In US men, prostate cancer is the leading malignancy diagnosed and the second leading cause of cancer death (1–3). Prostate cancer affects approximately 144 per 100,000 men aged 50 to 54 years; this number increases rapidly with age to 930 per 100,000 men aged 70 to 74 years. Age-adjusted mortality rates predict that 23.6 of 100,000 men will die each year from complications resulting from prostate cancer (1).

The primary goal of prostate screening (eg, prostate-specific antigen [PSA] test, digital rectal examination) is to identify men for whom treatment would reduce morbidity and mortality. Guidelines for prostate screening differ across medical organizations (4); recently, the US Preventive Services Task Force (USPSTF) changed its guidelines. In 2008, USPSTF concluded that evidence was insufficient to assess the balance of benefits and harms of screening for prostate cancer in men aged younger than 75 years and recommended against screening for prostate cancer in men aged 75 years or older (5). In 2012, USPSTF recommended against PSA-based screening for average-risk men of all ages (6). The American Urological Association recently recommended shared decision making for men aged 55 years or older regardless of risk (7). The American Cancer Society recommends that men make informed decisions by considering the available information, having a discussion with their physician, and exploring their own views on the benefits and side effects of screening and treatment (8). The rationale against testing for prostate cancer is that the cancer is often slow-growing and may never progress enough to threaten one’s life. Thus, since treatment carries risks, finding and treating early tumors may do more harm than good for some patients. Although stakeholders have not reached consensus on prostate screening recommendations, screening can provide an opportunity for early detection, when cancer treatment is most effective (9,10).

Relatively little is known about prostate cancer and screening behaviors among American Indian and Alaska Native (AI/AN) men. Some studies indicate that AI/AN men have a lower incidence of prostate cancer than white men (11,12), yet AI/ANs have the highest risk of prostate cancer mortality of any racial group (11,13). A recent examination of prostate cancer mortality found that rates were higher for AI/ANs than for whites and that mortality declined from 1999 to 2009 for whites but not for AI/ANs (11). Another recent study found that the prostate cancer mortality-to-incidence ratio from 1999 to 2009 was higher among AI/ANs than among whites (14).

An examination of the 637 Indian Health Service Contract Health Services Delivery Area counties found lower age-adjusted prostate cancer incidence rates among AI/AN men than among white men in all regions except the Southern Plains. Rates of age-adjusted prostate cancer incidence among AI/AN men ranged from 81.3 in the Southwest to 164.1 in the Northern Plains. This range was smaller among white men, ranging from 132.1 in the Southwest to 165.4 in Alaska. The same study found higher rates of age-adjusted prostate cancer death in the Northern and Southern Plains and a lower death rate in the East among AI/AN men than among white men. Prostate cancer death rates among AI/AN men ranged from 22.4 in the Southwest to 41.2 in the Northern Plains and among white men, rates ranged from 23.4 in the East to 26.9 in the Pacific Coast (11).

Accurate estimates of prostate screening rates among AI/ANs are challenged by small, nonrepresentative samples and widespread racial misclassification in vital statistics and cancer registry data. Nonetheless, several studies suggest that AI/AN men have lower prostate screening rates than men of other races and ethnicities (12,15–17). Three of the 4 studies that compared prostate screening rates between AI/AN men and men of other races and ethnicities combined data from multiple years of the Behavioral Risk Factor Surveillance System (12,15,17), and the fourth study used the 2001 California Health Interview Survey (16).

Given the limited available research literature, we set out to examine prostate screening at 5 time points during a 12-year period among AI/AN men aged 50 to 75 years and to compare their screening rates with those of African American men and white men in the same age range.

Across all 5 waves of HRS data, data on 9,704 men were available for analyses, including 119 AI/ANs (n = 333 observations), 1,359 African Americans (n = 3,704 observations), and 8,226 whites (n = 24,292 observations) (Table 1). About half (45%) of participants had data from 4 waves, 15% from 3 waves, 25% from 2 waves, and 15% from 1 wave. Only 1 man had data from all 5 waves. Age at the first wave (Wave 3 in 1996) differed significantly according to race, but the differences were not large, with a mean (standard deviation) age of 59.2 (7.7) years for AI/ANs, 60.0 (7.3) years for African Americans, and 61.0 (7.7) years for whites.

From 1996 to 2008, prostate screening rates changed for each group: from 57.0% to 55.7% among American Indians/Alaska Natives, from 62.0% to 71.2% among African Americans, and from 68.6% to 71.3% among whites. (Table 2). Averaged across all waves, the age-adjusted likelihood of prostate screening was significantly lower among AI/ANs than among African Americans (RR = 0.81; 95% confidence interval [CI], 0.71–0.92) and whites (RR = 0.79; 95% CI, 0.70–0.90).

We found a significant time-by-race interaction (P = .01). In an examination of the main effect of time separately in each racial group, we observed a trend for increased prevalence of prostate screening with time in all 3 racial groups. However, the trend was not significant among AI/AN men (RR = 1.03; 95% CI, 0.98–1.07); but it was significant for African American men (RR = 1.03; 95% CI, 1.02–1.04) and white men (RR = 1.02; 95% CI, 1.01–1.02).

This nationally representative, population-based study of older men demonstrates that prostate screening rates in recent decades were markedly lower among AI/AN men than among African American and white men. From 1996 to 2008, rates of prostate screening within the previous 2 years were consistently 60.0% or below among AI/ANs aged 50 to 75 years, whereas they ranged from 62.0% to 73.1% among African Americans and from 68.6% to 73.3% among whites. During the 12-year period, screening rates showed a pattern of significant increase among whites and African Americans but not among AI/ANs. The disparity in screening between AI/AN men and white men was constant, while the disparity between AI/AN men and African American men increased. Our study suggests that there has been no meaningful improvement in closing the prostate screening gap between AI/AN men and their African American and white counterparts.

Our study is one of several studies to report disparities in cancer screening rates between AI/ANs and other racial/ethnic groups. Screening disparities have been reported for breast cancer, cervical cancer, and colorectal cancer (15–17). AI/ANs lag behind their same-age counterparts in all other racial groups for achieving Healthy People 2020 targets for almost all common cancer screenings (26). Consistent with previous findings of screening disparities, our findings show that AI/ANs had the lowest prostate screening rates of the 3 racial groups evaluated (12,15–17). Researchers continue to speculate on the possible reasons for these persistent disparities; recent studies have identified low socioeconomic status (15), lack of cancer care services (27,28), language differences (27), illness beliefs (27,29), limited knowledge of cancer care (27,29,30), negative attitudes toward cancer treatment (30), transportation difficulties (27), and perceived discrimination by the health care provider (31) as potential contributing factors.

Our results should be interpreted in the context of several limitations. First, our data on prostate screening were based on self-report and therefore may be subject to recall bias. Such bias might over- or underestimate screening prevalence. Second, our dependent variable was based on whether respondents received any kind of prostate examination, so we cannot make direct comparisons with other prostate screening studies on the basis of data that specified a type of screening test. Similarly, the time frame of reference in the HRS for receiving an examination was 2 years, so we cannot directly compare screening prevalence in our study with that of other national surveys that use a different time frame, such as the National Health Interview Survey. Third, although the overall HRS sample was large, the HRS does not oversample AI/ANs. Consequently, subgroup analyses included a much smaller sample of AI/AN men than men of other races. This smaller sample size of AI/AN men makes the results for AI/ANs less statistically reliable than those for African American men and white men, and our results should be carefully interpreted. Also, this small sample size prevents us from examining the regional differences in prostate screening receipt reported in previous work (11,12,15,17). Directions for future research should include improved surveillance methods and survey methodology for AI/ANs.

For AI/AN men, programs and policies will probably need to be tailored, because AI/AN populations are culturally diverse and face a wide range of barriers to health care access. From 2005 to 2015, the National Cancer Institute sponsored the Community Networks Program Centers in an effort to reduce cancer health disparities through community-based participatory engaged education, training, and research among underserved populations. This initiative was responsible for 3 centers that developed and implemented culturally and regionally tailored programs to reduce cancer disparities among AI/ANs.

In addition, policies are needed to address access barriers, particularly barriers to age-appropriate health screening. For example, urban AI/ANs have more limited access to Indian Health Services than AI/ANs who reside on tribal lands (32). Similarly, older adults may require additional support mechanisms, such as transportation, to facilitate access to health care facilities and other preventive services. Future research should assess regional differences as well as cultural barriers and facilitators to explain disparities in prostate cancer screening across racial groups.