Comparative performance data now ready for use

First round of hospital performance data released

Whether or not your hospital currently is participating in the National Voluntary Reporting Initiative, you can make good use of the just-released first round of hospital performance data. The data for 400 hospitals were posted on the Centers for Medicare & Medicaid Services (CMS) web site, and another 600 hospitals are expected to have data posted by February 2004.

As a quality manager, you should take a close look at these data, advises Kathleen Catalano, director of regulatory compliance at Addison, TX-based Provider HealthNet Services. "I looked at the data for one of the Texas hospitals, and they tell you exactly what the Joint Commission [on Accreditation of Healthcare Organizations] found when it performed the last triennial survey," she says.

The public can access this information as well, she notes. "You should know what’s out there about you facility," Catalano says. "It’s amazing how many patients and their families review all of this information from here." She suggests comparing what is on the site for your specific hospital with the information you receive from ORYX core measures if you are doing the same indicators. "I’d certainly print out the information by indicator and share it with the relevant medical staff committees and medical executive committee."

Here are ways to utilize the comparative performance data:

Apply a similar model to your own internal comparative performance reports.

For example, you can determine the following for each measure, says Patrice L. Spath, BA, RHIT, a health care quality specialist with Forest Grove, OR-based Brown-Spath & Associates:

— Top 10% of JCAHO-accredited hospitals scored equal to or higher than: ______________

— Top 50% of JCAHO-accredited hospitals scored equal to or higher than: ______________

— Our hospital scored: __________________

When reviewing the comparative data on the CMS web site, Spath recommends evaluating the following three factors:

1. The difference between your performance rate and the comparison groups and whether it is statistically significant.

2. Whether your performance results represent a multiyear trend or just a one-time occurrence.

3. The gap between your performance rate and the goals your organization has established for your performance.

Have a plan for action if your organization’s performance is not satisfactory.

The publicly available information won’t be able to answer all of the questions that need answering to determine the best course of action, she says.

The data often are not risk-adjusted, Spath explains, nor are the unique attributes of the comparative organizations clearly defined, such as rural vs. urban facilities, or specialty vs. general hospitals. "It’s impossible to control for all risk factors or consider all facility characteristics. But knowing what risk factors or characteristics have been accounted for in the data comparisons can help people interpret performance comparisons across organizations."

You may need to gather additional data that allow practitioners to look at other dimensions of risk, or facility characteristics that are not adequately addressed in the comparative database, she explains. "It will also be necessary for you to review the data definitions that were used to create the measures to appreciate what the performance numbers are telling you." These definitions can be found in the Technical Appendix on the CMS hospital quality web site (

Be ready to encounter resistance.

"When you discover performance improvement opportunities, you need to get people’s attention around the organization," Spath says.

However, she cautions that individuals may react to significant variation on comparative reports in many ways. Even if the variation is statistically significant, they may choose to ignore the findings, according to Spath. "Don’t be surprised if you encounter resistance when you present unfavorable performance measurement results. It’s common for people to challenge the data’s validity when faced with such findings."

You can’t just print out the reports on the CMS site and expect people to respond to obvious performance improvement opportunities, Spath cautions. "Quality managers must present the data in a way that answers the common questions that may arise when practitioners review the comparative data results."

Here are two common questions:

1. Does our organization have sicker patients than those in the organizations we are being compared to?

Quality managers should acquaint people with the data definitions for each performance measure, Spath says. She gives the following example: For the measure "beta-blocker on arrival for patients presenting with heart attack," patients with potential contraindications or reasons for not prescribing a beta-blocker on arrival are excluded from the performance rate. "Thus, only those patients who are considered candidates for a beta-blocker on arrival are included in the denominator. The issue of Our patients are sicker’ is not applicable to this measure."

The severity of a patient’s illness doesn’t significantly affect the performance rates for the measures being reported by CMS, according to Spath. For the most part, the denominators exclude patients who would not be candidates for the intervention being evaluated because of illness severity, she explains.

2. What are the performance rates for hospitals like ours, a rural hospital with only 50 beds?

The CMS data do not break down results by hospital size, Spath acknowledges. However, she suggests that you provide this information by creating a report that lists the hospital size and urban or rural designation next to each of the hospitals reporting data in their state or region.

Or create a report that only compares your hospital to similar-size hospitals in similar rural locations, Spath suggests. If you do this, it will be necessary to find out the size and location of other hospitals in the state or region. This information can be obtained from the American Hospital Association or state hospital associations.

"Remember, however, that quality medical care should not vary significantly based on hospital size or location," Spath cautions. If the measure represents appropriate care, such as "pneumonia inpatients who receive an oxygenation assessment, arterial blood gas, or pulse oximetry within 24 hours of hospital arrival," then your goal should be 100%, she says.

Use comparative data to provide information to others.

The ultimate success of your organization’s performance measurement system hinges on the ability to fairly assess important aspects of patient care, Spath says. "It should be simple so that everyone in the organization understands how performance is being evaluated."

An effective measurement system that includes comparative data should allow practitioners and administrative leaders to make objective and valid assessments of patient care quality, she says. In addition, the measures should provide people with an accurate source of information that enables them to oversee and improve patient care services on a continual basis.

Encourage your facility to participate in the initiative voluntarily.

Otherwise, participation most likely will become mandatory per CMS, Spath adds. "It’s best if hospitals can show support for comparative performance measurement initiatives and public reporting of results through a voluntary effort."

Curiosity and professional pride are the most common reasons practitioners want to see what their performance looks like compared to other organizations, she says. Quality managers should use these motivators when discussing the hospital’s participation in the reporting initiative. "These are carrots’ for practitioners," Spath says. "Of course, we can always resort to the argument, If we don’t do it voluntarily, the regulators will force us to share performance data with the public.’"

Using "sticks" such as that is less likely to create an environment that embraces the opportunity to compare performance, and people will be less likely to initiate actions when performance is found to be undesirable, according to Spath. "Whenever possible, use carrots instead of sticks. The quality manager’s job will be much easier if everyone agrees on the value of comparing performance across organizations."

[For more information about comparative performance data, contact:

Kathleen Catalano, Director of Regulatory Compliance, Provider HealthNet Services, 15851 Dallas Parkway, Suite 925, Addison, TX 75001. Telephone: (972) 701-8042, ext. 216. Fax: (972) 385-2445. E-mail:

Patrice L. Spath, BA, RHIT, Health Care Quality Specialist, Brown-Spath & Associates, P.O. Box 721, Forest Grove, OR 97116. Telephone: (503) 357-9185. E-mail: Web:

From Quality to Excellence: Using Comparative Data to Improve Health Care Performance is a 2002 book published by Brown-Spath & Associates. The cost is $40 plus $6 shipping. To order, go to

The National Voluntary Reporting Initiative, a joint effort led by the American Hospital Association, the Federation of American Hospitals, and the Association of American Medical Colleges, is available on the CMS web site:]