The U.S. Medicare program spends about $4 billion each year on procedures related to coronary artery bypass graft (CABG) surgery. Under the prospective payment system (PPS), hospitals are paid per-case for bypass surgery based on the patient's diagnosis-related group (DRG), regardless of the intensity of care provided to each patient, except for extraordinary outlier costs. Physicians, however, are reimbursed per procedure based on the Medicare Physician Fee Schedule, in which they are paid for every additional service they provide, including daily hospital visits and consultations.
In May 1991, the Health Care Financing Administration chose four hospitals to participate in the Medicare Participating Heart Bypass Center Demonstration, in which Medicare paid a single global payment that covered both hospital and physician services provided to heart bypass patients. Three hospitals were included in the demonstration at a later date.
Two recent studies supported by the Agency for Healthcare Research and Quality (HS09559) examined hospital costs for CABG surgery. In the first study, researchers investigated the impact on costs of care for bypass patients of a single global payment to hospitals and physicians. In the second study, the researchers analyzed data collected to evaluate the Medicare heart bypass demonstration to examine the preoperative correlates of CABG hospital costs.
Liu, C-F., Subramanian, S., and Cromwell, J. (2001, Summer). "Impact of global bundled payments on hospital costs of coronary artery bypass grafting." Journal of Health Care Finance 27(4), pp. 39-54.
Three hospitals had lower direct variable costs for CABG inpatient care when Medicare paid doctors and hospitals a single negotiated price for all inpatient care for heart bypass patients as part of the 1991 Medicare bypass demonstration project. These reduced costs held even after accounting for other factors affecting CABG costs, such as patient risk factors and postoperative complications. Direct variable costs included such costs as wages paid to nurses, costs of prescriptions, number of lab tests, length of intensive care unit stay, or routine nursing. By 1993, one hospital's total direct variable cost fell 27 percent (a reduction of $2,678 per patient), another fell 18 percent (reduction of 1,848 per patient), and a third fell 12 percent (a reduction of $1,124 per patient).
However, the patterns in cost reductions across major departments were different across hospitals. By 1993, costs for bypass patients with catheterization (diagnosis-related group or DRG 106) and without catheterization (DRG 107) were reduced at the first hospital by 55 and 56 percent, respectively, in routine nursing; 41 and 38 percent, respectively, in nursing intensive care unit (ICU); 33 and 32 percent, respectively, in operating room (OR) and recovery; 32 and 28 percent, respectively, in pharmacy; and 31 percent in catheter lab for DRG 106 patients.
The cost reductions by 1993 at the second hospital came from pharmacy (38 and 35 percent, respectively), nursing ICU (26 and 38 percent, respectively), and routine nursing (21 percent for DRG 106 patients). A third hospital experienced a significant 6 percent cost increase in OR and recovery for both DRGs 106 and 107 patients by 1993.
Hospital cost reductions may have reflected changes in patient care management. For example, all three hospitals introduced a new 24-hour protocol for postsurgical ICU stays and began using shorter acting anesthetic agents to promote early extubation in the ICU. As a result, all three hospitals showed a reduction in their ICU per-patient costs ranging from 31 to 41 percent for DRG 106 patients and from 29 to 38 percent for DRG 107 patients. These reduced costs apparently did not diminish CABG quality of care based on in-hospital and 1-year mortality rates. These findings were based on analysis of micro-cost and clinical data on every Medicare patient undergoing CABG at each hospital from 1991 through 1993.
Subramanian, S., Liu, C-F., Cromwell, J., and Thestrup-Nielsen, S. (2001, June). "Preoperative correlates of the cost of coronary artery bypass graft surgery: Comparison of results from three hospitals." American Journal of Medical Quality 16(3), pp. 87-91.
Preoperative factors that influence the cost of CABG surgery can be quite different among hospitals. Therefore, results from one hospital cannot be broadly generalized to others, concludes this study. Of three hospitals studied, patient age, urgent/emergent surgical priority, previous CABG, and chronic obstructive pulmonary disease (COPD) significantly contributed to CABG hospital costs. However, the cost impact of these factors varied among the three hospitals.
For teaching hospital A, the major factors influencing CABG costs were minority race (31 percent), COPD (26 percent), and preoperative insertion of an intra-aortic balloon pump or IABP (25 percent). For teaching hospital B, previous CABG (31 percent) had the most significant cost impact by far. For the nonteaching hospital, preoperative insertion of the IABP (36 percent) and previous CABG (21 percent) were the main correlates of cost.
The cost impact of patient risk factors also differed among hospitals. Recent heart attack, stroke, hypertension, and low ejection fraction (indication of ventricular dysfunction) significantly increased costs only for the nonteaching hospital. Previous coronary angioplasty increased marginal cost in teaching hospital B but not in the others. At teaching hospital A, minority race was a significant cost driver but not at the other two hospitals.
The mean direct variable costs for CABG surgery were $10,285 at teaching hospital A, $9,880 at teaching hospital B, and $10,363 at the nonteaching hospital. The average length of stay for the CABG procedure was 12.3, 11.6, and 10.3 days for teaching hospital A, teaching hospital B, and the nonteaching hospital, respectively. Given the differences in patient risk factors and their link to hospital CABG costs among the hospitals studied, the researchers question whether optimum reimbursement for CABG procedures can be set using risk-based contracts.