Healthcare systems with higher levels of physician and clinical integration are more profitable and efficient, according to the results of a four-year study of nine systems.
The study also concluded that systems can learn from one another's best practices as they move toward greater integration.
"No one system has more than three or four clinical areas where they are well integrated across operating units. They really are in the early stages of integration," said Stephen Shortell, one of the researchers. Shortell is a healthcare professor at Northwestern University, Evanston, Ill.
Clinical areas studied included cardiovascular, oncology, behavioral medicine, orthopedics, obstetrics and women's health. Measurements reviewed included total profit margin, cash flow and full-time-equivalent employees per adjusted admission. Researchers performed a statistical analysis on the financial measurements, which were expressed as economic ratios.
"The best payback is in the cardiovascular area," Shortell said.
For example, in open-heart surgery, Minneapolis-based Fairview Hospitals and Healthcare Services' average length of stay for patients is four days, Shortell said. Sentara Health System, Norfolk, Va., has an even shorter average of three days, he said. The national average is seven, he said.
Fairview and Sentara have been able to reduce length of stay, and costs, by integrating care across operating units and developing post-acute-care programs, Shortell said.
However, the study also found geographic limits to integration. For example, hospitals or operating units that were more than 50 miles away from other units received fewer benefits from clinical integration.
"Integration is more difficult in larger regional areas," Shortell said. "Physicians are less likely to refer patients across greater distances."
Another barrier to integration is lack of clinical information system investment, Shortell said. Healthcare systems spend about 2% of their operating budgets on information systems, compared with 7% in other industries, he said.
"Easy access to medical records hampers integration efforts," he said.
One key factor found that helped promote integration was a "participative and supportive" corporate culture that linked trustees, management and physicians, Shortell said.
"Bureaucratic-type organizations had statistically negative associations with clinical integration and physician-system integration," Shortell said.
David Anderson, another researcher and a principal of KPMG Peat Marwick's Chicago office, said corporate culture can be improved by seeking consensus on strategic objectives, modernizing governance and management, and specifying the role of physicians within the system.
The study also identified key integration areas. They include sharing common clinical procedures among operating units; sharing a consolidated laboratory and radiology group; coordinating tertiary services such as oncology or cardiology; and integrating medical records across units with a common identifier number (See chart).
Previous findings suggest that top executives should actively promote integration within their systems; that some hospital board members should be replaced with trustees who understand the benefits of coordinating operating units; that integration is slow and sometimes costly; and that total quality management programs are effective integration tools (Oct. 28, 1991, p. 8; Nov. 2, 1992, p. 14; March 8, 1993, p. 5).
Shortell said further research is needed to assess the relationship between clinical integration and outcomes, satisfaction, expenses and utilization.
In addition to Fairview and Sentara, systems participating in the last year of the study were Baylor Health Care System, Dallas; Evangelical Health Systems, Oakbrook, Ill.; Franciscan Health System, Aston, Pa.; Health Midwest, Kansas City, Mo.; Henry Ford Health System, Detroit; Sharp Health Care, San Diego; Sutter Health, Sacramento, Calif.; and Mercy Health Services, Farmington Hills, Mich.