The authors propose a method to generate information relevant to the decision tree that adds additional perspective to the characterization of health quality during survival. Their approach uses survival data to distinguish two attributes of utility: prolongation of life and quality of life (QOL). Health-state transition probabilities correspond to the prolongation of life and are modeled in a discrete-time transient semi-Markov process. Quality-of-life-state transition probabilities are derived from the assumptions of a simple recurrent Markov process. They reflect events within the health-state sojourn time that differentiate perceptions of pain and suffering over a short fixed time period. Outcomes for these two dimensions of utility are highly relevant to the assessment of medical technology that might prolong life at the cost of increased pain and suffering, implying a reduced QOL. The methods are demonstrated on a subset of follow-up data from the Beta-Blocker Heart Attack Trial (BHAT).