It's never been easy sorting out the pros and cons of taking estrogen supplements at menopause. Women have always had to weigh the many benefits -- reduced hot flashes, lower risk of heart disease, osteoporosis, colon cancer, and perhaps Alzheimer's -- against the modest but distressing risks, notably an increased chance of breast cancer and blood clots.

But lately, with every new study, it's gotten more complicated.

About a month ago, researchers studying 37,000 Iowa women reported what looked like reassuring news: Hormone replacement therapy was not linked to an increase in many common types of breast cancer. But it was linked to a higher risk of other breast cancers, though these had better prognoses to begin with.

Then, about two weeks ago, California researchers reported that one of the new ``designer estrogens,'' a drug called raloxifene (Evista), reduced breast cancer risk by an impressive 76 percent in 7,705 older women with osteoporosis.

So should women now toss out their Premarin and switch to the new stuff?

It depends on many factors.

It's no longer just: Estrogen, yes or no? It's estrogen, for how long? It's should you start at 50, when hot flashes and bone loss are worst, then stop before breast cancer risk kicks in?

Should you switch to low-dose estrogen? Or wait until age 65 to start? Or take raloxifene instead, which, unlike estrogen, can be used without another hormone called progesterone and does not raise the risk of uterine cancer?

``We're getting more and more information, but that's raising more and more questions, though it also means we have more choices,'' says Dr. Nananda Col, an internist at New England Medical Center.

Next week in the Archives of Internal Medicine, Col's team will publish a sophisticated decision-analysis technology to guide hormone replacement decisions.

She says the first question every woman should ask is why she would take any hormone at menopause. If the answer is to control hot flashes and vaginal dryness, estrogen may be the best bet -- for a few years, until symptoms have subsided.

But if osteoporosis is your main concern, the equation may change, notes Dr. JoAnn Manson, co-director of women's health at Brigham and Women's Hospital.

Both raloxifene and estrogen increase bone density a little (1 to 2 percent over several years, with estrogen somewhat more effective than raloxifene). But since raloxifene decreases breast cancer risk, it may be a better choice for women who are particularly worried about getting the disease.

On the other hand, if you don't want to take any estrogenic hormones, there are other options (besides exercise, which women should do anyway.) The prescription pill Fosamax and a nasal spray called Miacalcin can boost bone density. (And San Francisco researchers recently reported that parathyroid hormone does, too, but it's still experimental.)

The key point is that the most rapid bone loss occurs right at menopause, so women should not put off a decision on treatments to prevent osteoporosis. If you do choose to skip hormones or other drugs, it's important to get enough calcium -- about 1,500 milligrams a day -- and 400 to 800 IUs of vit D, too.

Suppose, though, that heart disease is your major concern. Then estrogen may be a good choice. But if you don't want to take it and your cholesterol is high, you have another option: ``statins,'' the cholesterol-lowering drugs.

For its part, estrogen raises HDL, or ``good'' cholesterol, lowers LDL, or ``bad'' cholesterol, and keeps total cholesterol under control. It also seems to prevent heart attacks. By contrast, raloxifene helps with total cholesterol, but not HDL.

Okay, say amidst all this, you're also petrified of breast cancer, as many women are, even though women tend to vastly overestimate their risk of breast cancer and underestimate their risk of heart disease, which kills six times as many women.

If you're at high risk for breast cancer, it makes sense to talk to your doctor about taking tamoxifen, a kind of anti-estrogen that appears to help prevent the disease.

And if you want to take regular estrogen, consider the following. In 1997, British researchers combined data from 51 studies and concluded that estrogen raises breast cancer risk about 35 percent if taken for five years or more.

But how much of an increase is that, really? ``Not very much for the average woman,'' says Dr. Eric Winer, director of breast oncology at Dana-Farber Cancer Institute.

If you're a 50-year old woman, your risk of developing breast cancer in the next 10 years is 2.67 percent. Taking hormones for five years raises this by 35 percent, to 3.6 percent. In other words, Winer says, ``it doesn't make sense to worry excessively if you take it for a couple of years to fight hot flashes.'' Furthermore, breast cancers are not all equally likely to be fatal or to be spurred by estrogen supplements.

In fact, one of the most puzzling things about the apparently reassuring Iowa study, is that hormone use was linked to some types of breast cancer but not others. The study of 37,000 women is the first to analyze cancer risk and hormone use by the type of breast tumor, notes Thomas A. Sellers, a co-author and epidemiologist at the Mayo Clinic.

But the results don't tally up neatly. The researchers found the greatest risk of cancer among hormone users was in tumors with the most favorable prognosis, that is, tumors that under the microscope are classified as tubular, mucinous, or papillary.

Many of these tumors are estrogen-driven, but so are many tumors with less favorable prognoses.

The Iowa study also doesn't fit with the overall evidence, notes Manson. In fact, not only is there a 35 percent increased risk of getting breast cancer if you take hormones for five or more years starting at menopause, there's a comparable increased risk of dying of it if you take hormones for 10 years or more.

That suggests, Manson says, that minimizing the duration of hormone use is a wise idea. One approach is to take hormones for five years or less at onset of menopause, then start again at 65, when the risk of heart disease begins to rise more steeply.

``Another strategy is to use hormones for less than five years around menopause, then switch to raloxifene,'' though follow-up data on raloxifene only extends about three years.

It may also make sense, she says, to start on low-dose estrogen (0.3 milligrams a day) instead of the usual dose (0.625 mg a day). Some studies have shown that low-dose estrogen protects bones as well as the standard dose. It's less clear how well lower dose estrogen affects cholesterol.

If you do try lower-dose estrogen, however, keep track of potential bone loss with bone density testing, says Dr. Isaac Schiff, chief of the Vincent Obstetrics and Gynecology service at Massachusetts General Hospital.

And what about waiting to take estrogen until you're 60 or 65, when preventing heart disease becomes more of a concern? It depends on your risk profile. If you have a family history of heart disease, if you're sedentary, overweight, diabetic, have high cholesterol and smoke, you are at higher risk of heart disease regardless of age, so it makes sense to consider estrogen.

The bottom line is that, because the research is rapidly evolving, read all you can, talk to your doctor and remember that you can -- and should -- keep re-assessing your decision as your health status, and age, change.