What does KEEPS stand for?

KEEPS stands for Kronos Early Estrogen Prevention Study. It was funded by the Kronos Longevity Research Institute and conducted at nine U.S. clinical centers between 2005 and 2012.

How was KEEPS different from the Women's Health Initiative?

Three key differences: KEEPS enrolled younger women (mean age 52.7 vs. 63 in WHI), used lower hormone doses (0.45 mg CEE vs. 0.625 mg), and measured a subclinical imaging endpoint (CIMT) rather than clinical events like heart attacks. These design choices made KEEPS a test of the timing hypothesis that WHI was not designed to address.

Did KEEPS prove that early HRT is safe for the heart?

Not exactly. KEEPS showed no cardiovascular harm over four years in healthy, recently menopausal women. But it was too small and too short to prove safety for clinical events like heart attack or stroke. It is best interpreted as absence of a harm signal, not proof of safety.

Why did KEEPS use carotid intima-media thickness as its primary endpoint?

CIMT can be measured noninvasively by ultrasound and tracks with atherosclerotic risk. Using a surrogate marker allowed a smaller trial (727 women) to run for a shorter period than would be needed to detect clinical events like myocardial infarction.

What was the coronary artery calcium finding in KEEPS?

In the oral CEE group, coronary calcium accumulation was lower than in the placebo group (p = 0.05 in per-protocol analysis). This suggested a possible cardioprotective effect of oral estrogen through hepatic lipid pathways, but the result was borderline and has not been confirmed in a larger trial.

Did KEEPS find any difference between the pill and the patch?

Both delivery routes improved symptoms, mood, and bone density. The oral CEE arm showed a stronger mood improvement signal and the borderline CAC benefit, while the transdermal arm had a theoretically lower VTE risk profile. Neither was clearly superior overall.

How do KEEPS results apply to women starting HRT today?

KEEPS supports the current clinical consensus: for healthy women within 10 years of menopause, low-dose HRT is a reasonable option for symptom management. It should not be prescribed for heart disease prevention. The 2022 NAMS position statement reflects this interpretation.

Were there any breast cancer signals in KEEPS?

No increase in breast cancer was observed over the four-year trial period. The trial was not large enough or long enough to rule out a small long-term risk. Breast cancer risk with HRT appears to depend on duration of use, type of progestogen, and individual factors.

What trial came after KEEPS to further test the timing hypothesis?

The ELITE trial (Early versus Late Intervention Trial with Estradiol, published 2016) directly compared early vs. late initiation of oral estradiol. It found CIMT benefit only in the early-start group, providing stronger evidence for the timing hypothesis than KEEPS alone.

Should I use KEEPS results to decide about HRT for myself?

KEEPS is one piece of a larger evidence base. Discuss your specific symptom burden, cardiovascular risk factors, family history, and preferences with your clinician. KEEPS is most relevant if you are within a few years of menopause and considering low-dose therapy for symptom relief.

KEEPS Trial: A Plain-English Overview of What It Established

By HealthRX.com Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

Clinical image for KEEPS Trial: A Plain-English Overview of What It Established Image: HealthRX.com clinical image

At a glance

| Detail | Value | |---|---| | Full name | Kronos Early Estrogen Prevention Study (KEEPS) | | N | 727 | | Population | Healthy women aged 42-58, within 6-36 months of last menstrual period | | Intervention arm 1 | Oral conjugated equine estrogens (o-CEE) 0.45 mg/day + cyclic micronized progesterone 200 mg x 12 days/month | | Intervention arm 2 | Transdermal 17-beta-estradiol (t-E2) 50 mcg/day + cyclic micronized progesterone 200 mg x 12 days/month | | Comparator | Matching placebo patches and pills | | Duration | 48 months | | Primary endpoint | Rate of change in carotid artery intima-media thickness (CIMT) | | Key result | No significant between-group difference in CIMT progression; improvements in mood, vasomotor symptoms, bone density, and sexual function in both active arms | | Published | 2014, Annals of Internal Medicine |

Why KEEPS Existed in the First Place

By the mid-2000s, millions of women had stopped hormone therapy almost overnight. The Women's Health Initiative (WHI) had reported in 2002 that combined estrogen-progestin therapy raised the risk of breast cancer, blood clots, and stroke. Prescriptions fell by roughly 80% within a few years.

But a persistent question lingered. The average WHI participant was 63 years old, well past menopause. Many clinicians suspected that the cardiovascular harms of HRT were a consequence of starting hormones too late, when atherosclerosis was already entrenched. Start them earlier, the thinking went, and estrogen might actually protect the arteries. This became known as the "timing hypothesis."

KEEPS was designed to test exactly that. The investigators recruited women who were within three years of their final period (mean age 52.7), gave them lower doses of hormones than those used in WHI, and tracked a subclinical marker of atherosclerosis (Harman et al., 2014).

Who Got In (and Who Didn't)

Eligibility criteria shaped the trial population in ways that matter for interpretation:

Age 42-58, with last menstrual period 6 to 36 months prior
No prior cardiovascular events, no diabetes requiring medication, no uncontrolled hypertension (systolic >150 or diastolic >95)
No current or recent hormone therapy use (washout required)
Baseline mammogram and endometrial assessment clear

These filters produced a cohort of comparatively healthy, younger postmenopausal women. That was the point: KEEPS wanted to study the window of opportunity before significant vascular damage accrued. But it also means results cannot be directly extrapolated to older, sicker populations.

A total of 727 women were randomized across nine clinical centers in the United States. Roughly 220-230 women were assigned to each of the three arms (o-CEE, t-E2, placebo).

What They Were Given

Both active arms received micronized progesterone (Prometrium 200 mg) for 12 days per month to protect the endometrium. The estrogen component differed:

Oral CEE arm: 0.45 mg/day of conjugated equine estrogens (half the dose used in WHI's 0.625 mg arm)
Transdermal estradiol arm: 50 mcg/day patch delivering 17-beta-estradiol

This design allowed a head-to-head comparison of two delivery routes and a comparison of each against placebo. The lower doses reflected prescribing trends that had already shifted toward "the lowest effective dose," in line with updated FDA labeling for menopausal hormone products.

The KEEPS Decision Matrix: How to Read This Trial's Results

One challenge with KEEPS is that the headline finding (no CIMT benefit) can obscure the secondary findings that changed clinical thinking. The table below maps each outcome category to what the trial actually showed and what it did not.

| Outcome domain | What KEEPS showed | What KEEPS did NOT show | |---|---|---| | Arterial wall thickness (CIMT) | No difference between either HRT arm and placebo over 48 months | That HRT is harmful to arteries when started early | | Coronary calcium (CAC) | o-CEE arm trended toward less CAC accumulation (p = 0.05 in per-protocol analysis) | A definitive cardioprotective benefit | | Vasomotor symptoms | Both active arms reduced hot flashes substantially | That one delivery route was clearly superior for symptom control | | Mood and depression scores | Both arms improved mood; o-CEE showed a stronger signal | That HRT should be prescribed as antidepressant monotherapy | | Sexual function | Improvements in desire and reduced pain in both arms | Long-term sexual health outcomes beyond 4 years | | Bone density | Increases in spine and hip BMD in both arms | Fracture reduction (trial was not powered to detect fractures) | | Breast cancer | No increase over 4 years | Long-term safety beyond the trial window | | Venous thromboembolism | No significant increase in either arm | That low-dose HRT carries zero clotting risk at scale |

This framework matters because KEEPS is often cited as either "HRT doesn't help the heart" or "HRT is safe early on." Both readings are incomplete. The more precise takeaway: in early postmenopause, low-dose HRT does not measurably slow arterial thickening over four years, but it also does not cause the harms that dominated the WHI headlines, and it produces real quality-of-life benefits.

Primary Endpoint: Carotid Intima-Media Thickness

CIMT is measured by ultrasound. It reflects the combined thickness of the inner two layers of the carotid artery wall. Increasing thickness correlates with atherosclerotic risk, though it is an imperfect surrogate for clinical events like heart attacks.

Over 48 months, the rate of CIMT progression was virtually identical across all three groups. The annualized rate was approximately 0.007 mm/year regardless of treatment assignment. Neither oral CEE nor transdermal estradiol slowed the thickening compared to placebo (Harman et al., 2014).

Several explanations have been proposed:

Healthy-participant effect. Baseline CIMT values were low. There may not have been enough progression to prevent.
Insensitivity of CIMT. Coronary artery calcium (measured by CT) may be a better atherosclerosis surrogate than carotid ultrasound.
Dose and duration. Four years of low-dose therapy may be too little time to produce a detectable vascular signal.
Sample size. With 727 women, the trial had limited power to detect small CIMT differences.

Secondary Endpoint That Made Headlines: Coronary Artery Calcium

The CAC sub-study told a more interesting story. Women in the o-CEE group showed lower accumulation of coronary calcium than the placebo group, with a borderline-significant p-value of 0.05 in per-protocol analysis. The transdermal arm did not show this trend.

This result was hypothesis-generating, not definitive. It suggested that oral estrogens (which undergo first-pass hepatic metabolism) might influence lipid-mediated pathways in ways that transdermal delivery does not. Follow-up data from the ELITE trial later provided stronger support for the timing hypothesis, showing that oral estradiol reduced CIMT progression when started within six years of menopause but not when started ten or more years after.

Quality-of-Life Outcomes: Where KEEPS Had Its Clearest Signal

The symptom and quality-of-life data from KEEPS were unambiguous and often get overlooked in cardiovascular-focused summaries.

Vasomotor symptoms. Both the o-CEE and t-E2 arms reduced hot flash frequency and severity compared to placebo. This was expected and confirmed that low-dose regimens remain effective for symptom management.

Mood. Women in the o-CEE arm showed statistically significant improvements in depression and anxiety scores. The transdermal arm trended in the same direction but did not reach significance. This finding has been cited in subsequent Endocrine Society guidelines as supporting evidence that HRT can improve mood during the menopausal transition.

Sexual function. Both active arms improved sexual desire scores and reduced dyspareunia compared to placebo. These outcomes are often the primary reason women seek HRT, and KEEPS confirmed that low-dose regimens deliver on that front.

Bone mineral density. Both arms increased BMD at the spine and hip. KEEPS was not designed or powered to detect fracture prevention, but the BMD improvements were consistent with what other trials of HRT have shown (Cauley et al., 2003).

Limitations the Authors Acknowledged

The KEEPS publication was transparent about several constraints:

No clinical cardiovascular endpoints. CIMT and CAC are surrogate markers. The trial could not answer whether early HRT prevents heart attacks or strokes.
Short duration. Four years may not be long enough to see vascular effects in a low-risk population.
Small sample size. Seven hundred twenty-seven women cannot detect rare events like VTE or breast cancer with statistical confidence.
Healthy cohort. The strict exclusion criteria mean results apply to healthy, recently menopausal women, not to the broader postmenopausal population.
High placebo response. Quality-of-life endpoints are susceptible to placebo effects, though the between-group differences were still significant.
Adherence. Roughly 25-30% of participants discontinued their assigned treatment before 48 months, a rate that complicates intention-to-treat analysis.

What KEEPS Changed in Clinical Practice

KEEPS did not produce a single practice-altering result. Its influence was more cumulative. Combined with ELITE and the WHI age-stratified reanalyses, it helped shift professional consensus toward three positions that are now reflected in the 2022 North American Menopause Society (NAMS) position statement:

HRT initiated within 10 years of menopause or before age 60 does not carry the same cardiovascular risk profile reported in WHI.
Low-dose regimens are effective for symptom management with a favorable short-term safety profile.
The choice between oral and transdermal delivery should be individualized based on risk factors (e.g., transdermal preferred in women with elevated VTE or metabolic risk).

KEEPS did not prove the timing hypothesis. But it removed one of the main objections to it: the fear that any HRT, at any time, would accelerate cardiovascular disease.

The Bottom Line for Patients

If you are a woman in early menopause considering hormone therapy, KEEPS tells you three things directly. Low-dose estrogen (oral or patch) will likely improve your hot flashes, mood, sexual comfort, and bone density. It will not, based on four years of data, cause the heart and cancer scares that made headlines in 2002. And it will not, on its own, protect your arteries in any measurable way. The decision to use HRT should rest on symptom burden and individual risk, not on hopes of cardiovascular prevention.

Frequently asked questions

›

References

Harman SM, Black DM, Naftolin F, et al. Arterial imaging and atherosclerosis in recently menopausal women: a randomized trial. Ann Intern Med. 2014;161(4):249-260. PubMed
Hodis HN, Mack WJ, Henderson VW, et al. Vascular effects of early versus late postmenopausal treatment with estradiol (ELITE trial). N Engl J Med. 2016;374(13):1221-1231. PubMed
Cauley JA, Robbins J, Chen Z, et al. Effects of estrogen plus progestin on risk of fracture and bone mineral density: the Women's Health Initiative randomized trial. JAMA. 2003;290(13):1729-1738. PubMed
The 2022 Hormone Therapy Position Statement of The North American Menopause Society. Menopause. 2022;29(7):767-794. PubMed
Stuenkel CA, Davis SR, Gompel A, et al. Treatment of symptoms of the menopause: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2015;100(11):3975-4011. PubMed