The role of Federal legislation on breast cancer disparities

每年，美国用于医疗保健的支出占国内生产总值（gdp）的比例都高于中国

Each year, the US spends a larger percentage of its gross domestic product on health care than any other nation.1 Yet, rather than achieving a higher standing among industrialized nations for such key indicators as life expectancy, the US position has declined2. Numerous investigators assert that this occurs, in part, because the nation's preoccupation with health care diverts resources from potentially more important targets such as the solution of social, psychological, behavioral and environmental problems.2"9' ¿'n In the present proposal, we suggest that difficulties posed by the magnitude and nature of US health care spending go beyond diversion. We hypothesize that federal laws and policies related to health care inadvertently promote racial disparities in mortality by favoring the acquisition of life saving innovations by those of higher socioeconomic status. The resulting premature loss of life may also be contributing to the declining US position relative to other industrialized nations. We have shown that Black;white disparities in US mortality increased after three lifesaving, disease-specific innovations with clear start dates, namely Surfactant for Respiratory Distress X licensed by the Food and_DmAdininistration FDA in 1 9 8 9 i h . A : t w e Anti-Retroviral Therapy for Human Immunodeficiency Virus (HIV disease) licensed by the FDA from December 1995 to March 1996 ; and mammography re-imbursement by Medicare in 1991. For RDS, mean (+/-standard deviation) 5 year pre- and post-innovation black:white Mortality Rate Ratios (MRRs) were 1.92 (+/-0.26) and 2.70 (+/-0.37) (p=0.005). The corresponding MRRs were 3.98 +/-0.51 and 7.98 +/-0.37 (p < 0.001) for HIV mortality (age-adjusted, 25-64 years) and 0.93 +/-0.05 and 1.04 +/-0.04 (p=0.003) for breast cancer (age-adjusted, 65-85+ year old women).12 Thus, after each innovation, rates declined less in blacks than whites, translating into as many as 18,995 premature deaths among blacks through 2004. These descriptive data are compatible with the hypothesis that Medicare law (which defines medical assistance (health insurance) as a cash benefit regardless of the extent to which cash is a barrier to service acquisition) and administrative policies of the Food and Drug Administration (basing drug marketing decisions on biological safety/efficacy without considering possible adverse social effects) contribute to disparities in black:white mortality, in part, by actively (Medicare) and passively (Food and Drug Administration) helping to assure that the benefits of life-saving innovations are more likely to accrue to persons of higher socioeconomic status. This is not to deny the beneficial effects of these programs, but rather to suggest that any health-related intervention may have unintended, adverse effects. The present collaborative study will focus on breast cancer in order to address the overarching hypotheses as they pertain to Medicare law. We will purchase Medicare claims data from the Center for Medicare and Medicaid Services (CMS) and SEER-Medicare data. We shall also use data from the Area Resource File and the National Cancer Institute's Cancer Information System (CIS) to locate medical resources and programs that support utilization of screening mammography. We shall then pursue the following aims and hypotheses: Specific Aim 1: To determine the importance of regional level characteristics on the utilization of mammography by elderly women in the years 1992 to 1995 and 2002 to 2005. We will use the Medicare data to identify patient level information (e.g., age, race, co-morbidity, regular source of care), and both ARF and CIS to identify regional level data. We will use multilevel logistic regression to model women clustered within regions and states. The hypothesis to be tested will be: H-l: Regional and state level characteristics will have a significant effect on mammography utilization after adjusting for individual factors. H-2: The effect of regional and state level characteristics on mammography utilization will be different for African American and White women. H-3: The region having the greatest equitability for screening mammography utilization will be more likely to have programs in place aiming to increase mammography utilization and reduce disparities. Specific Aim 2: To examine the association between screening mammography utilization and breast cancer survival of women age 67 and older. For the years 1992 through 2005, we will use Statistical Epidemiology and End Results (SEER)-Medicare data to take potential confounding (e.g., age at diagnosis) and effect modification (e.g., race/ethnicity) of this association into account. Using Cox proportional hazard regression, we will test the following hypotheses: H-4: Relative to moderately successful or relatively unsuccessful places, highly unsuccessful places will be more likely to have lower contextual socioeconomic status, low survival from breast cancer, and greater percentage increase in racial disparity in survival over time. H-5: Relative to places that are jTighIy_unsuccessful, moderately successful or relatively unsuccessful with regard to screening mammography utilization, exceptionally successful places will be more likely to have high survival from breast cancer, and to have greater percentage reduction in racial disparity in survival over time. Socioeconomic status is not given as a part of this hypothesis to allow for the possibility that, under present conditions, exceptionally successful places may be those that overcome contextual socioeconomic barriers.