[Skip to Content]
[Skip to Content Landing]
Invited Commentary
Dec 12 2011

What Is the Right Cancer Screening Rate for Older Adults?Comment on ”Prevalence of Cancer Screening in Older, Racially Diverse Adults“

Author Affiliations

Author Affiliations: Division of Geriatrics, University of California, San Francisco, and San Francisco VA Medical Center, San Francisco.

Arch Intern Med. 2011;171(22):2037-2038. doi:10.1001/archinternmed.2011.556

Cancer screening rates are often used to measure of the quality of medical care. To define high quality, there are established target rates for screening in younger persons. For example, the Healthcare Effectiveness Data and Information Set (HEDIS) and Veterans Affairs set target rates for screening persons aged 50 to 75 years for colorectal cancer, women aged 50 to 69 years for breast cancer, and women aged 21 to 64 years for cervical cancer.1 Age ranges for these quality measures are based on US Preventive Services Task Force guidelines. However, there are no quality measures that address appropriate target rates for cancer screening in persons older than 75 years.

This lack of target rates for cancer screening in older persons stems in part from the controversy surrounding screening in this population. Some have advocated for high breast and colorectal cancer screening rates in healthy older adults because the incidence and mortality of these cancers increase with age.2 In addition, cancer screening tests do not suddenly stop being effective at age 75 years, and many older adults live long enough to potentially benefit from early detection and treatment. For example, 80-year-old women in the United States have an average of 10 years of life remaining.3 Others have advocated for low cancer screening rates in persons older than 75 years because the benefit of continuing screening is uncertain and screening can cause substantial harm, especially in frail older adults. For example, false-positive screening results often lead to a cascade of invasive procedures that may result in complications and overtreatment of clinically insignificant disease.4,5

In this issue of the Archives, Bellizzi et al6 present cancer screening rates in persons 75 years or older in the United States based on the 2005 and 2008 National Health Interview Survey. Among persons 75 to 79 years old, 57% reported colorectal cancer screening ever, 62% reported breast cancer screening in the past 2 years, 53% reported cervical cancer screening in the past 3 years, and 57% reported prostate cancer screening in the past year. Screening rates remained at 35% to 50% for persons 80 years or older. Recalling that a physician had recommended the screening test was the strongest predictor of screening.

So does this study by Bellizzi et al6 identify a problem with the quality of care being provided to older persons? Are the screening rates identified in this study too high? It is difficult to make this determination based solely on calculations of the percentage of persons 75 years or older who report a cancer screening test during a recommended interval. On the one hand, one might argue that since 56% of US women aged 75 to 79 years will live 10 years or more, cancer screening rates around 60% may not be unreasonably high.3 On the other hand, since the risks and benefits of cancer screening are influenced by many patient factors beyond chronological age, it would be more informative to consider screening rates in the context of health status and history of screening when deciding if rates are appropriate. While at extremes in older age the likelihood of surviving long enough to benefit from potential detection and treatment of asymptomatic cancer becomes quite small and the likelihood of harm becomes quite large irrespective of other factors, Bellizzi et al6 were not able to calculate screening rates in subgroups older than 85 years owing to the small number participating in the survey.

Still, the data by Bellizzi et al6 raise the issue of whether quality measures should address the overuse of cancer screening. Currently, quality measures in cancer screening focus on increasing screening in persons younger than 75 years, but what about the problem of overscreening? Sending frail elderly persons with dementia for potentially harmful cancer screening tests is a serious problem, but quality measures have not addressed this.5,7 Perhaps overscreening has not been measured because it often requires more clinical details than measuring underscreening in younger persons who have less individual variability in health.8 For example, measures that attempt to define overscreening according to age alone do not distinguish potentially appropriate screening in a healthy 80-year-old man who has never been screened from potentially harmful screening in a frail 80-year-old man who has serious medical illness. Given the tremendous heterogeneity of the elderly population, we need to move beyond using age alone to define who is being overscreened.

Therefore, rather than trying to determine the “right” cancer screening rate in older age groups, more effort in quality measurement should be focused on subgroups in which there is little controversy that screening rates should be low and that clinicians should not recommend screening. For example, women of any age who have had a total hysterectomy for benign disease should not receive cervical cancer screening.9 Persons of any age with normal colonoscopy results should not receive additional colorectal cancer screening within 10 years.10 Persons of any age with serious, life-limiting illness that causes life expectancy to be less than 5 years should not receive cancer screening tests.5,8 Screening in these subgroups has no expectation of benefit and may lead to a cascade of procedures that cause serious harm. The goal of measuring screening rates in such subgroups should be to identify and discontinue policies and procedures that encourage potentially harmful overscreening.

In addition, future measures of the quality of care should include more than the percentage of persons who received cancer screening. While it is useful to determine screening rates among persons in whom screening will likely result in net benefit or net harm, between these 2 groups is a large number of older persons in whom screening offers small or uncertain net benefit. For these persons, the decision about whether the potential benefits of screening outweigh the risks is a value judgment that requires informed decision-making. Therefore, future quality measures should tell us the percentage of such persons who made an informed decision about whether to continue or stop cancer screening. Such measures of informed decision making might include improved patient knowledge about the realistic outcomes of cancer screening and clarity about the value placed on these outcomes. While arguments persist about what is the “right” rate of cancer screening in older persons, it seems clear that the rate of informed decision-making should approach 100%.

Back to top
Article Information

Correspondence: Dr Walter, Division of Geriatrics, San Francisco VA Medical Center, 4150 Clement St, 181G, San Francisco, CA 94121 (louise.walter@ucsf.edu).

Financial Disclosure: None reported.

Funding/Support: Dr Walter is supported by an R01 grant from the National Cancer Institute at the National Institute of Health (CA134425) and the San Francisco VA Medical Center. The funding sources had no role in the preparation, review, or approval of this editorial.

Disclaimer: The views expressed in this article are those of Dr Walter and do not necessarily reflect the position or policy of the Department of Veterans Affairs.

VA Office of Quality and Performance.  FY2010 Q4, Volume 2 Clinical Measures Specification Manual, 2010. VA Office of Quality and Performance Web site. http://www.va.gov/oig/52/reports/2011/VAOIG-11-00314-123.pdf. Accessed October 7, 2011
Badgwell BD, Giordano SH, Duan ZZ,  et al.  Mammography before diagnosis among women age 80 years and older with breast cancer.  J Clin Oncol. 2008;26(15):2482-248818427152PubMedGoogle ScholarCrossref
National Center for Health Statistics.  Life Tables of the United States, 2006. http://www.cdc.gov/nchs/data/nvsr/nvsr58/nvsr58_21.pdf. Accessed September 18, 2011
Schonberg MA, Silliman RA, Marcantonio ER. Weighing the benefits and burdens of mammography screening among women age 80 years or older.  J Clin Oncol. 2009;27(11):1774-178019255318PubMedGoogle ScholarCrossref
Walter LC, Eng C, Covinsky KE. Screening mammography for frail older women: what are the burdens?  J Gen Intern Med. 2001;16(11):779-78411722693PubMedGoogle ScholarCrossref
Bellizzi KM, Breslau ES, Burness A, Waldron W. Prevalence of cancer screening in older, racially diverse adults: still screening after all these years.  Arch Intern Med. 2011;171(22):2031-2037Google ScholarCrossref
Mehta KM, Fung KZ, Kistler CE, Chang A, Walter LC. Impact of cognitive impairment on screening mammography use in older US women.  Am J Public Health. 2010;100(10):1917-192320075325PubMedGoogle ScholarCrossref
Walter LC, Covinsky KE. Cancer screening in elderly patients: a framework for individualized decision making.  JAMA. 2001;285(21):2750-275611386931PubMedGoogle ScholarCrossref
Sirovich BE, Welch HG. Cervical cancer screening among women without a cervix.  JAMA. 2004;291(24):2990-299315213211PubMedGoogle ScholarCrossref
Goodwin JS, Singh A, Reddy N, Riall TS, Kuo YF. Overuse of screening colonoscopy in the Medicare population.  Arch Intern Med. 2011;171(15):1335-134321555653PubMedGoogle ScholarCrossref