Study: Top performers also can be top improvers

Lower baseline doesn’t guarantee improvement

In a finding that many would consider counterintuitive, research by Evanston, IL-based Solucient demonstrates that continuous improvement over time is strongly associated with top performance.

"This research refutes the assumption that poorly performing hospitals are most likely to be advantaged when measured by rate and consistency of improvement while very high performers experience difficulty in showing continual high rates of improvement," notes Jean Chenoweth, senior vice president of performance improvement and the 100 Top Hospitals programs at Solucient’s Center for Healthcare Improvement.

"I was delighted to find that link between the two," she adds.

The finding is the result of the latest research conducted about hospitals recognized through one of Solucient’s 100 Top Hospitals programs and the 100 Top Hospitals programs at Solucient’s Center for Healthcare Improvement.

Data from previous studies of the 100 Top Hospitals: Performance Improvement Leaders and the 100 Top Hospitals: National Benchmarks for Success were analyzed to determine the relationship between improvement and performance.

The research found a "robust and statistically significant" link between performance in the Benchmarks for Success and the Performance Improvement Leaders studies.

Specifically, it found that more than one-quarter of the variability on benchmark performance may be accounted for by sustained performance improvement. There also is a weaker but significant association between rank in the Performance Improvement Leaders study and the number of years a hospital has won the national study — that is, the more times a hospital has won the national study, the greater its improvement during the studied time period.

Tracking the trends

"We’ve been doing the 100 Top Hospitals study since 1993; and a couple of years ago, we tested out a trended database," Chenoweth recalls. "It took a year to build that database and normalize it so we could use the 100 top measures and apply them backward across time." The performance measures used by Solucient are:

  • risk-adjusted mortality index;
  • risk-adjusted complications index;
  • severity-adjusted average length of stay;
  • expense per adjusted discharge, case mix- and wage-adjusted;
  • profitability (operating profit margin);
  • proportion of outpatient revenue;
  • productivity (total asset turnover rate);
  • coding specificity.

By using six years of data, she explains, hospital executives can ask themselves and their staffs important questions.

"For example, if you have a high complication rate — but three years ago, it was even higher — you’d continue what you were doing. On the other hand, if you show high expenses and the previous year they were low, you need to ask what suddenly changed in terms of your cost of operations vs. those of other hospitals. Did they get better, or did you just get worse?" asks Chenoweth.

A culture of performance improvement

As Solucient started examining the data, it realized that since its measures covered clinical, efficiency, financial, and coding data, "We could tell if the hospital was improving only clinically, but not financially, or vice versa, and thus whether across the organization there was actually a culture of performance improvement [PI]," Chenoweth says. "That goes to the heart of one of our findings — that ranking and performance improvement appeared to be linked together."

Her team didn’t believe that initially, she admits. "We thought that if you looked for the most improved hospitals, the best performers would be the big losers, but it wasn’t that way at all."

Ultimately, it became clear "the reason hospitals that are already high performers also seem to be able to improve quickly is that they have that culture already built. Across the pattern of the organization they do PI well — they are used to it," Chenoweth says,

For example, if a hospital tries to improve financially but is not working on other areas, its performance graph will look like a zigzag. "You can’t just improve in one area," she asserts. "We’ve just never shown it in data before."

The top performers/improvers also had something else in common: areas in which they are concentrating.

"We did an analysis of where the top 100 best of the best were placing their emphasis. And we found they tended to focus on cardiovascular technologies, adopting new stroke treatment, and new technologies for neurosurgery," Chenoweth notes.

"Think about it: Cardio is a cash cow; stroke is high volume, neurosurgery is one of the few growth areas for the future. These hospitals have a sense of what the community is going to need, what makes business sense, and a sense of the growth of their population. They are being very selective," Chenoweth adds.

Does improvement have limits?

So, after years and years of improvement, can a hospital come so close to its goals that significant improvement is no longer possible? Chenoweth is not sure.

"I don’t think anyone knows, but technology changes so rapidly and readily it would be hard to find a case where no improvement could be generated," she asserts. "Hospitals are very complex and mostly dependent on the behavior of human beings, so tell me — how can there not be improvement?"

In any organization, she continues, improvement is always an option.

"You can always do something better," insists Chenoweth. "In every single top 100 hospital, I can find areas where they perform poorly — no matter how many times they have won. They do tend to be early adopters [of new PI techniques], but not in everything."

She concludes by pointing out how rare PI success still is. "To give you a sense of how rare it is: Only 25% of all the U.S. hospitals significantly reduced mortality or length of stay over five years," Chenoweth notes. "Only 2% actually reduced expenses; only 4% were able to improve their profitability significantly."

PI is an enormous challenge to the health care industry, she continues, making it that much more important to be able to measure it and to measure the impact of consistently improving, and to show that it actually leads to performing at national levels.

"This is very doable; but you have to use business tools to get there, and you have to measure your performance," Chenoweth adds.

Need More Information?

For more information, contact:

• Jean Chenoweth, Senior Vice President, Performance Improvement, Solucient, 1007 Church St., Suite 700, Evanston, IL 60201. Phone: (800) 366-7526. E-mail: