July 2005 Volume 69	Number 7

Public Reporting of Performance Data

Karin Bierstein, J.D., M.P.H.
Assistant Director of Governmental Affairs (Regulatory)


ayment, as in pay-for-performance, is a quality improvement incentive that is still in the experimental phase. Public reporting of performance data — call it publicity for performance — is a more traditional motivator. It also is considerably easier to implement, since it simply allows the facts to speak for themselves.

Two major new public reporting systems are the Centers for Medicare & Medicaid Services’ (CMS’) Hospital Compare and the State of California’s Report Card on Medical Group Quality. Both are comparative databases built on voluntary reporting, and both are accessible through patient-oriented Web sites. In theory they will help hospitals and group practices identify targets for improvement and allow patients to rank practices according to their relative scores on a set of performance measures. In all likelihood, their experience also will influence the types of measures that will be used in pay-for-performance programs.

Reporting Comparative Hospital Data
Hospital Compare is the result of a partnership between CMS and the Hospital Quality Alliance, a coalition of organizations that includes the American Hospital Association, the American Medical Association and the Joint Commission on Accreditation of Healthcare Organizations. It is based on just three common clinical conditions: acute myocardial infarction, heart failure and pneumonia and 17 performance measures such as administration of aspirin and beta-blockers upon arrival and initial antibiotics within four hours. Table 1 lists the 17 measures, several of which apply to more than one condition.

The selection of measures is by necessity somewhat arbitrary, although the evidence for the effectiveness of the interventions is considered strong. Ten of them were included in the section on incentives to hospitals in the Medicare Modernization Act of 2003 (MMA). Seven more were added by consensus, with data reporting beginning only in 2005. The Hospital Quality Alliance is considering new measures related to patient safety — including infection control, which will be of interest to anesthesiologists if it includes surgical wound infection preventive measures — to add to the database.

The data are reported voluntarily to Medicare’s 53 Quality Improvement Organizations (QIOs) by nearly 4,200 hospitals (99 percent) across the country. For more information on the data collection and analysis, see the discussion under the “Professionals” tab on the CMS Hospital Compare Web site at <www.hospitalcompare.hhs.gov>.

The data are ultimately reported on the Hospital Compare Web site in the manner shown in Figure 1.

Figure 1: Comparison of Three Atlanta Hospitals on One Pneumonia Care Measure Rate Calculation: Numerator: Number of pneumonia inpatients receiving first dose of antibiotics within four hours after arrival at the hospital. Denominator: Inpatients aged ≥ 29 days with a principal diagnosis of pneumonia, or septicemia or respiratory failure and another diagnosis of pneumonia, who received antibiotics within four hours after arrival.

Patients using the Web site may select any or all of the 17 performance measures and any or all of the hospitals in the state, city or ZIP code of interest. The Web site, which went live on April 1, 2005, does a decent job of explaining the nature of the interventions and the limitations of the information to patients, advising them that:

“Hospitals, doctors, scientists and other health care professionals agree that these quality measures give a good snapshot of how well hospitals provide these specific types of care. Hospitals should try to give all of their patients the recommended care when it is appropriate. The goal for each measure is 100 percent.

A hospital’s quality is more than just its scores on these measures. Hospitals provide care for other illnesses and conditions for which measures are still under development. A hospital should be able to tell you what steps it is taking to improve its care. The information you will find on this Web site is intended to help you start a conversation with your physician or hospital about how you can best get the care you need.”

To treat hospital scores as a starting point for discussion only is appropriate. The measures cover a very small set of patients and interventions or “processes”; there is no information on outcomes of care. Nor is there any external auditing of the data supplied. Nevertheless the hope is that public disclosure of how one hospital compares to another in a growing number of domains will lead hospitals to do what they must in order to meet or exceed national and regional benchmarks.

Although reporting is optional, compliance is at the 99-percent level because hospitals that do not supply data on the 10 MMA measures would be seeing a 0.4-percent reduction in their Medicare payment update, as noted in the article by Alexander Hannenberg, M.D., “Pay for Performance: More Opportunity Than Threat,” on page 3 of this issue. Thus the Hospital Compare program uses dollars to inspire public reporting as an interim step in improving quality.

Reporting Comparative Medical Practice Data
In California the state Office of the Patient Advocate (OPA) runs a Web site that allows consumers and professionals to see how large medical group practices compare on a small set of ambulatory care and patient satisfaction measures (see Figure 2 below).

Figure 2: California Office of the Patient Advocate Ratings for San Francisco Group Practices

This is a pure publicity-for-performance program, although the clinical measures used were generated by the largest physician pay-for-performance program in the country: the one sponsored by the Integrated Healthcare Association (IHA), a collaborative leadership group of California health plans, physician groups and health care systems, plus academic, consumer, purchaser, pharmaceutical and new technology representatives. Six California health plans, including Aetna and Blue Cross/Blue Shield, participate in the IHA P4P program, which thus covers nearly 7 million patients and 45,000 doctors in 215 groups. Last year it paid out about $50 million to physicians based on their performance on the P4P standardized measurement set.

The OPA’s Web site offers patients — and, yes, payers — comparative information on group performance in two domains. The ratings are based on aggregate scores derived from specific measures contained in the Health Plan Employer Data and Information Set (HEDIS) maintained by the National Committee for Quality Assurance. Detailed information, including relative performance on the 50+ individual measures used in the OPA Quality Report Card, also is accessible on the OPA Web site. The ambulatory care measures are limited to preventive or chronic care services. Known as the “Getting the Right Medical Care” measures, they relate to:

• Asthma medication

• Childhood immunizations

• Screening for breast cancer

• Screening for cervical cancer

• Testing blood sugar.

The second domain in the OPA Report Card is patient satisfaction, with data obtained through surveys. Individual (“Patient Experience of Care“) scores are combined in an overall “Excellent,” “Good,” “Fair” or “Poor” rating. The four categories of measures are:

• Communicating with patients

• Getting treatment and specialty care

• Timely care and service

• Coordinating patient care.

All of the above measures are explained on the OPA Web site, where patients can also check the comparative performance of their HMO. In fact the HMO comparisons are the principal objective of visits to the OPA Web site, according to a University of California-Davis evaluation commissioned by OPA.

Is the information on medical groups useful? UC-Davis found that patients particularly valued data on the ease of access to specialists within a group and on timeliness of appointments or tests. (Is it not interesting, incidentally, that satisfying the demand for instant service has become a major quality indicator in health care?) The participating medical groups — whose number is growing annually — use the Report Card primarily for benchmarking performance with similar practices but have not noticed much impact on market share. Some groups are using the Report Card in their negotiations with payers.

Is Any of This Relevant to Anesthesiologists?
Until anesthesia performance measures are developed and then accepted by this specialty and others, as well as by the P4P industry, we have nothing to report. Fortunately ASA leadership and the committees on Performance and Outcomes Measurement, Economics and Practice Parameters are beginning to identify measures that will make sense for anesthesiology and pain medicine. As Dr. Hannenberg writes in his P4P article, “Maintenance of perioperative normothermia and normoglycemia and the timely administration of antibiotic prophylaxis are among those measures with strong acceptance among medical specialties and federal agencies,” and they are activities for which anesthesiologists can assume the responsibility — and the credit. There are others. Both private and governmental payers are interested in working with us to create incentives for quality improvement — or continued excellence — in anesthesia and pain medicine as soon as we are ready.