Can life expectancy and QALYs be improved by a framework for deciding whether to apply clinical guidelines to patients with severe comorbid disease?

BACKGROUND: Guidelines with short-term harms and long-term benefits are often applied to chronically ill patients who may not benefit. The payoff time framework has been proposed (i.e., do not apply a guideline if a patient's life expectancy (LE) is shorter than when a guideline's cumulative incremental benefits first exceed its cumulative incremental harms), but its health impact is unclear.
OBJECTIVE: To investigate whether the payoff time framework improves LE and/or quality-adjusted life-years (QALY) for chronically ill patients.
METHODS: I evaluate impact of the payoff time framework on LE and QALYs, assuming (1) high and constant background mortality rate from chronic illness (≥ 10% per year), (2) immediate guideline-related harm with probability < 1, and (3) constant guideline-related benefit that occurs over an extended time. I apply the framework to questions of whether to screen chronically ill 50-year-old women for colorectal cancer using colonoscopy, and whether to advocate intensive glucose control for chronically ill diabetics.
RESULTS: If a guideline's payoff time is greater than a patient's LE, then withholding that guideline will increase LE and QALYs for that patient. For a 50-year-old chronically ill woman with background mortality > 0.15 per year (corresponding to LE < 6.5 years), withholding CR screening will increase LE. For a diabetic with background mortality > 0.11 per year (corresponding to LE < 9.4 years), withholding CR screening will increase QALYs.
CONCLUSION: The payoff time framework may indicate when withholding a guideline with short-term harms and long-term benefits may increase LE and/or QALY.