Discussion of stopping cancer screening may go best when focused on lack of benefit instead of limited life expectancy
Survey respondents' most preferred phrase was “your other health issues should take priority”; the least preferred choice was “the doctor does not give an explanation.”
In discussions about stopping cancer screening, patients may prefer that clinicians use phrases that reference other health priorities and guidelines rather than referencing life expectancy or offering no explanation at all, a recent survey of older adults found.
In November 2016, 881 English-speaking adults ages 65 years and older (mean age, 73.4 years; 55.2% female) completed the national cross-sectional survey, which used best-worst scaling techniques to test preferences among 13 phrases a clinician may use to explain why a patient should not undergo a routine cancer screening test. Phrases were found through literature review and from previous qualitative interviews with older adults and primary care physicians.
Participants were randomized to questions about screening for prostate cancer, breast cancer, or colorectal cancer. They were given 13 choice tasks, each displaying four of the 13 phrases, and were asked to choose one best phrase and one worst phrase for each task. Researchers calculated a standardized score for each phrase by dividing the sum of assigned values (1 for each time a phrase was chosen as best, and −1 for each time it was chosen as worst) by the number of times the phrase was presented in the survey. Scores, which indicate the relative strength of preference for a phrase, ranged from −1.0 (least preferred) to 1.0 (most preferred).
Results were published as a research letter on June 28 by JAMA Oncology.
The most preferred phrase was “your other health issues should take priority” (mean score, 0.41; 95% CI, 0.39 to 0.43), and the least preferred choice was “the doctor does not give an explanation” (score, −0.42; 95% CI, −0.44 to −0.40). Other preferred phrases referenced guidelines, older age, the lack of benefit, and the high risk for harm, while less preferred phrases involved life expectancy, the discomfort or inconvenience of a screening test, and the clinician not initiating a discussion of cancer screening. There were minimal differences in preferences across cancer screening types.
The authors noted that although the survey study relied on a hypothetical scenario, it allowed them to collect perspectives from a national sample.
“Clinical practice guidelines increasingly advocate using life expectancy to inform cancer screening recommendations, and our findings highlight the importance of communicating these guidelines in language that is acceptable to and preferred by patients,” they wrote. “In particular, framing the discussion around lack of benefit, without necessarily mentioning life expectancy, may be a more appealing communication strategy.”