Am Fam Physician. 2002 Feb 1;65(3):494-495.
Making decisions about health maintenance measures, specifically the use of cancer screening tests, in patients older than 75 years is difficult because of the lack of randomized, controlled trials in this patient population. Whether cancer screening is as effective in the elderly as it is in younger patients remains unclear; if screening is as effective, it is not certain what factors change the probability of risk and benefit in an individual patient. Walter and Covinsky have developed a framework for making cancer screening determinations in elderly patients. Instead of relying solely on age, this framework ties decisions on screening to life expectancy, risk of death from a cancer, and screening outcomes based on published data.
The first step is to determine the individual patient's risk of dying from a certain cancer. This risk depends on the patient's life expectancy and the age-specific mortality rates for the cancer. Determining life expectancy necessitates taking into account whether the patient is “average” for his or her age, or whether he or she could be considered in the upper or lower quartile for life expectancy. Comorbid conditions and functional impairments must be considered. Once life expectancy is determined, information about population-based estimates of the risk of cancer death can be used to calculate the risk of dying from the specific cancer in the remaining years of the patient's life (http://www.jama.ama-assn.org/issues/v285n21/fig_tab/jsc00476_html).
The next step is to determine the benefit of the screening measure (the number of patients that need to be screened to prevent one cancer death). For example, various decision models have shown that a geriatric patient who had regular and normal Papanicolaou smears before menopause does not benefit from continued screening for cervical cancer. The potential harms to cancer screening also must be considered. These harms include complications from diagnostic tests that might be ordered as a result of the screening examination, treatment of a clinically unimportant cancer, and psychologic suffering that could occur as a result of the screening.
Finally, the individual patient's preferences must be taken into account. The physician's role in helping a patient evaluate the risks and benefits of screening will vary. In some instances, the physician may need to consider the patient's values and the benefits and risks of screening, and then make a formal recommendation. In other instances, the physician and patient will consider all information and make a decision together. The authors recommend that consideration be given to omitting a cancer screening test if a patient, especially one with dementia, becomes agitated or frightened by a screening test. A patient who would refuse further treatment for the condition in question should not be screened.
In an accompanying editorial, Welch discusses some of the problems involved in screening. For example, screening may miss aggressive cancers, and there is a risk of false-positive results. He concludes that physicians and health care organizations must get away from the belief that achieving a 100 percent screening rate is a good goal. It may be that focusing on screening detracts from treatments that would offer more benefit to elderly patients.
Walter LC, Covinsky KE. Cancer screening in elderly patients. A framework for individualized decision making. JAMA June 6, 2001;285:2750–6, and Welch HG. Informed choice in cancer screening [Editorial]. JAMA. June 6, 2001;285:2776–8.
Copyright © 2002 by the American Academy of Family Physicians.
This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP. Contact firstname.lastname@example.org for copyright questions and/or permission requests.
Want to use this article elsewhere? Get Permissions