BACKGROUND: The ability to predict prolonged length of stay (LOS) is essential to control escalating hospital costs. Operative mortality is a poor predictor of LOS; morbidity as defined by hospitalization for > 14 days after coronary artery bypass graft surgery (CABG), appears to be responsible for increasing costs. The purpose of this study was to measure preoperative predictive indicators of increased LOS with an eventual plan to offer alternative cost-benefit therapeutic options. METHODS AND RESULTS: Nine hundred twenty-four consecutive patients (age, 60-86 years) undergoing CABG were retrospectively studied by means of the Cox proportional hazards model. Seventeen variables, excluding death, were analyzed and quantified as to importance, and point totals were calculated for each patient. Scores were 12 for congestive heart failure and intra-aortic balloon assist device; 10, creatinine > 2; 6, intra-aortic balloon assist device only; 5, congestive heart failure only; 3, obesity; 6, age > 75 years; 3, age 70-75 years; and 2, 65-69 years. CONCLUSIONS: Increasing index score directly correlated with an exponential increase in LOS. These data substantiate the hypothesis that a mathematical model can predict LOS in CABG patients and may offer rational alternative strategies in delivering cost-effective health care.
BACKGROUND: The ability to predict prolonged length of stay (LOS) is essential to control escalating hospital costs. Operative mortality is a poor predictor of LOS; morbidity as defined by hospitalization for > 14 days after coronary artery bypass graft surgery (CABG), appears to be responsible for increasing costs. The purpose of this study was to measure preoperative predictive indicators of increased LOS with an eventual plan to offer alternative cost-benefit therapeutic options. METHODS AND RESULTS: Nine hundred twenty-four consecutive patients (age, 60-86 years) undergoing CABG were retrospectively studied by means of the Cox proportional hazards model. Seventeen variables, excluding death, were analyzed and quantified as to importance, and point totals were calculated for each patient. Scores were 12 for congestive heart failure and intra-aortic balloon assist device; 10, creatinine > 2; 6, intra-aortic balloon assist device only; 5, congestive heart failure only; 3, obesity; 6, age > 75 years; 3, age 70-75 years; and 2, 65-69 years. CONCLUSIONS: Increasing index score directly correlated with an exponential increase in LOS. These data substantiate the hypothesis that a mathematical model can predict LOS in CABG patients and may offer rational alternative strategies in delivering cost-effective health care.
Authors: Eric L Eisenstein; Charles F Bethea; Lawrence H Muhlbaier; Marie Davidian; Eric D Peterson; Judith A Stafford; Daniel B Mark Journal: J Med Syst Date: 2005-04 Impact factor: 4.460