Colorectal cancer screening in the United States: Trends from 2008 to 2015 and variation by health insurance coverage

Abstract

Regular colorectal cancer (CRC) screening is recommended for reducing CRC incidence and mortality. This paper provides an updated analysis of CRC screening in the United States (US) and examines CRC screening by several features of health insurance coverage. Recommendation-consistent CRC screening was calculated for adults aged 50-75 in 2008, 2010, 2013 and 2015 using data from the National Health Interview Survey. CRC screening prevalence in 2015 was described overall and by sociodemographic subgroups. CRC screening by health insurance coverage was further examined using multivariable logistic regression, stratified by age (50-64 years and 65-75 years) and adjusted for age, race/ethnicity, sex, education, income, time in US, and comorbid conditions. Recommendation-consistent screening increased from 51.6% in 2008 to 58.3% in 2010 (p < 0.001). Use plateaued from 2010 to 2013 but increased to 61.3% in 2015 (p < 0.001). In 2015, adults aged 50-64 years with traditional employer-sponsored private insurance were more likely to be screened (62.2%) than those with traditional private direct purchase plans (50.9%) and the uninsured (24.8%) (p < 0.01, respectively). After multivariable adjustment, differences between traditional employer-sponsored private insurance and the uninsured remained statistically significant. Adults aged 65-75 with Medicare and private insurance were more likely to be screened (76.3%) than those with Medicare, no supplemental insurance (68.8%) or Medicare and Medicaid (65.2%) (p < 0.001). After multivariable adjustment, the differences between Medicare and private insurance and Medicare no supplemental insurance remained statistically significant. CRC screening rates have increased over time, but certain segments of the population, especially the uninsured, continue to screen below recommended levels.

Keywords: Colorectal cancer; Insurance coverage; Screening.

Fig. 1.

Percentage of Adults Aged 50–75 Receiving Recommendation-Consistent Colorectal Cancer Screening, 2008–2015 Estimates were age-adjusted using the 2000 U.S. standard population by 5-year age groups. Recommendation-consistent screening was defined as home FOBT within the past year, sigmoidoscopy within the past 5 years and fecal occult blood testing (FOBT) within the past 3 years, or colonoscopy within the past 10 years. Percentages were weighted to account for the complex design of NHIS. Error bars are 95% confidence intervals. Numbers in bold text are significantly different (p < 0.05) from the previous assessment year. Overall differences between 2008 and 2015 were statistically significant (p < 0.05) for screening within recommendations, colonoscopy, and FOBT.

Fig. 2.

a: Recommendation-consistent colorectal cancer screening in 2015 by type of health insurance among adults ages 50–64 years. Estimates were age adjusted to the 2000 United States standard population by 5-year age groups. Screening within recommendations was defined as home FOBT within the past year, sigmoidoscopy within the past 5 years and FOBT within the past 3 years, or colonoscopy within the past 10 years. Employer-sponsored insurance also includes insurance obtained through a union. Directly purchased private health insurance includes private plans obtained through the Health Insurance Marketplace, school, or other means. Other public plans include insurance classified as public only, other government only and SCHIP. Military coverage includes coverage through TRICARE, VA, and Champ-VA. Military coverage is restricted to civilians only and includes those with military coverage and Medicare disability. Predicted probabilities (adjusted percentages) were derived from multivariable logistic regression analyses adjusting for age, sex, race/ethnicity, education, income as a proportion of the federal poverty line, comorbidity, and years in the US. Error bars represent 95% confidence intervals. All estimates were weighted to account for the complex survey design of NHIS. * Statistically significant difference (p < 0.05) in age-adjusted analysis compared to traditional employer-sponsored private insurance. + Statistically significant difference (p < 0.05) in multivariable-adjusted analyses compared to traditional employer-sponsored private insurance. b: Recommendation-consistent colorectal cancer screening in 2015 by type of health insurance among adults aged 65–75 years Estimates were age adjusted to the 2000 United States standard population by 5-year age groups. Screening within recommendations was defined as home FOBT within the past year, sigmoidoscopy within the past 5 years and FOBT within the past 3 years, or colonoscopy within the past 10 years. Medicare and private includes adults with only private insurance. Medicaid or any public includes other state-sponsored health plans. Predicted probabilities (adjusted percentages) were derived from multivariate logistic regression analyses adjusting for age, sex, race/ethnicity, education, income as a proportion of the federal poverty line, comorbidity, and years in the US. Error bars represent 95% confidence intervals. All estimates were weighted to account for the complex survey design of NHIS. *Statistically significant difference (p < 0.05) in age-adjusted analysis compared to Medicare and private. + Statistically significant difference (p < 0.05) in multivariable-adjusted analyses compared to Medicare and private.