Although women's health is rapidly becoming part of the mainstream of medicine, this update focuses on important and provocative subjects: domestic violence, medical abortion, nutrition, hormonal therapy, the problems of elderly women, and issues of women in medical academic careers.

Domestic Violence

Domestic violence is now recognized as a major hazard to the health of women. Through both the medical literature and the mass media, Americans have become aware of the frequency with which women are abused.

As many as 4 million women in the United States are physically abused each year. Although it is known that medical services are often used after an episode of abuse, the characteristics of battered women who present to physicians have only recently been examined. One risk factor, household crowding, may account for the higher rates of violence seen among black Americans. Domestic violence, however, crosses racial lines.

The Characteristics of Battered Women Seen in Primary Care

McCauley J, Kern DE, Kolodner K, Dill L, Schroeder AF, DeChant HK, et al. The "battering syndrome": prevalence and clinical characteristics of domestic violence in primary care internal medicine practices. Ann Intern Med. 1995; 123:737-46.

This study focused on the prevalence of and risk factors for domestic violence among women who were mostly white, middle-class, married, and medically insured.

Women visiting four primary care internal medicine practices in the Baltimore area were surveyed. Of the 3203 patients seen during the survey period, 2392 were recruited. A patient was considered to be a current victim of domestic violence if she answered "yes" to either of the following questions: "Within the last year, have you been hit, slapped, kicked, or otherwise physically hurt by someone?" or "Within the last year, has anyone forced you to have sexual activities?" and if she identified the abuser as a husband, ex-husband, boyfriend, or relative.

Of the women surveyed, 108 (5.5%) reported having experienced domestic violence in the previous year. About half of these had experienced "severe" violence—they were threatened or hurt with a weapon; burned; choked; or hit, kicked, or hurt with resultant broken bones, head injuries, or internal injuries. When women were asked if they had experienced violence in their lifetime, 639 (32.7%) answered "yes"; 308 (15.8%) women reported having been sexually abused as an adolescent or child. Risk factors for current domestic violence and the relation between these risk factors and current violence are shown in Table 1.

These data bear out the results from previous studies that had less generalizable findings. The risk factors are important but are neither sensitive nor specific. Therefore, because domestic violence is so prevalent and because physicians have the ability to address it, we feel that universal screening for violence is appropriate for all women.

Violence against women is a complicated problem. Male perpetrators seek power and control by intimidating women, isolating them from social support, instilling in them low self-esteem through emotional abuse, controlling their economic status, and threatening to harm their children. This can account for the real and perceived threats that can cause a woman not to leave an abusive relationship. In reality, leaving can be a dangerous thing to do. Planning to leave involves overcoming real barriers: finding money on which to live, obtaining transportation away from the home, and finding a safe place to go. These barriers are often complicated by the woman's depression and sometimes by her hopes for change: An episode of domestic violence is often followed by contriteness on the part of the perpetrator. However, violence tends to worsen over time.

The physician's role is to diagnose domestic violence and then offer support Table 2 [1]. The physician must let the patient know that violence against her is unacceptable and, because escape can be so difficult, must be careful not to shame or blame her for being in a violent situation. Physicians must also abandon a "quick-fix" attitude and use patience instead. They should leave the door open so that patients can return for continued support and future conversations. Finally, referring victims of abuse to local domestic violence services is critical.

Domestic violence is seen with remarkable frequency in primary care internal medicine practice. All women should be asked about domestic abuse, and physicians should treat abuse as a major potential health hazard to all women.

The Reproductive Years

Women's reproductive years are marked by unique health issues. Many of these issues have been delegated to obstetricians and gynecologists, but several must be addressed by primary care physicians. Among these are new possibilities for medically induced abortion and the need for good nutrition for women who are considering having children.

In recent years, access to abortion services has grown more difficult. First, the practice of surgical abortion has become concentrated in cities and away from hospitals. In 1973, 46% of abortions were done in freestanding clinics. By 1988, that Figure had increased to 86%. This means that abortion services are isolated from routine medical care. Second, there is an increasing shortage of physicians trained to do first-trimester surgical abortions Table 3 [2-4].

Surgical abortion is a relatively safe procedure in the United States; the rate of major complications is about 0.1%. In developing countries, however, surgical abortion accounts for as many as 19% of maternal deaths [5]. Because of these factors, medical methods with which to terminate pregnancy have been developed. In Europe, mifepristone (RU486) is the mainstay of medical abortion, and as many as 46% of French women seeking abortions use it. In the United States, however, the drug remains illegal. As a result, other drugs—including methotrexate—have been used.

Methotrexate and Misoprostol Were Effective in Terminating Early Pregnancy

Hausknecht RU. Methotrexate and misoprostol to terminate early pregnancy. N Engl J Med. 1995; 333:537-40.

This study sought to evaluate the efficacy of a single low dose of methotrexate followed by intravaginal misoprostol for the medical termination of early pregnancy.

Of 209 women who were seeking to terminate a pregnancy, 178 were eligible for the study and agreed to be included. Each had an intrauterine pregnancy of no more than 63 days' gestation as measured by ultrasonography, and none had known liver or kidney disorders, asthma, hematologic conditions, or emotional instability. At the first visit, methotrexate (50 mg/m² body surface area) was administered. Patients returned 5 to 7 days later for the intravaginal placement of misoprostol, 800 µg. Acetaminophen with codeine was prescribed for pain. Seven days later, patients had pelvic examinations and ultrasonography. If incomplete abortion or a retained gestational sac was found, vacuum aspiration or repeated administration of misoprostol was offered. Successful abortion was defined as a complete termination of pregnancy within 7 days after the first or second administration of misoprostol.

Abortion was successful in 153 women (86%) after a single administration of misoprostol. One hundred seventy-one women (96%) had a successful abortion after two doses of misoprostol. The onset of heavy vaginal bleeding and cramps occurred less than 6 hours to 4 days after the administration of misoprostol and occurred within 24 hours in 88% of women. Bleeding persisted for 1 to 7 days but lasted for 4 days or less in 98% of women. In other studies of these drugs, however, bleeding has lasted longer [6]. Three fourths of women rated their pain as 2 or less on a scale of 0 to 4. Other symptoms included stomatitis without hematologic changes, diarrhea, and fever (in one woman). Sixty-three percent of women had previously had abortions; they "overwhelmingly" preferred the medical option.

The combination of methotrexate and misoprostol is safe and effective for the termination of early pregnancy, and these drugs are currently available in the United States. The effectiveness of these two drugs does seem to decrease, however, as gestation progresses. The cost of this type of medical abortion is about the same as that of surgical abortion because of the need for ultrasonography.

Perhaps the greatest advantage of medical abortion is that a woman can choose to terminate her pregnancy within the context of day-to-day medical care; she is not forced to go to a freestanding abortion clinic. Moreover, the physician need not be an obstetrician or gynecologist. However, the physician must know how to interpret ultrasonographic data to be able to deal with the few incomplete abortions that do occur. Special training is now being offered to all physicians through the National Abortion Federation (1436 U Street NW, Suite 103, Washington, DC 20009; 202-667-5881).

Nutrition before Pregnancy

Another important issue for internists concerns the woman who is seeking to have a healthy child. Because it is critical to remember that nutritional benefits are maximized if they are present at the time of conception, all women who are seeking to become pregnant need counseling about nutritional factors.

The best-studied of these nutritional factors is folate intake [7, 8]. High intake of folate dramatically decreases the incidence of neural tube defects, which occur during the first 2 weeks of pregnancy—usually before the pregnancy is recognized. The Centers for Disease Control and Prevention now recommend folate supplementation. An intake of 0.4 mg per day is adequate, and this is the amount typically included in a multivitamin tablet.

Calcium intake has been shown to reduce the incidence of pregnancy-induced hypertension and preeclampsia [9]. Current recommended daily allowances are about 1200 to 1600 mg per day for young women and 1200 to 1500 mg per day for pregnant women. However, surveys have shown that the typical young woman consumes only about 600 to 700 mg per day. Calcium supplementation in the form of a multivitamin is frequently needed to supplement the average U.S. diet.

In contrast, vitamin A, a member of the retinol family, produces adverse effects at high doses [10]. Doses in excess of 10 000 IU/d (the amount present in two prenatal vitamin tablets) are associated with several birth defects, including craniofacial abnormalities, central nervous system defects, and hearing defects. Vitamin A is also present in high doses in liver; thus, women hoping to become pregnant should probably avoid eating this food.

Zinc Supplementation Improved Pregnancy Outcome

Goldenberg RL, Tamura T, Neggers Y, Copper RL, Johnston KE, DuBard MB, et al. The effect of zinc supplementation on pregnancy outcome. JAMA. 1995; 274:463-8.

Zinc represents an intriguing new nutritional finding. Zinc catalyzes more than 200 human enzymes and serves as a structural component to proteins, hormones, and nucleotides. It is present in high levels in seafood, meat, nuts, and milk. Studies done in animals have shown that a zinc-deficient diet is associated with central nervous system malformations.

Five hundred eight medically indigent but otherwise healthy pregnant black U.S. women were enrolled in this randomized trial. All had plasma zinc levels that were lower than the median levels in the population, and all presented before 22 weeks' gestation. They were randomly assigned to receive either zinc supplementation (25 mg) or placebo in addition to a multivitamin that did not contain zinc. Plasma zinc concentrations were measured throughout the remainder of the pregnancies. Primary outcomes included the child's birthweight, gestational age, and cranial circumference. Subgroup analyses were done for women whose body mass indices were higher and lower than the median body mass index (26 kg/m²).

The women in the two groups had similar baseline characteristics, including zinc levels (9.6 µmol/L) and fetal growth variables as measured by ultrasonography before randomization. Compared with infants born to women in the control group, infants born to women in the zinc supplementation group were an average of 126 g heavier (P = 0.03) and had a mean cranial circumference that was 0.04 cm greater (P = 0.02). In women whose body mass indices were less than 26 kg/m² body surface area, zinc supplementation was associated with a mean infant birthweight that was 248 g higher and a mean infant cranial circumference that was 0.7 cm greater than in the control group. Zinc supplementation had no apparent adverse effects on mother or baby.

In women with relatively low plasma zinc concentrations, zinc supplementation is associated with enhanced infant birthweights and cranial circumferences; the effect occurs predominantly in women with low body mass indices.

Women of childbearing age, especially those seeking to become pregnant, should be advised to increase their intake of folate, calcium, and zinc and to avoid high doses of vitamin A. A well-balanced diet with an emphasis on milk products or a multivitamin supplement with 5000 IU of vitamin A or less would suffice.

Women in Academic Medicine

The status of women in academic medicine is marked by good news and bad news. The good news is that women now make up 20% of academic faculties, and a recent study showed that the salaries of female physicians were generally equal to those of men, after adjustment for specialty and practice setting [11].

The bad news is threefold. First, women continue to choose less lucrative specialty options. Second, black women comprise only 2% of all practicing physicians and, together with their black male colleagues, make up only 4% of all physicians—a Figure thathas remained stable since 1910. Third, although more women are now entering academic practice, their status remains disproportionately low.

Women Rank below Men in Academia

Tesch BJ, Wood HM, Helwig AL, Nattinger A. Promotion of women physicians in academic medicine. Glass ceiling or sticky floor? JAMA. 1995; 273:1022-5.

This cohort study attempted to assess possible explanations for the finding that the percentage of female medical school faculty members who hold the rank of associate or full professor remains lower than the percentage of male medical school faculty members who hold these ranks.

Male and female medical school faculty members who were first appointed between 1979 and 1981 were identified through the American Medical Colleges Faculty Roster System. Each female faculty member was matched with two male controls from the same faculty. Persons who remained on any medical school faculty in 1991 were surveyed; the response rate was 53.7%. The analyses included 416 faculty members: 153 women and 263 men. Respondents were asked about their present rank, training, academic resources, work and family responsibilities, publications, and extramural funding.

About 50% of men and women left academic medicine during the study period. After a mean of 11 years on the medical school faculty, 59% of women and 83% of men had reached the rank of associate or full professor (odds ratio, 0.28; 95% CI, 0.18 to 0.45). Five percent of women and 23% of men had achieved the rank of full professor (odds ratio, 0.19; CI, 0.09 to 0.40). Women and men reported having similar preparation for academic careers, including board certification, fellowship training, time devoted to research during fellowship, and a mentor. However, academic and funding resources—particularly grant support, office and laboratory space, and protected time for research—were less available to women during their first faculty appointments.

In their current positions, women worked about 10% fewer hours than men, devoted more of their time to clinical care, and were more likely to be in primary care specialties. Academic productivity, including number of years of extramural grant support and number of publications, was greater for men than for women. However, after adjustment for academic productivity factors, women were still significantly less likely to have been promoted to an associate or full professorship (odds ratio, 0.37; CI, 0.21 to 0.66).

Female physicians who are members of medical school faculties are promoted more slowly than their male colleagues. These differences in rank are not completely due to differences in productivity or attrition from academic medicine.

The Mature Years

Health care issues change as women age. The prevalence of violence decreases with age, and childbearing issues become less meaningful. But primary and secondary prevention issues grow more important. Among the major concerns are obesity, cardiovascular disease, osteoporosis, and falling.

Women's J-Shaped Mortality Curve Disappeared after Adjustment for Smoking

Manson JE, Willett WC, Stampfer MJ, Colditz GA, Hunter DJ, Hankinson SE, et al. Body weight and mortality among women. N Engl J Med. 1995; 333:677-85.

The U.S. Department of Agriculture estimates that the consumption of snack food has doubled in the past 15 years and that about one third of adults are now obese. Obese women outnumber obese men. Black and Hispanic women have especially high risks for obesity; prevalence rates have reached 70% in some studies [12]. Obesity is clearly associated with cardiovascular disease, hypertension, hypercholesterolemia, non-insulin-dependent diabetes, biliary tract disease, and cancers of the breast, colon, and endometrium.

However, a debate has continued for years on the health risks that may be presented by extreme leanness. Some studies have shown a higher mortality rate among extremely lean persons, but the flaws of these studies have included failure to account for smoking and failure to account for leanness caused by previous disease. This cohort analysis attempted to account for those confounders and to determine the relation between body weight and overall mortality rates.

A cohort of 115 195 women enrolled in the prospective Nurses' Health Study was followed for 16 years. Body mass index and mortality rates were measured, and participants responded to questionnaires every 2 years. During the 16 years of the study, 4726 of the women died: Eight hundred eighty-one died of cardiovascular disease, 2586 of cancer, and 1259 of other causes.

In analyses done only according to age and body mass index, a J-shaped mortality curve was produced. When smoking was factored out, however, no increase in mortality was seen among the leaner women. In multivariate analyses of women who had never smoked and who had recently had stable body weights—analyses from which the first 4 years of follow-up were excluded—the relative risks for death from all causes increased with increasing body mass index (P for trend less than 0.001) (Figure 1).

View larger version (16K): [in this window] [in a new window]   Figure 1. The influence of controlling for smoking on the shape of the curve describing the relation between body mass index and the relative risk for death from all causes. Reprinted with permission from Manson et al. N Engl J Med. 1995; 333:677-85.

Body mass index and all-cause mortality rates were directly related in these middle-aged women. The lowest mortality rates were seen among women who weighed at least 15% less than the U.S. average for women of similar age and among women whose body weights had been stable since early adulthood.

Obesity is a major health hazard, and physicians need to consider how to approach it. Four issues are important. First, because weight loss is extremely hard to sustain, expectations must be tempered. Some surveys have shown that as few at 5% of persons who lose weight keep it off for more than 5 years. A reasonable goal is to reduce an obese patient's weight by 10% to 15%, at most. Attempts to reach more ambitious goals are likely to fail. Second, because of the likelihood for weight rebound, obesity needs to be viewed as a chronic condition that should be monitored, like diabetes or hypertension. Third, as shown in some recent studies, drug therapy is becoming more important in the treatment of obesity. The best results will probably come from continual behavior modification combined with drug therapy. Finally, clinicians should be aware of promising research done in animals on such substances as leptin and glucagon-like peptide, research that may result in the development of more powerful antiobesity agents.

Behavior modification works best when it is combined with an exercise program and a low-fat diet tailored to each woman's personal dietary preferences. Because of the danger of rebound, behavior modification must continue after weight loss, and patients should probably attend sessions at least twice a month after they reach their target weights.

In the United States, three drugs are now approved for weight loss—phentermine, fenfluramine, and dexfenfluramine [13]. Phentermine is a norepinephrine reuptake inhibitor; fenfluramine and dexfenfluramine are serotonin reuptake inhibitors. These drugs are most effective when used in conjunction with behavior modification programs; their long-term adverse effects are not known. Reports have shown that a higher-than-expected frequency of pulmonary hypertension is associated with the use of fenfluramine and dexfenfluramine, but pulmonary hypertension is a rare complication [14]. Certainly, women who have imminent health risks associated with obesity, such as non-insulin-dependent diabetes, are good candidates for drug therapy. At present, the role of drugs for overweight women who are not obese is controversial.

No Definite Benefits Yet Shown from Dehydroepiandrosterone

Casson PR, Buster JE. DHEA administration to humans: panacea or palaver? Seminars in Reproductive Endocrinology. 1995; 13:247-56.

Obesity is frequently the subject of attention from the mass media. Another topic often covered by the press is the use of dehydroepiandrosterone (DHEA; 3 ß-hydroxy-5-androstene-17-one), which has been touted as a "youth drug." Some have claimed that it increases longevity; improves cognitive abilities; and decreases the incidence of cancer, heart disease, infection, diabetes, and osteoporosis. Sales of DHEA are increasing both in and outside of the medical community, and the content of the compound—which is made by individual pharmacies—is not standardized.

Dehydroepiandrosterone and its sulfated congener (DHEA-S) are steroids produced primarily in the adrenal glands. The hormones are present in high levels in women's plasma, but their clinical effects are unclear. These authors reviewed the existing evidence about these clinical effects, both in humans and in animals.

Observational studies have documented that DHEA levels peak when persons are in their twenties and decrease to 10% of peak by the time patients are in their seventies. Supplementation can restore the adrenal androgen levels to those of young adulthood without producing adverse androgenic effects.

Dehydroepiandrosterone appears to play a role in immune modulation, and treatment may reverse some of the decrease seen in immune response with age. One small, short-term trial done in postmenopausal women showed an increased number of T-lymphocyte killer cells and increased cytotoxicity. Dehydroepiandrosterone may be cardioprotective, but high doses in women have resulted in decreased high-density lipoprotein cholesterol levels and increased low-density lipoprotein cholesterol levels. Low-dose DHEA may attenuate age-related increases in insulin resistance. Evidence is sparse and inconsistent about the relation of DHEA to improved cognition, increased rapid eye movement (REM) sleep, self-reported improved well-being, and changes in bone metabolism.

Unsubstantiated claims and the widespread availability of DHEA in the United States have led to the increasing use of this hormone. Although the evidence about the clinical risks and benefits of DHEA is limited, preliminary results of studies done in animals and humans indicate that carefully designed investigations are warranted.

The increasing use of DHEA by persons in the United States emphasizes that health care providers need to stay familiar with the alternative therapies that their patients use. As many as 75% of patients who seek alternative healing methods have physicians who are unaware that they are doing so. Physicians need to learn as much as possible about the risks and benefits of alternative treatments to counsel their patients effectively and decrease the fragmentation of health care that results when patients turn to multiple providers.

Alendronate Prevented Fractures and Loss of Height in Women with Osteoporosis

Liberman UA, Weiss SR, Broll J, Minne HW, Quan H, Bell NH, et al. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. The Alendronate Phase III Osteoporosis Treatment Study Group. N Engl J Med. 1995; 333:1437-43.

Osteoporosis and the fractures associated with it are major causes of illness and death in older women. New developments in an old class of drugs seem to have brought an effective treatment for women with osteoporosis. Bisphosphonates are synthetic analogues of inorganic pyrophosphate, which is an endogenous regulator of bone turnover. Alendronate inhibits bone resorption more effectively than it inhibits bone mineralization. Preliminary studies have shown that alendronate has a positive effect on bone mineral density in women, but these studies did not specifically look at the incidence of fractures. This randomized clinical trial tested the effects of oral alendronate on bone mineral density and on the incidence of fractures and height loss.

Women were recruited by centers in the United States, Australia, Canada, Europe, Israel, Mexico, New Zealand, and South America. All had been menopausal for at least 5 years, and all had osteoporosis as defined by bone mineral density measurements. Women whose osteoporosis was caused by a known disease were excluded. A total of 881 women were randomly assigned to receive one of three regimens: placebo; alendronate, 5 or 10 mg daily for 3 years; or alendronate, 20 mg daily for 2 years and then 5 mg daily for 1 year. Primary end points were bone mineral density, occurrence of new vertebral fractures, progression of vertebral fractures, progression of vertebral deformities, and loss of height.

At 3 years, the mean (± SE) differences in bone mineral density between the women receiving 10 mg of alendronate daily and those receiving placebo were 8.8% ± 0.4% in the spine, 5.9% ± 0.5% in the femoral neck, 7.8% ± 0.6% in the trochanter, and 2.5% ± 0.3% in the total body (P < 0.001 for all comparisons). Overall, treatment with alendronate was associated with a 48% reduction in the percentage of women with new vertebral fractures (3.2% in the alendronate group compared with 6.2% in the placebo group; P = 0.028; number needed to treat, 33 for 3 years) and reduced loss of height. Adverse effects were similar in all groups. It should be noted that alendronate is currently indicated only for the treatment of established osteoporosis. It is important to carefully follow treatment guidelines because episodes of erosive esophagitis have recently been reported in women who have not taken alendronate exactly as recommended.

Nevertheless, daily treatment with alendronate progressively increases bone mass in the spine, hip, and total body. It reduces the incidence of vertebral fractures, the progression of vertebral deformities, and loss of height in postmenopausal women with osteoporosis.

Exercise Reduced the Incidence of Falls in Elderly Patients

Province MA, Hadley EC, Hornbrook MC, Lipsitz LA, Miller JP, Mulrow CD, et al. The effects of exercise on falls in elderly patients. A preplanned meta-analysis of the FICSIT trials. Frailty and Injuries: Cooperative Studies of Intervention Techniques. JAMA. 1995; 273:1341-7.

Each year, about 30% of persons older than 65 years of age sustain a fall. About 10% to 15% of falls result in serious injury, and as many as 1% result in hip fracture. This meta-analysis sought to determine whether short-term exercise reduces the incidence of falls and fall-related injuries in elderly persons.

A planned meta-analysis was done on the seven randomized, controlled FICSIT (Frailty and Injuries: Cooperative Studies of Intervention Techniques) clinical trials that were done in seven cities and that assessed the efficacy of exercise in reducing the incidence of falls and frailty in elderly patients. All of the trials included an exercise component that lasted 10 to 36 weeks. Follow-up on falls and injuries in the 2328 patients was obtained for as long as 4 years. Training was done in one or more of the following areas: endurance, flexibility, balance platform, tai chi (dynamic balance), and resistance. Several treatment arms included additional nonexercise components, such as behavioral training, changes in medication, education, functional activity, and nutritional supplements.

The adjusted fall risk ratios were 0.90 (CI, 0.81 to 0.99) for treatment arms that included general exercise and 0.83 (CI, 0.70 to 0.98) for those that included balance.

Treatments that include exercise reduce the risk for falls in elderly adults. Exercise seems to improve gait, gait speed, muscle mass, strength, reaction time, balance, and bone mass. It thus makes sense that it should provide a reduction in the incidence of falls and subsequent injuries. Improved gait speed allows for more forward motion, thus minimizing the risk for collapsing on a hip. Reaction time and strength allow a person to grab an object more effectively to prevent a fall. Increased muscle mass acts as a shock absorber when a fall occurs, and increased bone mass minimizes the risk for fracture if a person falls. In addition, exercise decreases depression and improves mood, functional status, and cardiovascular health [15-17].

Because exercise is not a salable product, physicians are the only ones likely to promote it. It is estimated that only about 38% of women exercise regularly. As with diet, exercise prescriptions must be specific and tailored to the particular needs of each patient, and goals must be set for each patient. For example, younger women profit most from aerobic exercise, whereas older women need to improve gait and build strength. To simply prescribe "exercise" is like prescribing "any antibiotic" for an infection.

Conclusions

Growth in the study and promotion of women's health has recently been rapid. Work done in this discipline has pointed out many important but seldom discussed hazards to the health of U.S. women, particularly those in vulnerable groups, such as the elderly, the poor, and members of minorities. For example, the press has given much attention to heart disease and breast cancer, but few recognize the disproportionate risks black women have for these conditions. Compared with white women, they are twice as likely to die from stroke, 3 times as likely to die from complications of diabetes, 5 times as likely to die as a result of violence, and 10 times as likely to die of complications of the acquired immunodeficiency syndrome.

The study of women's health raises fundamental questions about the effects on health of demographic and economic shifts in the United States. For example, 72% of persons 85 years of age and older are women, and the number of persons in this age group is expected to double by the year 2020. Women in this age group are usually poor; they make up about 75% of nursing home residents, and most of the rest live alone. Poverty and social isolation can adversely affect health by aggravating disability and chronic illness.

Poverty is another health-related issue that has been raised by advocates of women's health. It has increased dramatically, especially among women in minority groups. A major cause of this is changes in the structure of the family. In 1950, 90% of children were born to married couples. In the 1990s, about 50% of children are born to single mothers, most of whom are poor. Overall, working women still earn 30% less than men, and they tend to have jobs that are demanding and highly stressful and over which they have little control. Moreover, women are still the primary childrearers, housekeepers, and caregivers of elderly parents. In addition, 14 million women in the United States have no medical insurance.

A final health risk for women in the United States is the fragmentation of primary care. Women often see at least two specialists for primary care, such as an internist and a gynecologist, neither of whom may have been specifically trained to provide truly comprehensive services. This is borne out by surveys showing that 40% of women have changed physicians because of dissatisfaction, 17% have had their symptoms "trivialized" by a physician, and 30% have been "talked down to" by a physician.

We believe that every woman has the right to comprehensive and compassionate primary care. Toward that end, we believe that all primary care physicians, in addition to honing their listening skills with their female patients, should learn to more effectively use the routine examinations, diagnostic tests, and treatments that women should receive.

Dr. Ness: Graduate School of Public Health, University of Pittsburgh, 130 DeSoto Street, Pittsburgh, PA 15261.

Dr. Roberts (Series Editor): York Health System, York, PA 17403.