Not anymore. Critical-care physicians—who monitor patients using an electronic system called an e-ICU—“have full privileges to do whatever they need to do in order to take care of the patient right now,” says John Hensing, Phoenix-based Banner's executive vice president and chief medical officer. The e-ICU system is installed at 80% of intensive-care beds; the remaining beds will be added in 2010.
How did Banner executives handle this politically sticky situation with local attending physicians? They presented data. They compared how many patients would have been expected to die versus how many actually died, concluding that the e-ICU saved 475 lives in 2008 and 560 lives in 2009.
The way the e-ICU was implemented at Banner demonstrates how the system's centralized approach to running 22 hospitals works. The corporate board is responsible for quality. Where they exist, local boards operate in an advisory capacity.
Responsibility for day-to-day management also is centralized at the system level. For example, hospital-level executives for medicine, finance and human resources report directly to the system—and not the CEO of the hospital. The hospital-level chief nursing officer is the only position that reports directly to the hospital CEO, Hensing says.
The centralized approach to quality and efficiency improvement at Banner may have helped it land on the list of the top 10 systems based on the second annual 100 Top Hospitals: Health System Benchmarks study from Thomson Reuters. Released exclusively to Modern Healthcare
, the study evaluated 255 systems, comprising 1,903 hospitals and representing 6.7 million Medicare discharges. Thomson Reuters also released a list of the top quintile, totaling 51 systems.
The systems were ranked on eight measures of clinical quality, efficiency and patient satisfaction.
Banner was one of five newcomers to the top 10 system list, which also includes one other organization from the Western region: Scripps Health, San Diego. The other eight members of the list were from the Midwest.
The Midwest also dominated the top quintile, although to a lesser degree than last year. In the 2010 study, 28 Midwest systems, or 55%, are in the top quintile, down from 38, or 76%, in the 2009 study.
At the same time, the Western region also gained ground in the top quintile—nearly doubling to 11 systems, or 22%, in the 2010 study, compared with six systems, or 12%, in the 2009 study.
The Northeast also increased its presence in the top quintile to seven systems, or 14%, in the 2010 study, compared with two systems, or 4%, in the 2009 study.
The number of Southern systems increased to five systems, or 10%, from four systems, or 8%.
The health systems study used data for 2007 and 2008 from the Medicare Provider Analysis and Review data set as well as the CMS Hospital Compare data set published in the third quarter of 2009.
To be included in the study, systems had to have at least two acute-care hospitals with a minimum of 25 beds each. The average number of hospitals for all systems in the study was 7.5.
To make the list of the top 10, systems must score at least as well as the median level of performance on each of the eight measures evaluated in the study, while also outranking their peers overall.
The top systems performed better than peer systems on all measures, although the difference between them was quite small for some measures. For example:
- The risk-adjusted mortality index for the top-10 systems was 0.87, compared with 1.01 for peer systems (A lower score is better).
- The risk-adjusted complications index was 0.87 for the top systems, compared with a score of 1.00 for peer systems (a lower score is better).
- The core measures mean percentage was 95.2% at the top systems, compared with 93.3% at peer systems. The score is an aggregate based on process-of-care standards for heart attack, heart failure, pneumonia and surgical-infection prevention.
- The risk-adjusted patient-safety index was 0.94 for the top systems, compared with 0.99 for peer systems (a lower score is better). The measure is based on patient safety indexes from the Agency for Healthcare Research and Quality.
- The 30-day mortality rate was 12.3% at the top systems, compared with 12.7% for the peer group.
- The 30-day readmission rate was 20.6% for the top systems, compared with 20.7% at peer systems.
- The average length of stay was 4.7 days at the top systems, compared with 5.3 days for peer systems.
- The HCAHPS (CMS' Hospital Consumer Assessment of Healthcare Providers and Systems) patient-satisfaction score was 258 at top systems, compared with 255 at peer systems.
One reason winning systems perform well: Like Banner, they tend to have centralized oversight of performance improvement.
“Where the health system boards have taken on responsibility for quality and are actively insisting on standards and goals across the system and monitoring progress, you will see much greater alignment and much faster improvement across the hospitals in the system,” says Jean Chenoweth, senior vice president for performance improvement and the 100 Top Hospitals programs at Thomson Reuters.
Accountability for quality and efficiency also resides centrally at 11-hospital Advocate Health Care in Oak Brook, Ill., which ranked among the top-10 systems for the second year.
An example of the system-based approach is Advocate Health Partners' clinical integration program—a collaboration between Advocate and 3,400 physicians to improve quality and efficiency. Physicians receive payments based on the performance of their practice group on metrics for quality and efficiency.
“If you are a solo doctor, it is just yourself, but if you are in a three-person group, it is the three of you,” says Lee Sacks, executive vice president and chief medical officer at Advocate Health Care. Managed-care plans under contract with Advocate contribute money to the incentive payments.
In 2010, the clinical integration program will score physicians on 41 measures, including 30-day readmission rates, asthma-care outcomes, generic prescribing, provision of smoking-cessation counseling and patient satisfaction.
“Physicians and system leadership have the same goals on outcomes, and they just reinforce each other and help us move forward,” Sacks says.