class of insulin resistance drugs.  In fact, the next month they easily introduced 2 newer agents to take their place. Avandia and Actose.

So, it creates mass confusion for millions of women and a lot of very pompous “I told you so’s.”  “Do I stop taking these now?”  I will bet that every Gynecologist office across America received at least 20 calls per day asking for advice, clarification or solace. “Now what do I do?”

Those that are taking bio-identical hormones (Natural hormones) such as BiEst and Natural Progesterone -- don’t worry….

Remember your body has been producing these in prodigious amounts since you were 13 years old.  They start running out between age 45 and 50.  What your body is not running out of is Premarin (horse estrogen) and Progestins (Provera).

Finally, if this is really as “dangerous” as advised, then shouldn’t we be hearing something about stopping ALL birth control pills?  They are, of course, estrogen and a progestin.  Almost the same drugs.  The side effects that are associated with these drugs are the same.

summarizing

To sum up: Premarin is a xeno-estrogen.  Estrogen is not equal to Premarin.   Provera is not Progesterone.  So the study did not prove that Bio-identical, Natural Estrogens and Progesterone would have the desired protective effect on cardiovascular disease.  To the writer's credit, and well hidden from any media coverage, these hormonal and methodological differences were briefly noted.

And, we are finally beginning to see even the original investigators imploring women and physicians to see the results in proper perspective.

... Prof Susan Johnson, an investigator in the study's clinical trials, insisted yesterday that HRT's benefits were unequivocal for women who suffered hot flushes, mood swings, irritability, night sweats and fatigue.

Around a third of these stopped taking the drug because of the WHI results, Prof Johnson told the American Association for the Advancement of Science's annual meeting in Seattle.

However, HRT "continues to be an excellent choice for the treatment of menopause-related symptoms", she said.

So here are all the positive reasons to be considering bio-identical hormone replacement (BHRT) with estradiol and estriol.  We still consider this to be a high benefit to risk ratio. With hundreds of functions and benefits, the answer is simple -- immensely improved quality of life, joy and happiness.

And a to keep this topical, this study has been used to "spin off" a number of papers that followed.   But a more sober assessment comes from, of all places, the New South Wales Breast Cancer Institute:

The director of the New South Wales Breast Cancer Institute has played down the risks associated with women taking the birth control pill or undergoing Hormone Replacement Therapy (HRT).

But Professor Boyages says research indicates that HRT might increase the chances of developing breast cancer by less than 1 per cent over a five-year period.

He says it is important for women to understand both the risks and the benefits associated with HRT.

"I think these risks are acceptable risks that in our western way of life are worth taking, particularly around oral contraceptives," he said.

a new look at risk analysis

I highly recommend you purchase and read the recently published book, Calculated Risks by Gerd Gigerenzer.

Let's look at an exaggeration first.  It is much easier to see these graphically.  Let's look at the difference between 10 out of 100 (10/100) and 20 of out 100 (20/100) ... A double:

10 in 100 20 in 100

Ok.  Now let's look at 10/1,000,000 vs. 20/1,000,000  -- still a double.  But the absolute difference here is highly enlightening -- you can't see it --- because the absolute difference (as opposed to the relative difference) is almost ... indistinguishable.

 

10 in 1 million 20 in 1 million

 

Now to the study. If you see an increase in problems from 10 to 13 per 10,000 that would (and will) be reported about as a 30% increase. Unacceptable! 

However, the real increase is only 0.03%...   See, 10/10,000 is 0.10% and 13/10,000 is 0.13%.  That is, in reality, an increase of 0.03%.  3 one hundredths per cent!  Rather like a photo finish at the Olympics – too close to call.

Not convinced?  Try this:  10 per 1,000,000 to 13 per 1,000,000 is still 30% change.  But the real number is ….  0.00003%.  That’s 3 one hundred thousandths.  That is insignificant.  Would still be reported out as a dramatic 30% change.

Want an even more compelling brief?  This is from The July 2002 Scientific American quoting a JAMA brief on statistical reporting.

MEDICAL REPORTING
Only the Best

Advertising campaigns routinely hype the most flattering claims to sell their products. Evidently so do papers in medical journals. Researchers at the University of California at Davis School of Medicine found too much emphasis given to favorable statistics in five of the top medical journals: the New England Journal of Medicine, the Journal of the American Medical Association, the Lancet, the Annals of Internal Medicine and the BMJ. The study, which looked at 359 papers on randomized trials, found that most researchers furnish a statistic only for "relative risk reduction"--the percentage difference between the effect of the treatment and a placebo. Just 18 included the more straightforward absolute risk reduction. If a treatment reduced the absolute risk from, say, 4 to 1 percent, it appears more impressive to present only the relative reduction of 75 percent. Researchers also failed to show other statistics that provide a more nuanced picture of the results of clinical trials. The article is in the June 5 Journal of the American Medical Association.(full pdf) --Benjamin Stix   [emphasis added for clarity - ed.]
 

These are the real numbers from this study.  You tell me which sells headlines and which is used to describe the actual change. 

the data

Absolute vs. Relative Statistics

JAMA July 17, 2002 Vol 288, No 3:321-33 (full pdf)

 

Estrogen

Placebo

Difference

 
Alive at end 7968 7608    
Lost to follow-up 307 276    
Deaths 231 218    
Total Numbers 8506 8102    
         
CHD 164 122 34.43% relative
  1.93% 1.51% 0.42% absolute
Stroke 127 85 49.41% relative
  1.49% 1.05% 0.44% absolute
Thromboembolic 151 67 125.37% relative
  1.78% 0.827% 0.95% absolute
Breast Cancer 166 124 33.87% relative
1.95% 1.53% 0.42% absolute
Endometrial 22 25 -12.00% relative
  0.26% 0.31% -0.05% absolute
Colorectal 45 67 -32.84% relative
  0.53% 0.83% -0.30% absolute
         
Hip fractures 44 62 -29.03% relative
  0.52% 0.77% -0.25% absolute
         
Vertebral fracture 41 60 -29.03% relative
  0.48% 0.74% -0.26% absolute
         
Osteoporosis 579 701 -17.40% relative
  6.81% 8.65% -1.85% absolute
         

critical benefits glossed over

On first analysis, by looking at the aggregate total numbers, you see that there was a decrease in endometrial cancer, colorectal cancer, hip fractures, vertebral fractures and osteoporosis. Did they look for changes in quality of life?

In an even more detailed analysis, looking more closely at the numbers year by year and calculating the absolute percentage changes we see results completely different than concluded in the study.  In fact, the absolute incidence of all disease conditions above, in the hormone treated group (except invasive breast cancer), actually declined against the baseline "placebo" group.  Quite eye opening.

I truly hope this helps you to begin looking at these public pronouncements and studies and misrepresentations of statistical analyses in a more meaningful fashion.

Feel free to write me with any comments.

Philip Lee Miller, MD
Director and Founder, California Age Management Institute