Impact of Relative Conditional Survival Estimates on Patient Prognosis After Percutaneous Coronary Intervention
Some aspects of prognosis are not reflected by cumulative survival estimates. These aspects include information on the time already survived by the patient and the patient’s survival compared with the general population. Conditional survival (ie, conditional on having survived a certain period of time already) and relative conditional survival (ie, compared with the general population) do incorporate these aspects. We investigated these measures of prognosis in patients undergoing percutaneous coronary intervention.Methods and Results—
We studied 17 903 consecutive patients undergoing percutaneous coronary intervention between 2000 and 2014. Cumulative survival was estimated for patients with ST-segment–elevation myocardial infarction (n=5996, 853 deaths), non–ST-segment–elevation myocardial infarction (n=5371, 901 deaths), and stable angina pectoris (n=6536, 965 deaths) in 4 age categories. One-year conditional and relative conditional survival up to 10 years post–percutaneous coronary intervention was calculated. The results demonstrated that 1-year cumulative survival for patients with ST-segment–elevation myocardial infarction aged ≥76 years was 83%. One-year conditional survival, conditional on surviving the first month, was 92% in this group, and relative conditional survival (relative to the general population) was 99%. In younger age categories and in patients with non–ST-segment–elevation myocardial infarction and stable angina pectoris , similar patterns were found albeit less pronounced. Five-year relative conditional rendered similar results.Conclusions—
Relative conditional survival provides a comprehensive picture of patient prognosis, particularly for older patients with ST-segment–elevation myocardial infarction. Although, as expected, their cumulative survival is low, once they survive the first month after percutaneous coronary intervention, their prognosis is comparable to that of the general population. Therefore, relative conditional survival estimates provide an important, meaningful addition when discussing prognosis with patients.