An Internet-based consumer health information service has devised its own formula for rating all U.S. hospitals' performance by product line and posting the results on the World Wide Web.
The service, which debuted earlier this month, also has a method for determining who among the nation's 640,000 physicians should be judged better than the rest.
In addition, the Web site has selected performance information from the National Committee for Quality Assurance and fashioned it into a rating system of one to five stars for health plans.
By compiling consumer guides from publicly available criteria, the prime movers behind HealthGrades.com hope to attract the Internet traffic that will lure advertisers, sponsors, health-related merchandisers and marketing deals.
All this comes from a publicly traded company that was mainly in the physician practice management business less than a year ago.
Specialty Care Network, a Lakewood, Colo.-based operator of musculoskeletal specialist groups, branched into the on-line information business last October when it launched a predecessor Web site, HealthcareReportCards.com, which published hospital ratings in a few clinical specialties.
Meanwhile, Specialty Care was bailing out of the troubled practice management arena, restructuring 20 contracts so physicians could buy back their practices and the company could focus on its Internet strategy, said Sarah Loughran, vice president of content.
In addition to expanding the number of services rated in hospitals and introducing physician and health plan grades, the company plans to add dentist ratings soon and eventually include ambulatory surgery, long-term care and home healthcare, Loughran said.
The hospital database in HealthGrades uses a five-star rating system to grade performance in cardiac surgery, cardiology, orthopedic surgery, back and neck surgery, pulmonary/respiratory treatment, vascular surgery and obstetrics.
Data for all areas except obstetrics are taken from clinical outcome reports that hospitals submit to Medicare annually. Obstetrics statistics are available from only 18 states that release data for all healthcare payers.
The ratings of all Medicare-reported services except orthopedics are based on mortality data that are adjusted for the severity of patients' illnesses. Orthopedic procedures are rated based on major complications.
The best-rated hospitals have a significantly lower incidence of deaths and complications than expected given such factors as patient demographics and patients' conditions when admitted (See chart). The ratings consider each hospital service separately, and mortality track records are measured during hospitalization and one month and three months after discharge.
Thus a facility can get widely variable grades depending on specialty, procedures within each specialty and differences in death rates for the same procedure during three time periods.
The American Hospital Association doesn't oppose the HealthGrades effort or any other rating program, but it's concerned about how people may use the information, said Richard Wade, senior adviser for communications. "Consumers shouldn't use any of those rating systems as the only source of making a decision about selecting a hospital or a doctor," he said.
Loughran said the site advises users that the information should not be used alone to determine quality of care, a proviso Wade said he applauded.
In the not-too-distant past, the use of Medicare mortality data and the notion of assigning stars akin to restaurant reviews sent howls of protest through the hospital industry.
HCFA started releasing Medicare mortality data in 1987 but stopped doing so in 1993 after complaints that the reporting process was not consistent year to year and that death rates were not adjusted for differing severity of illnesses and other factors beyond hospitals' control.
The Joint Commission on Accreditation of Healthcare Organizations in 1994 sought to use a rating system of one to five stars to publicly disclose results of accreditation in two dozen areas of performance. But it eventually agreed to use numerical scores after hospital representatives complained that stars oversimplified a complex process and would send the wrong message to consumers.
Wade said stars are useful only if the rating organization tells consumers how the ratings were figured, and the JCAHO did not do that. The HealthGrades site explains its methodology in detail and includes hyperlinks in charts for consumers to get additional information.
Each specialty also is adjusted for a separate set of risk factors, post-surgical complications, secondary conditions and other factors to allow comparisons. The statistical crunching was developed with help from an expert at the Medical University of South Carolina in Charleston.
Compared with the hospital rating process, the system for rating physicians is rudimentary.
No stars or relative grades are issued. Instead, physicians pass or fail based on the minimum requirements for designation as a "leading physician" by specialty. Criteria include board certification in the physician's area of practice, number of years in practice, hospital affiliation and any sanctions by Medicare or a state medical board.
Slightly less than half the nation's physicians make the list, but there's wide variation according to specialty, said Loughran. For example, a higher percentage of obstetricians make the list than plastic surgeons, she said.
Board certification is the biggest sticking point, disqualifying physicians who are board-certified but not specifically enough. For example, a practicing cardiologist would be excluded if he or she was certified only in internal medicine, Loughran said.
Physicians also have to be on the staff of at least one hospital rated three or more stars in their specialty.
A top American Medical Association official said he supported the concept but not the execution.
"The whole idea of devising some way of recognizing quality in physician performance is something the public is demanding," said Randolph Smoak Jr., M.D., chairman of the board of the American Medical Accreditation Program.
But Smoak said the HealthGrades physician site doesn't achieve that end. He criticized the criteria as too narrow to yield the best practitioners and subject to technical errors that could disqualify good doctors.
He also criticized the information-gathering from the databases HealthGrades said it mines for criteria.
The Internet company said physicians could be excluded "if their data were not included in the databases used, or because the data record for the physician contained apparent inaccuracies caught by the data integrity checks."
Those inaccuracies could be as simple as using an initial for a first name instead of the full name, Smoak said.
After reviewing results in his home state, Smoak said some physicians who haven't practiced in years were on the list, while some "well-recognized, outstanding physicians" were not.