Defining Priorities for Reducing Disparities in Cancer Mortality Free

Improvements in screening, diagnosis, and treatment for many cancers have led to decreases in cancer death in recent years (1). However, this decrease has not been the same in all racial/ethnic groups or for all cancer sites, and major disparities continue to exist between racial/ethnic groups. From 2008 to 2012, the highest age-adjusted mortality rates in the United States men were in Blacks (BL) with a rate of 261.5 per 100,000. This rate is higher than non-Hispanic White (NHW) men (210.6), Hispanic (HI) men (148.0), American Indian/Alaskan Native (AIAN) men (140.3), and Asian/Pacific Islander (API) men (128.4). Cancer death rates in women are generally lower, with the highest mortality rates being in BL women (166.3 per 100,000). Mortality rates per 100,000 are lower in NHW women (149.2), AIAN women (100.9), HI women (99.4), and API women (91.2). Based solely on these numbers, it is clear that BL men and women suffer disproportionately from cancer mortality compared with other racial/ethnic groups. However, each racial/ethnic group, including NHW, suffers from disparities in cancer mortality for specific cancer sites. An objective quantification of these disparities may be of use in targeting research activities and resources needed to address these disparities.

Two quantities were computed to characterize disparities in rates and trends that may identify the most prominent disparities in mortality for BL, HI, API, and AIAN compared with NHW. First, cancer mortality rates for BL, HI, API, and AIAN were compared with NHW by a simple ratio of the mortality rate in each group divided by the rate in NHW and reported as a relative percentage. 2008–2012 Surveillance Epidemiology and End Results (SEER) 18 area age-adjusted standardized mortality rates per 100,000 population were used to generate this ratio (2). A value of 100% indicated that NHW and BL, HI, API, or AIAN had equal rates; a value >100% indicated that BL, HI, API, or AIAN had a higher rate than NHW; a value <100% than indicated that BL, HI, API, or AIAN had a lower rate than NHW. Second, the difference in mortality rate change between NHW and BL, HI, API, or AIAN was computed. The difference in average annual percent change in mortality for NHW versus BL, HI, API, or AIAN was obtained using data from the Joinpoint Regression Program Version 4.2.0 with SEER 13 area data for the period 2003–2012 (2). A value of zero indicated that the change in rates for BL, HI, API, or AIAN was equal to that of NHW; a value greater than zero indicated that BL, HI, API, or AIAN experienced a less favorable change in rates than NHW; a value less than zero indicated that BL, HI, API, or AIAN experienced a more favorable change in rates than NHW. Note that all rates being compared may have dropped during the 2003–2012 period, and this metric reflects only the change for BL, HI, API, or AIAN relative to NHW.

Figures 1 and 2 and Supplementary Tables S1 and S2 summarize the rates and trends of BL, HI, API, and AIAN compared with NHW, and a series of priority cancers were identified. Priority 1 cancers have rates elevated 200% or more relative to NHW and demonstrate an unfavorable trend in rates compared with NHW. Priority 1 included stomach cancers in BL men, BL women, and HI women, and myeloma in BL women. Priority 2 cancers are those with mortality rates elevated between 100% and 200% and exhibit unfavorable trends relative to NHW. These include cancers of the stomach in HI and AIAN men; colon/rectum in BL men; thyroid in BL and HI women; uterine corpus in BL women; and kidney in AIAN. Priority 3 cancers include those with greater than 200% greater mortality but exhibit a decreasing trend in rates relative to NHW populations. These include cancers of the stomach in API men and women; prostate in BL men; and cervix in BL women. These cancers reflect some of the most pressing problems in cancer disparities and should be prioritized in future cancer research and control activities.

Priority 4–5 cancers are also relevant because they represent elevated rates or unfavorable trends for BL, HI, API, or AIAN compared with NHW. The impact of these cancers on disparities can be assessed by viewing the data in Supplementary Tables S1 and S2. Finally, it is important to note that NHW individuals also experience unfavorable mortality rates and trends relative to all other groups. Cancers in Priority Group 6 represent those for which NHW suffer disproportionately. For example, mortality rate trends in oral cavity/pharyngeal and esophageal cancers in men have been less favorable in NHW compared with other racial/ethnic groups (Fig. 1). As noted above, NHW cancer mortality rates overall are second only to those in BL, and NHW represent the majority of the U.S. population. Disparities affecting NHW also require attention because the reduction in mortality of these cancers will have a very large impact on overall cancer mortality in the general population.

The quantification of cancer disparities is dramatically easier than finding solutions to these problems. Disparity reduction may require discovery of basic mechanisms of carcinogenesis, improved early detection methods, or optimal implementation of existing methods. Although efforts in all of these areas are needed for most cancers, Table 1 provides examples of some opportunities for disparity in mortality reduction for each of the high-priority cancers identified. The table does not consider improved treatment modalities, which are needed for all cancers. In each example cited, activities that account for the unique biologic and social mechanisms of each race/ethnic group may need to be considered to optimize interventions for each population at risk.

First, basic discovery about the mechanism of disease is needed for priority cancers. These efforts include discovery of effective modalities for prevention and early detection, including genomic and biomarker-based strategies. Second, prevention strategies exist for some cancers but need improvement to reach their full potential for all people (e.g., early detection of prostate, stomach, and thyroid cancers). Finally, there are strategies thought to be effective that require better dissemination to those who can benefit (e.g., colonoscopy in individuals over the age of 50 years and human papillomavirus vaccination). Given that BL, HI, and AIAN have high rates of obesity, opportunities for prevention of many of the priority cancers could focus around obesity reduction. The priority cancers identified here that are also associated with obesity include colorectal, kidney, thyroid, and uterine corpus cancers.

The causes of disparities across race/ethnicity groups are complex and multifactorial: differences in biologic aggressiveness of tumors, delays in diagnosis because of limitations to health care access, behavioral and lifestyle factors, and receipt of timely and appropriate care are likely to underlie disparities in cancer mortality. The information presented here focuses on disparities in rates and trends, not on the population impact in terms of preventable fraction of cancer overall. It also does not consider the cost–benefit relationship of investment in modalities designed to limit mortality in minority groups. However, enumeration of cancer rates and trends presented here may be used to define priorities for basic, prevention, or treatment research aimed at reducing disparities in cancer mortality.

See all articles in this CEBP Focus section, “Multilevel Approaches to Addressing Cancer Health Disparities.”