Synthesis, Findings, and Recommendations
As discussed in Chapter 1, the committee developed a series of questions to focus its deliberations and to ensure that it responded to its charge (see Box 1-4). In this chapter, the committee synthesizes the previous chapters of the report in response to those questions and presents its key findings and recommendations.
The committee approached women’s health as a concept that has expanded beyond a narrow focus on the female reproductive system to encompass other conditions that create a significant burden in women’s lives. The committee focused on health conditions that are specific to women, are more common or more serious in women, have distinct causes or manifestations in women, or have different outcomes or treatments in women. Numerous conditions could be included in such a list. The committee could not review all such conditions and, therefore, highlights a number of such conditions as examples that are specific to women because of differences in prevalence, severity, preferred treatment, or understanding or because the condition is prominent in women or there is a research need regarding women, whether or not there are sex differences.
IS WOMEN’S HEALTH RESEARCH STUDYING THE MOST APPROPRIATE AND RELEVANT DETERMINANTS OF HEALTH?
The committee noted the growing understanding of the full spectrum of determinants of health and illness that has resulted from research over the last 20 years. Determinants can be related to a woman’s biologic makeup, behaviors, and the social, cultural, and environmental contexts in which genetic vulnerabilities and individual traits and behaviors are developed and expressed. As discussed in Chapter 2, the committee concluded that more attention should be given to the
social and behavioral determinants of health than has been given in the past to achieve greater gains in women’s health.
The committee initially planned to discuss health determinants in relation to each separate condition under consideration. However, the pervasive impact of social and behavioral determinants became apparent in that the same risk factors (such as smoking, eating habits, lack of physical activity, sexual risk behavior, and alcohol use) played a role in most of the conditions under consideration; therefore, the committee reviewed them in a separate chapter and provided evidence of their associations with a variety of conditions. For example, there is substantial evidence of links between smoking, alone and in conjunction with oral contraceptive use, and breast cancer, lung cancer, and cardiovascular disease (Burkman et al., 2004). There is growing evidence, from large cohort studies of women, on the role of eating habits and physical activity in the health of women.
As highlighted in the ecologic model presented in Chapter 2, smoking, eating habits, physical activity, and other behaviors are shaped by cultural and social contexts, including factors associated with social disadvantage. The marked differences in condition prevalence and mortality in women who experience social disadvantage are associated with race and ethnicity, lack of education, low income, and other factors such as differential exposure to stressors and violence, which are more common in more disadvantaged communities. Such exposures are related to outcomes as varied as injury and trauma, depression, arthritis, asthma, heart disease, human immunodeficiency virus (HIV) infection, and other sexually transmitted infections (Campbell et al., 2002; Coker et al., 2000; Ozer and Weinstein, 2004; Tjaden and Thoennes, 1998). Although the impact of social and community factors has been documented, little research on how to modify these factors to improve women’s health has been conducted. Even less research has been conducted on the effects of those social and community factors in specific groups of women.
Greater support for research on social and behavioral determinants is needed, particularly research on how to modify them to improve health. Social and behavioral factors are important determinants of health for men as well as women, but the underlying factors probably differ by sex, and interventions tailored to women may be more effective than general treatments. For example, some studies show that women have a more difficult time with smoking cessation than men (Gritz et al., 1996) and face unique barriers to cessation (Schnoll et al., 2007). Smoking-cessation treatments designed to address barriers specific to women may be more effective, but they require empirical validation. Many intervention programs show promise, but most have not been well tested, and the health effects of smoking, such as cardiovascular disease and lung cancer, are still the leading killers of women. Similarly, interventions to improve eating habits, increase physical activity, decrease sexual risk behavior, and decrease substance abuse have not been as successful as they need to be to improve women’s health.
IS WOMEN’S HEALTH RESEARCH FOCUSED ON THE MOST APPROPRIATE AND RELEVANT DISEASES, DISORDERS, CONDITIONS, OUTCOMES, AND END POINTS?
Several issues surfaced with the assessment of the question of whether women’s health research has focused on the most appropriate and relevant health conditions. One was whether to look at progress in relation to women’s overall health and well-being rather than listing progress by condition. Although the committee preferred the former, much of the knowledge regarding women’s health has been conceptualized and conducted in relation to specific conditions. That approach reflects the organization of the National Institutes of Health (NIH), health-professions schools, specialties, and professional organizations. As a result, the committee examined progress in individual conditions in Chapter 3.
A second issue was the selection of conditions to discuss. Which conditions fall under the heading of women’s health, and which conditions should the committee discuss? A narrow view would include only conditions that are unique to women. Although those are important, such a narrow view defines women primarily by their reproductive function. The committee chose a broader perspective: to consider all conditions that affect women disproportionately, that have different risk factors or that present differently in women, or that are treated differently in women and men. Applying the broader definition includes almost all conditions. The committee could not review all such conditions and, therefore, highlights a number of such conditions as examples that are specific to women because of differences in prevalence, severity, preferred treatment, or understanding or because the condition is prominent in women or there is a research need regarding women, whether or not there are sex differences.
Chapter 3 reviews and categorizes the progress in some of the most prevalent and problematic of the conditions for women, and they are presented in Table 6-1. A large amount of research on those conditions has been conducted, and progress has been made, especially with respect to some of the leading killers of women. That progress, however, has not been seen for all conditions, in particular for nonfatal conditions that affect women’s quality of life. Women’s health research has been focused on some appropriate and relevant health conditions, but more conditions need to be studied. Some of the research findings and the progress that has been made against a number of conditions are summarized briefly below.
Conditions on Which Research Has Contributed to Major Progress
Through its review of the literature, the committee identified three conditions in which scientific research led to major progress with respect to improvements in prevention, diagnosis, or treatment that resulted in substantial reductions in incidence or mortality: breast cancer, cardiovascular disease, and cervical cancer.
Research advances in breast cancer resulted in a decrease in mortality over
TABLE 6-1 Conditions Discussed by Committee, Categorized by Extent of Progress
Conditions on Which Research Has Contributed to Major Progress
Conditions on Which Research Has Contributed to Some Progress
Conditions on Which There Has Been Little Progress
Maternal Morbidity and Mortality
Alcohol and Drug Addiction
Gynecological Cancers Other than Cervical Cancer
Non-Malignant Gynecological Disorders
the last 20 years. Consumer demand and involvement and increased funding activity spurred research that elucidated key elements of the underlying pathophysiology of breast cancer, that developed more sensitive detection methods, that increased identified biomarkers used in diagnosis and selection of targeted treatments, that characterized risk factors, and that improved treatment options. A substantial drop in breast-cancer incidence occurred following the reporting of the findings from the Women’s Health Initiative (WHI) and the later reduction in the use of menopausal hormone therapy, which had both harmful and beneficial effects. The WHI demonstrated that a given drug or treatment—in this case, conjugated equine estrogen with or without progestin—could have adverse effects on some conditions (for example, breast cancer, stroke, embolism, and gall bladder disease) while having beneficial effects on others (for example, reducing the risk of bone fracture) and that women needed to balance risks and benefits, taking into account a variety of outcomes.
Mammography provides another example of the need to balance risks and benefits. Recent mammography guidelines try to balance the benefit of early detection in reducing mortality with the risks of overtreatment and anxiety caused by false positives. The risk–benefit information is potentially useful to women in making decisions about the use of mammography, but problems in communicating the findings have decreased its impact.
Research on breast cancer at the molecular, cellular, and animal levels in con-
junction with clinical trials and observational studies in women have combined to improve outcomes, including quality of life after treatment. Disparities remain, however, in that black women have a higher mortality from breast cancer than white women despite a lower incidence; this highlights the need for research on groups of women that have the highest risks.
Although cardiovascular disease remains the leading cause of death of women and of men, women’s age-adjusted mortality from coronary heart disease declined from 263.3 to 134.4 deaths per 100,000 from 1980 to 2000 (Ford et al., 2007). The first major breakthrough was the recognition that women are susceptible to cardiovascular disease just as men are. Sex differences in the underlying physiology and manifestation of cardiovascular disease have been documented. Data from women-only studies have contributed to the knowledge base of cardiovascular disease in women. The decline in mortality from cardiovascular disease occurred first in men and later in women. About half the decline in women is attributable to behavior change, including a drop in smoking, and the other half is attributable to new clinical treatments that emerged from research, such as secondary preventive therapy and initial treatment for acute myocardial infarction or revascularization (Ford et al., 2007). Major progress has been made in increasing awareness of cardiovascular disease in women through increasing physician’s awareness of cardiovascular disease in women and through public-health campaigns. However, the long period during which cardiovascular disease was studied only in men has delayed greater progress. The lack of effective interventions for changing risk factors in women has limited decreases in the incidence of and mortality from cardiovascular disease in women.
Reductions in the incidence of and mortality from cervical cancer began in the 1960s and has continued as diagnosis and screening have improved further. In addition, a major transformation in research on cervical cancer and in the approach to cancer in general occurred with the development of a vaccine that prevents infection by several strains of human papilloma virus (HPV) that cause most cervical cancers. The vaccine was developed by using the findings of research on the basic biology of the virus and its relationship to cervical cancer in human cells and animals, and of studies of cervical cancer’s etiology. Success with that multipronged approach highlights how information from different types of research builds and combines to yield progress against a condition. Although overall gains have been seen in mortality from cervical cancer, rates remain higher in black and Hispanic women than in white and Asian women. The differences highlight the need to focus translation and communication activities to the populations at greatest risk for the condition.
Conditions on Which Research Has Contributed to Some Progress
The committee identified depression, HIV/acquired immunodeficiency syndrome (AIDS), and osteoporosis as conditions on which some progress has been made through women’s health research.
The incidence of depression is about twice as high in women as in men, and the consequences of depression, such as reduced educational attainment, are greater in women (Berndt et al., 2000). Some aspects of depression, such as postpartum depression, are more common to women. Advances have been made in the treatment of depression in the last 20 years (such as the development of combination norepinephrine-uptake inhibitors and serotonin-receptor antagonists with nonpharmaceutical interpersonal psychotherapy and cognitive–behavioral therapy) although their impact has not been maximized because of inadequate translation, particularly in relation to primary providers and the fragmentation of mental health services from other health-care delivery.
There have been advances in the treatment of HIV/AIDS in the last 20 years, including a decline in maternal–fetal transfer of HIV/AIDS because of treatment of pregnant women with antiretroviral agents. Much of the treatment research has been in men. The rapid development of treatments has benefited women despite the focus of the research on men, but the predominance of male-focused studies has limited some of the benefits for women. For example, issues related to the toxicity of treatments in women are only now being identified through women-based research. HIV/AIDS is becoming much more a women’s issue, especially in black women, who make up 64% of the women living with HIV/AIDS.
Over the last 20 years there have been advances in the knowledge of the basic science underlying osteoporosis and in the diagnosis and treatment of osteoporosis. That includes the identification of genes whose expression affects the risk of osteoporosis (Huang and Kung, 2006). Recent trends show a decrease in the incidence of hip factures in the United States and Canada (Brauer et al., 2009; Leslie et al., 2009). Osteoporosis remains, however, a condition that greatly impacts the quality-of-life of a large number of women, particularly as they age.
Conditions on Which Little Progress Has Been Made
The committee identified a number of conditions on which little progress has been made in reducing incidence, morbidity, or mortality. Unintended pregnancy and autoimmune disease are highlighted below as examples of conditions on which there has been little progress.
Effective contraceptives are available to prevent pregnancy, and research has identified several risk factors for unintended pregnancy, but rates of unintended pregnancy have remained steady for decades. That unintended pregnancy continues to occur at a high rate points to the need for research on how to improve knowledge of and compliance with contraceptive regimens, the need to develop
contraceptives that are more acceptable to cultures in which unintended pregnancy occurs with greater frequency, and the need for social and community-level interventions to decrease unintended pregnancy.
Another common condition in women that has seen little progress is auto-immune disease, which constitutes about 50 diseases. Most of the diseases are more common in women than in men, and as a group they are the leading cause of morbidity in women, greatly affecting quality of life. Despite their prevalence and morbidity and the availability of drugs to treat some of the symptoms, little progress has been made in understanding the conditions better, in identifying the risk factors, or in developing diagnostic tools, better treatments, or cures.
Looking at the set of conditions on which little progress has been made—including unintended pregnancy; autoimmune diseases; alcohol- and drug-addiction disorders; lung, ovarian, endometrial, and colorectal cancer; nonmalignant gynecologic disorders; and Alzheimer’s disease—the committee tried to identify characteristics or explanations for the lack of progress. Several of the conditions have not had much attention and research, but other conditions have had considerable research. For example, there has been little progress in lung cancer despite a large amount of research, including some research on sex differences. The committee identified a number of potential reasons for the lack of progress, including degree of attention from government agencies, consumer advocate groups, and Congress and resulting research funding; availability of interested researchers trained in a given field; adequacy of understanding of the underlying pathophysiology and natural history of a condition; adequacy of understanding of behavioral factors associated with decreased risk; availability of sensitive and specific diagnostic tests and screening programs to identify individuals who are at risk for or have a condition; and nonmedical barriers associated with political or social concerns. Many of those reasons played a role in connection with specific conditions, but none emerged as a potent force that affected all the conditions. The committee did note, however, that many of the conditions that showed less progress were associated with morbidity and not mortality.
IS WOMEN’S HEALTH RESEARCH STUDYING THE MOST RELEVANT GROUPS OF WOMEN?
Women are not a homogeneous group, and differences—such as in age, race, ethnicity, social class, education, and area of residence—could lead to different health profiles and needs. Even when there has been progress in women’s health, groups of women often have not improved or have shown more modest gains. More often, even when women are analyzed separately, there is no analysis of differences among groups of women, and groups of women at higher risk of having or dying from a condition (for example, those of lower socioeconomic status and members of racial and ethnic minorities) are the least represented in the studies. In the present report, when data were available, the committee pointed out dis-
parities in knowledge regarding women of diverse backgrounds in determinants of health (Chapter 2), in progress in relation to specific conditions (Chapter 3), and in communication approaches (Chapter 5). The problems of adequate representation of women in research, discussed in Chapter 4, are compounded by challenges in having sufficient power to analyze groups of women separately. Not every study can have adequate numbers to allow such analysis, but the portfolio of studies funded by an agency should contain sufficient data to inform the understanding of vulnerable populations.
ARE THE MOST APPROPRIATE RESEARCH METHODS BEING USED TO STUDY WOMEN’S HEALTH?
In addition to reviewing the determinants and conditions that have been researched and have led to advances in women’s health, the committee looked more broadly at the characteristics of research to evaluate whether particular methods or approaches yielded information that resulted in more progress and whether any particular components have contributed more to improving women’s health.
Research can broadly be categorized as basic research, which involves the study of cellular and molecular components in animal and in vitro systems; population-based or observational studies; clinical research, which is conducted in human subjects; and health-services and health-policy research. Studies vary in size and complexity. Methods can be observational or experimental (such as randomized clinical trials), and the focus of research can range from cells to populations to cultural and societal factors. Chapters 2 and 3 demonstrate the diversity of women’s health research and how the diverse research portfolio has contributed to progress in improving women’s health, as in the cases of breast and cervical cancer.
The committee found that observational studies of women’s health have provided information for identifying associations and are especially useful for generating hypotheses for further testing. Observational studies—such as the dietary component of the WHI, the Nurses’ Health Study, and the Study of Women’s Health Across the Nation (SWAN)—are sometimes coordinated in multiple research centers to accrue large and diverse samples. The postmenopausal hormone therapy and the calcium and vitamin D components of the WHI were randomized clinical trials that were based on the findings from such observational research. In addition, the observational studies led to animal and in vitro studies aimed at elucidating the pathophysiology of conditions and identifying potential treatments.
Different study types have different strengths and limitations, and results from diverse study designs can combine to provide extremely useful information that is directly relevant to the health of women and that has led to clear improvements in women’s health. The committee recognizes that there are drawbacks to different study types. For example, observational and clinical studies can be
expensive, can have subject attrition, can be of long duration, and, in the case of observational epidemiologic studies, can have difficulty in finding appropriate comparison populations and in controlling for potential confounders. Large human studies, such as the WHI, are further hindered by complex study designs and associated pitfalls. Despite those drawbacks, the committee concluded that information from large, complex observational and clinical studies could not be obtained with other study designs and are integral to progress in women’s health. New study designs that yield similar levels of certainty would be valuable. Smaller studies, in contrast, provide different information, can often be better controlled, and are potentially faster (depending on the end points studied) and less expensive. Internal validity (for example, the ability to establish causal relationships) is generally stronger in such studies, but this may be at the cost of external validity (that is, generalizability). Smaller studies are important to provide information on which to base large studies and to test specific hypotheses.
Although women are now routinely included in clinical research, persistent problems limit the usefulness of the data and their applicability to women’s health. Too often, the initial design is not optimal for obtaining data on women, because of problems in the inclusion criteria and end points assessed that may be less relevant to women’s health given what is known about sex differences in the etiology and symptoms of disease. Sample size and the ability to recruit women in studies in numbers that are adequate for appropriate analyses can be a challenge. Notably, there is a continuing lack of analysis and reporting of data by sex to determine whether there are sex differences.
Methods and techniques that could facilitate sex-specific or group analysis should be explored. New analytic techniques, such as Bayesian analysis, appear promising and should be investigated and refined and expanded as appropriate. As outlined in NIH guidelines, clinical research should take advantage of previous epidemiologic studies, smaller human studies, and available incidence, prevalence, and mortality data to identify where sex, racial, and ethnic differences have been documented (NIH, 2001). Research should focus on groups at the highest risk for morbidity and mortality from a condition.
To improve women’s health, sex- and gender-specific analyses must be published and used in drug development and clinical guidelines. Even when sex-specific analyses are conducted by researchers, the results of the analyses are often not included in publications, because of page limitations and journal restrictions. Such practices are a barrier to progress in women’s health.
Women tend to report worse overall health than do men and tend to rank quality of life as very important when considering their health (Rieker and Bird, 2005). Nevertheless, the end points assessed in studies are often incidence and 5-year survival rates; a smaller number of studies look either at morbidity or quality-of-life end points after treatment and survival or at diseases, such as autoimmune disease and benign gynecologic disorders, that result in morbidity and affect quality of life rather than resulting in death. Health-related quality of life
(HRQoL) encompasses morbidity and functioning and is affected by a variety of factors, including limitations attributable to specific symptoms, the constellation of symptoms associated with a specific condition, and the combined effects on overall well-being. Although the effect of treatments on women’s HRQoL is increasingly recognized as an important outcome, the ability to infer that effect is limited by the ability to compare measures among trials, treatments, and conditions, especially in a sex-specific manner. Often, health and treatment decisions are driven by HRQoL considerations, so having unifying metrics for the assessment of quality of life that can be used to compare clinical interventions across the spectrum of conditions and patients is important. Including a common metric for women’s health HRQoL in studies would be an important first step in comparing treatments and interventions and would assist women in making health-related decisions.
ARE THE RESEARCH FINDINGS BEING TRANSLATED IN A WAY THAT AFFECTS PRACTICE?
Although it is not unique to women’s health, there is clearly a major problem in the translation of research findings into practice; many findings take 1 of 2 decades to be incorporated into practice (Bansal and Barnes, 2008; Sussman et al., 2006). There are some instances of success in women’s health. Notably, a decrease in the incidence of breast cancer has been attributed to the decrease in prescriptions for hormone-replacement therapy after the WHI trial (Jemal et al., 2007). It is partly attributable to effective processes for translating findings to practice, but the decrease in use also resulted from effective communication of the findings directly to women, who then questioned their doctors about the risks posed by using the medication (Bluming and Tavris, 2009).
Such success is the exception rather than the rule. More commonly, research findings are not translated into practice quickly or effectively. The problem of inadequate translation reflects the whole set of obstacles to optimal health care that occur in the current system. The committee identified a number of barriers to the translation of research on women’s health, of which some are more of a problem for women and others are problems in all health research. The nature of scientific research is itself a barrier to translation. The advancement of scientific knowledge is an iterative process in which one research result builds on previous work and inconsistent or contradictory results are often published before a clear picture emerges from a multitude of different types of studies. Inconsistent results can lead to uncertainty in best practices, including contradictory guidelines and suspicion of new practices, and can delay adoption of scientific findings.
Social and cultural factors can be barriers to the translation of scientific findings regardless of scientific consensus, especially in connection with sexual mores, as occurred with a vaccine against HPV that has the potential to prevent most cases of cervical cancer. Socioeconomic status, immigration status, and
language barriers can delay the adoption of research findings in some populations of women.
Health-care providers themselves and quality of care can be barriers to the translation of new findings. A particular concern for women with respect to health-care providers is the fragmentation of care that results when women see multiple providers for different health concerns; primarily care is often provided by an obstetrician or gynecologist during their reproductive years. The healthcare systems in place in the United States can also present barriers to the translation of results of women’s health research. Performance measures (such as the Healthcare Effectiveness Data and Information Set) used to evaluate the quality of care do not include many conditions that are more common to women, and sex-based differences in care are rarely analyzed (Bird et al., 2007; McKinley et al., 2002; Weisman, 2000), so incentives to include the latest research findings on women might not be present. In addition, in the case of both men and women, if new treatments or practices are not reimbursed by government and private insurers, they are less likely to be adopted into widespread practice. Patients themselves can be barriers to adoption of research findings. Faced with a multitude of research findings and complex decisions related to their own health, patients can have a difficult time in weighing new options.
ARE THE RESEARCH FINDINGS BEING COMMUNICATED EFFECTIVELY TO WOMEN?
As is the case with translation, the problems of communication are not limited to women’s health issues. Communication of research findings in women’s health, however, is facilitated by women’s tendency to seek health-related information, including engagement in more health-related online activities than men (Fallows, 2005). The complexity of results and the sometimes inconsistent or contradictory results present challenges to the communication of findings and highlight the importance of having a clear, concise message. The emergence of the Internet and the World Wide Web has increased the amount of and access to health-related information for the general public but has also added to the confusion about findings and to concerns about the validity of the available information (Arora et al., 2008). Communication is further complicated by competing forces, especially when some health messages are competing with the marketing forces of industries.
Lessons learned from the WHI, mammography, and cardiovascular disease highlight both the challenges to and the successes in the translation and communication of research findings. The results of the WHI were disseminated to women, both study participants and the general public, through the mass media and health-care providers. The findings quickly translated into changes in prescribing practices, but that occurred amid much confusion (French et al., 2006). Cardiovascular disease was long thought to be a male disease despite being the
leading cause of death of women. Increased research into cardiovascular disease has provided information specific to the prevention and treatment of heart disease in women. Concerted efforts by government agencies and nonprofit organizations to disseminate information, including research findings, on cardiovascular disease in women have improved awareness of cardiovascular disease in women and helped in the translation of the research findings into practice, including the development of women-specific guidelines. The combination of all those efforts has contributed to decreased mortality from cardiovascular disease in women. The broad availability of affordable screening, the effectiveness of mammography screening in reducing breast-cancer mortality, and more recently the most effective age and frequency for that screening have been controversial issues. Advocacy groups have played a large role in successfully pushing Congress and government agencies to fund screening programs and to increase the awareness of breast cancer in the general public. And the implementation of mandated national standards for mammography improved the quality of mammography and patient care.
GAPS IN WOMEN’S HEALTH RESEARCH
In addition to the many successes in women’s health research, the committee identified gaps, including inadequate study of some conditions and at-risk populations, insufficient focus on determinants of specific health outcomes or conditions and of general health, inadequate attention to function and quality of life, and gaps in the translation of research findings into practice and in communication of the findings to the public. With respect to specific conditions, there are few data on ovarian and endometrial cancer, for which effective screening methods are lacking, and little is known about the causes and prevention of preeclampsia during pregnancy, which is a major cause of maternal morbidity and mortality. Conditions that affect elderly women, including frailty, have received relatively little attention. Research on prevention of and treatment for Alzheimer’s disease, obesity, and diabetes has rarely examined sex differences. There is little information on autoimmune diseases that affect women disproportionately, such as rheumatoid arthritis, lupus, and other potentially related immune system disorders. And conditions that affect women’s quality of life—such as endometriosis, chronic pain, chronic fatigue syndrome, and fibroids—are under studied.
Even in the case of health outcomes in which research has enabled advances, such as cardiovascular disease and breast and cervical cancer, challenges remain. New guidelines for the care of women have been developed, but there are still questions. Health-services research can address questions of implementation, such as whether the specialty of a clinician (for example, cardiologist vs internist) or the type of care facility (for example, women’s health center vs general clinic) affects the likelihood or quality of implementation. Among the unresolved issues
in breast cancer are explanations of differential risk of onset, treatment success, and survival among different groups of women.
Some of the identified gaps apply to many or most of the conditions that affect women. There has been relatively little attention to the causes and treatment of comorbidities. The issue of co-occurring conditions and the need to evaluate risk and benefit tradeoffs among multiple outcomes and end points cut across both mental and physical conditions. Large-scale studies—such as the WHI, which included both observational and experimental arms and considered multiple end points—provided some of the needed data from that perspective. Its overall history demonstrates the limitations of using only observational studies even if they are based on a good understanding of the molecular and cellular basis of sex differences, and also shows their synergistic value in developing targeted prevention and treatment strategies
Genetic and pathophysiologic studies are increasingly providing information on risk factors for conditions and potential treatment options on an individual level. Clinicians now need better tools to make personalized decisions about prevention, screening, and treatment. Those issues are not peculiar to women’s health but are relevant to conditions that affect women especially.
Although genetic information is useful, it will not map perfectly onto health without accounting for environmental exposures that can affect gene expression in addition to affecting health directly. More information on factors in the physical and social environment is needed, including an understanding of how they may result in health disparities for disadvantaged groups. Research has provided more information on sex differences in the pathophysiology of conditions than on gender or sex differences in social determinants of health and illness. Both types of knowledge are necessary to prevent conditions in women and treat them. Social factors also play a role in the translation of knowledge into practice by health professionals and by women themselves. In some cases, particularly reproductive health, strong data supporting the safety and efficacy of treatments may be insufficient to fuel their use if there is social or political opposition on nonmedical grounds. Advances in women’s health may require attention to such obstacles in addition to those inherent in the research.
Having evaluated progress in women’s health research and considering whether the research has looked at the right determinants, has studied the right outcomes with the right methods, and has translated the resulting knowledge into practice and communicated it to women, the committee agreed on a set of primary findings and recommendations.
COMMITTEE’S KEY FINDINGS AND RECOMMENDATIONS
Substantial progress has been made since the expansion of investment in women’s health research. Research findings have changed the practice of medicine and public-health recommendations in several prominent contexts, includ-
ing changes in standards of care for women. There have also been decreases in mortality in women from breast cancer, heart disease, and cervical cancer. In other contexts, however, there has been less progress, including research on other conditions that affect women and identification of ways to reduce disparities among subpopulations of women.
Several barriers to further progress in improving the health status of women were identified. For example, there has been inadequate attention to the social and environmental factors that, along with biologic risk factors, influence health. There also has been inadequate enforcement of requirements that representative numbers of women be included in clinical trials and that women’s results be reported. A lack of taking account of sex and gender differences in the design and analysis of studies, and a lack of reporting on sex and gender differences has hindered identification of potentially important sex differences and slowed progress in women’s health research and its translation to clinical practice. The committee recommends that all published scientific reports that receive federal funding and all medical product evaluations by the Food and Drug Administration present efficacy and safety data separately for men and women.
Poor communication of the results of women’s health research has in many cases led to substantial confusion and may affect the care of women adversely. Research findings will have a greater impact if they are coupled with a well thought-out plan for communication and dissemination. Development of a plan for communication and dissemination should be a standard component of federally sponsored women’s health research and the clinical recommendations that are made on the basis of that research.
The committee’s specific findings and recommendations follow.
Investment in women’s health research has afforded substantial progress and led to improvements in women’s health with respect to such important conditions as some cancers and heart disease. Greater progress in women’s health has occurred in conditions characterized by multipronged research involving molecular, animal, and cellular data; in observational studies to identify effects in the overall population; and in clinical trials or intervention studies from which evidence-based conclusions on treatment effectiveness can be drawn.
US government agencies and other relevant organizations should sustain and strengthen their focus on women’s health, including the spectrum of research that includes genetic, behavioral, and social determinants of health and how they change during one’s life. In addition to conducting women-only research as appropriate, a goal should be to integrate women’s health research into all health research—that is, to mainstream women’s health research—in such a
way that differences between men and women and differences between subgroups of men and women are routinely assessed in all health research. Relevant US government agencies include the Department of Health and Human Services and its institutes and agencies—especially the National Institutes of Health, the Centers for Disease Control and Prevention, the Food and Drug Administration, the Agency for Healthcare Research and Quality, and the Substance Abuse and Mental Health Services Administration—and such others as the Department of Veterans Affairs, the Department of Defense, and the Environmental Protection Agency.
Women who experience social disadvantage as a result of race or ethnicity, low income, or low educational level suffer disproportionate disease burdens, adverse health outcomes, and barriers to care but have not been well represented in studies of behavior and health.
The National Institutes of Health, the Agency for Healthcare Research and Quality, and the Centers for Disease Control and Prevention should develop targeted initiatives to increase research on the populations of women that have the highest risks and burdens of disease.
The incidence, prevalence, morbidity, or mortality associated with a number of conditions—for example, unintended pregnancy, maternal mortality and morbidity, nonmalignant gynecologic disorders, alcohol- and drug-addiction disorders, autoimmune diseases, and lung, ovarian, and endometrial cancer—have not improved. Most of those conditions substantially affect the quality of life of those who experience them. The major focus of health research has been on reducing mortality; a singular focus on mortality, however, can divert attention from other health outcomes despite the high value that women place on quality of life.
Research should include the promotion of wellness and quality of life in women. Research on conditions that have high morbidity and affect quality of life should be increased. Research should include the development of better measures or metrics to compare effects of health conditions, interventions, and treatments on quality of life. The end points examined in studies should include quality-of-life outcomes (for example, functional status or functionality, mobility, and pain) in addition to mortality.
Social factors and health-related behaviors and their interactions with genetic and cellular factors contribute to the onset and progression of multiple diseases; they act as pathways that are common to multiple outcomes. Considerable progress has been made in understanding the behavioral determinants of women’s health, but less is known about how to change them and about the broader determinants of women’s health that involve social, community, and societal factors.
Cross-institute initiatives in the National Institutes of Health—such as those in the Division of Program Coordination, Planning, and Strategic Initiatives—should support research on common determinants and risk factors that underlie multiple diseases and on interventions on those determinants that will decrease the occurrence or progression of diseases in women. The National Institutes of Health’s Office of Research on Women’s Health should increase collaborations with the Office of Behavioral and Social Sciences Research to design and oversee such research initiatives.
Limitations in the design, analysis, and scientific reporting of health research have slowed progress in women’s health. Inadequate enforcement of recruitment of women and of reporting data by sex has fostered suboptimal analysis and reporting of data on women from clinical trials and other research. That failure has limited possibilities for identifying potentially important sex or gender differences. New methods and approaches are needed to maximize advances in promoting women’s health.
Government and other funding agencies should ensure adequate participation of women, analysis of data by sex, and reporting of sex-stratified analyses in health research. One possible mechanism would be expansion of the role of data safety monitoring boards to monitor participation, efficacy, and adverse outcomes by sex.
Given the practical limitations in the size of research studies, research designs and statistical techniques should be explored that facilitate analysis of data on sociodemographic subgroups without substantially increasing the overall size of a study population. Conferences or meetings with a specific goal of developing consensus guidelines or recommendations for such study methods (for example, the use of Bayesian statistics and the pooling of data across study groups) should be convened by the National
Institutes of Health, other federal agencies, and relevant professional organizations.
To gain knowledge from existing studies that individually do not have sufficient numbers of female subjects for separate analysis, the director of the Office of the National Coordinator for Health Information Technology in the Department of Health and Human Services should support the development and application of mechanisms for the pooling of patient and subject data to answer research questions that are not definitively answered by single studies.
For medical products (drugs, devices, and biologics) that are coming to market, the Food and Drug Administration should enforce compliance with the requirement for sex-stratified analyses of efficacy and safety and should take those analyses into account in regulatory decisions.
The International Committee of Medical Journal Editors and other editors of relevant journals should adopt a guideline that all papers reporting the outcomes of clinical trials report on men and women separately unless a trial is of a sex-specific condition (such as endometrial or prostatic cancer). The National Institutes of Health should sponsor a meeting to facilitate establishment of the guidelines.
The translation of research findings into practice can be delayed or precluded by various barriers—the complexity of science and research and challenges in communicating understandable and actionable messages, social or political opposition to advances for nonmedical reasons, fragmentation of health-care delivery, health-care policies and reimbursement, consumer confusion and apprehension, and so on. Many of those barriers are seen in connection with translation of research in general, but some have aspects that are peculiar to women, and few studies have been conducted to examine how to increase the speed or extent of the translation of findings related specifically to women’s health into clinical practice. Methods of translation that have been used and that warrant evaluation for translating research findings in women include clinical-practice guidelines, mandatory standards, reimbursement practices, laws (including public-health laws), health-professions school curricula, and continuing education.
Research should be conducted on how to translate research findings on women’s health into clinical practice and public-health policies rapidly. Research findings should be incorporated at the practitioner level and at the overall public-health systems level through, for example, the use of education programs targeted to practitioners and the development of guidelines.
As programs and guidelines are developed and implemented, they should be evaluated to ensure effectiveness.
The public is confused by conflicting findings and opposing recommendations that emerge from health research, including women’s health research. Conflicting results and work to resolve disagreements are part of the scientific process, but that iterative aspect of scientific discovery is not clearly conveyed to, or understood by, the public. The resulting uncertainty and distrust of research may affect women’s care adversely. Relevant knowledge from studies of communication often is not used by researchers, funders, providers, and public-health professionals to target health messages and information to women.
The Department of Health and Human Services should appoint a task force to develop evidence-based strategies to communicate and market health messages that are based on research results to women. In addition to content experts in relevant departments and agencies, the task force should include mass-media and targeted-messaging and marketing experts. The strategies should be designed to communicate to the diverse audience of women; to increase awareness of women’s health issues and treatments, including preventive and intervention strategies; and to decrease confusion regarding complex and sometimes conflicting findings. The goals of the task force should be to facilitate and improve the communication of research findings by researchers to women. Strategies for the task force to consider or explore might include
requiring a plan for the communication and dissemination of findings in federally funded studies to the public, providers, and policy makers; and
establishing a national media advisory panel of experts in women’s health that would be readily available to provide context to reporters, scientists, clinicians, and policy makers at the time of release of new research reports.
Arora, N. K., B. W. Hesse, B. K. Rimer, K. Viswanath, M. L. Clayman, and R. T. Croyle. 2008. Frustrated and confused: The American public rates its cancer-related information-seeking experiences. Journal of General Internal Medicine 23(3):223–228.
Bansal, A. T., and M. R. Barnes. 2008. Genomics in drug discovery: The best things come to those who wait. Current Opinion in Drug Discovery and Development 11(3):303–310.
Berndt, E. R., L. M. Koran, S. N. Finkelstein, A. J. Gelenberg, S. G. Kornstein, I. M. Miller, M. E. Thase, G. A. Trapp, and M. B. Keller. 2000. Lost human capital from early-onset chronic depression. American Journal of Psychiatry 157(6):940–947.
Bird, C. E., A. M. Fremont, A. S. Bierman, S. Wickstrom, M. Shah, T. Rector, T. Horstman, and J. J. Escarce. 2007. Does quality of care for cardiovascular disease and diabetes differ by gender for enrollees in managed care plans? Women’s Health Issues 17(3):131–138.
Bluming, A. Z., and C. Tavris. 2009. Hormone replacement therapy: Real concerns and false alarms. Cancer Journal 15(2):93–104.
Brauer, C. A., M. Coca-Perraillon, D. M. Cutler, and A. B. Rosen. 2009. Incidence and mortality of hip fractures in the United States. Journal of the American Medical Association 302(14):1573–1579.
Burkman, R., J. J. Schlesselman, and M. Zieman. 2004. Safety concerns and health benefits associated with oral contraception. American Journal of Obstetrics and Gynecology 190(4 Suppl.): S5–S22.
Campbell, J., A. S. Jones, J. Dienemann, J. Kub, J. Schollenberger, P. O’Campo, A. C. Gielen, and C. Wynne. 2002. Intimate partner violence and physical health consequences. Archives of Internal Medicine 162(10):1157–1163.
Coker, A. L., P. H. Smith, L. Bethea, M. R. King, and R. E. McKeown. 2000. Physical health consequences of physical and psychological intimate partner violence. Archives of Family Medicine 9(5):451–457.
Fallows, D. 2005. How Women and Men Use the Internet: Women Are Catching Up to Men in Most Measures of Online Life; Men Like the Internet for the Experiences It Offers, While Women Like It for the Human Connections It Promotes. http://www.pewinternet.org/~/media//Files/Reports/2005/PIP_Women_and_Men_online.pdf.pdf (accessed August 15, 2009).
Ford, E. S., U. A. Ajani, J. B. Croft, J. A. Critchley, D. R. Labarthe, T. E. Kottke, W. H. Giles, and S. Capewell. 2007. Explaining the decrease in US deaths from coronary disease, 1980–2000. New England Journal of Medicine 356(23):2388–2398.
French, L. M., M. A. Smith, J. S. Holtrop, and M. Holmes-Rovner. 2006. Hormone therapy after the Women’s Health Initiative: A qualitative study. BMC Family Practice 7:61.
Gritz, E. R., I. R. Nielsen, and L. A. Brooks. 1996. Smoking cessation and gender: The influence of physiological, psychological, and behavioral factors. Journal of the American Medical Women’s Association 51(1-2):35–42.
Huang, Q. Y., and A. W. C. Kung. 2006. Genetics of osteoporosis. Molecular Genetics and Metabolism 88:295–306.
Jemal, A., E. Ward, and M. J. Thun. 2007. Recent trends in breast cancer incidence rates by age and tumor characteristics among US women. Breast Cancer Research 9(3):R28.
Leslie, W. D., S. O’Donnell, S. Jean, C. Lagace, P. Walsh, C. Bancej, S. Morin, D. A. Hanley, and A. Papaioannou. 2009. Trends in hip fracture rates in Canada. Journal of the American Medical Association 302(8):883–889.
McKinley, E. D., J. W. Thompson, J. Briefer-French, L. S. Wilcox, C. S. Weisman, and W. C. Andrews. 2002. Performance indicators in women’s health: Incorporating women’s health in the Health Plan Employer Data and Information Set (HEDIS). Women’s Health Issues 12(1):46–58.
NIH (National Institutes of Health). 2001. NIH Policy and Guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research—Amended, October, 2001. http://grants.nih.gov/grants/funding/women_min/guidelines_amended_10_2001.htm (accessed August 15, 2009).
Ozer, E. J., and R. S. Weinstein. 2004. Urban adolescents’ exposure to community violence: The role of support, school safety, and social constraints in a school-based sample of boys and girls. Journal of Clinical Child and Adolescent Psychology 33(3):463–476.
Rieker, P. P., and C. E. Bird. 2005. Rethinking gender differences in health: Why we need to integrate social and biological perspectives. Journals of Gerontology Series B Journal of Gerontology: Psychological Sciences and the Journal of Gerontology: Social Sciences 60(Special Issue 2): S40–S47.
Schnoll, R. A., F. Patterson, and C. Lerman. 2007. Treating tobacco dependence in women. Journal of Women’s Health (2002) 16(8):1211–1218.
Sussman, S., T. W. Valente, L. A. Rohrbach, S. Skara, and M. A. Pentz. 2006. Translation in the health professions: Converting science into action. Evaluation and the Health Profession 29(1):7–32.
Tjaden, P. G., and N. Thoennes. 1998. Prevalence, Incidence, and Consequences of Violence Against Women: Findings from the National Violence Against Women Survey, Research in Brief. Washington, DC: US Department of Justice, Office of Justice Programs, National Institute of Justice.
Weisman, C. S. 2000. Advocating for gender-specific health care: A historical perspective. Journal of Gender-Specific Medicine 3(3):22–24.