Evidence that cardiovascular disease (CVD) and its risk factors are common in women has grown rapidly in recent decades. During the same period, numerous studies have demonstrated gender disparities in the detection, prevention, care, and outcomes of CVD as well as in conditions that contribute to CVD, such as diabetes (
Correa-de-Araujo and Clancy 2006
, Mosca et al 2005
, Bird et al 2003
). Yet recognition of CVD prevalence and its risk factors among women remains low and gender disparities remain “below the radar” for many key stakeholders. Consequently, gender disparities are not routinely assessed or acted upon. The set of four studies on gender differences in the quality of CVD and diabetes care in this issue of Women’s Health Issues offer an unprecedented look at care associated with an increasingly important, but relatively understudied group of stakeholders and settings of care—namely, managed health care plans. Managed health care plans are currently responsible for the care for more than half of the US population. They are also in a position to employ tools and technology to disseminate and reward evidence-based practices to address disparities in care. Unfortunately, the studies show that, as in other settings, gender disparities in managed health care plans have been and continue to be common. Taken together, the studies make a compelling case that routine assessment of and action on gender disparities in care for CVD and diabetes is warranted.The fact that gender disparities and the need to address them has not been a high priority for most health plans is understandable given the existent gaps in the gender disparities literature. For example, most studies to date have focused on gender disparities related to cardiac care and outcomes for acute cardiac events involving hospitalization, such as an acute myocardial infarction. Relatively few studies have examined gender disparities in ambulatory care settings where most primary and secondary preventive care for CVD and diabetes takes place. Moreover, most have involved Medicare fee-for-service patients >65 years old, rather than the younger population of commercial enrollees who constitute the vast majority of managed health care plan members. Such gaps raise questions about generalizability to and wisdom of plans devoting resources to reporting performance data separately by gender given they are already investing considerable resources into increasing overall quality of care.
The current studies largely redress previous gaps in the literature and clarify that considering gender differences in care can contribute to rather than compete with plans’ efforts to improve overall quality of care. These studies include analyses of gender disparities in large samples of enrollees in commercial and Medicare plan ambulatory settings in 1999 (
Bird et al 2007
) and more recently in 2005 (Chou et al 2007
, Chou et al 2007
, Chou et al 2007
). Although the papers examined data from different plans and periods of time and used somewhat different analytic models, results were consistent across papers and with previous studies (Corbelli et al 2003
, Correa-de-Araujo et al 2006
, Kim et al 2006
).Using data from 1999,
Bird et al 2007
assessed 11 screening, treatment, and intermediate outcome measures of quality care for patients with CVD or diabetes in 19 Medicare and commercial plans covering >2 million individuals affiliated with a single insurer. Even after adjustment for age, race/ethnicity, socioeconomic status (SES), and individual plan effects, gender differences were common, with more than three fourths of the observed differences favoring men among both Medicare and commercial enrollees. The authors found that the greatest disparities occurred in intermediate outcome measures, which also had poor overall performance. For example, the percentage of diabetic Medicare plan enrollees with their low-density lipoprotein cholesterol (LDL-C) level controlled was low overall (<40%), but even lower for women, who were 19% less likely to be controlled compared with men.In 2005,
Chou et al 2007
collected individual patient-level data on 7 quality indicators from 31 health plans representative of commercial plans in general. Data were also stratified by age, and presumed SES and race/ethnicity. For 6 of the measures, men had better outcomes than women. Again, lipid control in men with diabetes or CVD was clinically and statistically better than in women. The only measure to (slightly) favor women was blood pressure control. For the remainder, differences were small and varied as much or more by SES or race/ethnicity as by gender.Chou et al 2007
examined the feasibility of collection of gender-stratified results of CVD-related performance measures from a broad national spectrum of representative health plans and assessed this data by individual plan. Gender-related disparities were common and varied widely in magnitude both by plan and specific measure, but when present, almost uniformly favored men. There were no obvious patterns of performance except for the occurrence of greater gender differences in intermediate outcomes (e.g., lipid control) than in identification of risk (e.g., lipid screening). The collection and reporting of these types of quality indicators by gender was shown feasible with current methods, but reporting of the data at the plan level was judged a disappointingly low priority based on interviews with several of the plans’ quality improvement representatives.Finally,
Chou et al 2007
evaluated 6 diabetes management quality measures by both gender and race/ethnicity (African American vs. white) and found that, among diabetics, women compared with men and African Americans compared with whites consistently suffered gaps in performance measures and were significantly less likely to achieve adequate control of their diabetes and lipids. In particular, LDL-C control was poorest among African American women, followed by African American men, and diabetic control poorest in African American men followed by African American women. Three of the 4 evaluated screening performance measures were poorer for African Americans.Collectively, these findings make a convincing case that gender disparities have been and continue to be common within managed health care settings. The finding that large disparities were most often observed for intermediate control outcomes, such as control of hyperlipidemia, whereas gender differences on process measures, such as lipid profiles in diabetics, tended to be small, is particularly concerning. As the authors note, these findings suggest that plans are more effective at screening than at achieving the desired intermediate outcomes in women as compared with men. It is also troubling that this basic pattern remained relatively constant whether one focused on HEDIS 2000 data (
Bird et al 2007
) or HEDIS 2006 data (e.g., Chou et al 2007
), suggesting a stable situation in which, even when care for men and women appear to be similar, intermediate outcomes often are not. Results revealed by examining performance by gender such as these can permit plans to prioritize interventions and monitor improvement over time to make sure disparities are narrowing (Trivedi et al 2006
). However, as suggested in the papers, even relatively small gender disparities (e.g., fewer than 5 percentage points) observed for several measures warrants attention and action given the substantial number of people that translates into when weighted across all enrollees.These studies also remind us that, although “gender disparities” are sometimes assumed to refer only to instances where women receive worse care than men, they also include differences that disadvantage men. For example, although women with CVD risk factors generally were less likely than men to have their LDL-C controlled (
Bird et al 2007
, Chou et al 2007
, Chou et al 2007
, Chou et al 2007
), they were more likely than men to have their blood pressure controlled—at least among commercial plans. Interestingly, the gender disparity in blood pressure control observed among enrollees <65 years old in commercial plans disappeared or even reversed among older adults in Medicare plans.Delineating instances where care differs for men and women, particularly those that may be counterintuitive, are critical to developing a better understanding of how and why care may differ for men and women. For example, instances in which outcomes differ by gender but processes of care appear to be equivalent may reflect important gender differences in behavior or physiology (
Correa-de-Araujo 2006
). Because causal factors and effective interventions may vary across settings, the availability of such data will also be crucial to allowing quality improvement teams to identify the specific factors underlying gender differences in health outcomes and plan performance, and to initiate appropriate interventions within their respective organizations (Correa-de-Araujo and Clancy 2006
). Sharing knowledge between provider organizations about best practices for improving quality of care for women can help to accelerate progress.In addition to demonstrating differences in the quality of care for men and women, the papers also highlight the fact that gender differences in quality of care can vary in important ways depending on members’ race/ethnicity or SES. For example, although women were more likely than men to have poorly controlled cholesterol overall, a more complicated pattern was revealed when gender differences within and between racial subgroups was considered (e.g.,
Chou et al 2007
). Thus, to most effectively target interventions, plans need to assess whether and how gendered patterns of performance vary between racial/ethnic and SES groups. Although plans still lack self-reported race/ethnicity (and SES) data for most members, they can obtain accurate estimates cost efficiently by using newer indirect methods (Fiscella and Fremont 2006
, Elliot et al 2007
, Fremont et al 2005
) such as those used by plans in the National Health Plan Collaborative to Reduce Disparities and Improve Quality sponsored by Agency for Healthcare Research and Quality and the Robert Wood Johnson Foundation (Lurie et al in press
) or, in the National Committee for Quality Assurance’s pilot project that provided data for analyses described in Chou et al 2007
.Admittedly, the sheer logistics of sifting through performance measures stratified by gender, race/ethnicity, and SES can seem daunting to plans given resources they already allocate to analysis and reporting of overall performance scores. Yet, although we are sympathetic to the challenges these changes could create for plans, we are also clear that the status quo is not a viable or necessarily less expensive option for plans. Indeed, if they are to achieve the level of quality they are committed to and those paying for care are increasingly demanding, they need greater clarity on factors contributing to gaps in quality for different enrollee groups and more careful targeting of the interventions to those subgroups. These challenges should not be faced by plans alone, but should also be a focus of academic researchers, policy makers, and employers.
In conclusion, measuring, reporting, and benchmarking quality measures is crucial to improving quality of care and reducing disparities in care. The results reported in this issue of Women’s Health Issues provide strong evidence that gender disparities are common in health plans and that routine assessment and reporting of selected quality indicators by sex/gender is needed. Although improvements in the quality of cardiovascular care for both men and women are needed, priorities for intervention may vary. Because women have different experiences with care and on average have more frequent interactions with the health system than men, the effectiveness of quality interventions may also have vary by gender. Stratifying quality indicators by gender will allow plans to specifically evaluate the effectiveness of improvement interventions among women and men. Although this approach will undoubtedly create some new challenges for plans, the status quo is no longer a viable option. Key factors influencing health disparities remain unexplored and how the factors are interrelated is poorly understood. Consequently, systematic data collection to identify gaps and eliminate disparities needs to be broadened to capture additional information about SES and cultural factors as well as the determinants of health and disease both at community and individual levels.
Acknowledgments
The views expressed in this article do not necessarily represent the views of the RAND Corporation, the Agency for Healthcare Research and Quality, the U.S. Department of Health and Human Services, or the U.S. Federal government.
References
- Improving women’s quality of care for cardiovascular disease and diabetes.Women’s Health Issues. 2003; 13: 150-157
- Does quality of care for cardiovascular disease and diabetes differ by gender for enrollees in managed care?.Women’s Health Issues. 2007; 17: 131-138
- Gender disparities in the quality of cardiovascular disease care in private managed care.Women’s Health Issues. 2007; 17: 120-130
- Gender differences in cardiovascular disease care in commercial and Medicare managed care plans.Women’s Health Issues. 2007; 17: 139-149
- Gender and racial disparities in the management of diabetes mellitus among Medicare patients.Women’s Health Issues. 2007; 17: 150-161
- Gender bias in lipid assessment and treatment following percutaneous coronary intervention.Journal of Gender-Specific Medicine. 2003; 6: 21-26
- Serious gaps: How the lack of sex/gender-based research impairs health.Journal of Women’s Health. 2006; 15: 1116-1122
- Catalyzing quality of care improvements for women.Women’s Health Issues. 2006; 16: 41-43
- Gender differences across racial and ethnic groups in the quality of care for acute myocardial infarction and heart failure associated with comorbidities.Women’s Health Issues. 2006; 16: 44-55
Elliot, M., Fremont, A., Lurie N., Morrison, P., Pantoja, P., & Abrahamse, A. (2006). A New Method for Estimating Racial/Ethnic Disparities where Administrative Records Lack Self reported Race/Ethnicity. Population Association of America Annual Meeting. Los Angeles.
- Use of geocoding and surname analysis to estimate race and ethnicity.Health Services Research. 2006; 41: 1482-1500
- Use of indirect measures of race/ethnicity and socioeconomic status in managed care settings to identify disparities in cardiovascular and diabetes care quality.Health Affairs. 2005; 24: 516-526
- Gender disparities in lipid management: The presence of disparities depends on the quality measure.American Journal of Managed Care. 2006; 12: 133-136
Lurie, N., Fremont, A., Martin, C., Somers, S. A., Coltin, K., Gavras, J., et al. (In press). The National Health Plan Collaborative to reduce disparities and improve quality. The Joint Commission Journal on Quality and Safety.
- Opportunity for intervention to achieve American Heart Association Guidelines for optimal lipid levels in high-risk women in managed care setting.Circulation. 2005; 111: 488-493
- Relationship between quality of care and racial disparities in Medicare health plans.Journal of the American Medical Association. 2006; 296: 1998-2004
Article info
Publication history
Published online: May 21, 2007
Accepted:
April 12,
2007
Received:
April 12,
2007
Identification
Copyright
© 2007 Jacobs Institute of Women’s Health. Published by Elsevier Inc. All rights reserved.
