Reply to FDA removed black box warning for HRT

Forum Index » Perimenopause, Menopause, and Beyond

Reply to "FDA removed black box warning for HRT"

Subject:

Emoticons





More smilies

Text Color: Font: Close Marks

[quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous]For those who want data and official medical positions, here's some research on HRT and type 2 diabetes. Summarizing, it shows that across multiple large randomized trials and meta-analyses, HRT with estrogen (with or without progestin) reduces the risk of developing type 2 diabetes on the order of 20–30% and improves insulin resistance and glucose control:

Margolis KL et al., Diabetologia 2004 — 15,641 postmenopausal women randomized and placebo controlled: https://link.springer.com/article/10.1007/s00125-004-1448-x

Mauvais-Jarvis et al. 2017 – Endocrine Reviews
https://academic.oup.com/edrv/article/38/3/173/3063786

Note that HRT does not have an FDA-approved prevention indication.

The North American Menopause Society in their 2022 position statement said that they do NOT recommend starting HRT primarily to prevent diabetes, but that a diabetes-preventive effect is a real, evidence-based benefit in appropriate women (younger, early postmenopause): https://pubmed.ncbi.nlm.nih.gov/35797481/
[/quote]

Sure. But since we are being so detailed about the absolute of HRT it is also important to be detailed about the absolute benefits: “The cumulative incidence of treated diabetes was 3.5% in the hormone therapy group and 4.2%”

So there was 0.7% fewer diabetes cases in the HRT group. Hardly a stunning number on an individual level. These findings are probably most important to conclude that a woman with diabetes or at risk of it can take HRT without a negative impact on diabetes. [/quote]

This PP doesn't understand statistics. She doesn't know the difference between absolute difference and relative risk, so her post is seriously misleading.

Here's an example: Using seatbelts lowers your risk of death in a serious car crash by 50%. If you don't use a seatbelt and you get in a serious crash, your risk of dying is 43%. If you use a seatbelt, the risk drops to 21%. So by using a seatbelt, you cut your risk in half! Pretty good, right?

But the chances of anyone dying in a serious car crash isn't especially high. So the *absolute* difference between seat belt users and non-users -- 0.0005% -- looks absolutely tiny. That tiny-seeming number = cutting your chance of death in half.

So when PP says, 0.7% (absolute difference) is "hardly a stunning number on an individual level," she's completely misunderstanding what the appropriate "individual" number is. The "individual" number is the decrease in the relative risk an individual might have. That number is 21%, and that is the effect of HRT.

Whether you think a 21% reduction is a compelling number or not is up to you. [/quote]

Great now do the breast cancer and stroke risk. [/quote]

These were done upthread, but here are the numbers in one place:

Breast cancer (combined HRT): risks increase
Absolute increase: 0.5% over 5 years
Relative risk increase: 20% to 30% over 5 years
This is the equivalent of going from 23 per 1000 (non-HRT baseline) to ~28 per 1000 (with HRT) over 5 years. 
Net effect: 5 extra cases per 1000 over 5 years.

Breast cancer (estrogen only HRT)
No increased risk, and WHI actually showed a small reduction of risk. But estrogen-only applies only to women without a uterus.
Net effect: 0 or possibly protective.

Type 2 diabetes: risks decrease
Absolute decrease: ~1% to 3% over 5 years
Relative risk reduction: ~20% to 30% 
This is the equivalent of going from 50 cases per 1000 (non-HRT baseline for healthy women) to 35-40 per 1000 (with HRT) over 5 years. 
Net effect: 10 to 15 fewer cases over 5 years

Cardiovascular disease: risks decrease*:
Absolute decrease: 0.6% to 2% over 5 years
Relative risk reduction: 30 to 50% 
This is the equivalent of going from ~20–40 per 1000 (non-HRT baseline) to ~10–20 events per 1000 over 5 years
Net effect: 10-20 fewer serious coronary heart disease events (heart attacks and cardiac death) per 1,000 over 5 years
*in women who start HRT <60 or are within 10 years of menopause, the "early initiation" window. HRT is generally not recommended to initiate in older women.

But even this doesn't tell the whole story. Cardiovascular disease is the #1 killer of women -- far more common than breast cancer (35% of deaths vs. <3% ). So a risk relative reduction of 30-50% in coronary heart disease among early starters translates into vast population benefits. This is why trials show all-cause mortality falls by ~30-40% in early starters.

Stroke: data show no increased risk for women who begin HRT (transdermal estrogen, micronized progesterone) under 60 years of age or within 10 years of menopause. (There was a small increased stroke risk with the older formulations, and risks increased with age).

MOST IMPORTANT CAVEAT: we do NOT all have the same risks of diseases. Family history, BMI, BP, lifestyle, personal medical history, etc., all matter here. This is why the decision to take or not take HRT should never be one-size-fits all. So even with all the available data on risks and benefits, you really must discuss the decision with a health care provider who knows you, and understands your specific risk profile.

Hope this is helpful to anyone still reading. I'm going to bow out of this thread now, hoping that most readers can discern best-available data from strong opinions. Wishing you all well as we navigate our respective choices. [/quote]

Right - so the absolute risks AND absolute benefits are small. And we won’t get into how you are cherry picking studies, the limits of observational studies, and the fact that there are so many possible combinations of dosages and types of hormones administered that are not studied in detail.

[/quote]

How is 30% drop in all cause mortality a small absolute benefit? What is your background to take this tack?[/quote]

Where are you getting that from? Also yes, it is likely small in absolute terms (because it’s not like 50% of women between 50-60 die). [/quote]

Do you not understand how percentages work? 
[/quote]

Options

	Disable HTML in this message
	Disable BB Code in this message
	Disable smilies in this message

Review message

Search Recent Topics Hottest Topics