Previous studies have shown that a person's socioeconomic status (SES) (a proxy measure that can incorporate income, wealth, education, and occupation) is associated with cancer incidence and mortality. Examining variation in cancer rates by SES can help identify health disparities and target areas for cancer control activities. The Texas Cancer Registry (TCR) collects data on every newly diagnosed case of cancer in Texas, including personal and demographic data, but does not collect data related directly to SES. Using a county-level measure of SES determined by the 2000 US Census, we compared cancer incidence and mortality rates for selected cancer sites by counties categorized into Low, Intermediate, and High SES. The cancers examined in this analysis included lung, colorectal, female breast, prostate, cervical, and all cancers collected by TCR combined. Consistent with other studies, most incidence and mortality rates were lowest in the High SES counties. However, in general, the highest incidence and mortality rates were found in counties categorized as Intermediate SES, but patterns differed by cancer site and by race and ethnicity. This study provides additional evidence that geographically related SES is associated with cancer incidence and mortality.
As for most diseases, the cause of cancer is multifactorial, meaning that many factors, both external (eg, diet, tobacco, and radiation) and internal (eg, age, sex, and genetic make-up) contribute to carcinogenesis. Socioeconomic status (SES) represents a combination of income, wealth, education, and occupation and is commonly used as an indicator of underlying (and sometimes unmeasureable) characteristics that tend to be associated with many external exposures, individual health behaviors, and access to health care. These characteristics can be on the individual level or by geographic area and can directly or indirectly influence the health of an individual.
People at a lower SES tend to have poorer access to health care, lower educational levels, and a higher prevalence of many unhealthy behaviors. Socioeconomic factors have often been examined in relation to health, disease, and mortality differences, including cancer.1 Previous studies have indicated that the higher the SES, the lower the incidence and mortality rate for many different types of cancer, including lung, stomach, cervical, esophageal, oropharyngeal, and liver; however, these studies report higher rates of breast cancer and melanoma. These studies also indicate differences in cancer survival and in stage at diagnosis of cancer in relation to socioeconomic factors.2
Because individual-level data on SES are not always available for large cancer datasets, such as those collected by the Texas Cancer Registry (TCR), large population-based studies are often not available to directly investigate this relationship. For this reason, population-based studies of cancer incidence and mortality in relation to SES often rely on geographic SES measures linked to the cancer data.
This analysis examined the relationship between county-level SES and cancer incidence and mortality rates in Texas. The purpose of the analysis is to give a basic demonstration of disparities in cancer incidence and mortality rates by SES simply by comparing the characteristics of the county people live in. Because this is an ecologic study (ie, comparing the correlation between geographic area and cancer), causality cannot be inferred. Data from this report are meant to highlight possible gaps in cancer control measures and provide a basis to further investigate explanations for these disparities.
County Socioeconomic Status
The TCR does not collect information on income, education, or occupation for individual cancer patients; therefore, we are unable to measure directly the SES of each cancer patient. However, the TCR does collect the address at the time of diagnosis, and measures of SES at the population level are captured by the US Census Bureau.3 For this analysis, we used a single crude measure of SES that is consistent with a recent US study of SES and cancer: the county-level percent of population below the federal poverty level.2
Counties were divided into 3 categories to calculate separate incidence and mortality rates. Consistent with the previous study, the cutpoints chosen were as follows: < 10% of the county population below the federal poverty level (High SES); 10% to 19.9% below the federal poverty level (Intermediate SES); and ≥ 20% below federal poverty level (Low SES).
Cancer Incidence and Mortality Rates
The TCR is a population-based cancer surveillance (reporting) system that includes incident reports of certain benign, borderline, in situ, and malignant neoplasms occurring in Texas residents. Texas hospitals and cancer treatment centers are the primary sources of case reporting. Reports are also received from outpatient clinics, freestanding pathology laboratories, and other state central cancer registries when a Texas resident is diagnosed or treated at a facility outside of Texas. The data used in this report were abstracted primarily from medical records and pathology reports.
To calculate cancer incidence and mortality rates for each category, the TCR used 2001-2005 county-specific incident cases of cancer and deaths with corresponding population data from the US Surveillance, Epidemiology, and End Results (SEER) program. Cancer mortality data for 2001–2005 were extracted from electronic files provided by the Texas Department of State Health Services (DSHS) Center for Health Statistics and collected by the DSHS Vital Statistics Unit. These files contained demographic and cause-of-death information from Texas death certificates for all deaths occurring among Texas residents.
Rates were age-adjusted to the 2000 US standard population to eliminate the effect of differences in age structure among these populations and to allow the rates to be compared directly. Rates used in this report were calculated by using SEER*Stat software (version 6.4.4). This analysis was limited to invasive cancers only. The cancers examined in this analysis included lung, colorectal, female breast, prostate, cervical, and all cancers collected by TCR combined.
Incidence and mortality rates were calculated separately for white non-Hispanics, blacks, and Hispanics of any race. Numbers of the Asian/Pacific Islander and American Indian/Alaskan Native groups were insufficient to examine separately when further divided into socioeconomic groups. Race and ethnicity are associated with cancer incidence and mortality, and the population distribution varies considerably by region, even within counties in Texas. As a result, we found considerable evidence that overall rates were confounded by race and ethnicity. We were unable to adequately control for this confounding in calculating overall rates. Therefore, all rates by SES groups are presented stratified by race and ethnicity.
The Figure presents a county map of Texas showing how each county was categorized by SES. Most of the Low SES counties are found close to the Texas-Mexico border. The corresponding distribution of the population by SES groups varies substantially by race and ethnicity (Table 1) [PDF]. For the total population, 14% reside in High SES counties, 73% in Intermediate SES counties, and 13% in Low SES counties. More than 19% of white non-Hispanics live in High SES counties compared with only 7% of Hispanics and 9% of blacks. Conversely, 28% of Hispanics live in Low SES counties compared with 4% of blacks and 5% of white non-Hispanics.
All Cancer Sites
A total of 429,455 incident cancer cases and 169,374 cancer deaths were recorded by the TCR during 2001–2005. Examining all-site cancer rates by each race and ethnicity separately, cancer incidence rates are higher in the Low and Intermediate SES counties compared with the High SES counties (Table 2) [PDF]. However, among blacks and white non-Hispanics, the incidence rates were highest among those in the Intermediate SES counties. The pattern of mortality rates by race and ethnicity for each SES group is similar to incidence, with the exception that mortality rates in white non-Hispanics were lowest in the Low SES counties (Table 3) [PDF].
For lung cancer, incidence rates were higher in the Low and Intermediate SES counties for each race and ethnicity except for white non-Hispanics, where rates in the Low SES counties were roughly equal to those in the High SES counties (Table 2). Lung cancer rates were substantially lower overall among Hispanics, and although their rates were higher in the Low and Intermediate SES counties, not much disparity was seen between rates from each group. The greatest disparity in incidence rates was among blacks, with rates more than 50% higher for Low and Intermediate SES counties compared with High SES counties. Lung cancer mortality rates had similar patterns as incidence (Table 3). However, the mortality rate among white non-Hispanics was even lower in the Low SES counties, and a greater disparity was found in rates for the Intermediate and Low SES counties among Hispanics compared with the High SES counties.
Both blacks and white non-Hispanics had higher incidence rates in both the Intermediate and Low SES counties compared with the High SES counties (Table 2). Overall rates were lowest among Hispanics, and very little difference was found by SES group. Colorectal cancer mortality rates were highest in the Intermediate SES counties for each group (Table 3). However, only among blacks were the mortality rates in the Low SES counties higher than in the High SES counties.
Breast Cancer – Females
Breast cancer incidence rates were higher for each race and ethnicity in the Intermediate SES counties than in the High SES counties (Table 2). Comparing Low and Intermediate SES with High SES counties, we found the disparity in rates greatest among blacks. For breast cancer mortality, rates were still highest in the Intermediate SES counties, but less difference was found in rates for each race and ethnicity than for incidence rates (Table 3). Only among Hispanics was the mortality rate in the Low SES counties higher than in the High SES counties.
Overall, prostate cancer incidence and mortality rates were significantly higher among blacks. By SES, prostate cancer incidence rates were highest in the Intermediate SES counties for white non-Hispanics and blacks, but very little difference in rates was seen when Low SES counties were compared with High SES counties (Table 2). In Hispanics, the incidence rates increased with decreasing SES. Among white non-Hispanics, we found little difference in mortality rates by SES. For blacks and Hispanics, mortality rates increased with decreasing SES (Table 3) but were not statistically different from High SES counties.
The incidence rate of cervical cancer overall is substantially higher among Hispanics compared with blacks and white non-Hispanics. The incidence rates were slightly higher for Intermediate and Low SES counties compared with High SES counties in white non-Hispanics and Hispanics (Table 2). In blacks, the incidence rates were only higher in the Intermediate SES counties. For mortality, rates were unstable for blacks in High and Low SES counties because very few deaths occurred (11 and 13, respectively). Among white non-Hispanics and Hispanics, mortality rates were higher, but only slightly, in the Intermediate and Low SES counties (Table 3).
For this report, we applied an ecologic methodology to assess SES differences in cancer among Texas residents. We examined the rates separately for white non-Hispanics, blacks, and Hispanics and considered five cancer sites and all cancer sites collected by the TCR combined in relation to a single measure of SES, the percentage of persons below the federal poverty level. The overall pattern is generally consistent with patterns recognized in other state and national studies of SES and cancer. Cancer incidence and mortality rates tend to be higher in lower SES populations than in higher SES populations. However, in general, we found the highest incidence and mortality rates in counties that were categorized as Intermediate SES.
Although patterns differed by race or ethnicity for each specific cancer site, blacks generally showed a markedly greater disparity of incidence and mortality rates than did white non-Hispanics and Hispanics when High SES counties were compared with lower SES counties. Among white non-Hispanics, the Low SES counties actually had lower mortality rates than did the High SES counties for all cancer sites combined, lung, colorectal, and female breast cancer. We suspect this may be due to misclassification of overall county SES level by race and ethnicity. Additional research is needed to determine if other explanations can account for this uncharacteristic result.
Multiple explanations may be offered for the differences in cancer rates seen among SES groups. Lifestyle and behavior factors typically differ by SES: eg, smoking rates and the awareness of the health effects of smoking, diet and physical activity, and reproductive factors.4-6 In addition, we find substantial differences in access to health care, including lack of health insurance and access to physicians and health care facilities. Lack of health insurance has been shown to be associated with higher cancer mortality rates and a later stage at diagnosis.7-11 Persons without insurance are less likely to be screened for cancer, have a usual source of care, have had a health care visit in the past 12 months, and are more likely to delay or not get care due to cost.11,12 For Texas in 2007, estimates suggest 5,741,038 (27%) nonelderly adults (aged 18-64 years) were without health insurance, which is the highest percentage in the United States. By race and ethnicity, Hispanics had the highest uninsured rates with 41%, followed by blacks with 26%.13,14
An important limitation of this study is the potential for ecologic fallacy (ie, interpreting the area or aggregate county socioeconomic effects as individual-level effects). Many factors unrelated to SES could account for the differences in rates observed. In addition, we may expect that the effects of SES will be smaller in magnitude by using ecologic associations rather than individual associations by SES.15-20 However, in the absence of individual level SES measures in the TCR data, the ecologic associations may still be useful for cancer control planning, for resource allocations, and for improving access to cancer screening programs and other cancer control interventions.
Because of the ecologic approach, some misclassification by poverty level will occur. In this analysis, the poverty level was determined at the county level, which assumes that everyone of each race and ethnicity in a county falls into that category. Although obviously not true, at the ecologic level, this approach does capture certain common aspects of their experiences. For example, in medically underserved counties (http://bhpr.hrsa.gov/shortage/muaguide.htm), regardless of a person's SES, everyone in the county experiences this barrier to health care to some degree. We suspect a possible reason that we generally found the highest incidence and mortality rates among the Intermediate SES group versus the Low SES group is due to misclassification of SES level. Similarly, misclassification of either SES or ethnicity might also explain why white non-Hispanics in the Low SES counties tended to have lower incidence and mortality rates than those in the High SES counties.
This study provided additional evidence that SES plays a role in disparities of cancer incidence and mortality rates in data specific for Texas. Consistent with previous studies, we found that the poorer counties in Texas bear a greater burden of cancer than do higher SES counties. Continued effort is needed to increase access to health care, decrease risk factors for cancer, and increase awareness of cancer prevention measures. Additional research is needed to better understand the effect of SES on cancer incidence and mortality in Texas and the specific factors underlying SES that contribute to the disparities. Ultimately, this knowledge can help inform decision making in the implementation of cancer control measures.
The authors would like to acknowledge the technical assistance of George Lara, research specialist, for linking the County Attributes data from the 2000 US Census to the Texas Cancer Registry incidence and mortality datasets.
- Pamuk E, Makuc D, Heck K, Reuben C, Lochner K. Socioeconomic Status and Health Chartbook. Health, United States, 1998. Hyattsville, Maryland: National Center for Health Statistics. http://www.cdc.gov/nchs/data/hus/hus98cht.pdf [PDF].
- Singh GK, Hankey BF, Edwards BK. Area socioeconomic variations in US cancer incidence, mortality, stage, treatment, and survival, 1975-1999. NCI Cancer Surveillance Monograph Series, Number 4. Bethesda, Maryland: National Cancer Institute; 2003. NIH Publication No. 03-5417. http://seer.cancer.gov/publications/ses.
- Census 2000 data for the United States. US Census Bureau. http://www.census.gov/census2000/states/us.html. Accessed October 27, 2008.
- Lawrence D, Fagan P, Backinger CL, Gibson JT, Hartman A. Cigarette smoking patterns among young adults aged 18-24 years in the United States. Nicotine Tob Res. 2007;9(6):687-697.
- Hanson MD, Chen E. Socioeconomic status and health behaviors in adolescence: a review of the literature. J Behav Med. 2007;30(3):263-285.
- Phares TM, Morrow B, Lansky A, et al. Surveillance for disparities in maternal health-related behaviors--selected states, Pregnancy Risk Assessment Monitoring System (PRAMS), 2000-2001. MMWR Surveill Summ. 2004;53(4):1-13.
- Roetzheim RG, Pal N, Tennant C, et al. Effects of health insurance and race on early detection of cancer. J Natl Cancer Inst. 1999;91(16):1409-1415.
- Halpern MT, Bian J, Ward EM, Schrag NM, Chen AY. Insurance status and stage of cancer at diagnosis among women with breast cancer. Cancer. 2007;110(2):403-411.
- Brookfield KF, Cheung MC, Lucci J, Fleming LE, Koniaris LG. Disparities in survival among women with invasive cervical cancer: a problem of access to care. Cancer. 2009;115(1):166-178.
- Halpern MT, Ward EM, Pavluck AL, Schrag NM, Bian J, Chen AY. Association of insurance status and ethnicity with cancer stage at diagnosis for 12 cancer sites: a retrospective analysis. Lancet Oncol. 2008;9(3):222-231.
- Ward E, Halpern M, Schrag N, et al. Association of insurance with cancer care utilization and outcomes. CA Cancer J Clin. 2008;58(1):9-31.
- Potosky AL, Breen N, Graubard BI, Parsons PE. The association between health care coverage and the use of cancer screening tests. Results from the 1992 National Health Interview Survey. Med Care. 1998;36(3):257-270.
- Kaiser Family Foundation. http://www.statehealthfacts.kff.org. Accessed November 10, 2008.
- DeNavas-Walt C, Proctor BD, Smith, J. Income, Poverty, and Health Insurance Coverage in the United States: 2006. Washington, DC: US Census Bureau; 2007.
- Krieger N, Williams DR, Moss NE. Measuring social class in US public health research: concepts, methodologies, and guidelines. Annu Rev Public Health. 1997;18:341-378.
- Singh GK, Miller BA, Hankey BF. Changing area socioeconomic patterns in U.S. cancer mortality, 1950-1998: Part II – Lung and colorectal cancers. J Natl Cancer Inst. 2002;94(12):916-925.
- Singh GK, Miller BA, Hankey BF, Feuer EJ, Pickle LW. Changing area socioeconomic patterns in U.S. cancer mortality, 1950-1998: Part I – All cancers among men. J Natl Cancer Inst. 2002;94(12):904-915.
- Singh GK, Siahpush M. All-cause and cause-specific mortality of immigrants and native born in the United States. Am J Public Health. 2001;91(3):392-399.
- Smith GD, Wentworth D, Neaton JD, Stamler R, Stamler J. Socioeconomic differentials in mortality risk among men screened for the Multiple Risk Factor Intervention Trial: II. Black men. Am J Public Health. 1996;86(4):497-504.
- Ben-Shlomo Y, Smith GD. Commentary: socioeconomic position should be measured accurately. BMJ. 1999;318(7187):844-845.