The disparity of cervical cancer in diverse populations

Abstract

Significant disparities in cervical cancer incidence and mortality rates among minority groups have been documented in the United States, despite an overall decline in these rates for the population as a whole. Differences in cervical cancer screening practices have been suggested as an explanation for these disparities, as have differences in treatment among various racial and ethnic groups. A number of factors are attributed to these observed differences. As minority populations continue to grow in size over the next 50 years, persistent disparities will place an ever increasing burden on these populations and on the national healthcare system. Strategies to reduce cervical cancer disparities need to be employed in order to reverse these trends.

Introduction

Since the mid-1970′s, the incidence of cervical cancer has steadily declined in the United States (US), as have mortality rates [1]. These declines can largely be attributed to wide-scale implementation of cytological screening programs utilizing the Papanicolaou (Pap) test. While declines in incidence and mortality rates for cervical cancer can be observed across all racial and ethnic groups in the US from 1995 to 2004, marked disparities in these rates persist [1].

Analysis of average annual incidence rates for cervical cancer, reported by the Surveillance Epidemiology and End Results (SEER) database [1] from 2000 to 2004, indicate that Hispanic women are diagnosed with cervical cancer almost twice as often, and African American women more than 1.5 times as often as non-Hispanic white women (Fig. 1a). The incidence rate among Asian American/Pacific Islander (AA/PI) women is slightly higher than that of non-Hispanic white women, while American Indian/Alaskan Native (AI/AN) women have lower rates of diagnosis [1]. Similar differences are observed in the percent lifetime risk of being diagnosed with cervical cancer (Fig. 1b) [1]. While Fig. 1a states the annual incidence rate for AA/PI women as one group, considerable heterogeneity and range in incidence rates among subpopulations of AA/PI women exist. Analysis of the California Cancer Registry, 2000 to 2002, reveals that Chinese women have the lowest cervical cancer incidence rate (5.4 per 100,000) while Vietnamese women have the highest incidence rate (14.0 per 100,000) [2]. This variation can make national reported rates misleading, not just for AA/PI, but for other groups in which distinct subpopulations exist.

When the percentage of individuals diagnosed with cervical cancer among various racial and ethnic groups is examined by age, distinct trends can be observed in each group. For white women, the greatest percent prevalence is observed in those between 40 and 49 years of age, while Hispanic women and AA/PI women have broader prevalence peaks. African American women have a bimodal distribution of cervical cancer prevalence rates, as there is an initial peak between ages 35 and 45 and then rates increase again at age 60 (Fig. 1c) [1].

Differences exist regarding the stage of cancer at diagnosis among the various ethnic/racial groups. For example, white women are more likely to be diagnosed at an earlier stage than African Americans, American Indians, or Hispanics [3], [4]. Non-Hispanic African Americans have a higher percentage of cervical cancer diagnosis at later stages than Hispanics and non-Hispanic whites [5], [6], [7].

Later stage of diagnosis may partially explain why mortality rates do not always parallel incidence rates (Fig. 1a). The average annual cervical cancer mortality rate from 2000 to 2004 for African Americans was more than twice that for non-Hispanic whites (2.2 deaths per 100,000), while the rate for Hispanic women was 1.5 times greater than that for non-Hispanic white women; these are different ratios than those observed for incidence rates (Fig. 1a) [1]. Mortality rates for AA/PI women were slightly higher than those for white women, as were those for AI/AN women. However, when the AI/AN population is broken down to those who live in Contract Health Service Delivery Area (CHSDA) counties (geographic areas under the jurisdiction of Indian Health Services, often including a reservation or an area that borders a reservation), there is a marked difference; those that live in CHSDA counties have an annual mortality rate of 4.0 deaths per 100,000 persons, as compared with 1.1 per 100,000 for those living in non-CHSDA counties [1]. As previously noted, the heterogeneity within the AA/PI population can be observed when rates per subgroup are examined. For example, from 1999 to 2001, Japanese and Filipino women had the lowest mortality rates among this group of women (1.8–2.0 per 100,000 and 1.9–2.1 per 100,000, respectively), while Vietnamese women experience 4.3 to 4.5 annual cervical cancer deaths per 100,000 persons and Native Hawaiian women had the highest mortality rate of 6.1 to 7.0 per 100,000 [8].

Fig. 1b illustrates the lifetime risk of dying from cervical cancer for each group, which reflects similar trends as those seen in the mortality rates [1]. Hispanics and non-Hispanic whites have similar five-year survival rates (71% and 68%, respectively), and both have higher rates than those observed in African Americans (56%) [6]. African Americans have a significantly increased risk of death compared with non-Hispanic whites among patients with Stage 1 cervical tumors, but not at later stages [6]. For all types of cancers, African American women have a 9% lower incidence rate, but an 18% higher death rate than white women [9].

When mortality rates among minority groups are dissected into US versus foreign-born women, a distinct rise in cervical mortality rates among foreign-born women between 1985 and 1996 can be observed [10]. This difference is primarily seen among Hispanic and AA/PI women. The ratio of mortality rates between foreign-born and US-born Hispanic women was 4.11, and 1.40 for AA/PI women. Conversely, US-born African American women have higher mortality rates (7.31 per 100,000) compared with foreign-born African American women (5.86 per 100,000) during this time period, although this may not be statistically significant [10]. Incidence and mortality rates for Hispanic women may be even higher than reported, as speculation exists that these women go back to their countries of origin to receive “traditional” treatments or to die. Furthermore, all groups of women, including those without legal status, who may not seek or receive conventional medical help would not be reflected in the data.

Disparities may be influenced by the geographic distribution of minority groups in the US. For example, states in the Deep South, such as Arkansas and Mississippi, have 11% to 38% higher cervical cancer death rates than the overall rate in the US [1]. There are regional differences associated with cervical cancer mortality rates among AI/AN, with higher rates observed in the Northern Plains and Southwestern regions of the US [11]. Mortality rates for AI/AN populations may not be accurate. Studies have demonstrated incorrect racial coding on state death certificates, with 41% of AI/AN in the US coded as non-native, indicating an underestimation of cancer mortality rates [11].

A handful of factors may contribute to these marked disparities, including differences in screening and follow-up rates, treatment, behavioral risk factors, and underlying biological variations. The initiation of sexual activity at an early age and an increased number of sexual partners have been shown to be risk factors for cervical cancer [12]. Data collected from the National Youth Risk Behavior Survey indicate that a greater percentage of African American students have their first sexual intercourse before the age of 13 compared with Hispanic and white students (16.5% vs. 7.3% and 4.0%, respectively) and that they are more likely to have four or more partners during their life than Hispanic or white students (28.2% vs. 15.9% and 11.4%, respectively). Race- or ethnic-specific molecular variations, which could perhaps lead to increased susceptibility, have not been identified to date for cervical cancer, as they have for uterine cancer [13]. Differences in screening and follow-up rates and in treatment are discussed below.