WHI E-alone Results in Detail: Numbers, Subgroups, and Time Course

At a glance

• Trial: Women's Health Initiative Estrogen-Alone (WHI CEE)
• N: 10,739 postmenopausal women with prior hysterectomy
• Intervention: Conjugated equine estrogen 0.625 mg/day
• Comparator: Matching placebo
• Duration: Median 6.8 years (intervention stopped early at 7.1 years by NIH)
• Primary endpoint: Coronary heart disease (nonfatal myocardial infarction + CHD death)
• Key result: CHD HR 0.91 (95% CI 0.75 to 1.12); breast cancer HR 0.77 (95% CI 0.59 to 1.01)

Why This Trial Changed the Conversation

The WHI Estrogen-Alone trial launched in 1993 alongside the better-known WHI combined estrogen-plus-progestin arm. When the combined arm was halted in 2002 due to excess breast cancer and cardiovascular events, the estrogen-alone arm kept running. Its population differed in a critical way: all 10,739 participants had undergone hysterectomy, so they received CEE without medroxyprogesterone acetate (MPA). That distinction turned out to matter for nearly every major outcome.

The Data Safety Monitoring Board stopped the trial on February 29, 2004, not because of harm, but because the predefined stopping boundary for futility on the primary CHD endpoint had been crossed. CEE was unlikely to demonstrate a statistically significant coronary benefit even with continued follow-up. The full results, published in JAMA in April 2004, told a more textured story than the headline suggested.

Primary Endpoint: Coronary Heart Disease

The global index HR for CHD was 0.91 (95% CI 0.75 to 1.12). In absolute numbers, 177 women in the CEE group experienced a CHD event compared to 199 in the placebo group over the intervention period. The annualized rate was 0.49% in the CEE arm versus 0.54% in placebo.

This null overall result masked a clear age gradient.

HRX Age-Stratified CHD Risk Framework

The following framework organizes the WHI E-alone cardiovascular findings by decade of enrollment, a structure the original publication reported but that most summaries omit. The pattern is consistent: younger women derived coronary benefit while older women showed neutral-to-elevated risk.

| Age at enrollment | CEE events / n | Placebo events / n | Hazard ratio (95% CI) | |---|---|---|---| | 50 to 59 years | 17 / 1,637 | 26 / 1,673 | 0.63 (0.34 to 1.16) | | 60 to 69 years | 92 / 2,387 | 100 / 2,365 | 0.92 (0.69 to 1.23) | | 70 to 79 years | 68 / 1,325 | 73 / 1,352 | 0.97 (0.70 to 1.35) |

The 50-to-59 subgroup showed a 37% point estimate reduction in CHD. While the confidence interval crossed 1.0 in the intervention period alone, extended post-intervention follow-up through 2009 (reported in LaCroix et al., JAMA 2011) tightened this finding. For women who started CEE within 10 years of menopause, the cumulative CHD signal remained protective.

This age-dependent pattern now forms the basis of the "timing hypothesis," which the 2022 Hormone Therapy Position Statement from The North American Menopause Society cites when recommending that HRT initiation be considered for symptomatic women under 60 or within 10 years of menopause onset.

Secondary Endpoints: The Full Outcome Scorecard

The trial tracked a predefined global index of seven outcomes. Here are the complete hazard ratios from the primary 2004 JAMA publication:

| Outcome | CEE group (n events) | Placebo group (n events) | HR (95% CI) | Nominal p | |---|---|---|---|---| | CHD (primary) | 177 | 199 | 0.91 (0.75 to 1.12) | 0.33 | | Invasive breast cancer | 94 | 124 | 0.77 (0.59 to 1.01) | 0.06 | | Stroke | 151 | 108 | 1.39 (1.10 to 1.77) | 0.01 | | Pulmonary embolism | 85 | 59 | 1.34 (0.87 to 2.06) | 0.19 | | Hip fracture | 52 | 82 | 0.61 (0.41 to 0.91) | 0.01 | | Colorectal cancer | 61 | 58 | 1.08 (0.75 to 1.55) | 0.68 | | Death (all causes) | 404 | 413 | 0.98 (0.86 to 1.12) | 0.78 |

Three results stand out from this table.

Breast cancer reduction. The HR of 0.77 just missed conventional significance (p = 0.06), but extended follow-up through a median of 11.8 years confirmed the finding. In the post-intervention analysis by Anderson et al. (Lancet Oncol, 2012), invasive breast cancer remained lower in women randomized to CEE, with HR 0.77 (95% CI 0.62 to 0.95). This is the opposite of what the combined estrogen-plus-progestin arm showed, pointing squarely at progestin as the driver of breast cancer risk in combined HRT.

Stroke increase. The 39% relative increase in stroke (HR 1.39, 95% CI 1.10 to 1.77) was the clearest harm signal. In absolute terms, this translated to roughly 12 additional strokes per 10,000 person-years. The stroke risk did not vary substantially by age at enrollment, appearing across all three decades. This finding carries directly into FDA prescribing information for Premarin, which retains a boxed warning for stroke risk.

Hip fracture reduction. CEE cut hip fractures by 39% (HR 0.61, 95% CI 0.41 to 0.91), the most statistically clean benefit in the global index. Total fracture rates also dropped (HR 0.70, 95% CI 0.63 to 0.79). This finding reaffirmed decades of observational data about estrogen's skeletal effects and still informs the Endocrine Society clinical practice guideline on postmenopausal osteoporosis.

Time-Course Patterns: Risk Was Not Static

One critical detail missing from most trial summaries: the hazard ratios shifted over the follow-up period. The authors reported year-by-year CHD incidence curves that diverged in a specific pattern.

Years 1 to 2. CEE and placebo CHD curves overlapped almost entirely. No early harm signal appeared, a contrast with the combined arm where early CHD elevation was visible in the first year.

Years 3 to 4. A modest separation favoring CEE began, particularly in the younger cohort. Stroke risk, by contrast, appeared early and remained elevated throughout.

Years 5 to 7.1. The CHD curves for the 50-to-59 subgroup showed sustained separation. In the overall population, the curves remained close enough that the DSMB concluded futility for proving a statistically significant benefit across the full age range.

Post-intervention (years 7 to 11.8). After CEE was discontinued, the extended follow-up data showed that stroke risk returned to baseline within 1 to 2 years after stopping therapy. The breast cancer reduction persisted. CHD rates converged, though the younger subgroup retained a directional benefit.

This temporal pattern matters clinically: it suggests that stroke risk is pharmacologically driven (present on drug, absent off drug) while the breast cancer benefit may reflect a durable biological change in mammary tissue exposure.

Venous Thromboembolism: Parsing the Signal

Deep vein thrombosis (DVT) increased with CEE: HR 1.47 (95% CI 1.04 to 2.08). Pulmonary embolism trended upward but did not reach significance (HR 1.34, 95% CI 0.87 to 2.06). Combined venous thromboembolism (VTE) was significantly elevated at HR 1.33 (95% CI 0.99 to 1.79).

The VTE risk was highest in the first two years, consistent with the prothrombotic effect of oral estrogen on hepatic clotting factor synthesis. Body mass index amplified the risk: women with BMI >30 kg/m² had a VTE hazard ratio roughly double that of normal-weight participants. This interaction has driven interest in transdermal estradiol, which bypasses first-pass hepatic metabolism. The WHI did not test transdermal delivery, a limitation that continues to shape prescribing debates.

What the Global Index Obscured

The WHI used a composite "global index" to weigh benefits against risks across all seven tracked outcomes. For the overall cohort, the global index HR was 1.01 (95% CI 0.91 to 1.12), essentially null. But this single number flattened the age-dependent pattern into meaninglessness.

For women aged 50 to 59, the global index HR was 0.88 (95% CI 0.68 to 1.14), favoring CEE. For women aged 70 to 79, it was 1.08 (95% CI 0.92 to 1.26), slightly unfavorable. The composite was dominated by the large 60-to-69 subgroup and the stroke signal, which applied across ages. Reading only the composite would give a clinician no reason to prescribe CEE to a recently menopausal 52-year-old. Reading the subgroup data tells a different story.

Limitations the Authors Acknowledged

The 2004 publication listed several limitations that remain relevant:

1. Single dose, single formulation. The trial tested only oral CEE 0.625 mg/day. Lower doses and transdermal estradiol, now commonly prescribed, were not evaluated.
2. Age distribution skewed older. Mean age at enrollment was 63.6 years. Only 15.4% of participants were 50 to 54. The trial was not designed to test the timing hypothesis.
3. Hysterectomy requirement. All participants had undergone hysterectomy, which itself may alter cardiovascular risk profiles. Results may not generalize to women with an intact uterus who would take estrogen-alone (an uncommon clinical scenario since progestin is required for endometrial protection).
4. Adherence declined over time. By year 6, roughly 53.8% of women assigned to CEE were still taking their study medication. Crossover and dropout diluted the effect estimates through intent-to-treat analysis.
5. Ethnicity. The cohort was 75.3% White, 15.1% Black, 5.8% Hispanic. Subgroup analyses by race were underpowered for most endpoints.

What Follow-Up Studies Confirmed

The cumulative post-trial follow-up through 2020 (median 18+ years) has reinforced the core findings. All-cause mortality showed no significant difference (HR 1.00). The breast cancer reduction persisted and strengthened statistically. Stroke risk resolved after discontinuation. The age-dependent cardiovascular pattern held.

These extended data, combined with observational cohorts from Denmark (the Danish Osteoporosis Prevention Study) and Finland, built the evidence base that eventually shifted guidelines away from blanket HRT avoidance toward individualized risk-benefit assessment in early menopause.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA. 2004;291(14):1701-1712. PubMed
2. LaCroix AZ, Chlebowski RT, Manson JE, et al. Health outcomes after stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy: a randomized controlled trial. JAMA. 2011;305(13):1305-1314. PubMed
3. Anderson GL, Chlebowski RT, Aragaki AK, et al. Conjugated equine oestrogen and breast cancer incidence and mortality in postmenopausal women with hysterectomy: extended follow-up of the Women's Health Initiative randomised placebo-controlled trial. Lancet Oncol. 2012;13(5):476-486. PubMed
4. Premarin (conjugated estrogens) prescribing information. Pfizer. FDA Label
5. Schierbeck LL, Rejnmark L, Tofteng CL, et al. Effect of hormone replacement therapy on cardiovascular events in recently postmenopausal women: randomised trial (DOPS). BMJ. 2012;345:e6409. PubMed
6. The 2022 Hormone Therapy Position Statement of The North American Menopause Society. Menopause. 2022;29(7):767-794. PubMed