Online rankings rankle hospitals
Insurers offering data to consumers
Despite objections from some doctors and hospitals, the state's largest health insurers soon will enable their members to do something they've never been able to do: log onto the Internet to search for the best-rated hospitals for dozens of specific surgeries and illnesses.
The Internet sites, which are run by the insurance companies, aim to tap into a basic desire of many patients who have no idea how to objectively pick a hospital and may now be relying on recommendations from relatives, friends, or their doctor. Some in the medical community, however, are unhappy about the new rating sites. Critics say the system doesn't take into account many details that are important for an accurate portrayal of the quality of care.
Dr. Thomas Lee, network president for Partners HealthCare, the parent organization of Massachusetts General and Brigham and Women's hospitals, said consumers have a right to information about which hospitals perform best, but that the data need to be better. "Are these programs good enough to advise someone where to go for their back operation or for their brain tumor?" he asked. "Our feeling is no."
The insurance company websites are being driven by employers who are trying to direct consumers to hospitals that provide the best medical care with fewer complications and deaths. Their belief is that high-quality care leads to fewer medical errors, repeat procedures, and lower costs. By making rankings public, employers hope poorly performing hospitals will be motivated to improve.
This year, Tufts Health Plan plans to launch a service allowing its 786,000 members to search for top-ranked hospitals on a database on its website. Since December, Harvard Pilgrim Health Care has allowed its 804,000 members to rank hospitals on 100 surgeries and conditions. And Blue Cross & Blue Shield of Massachusetts said in February that it would offer a hospital ranking service to its 2.5 million members starting in September.
"Employers are asking, `What are we getting for what we're paying?' " said Dr. John Fallon, chief physician executive for Blue Cross. "The market is demanding this."
Sheila Berard, 30, of Stoughton, used the Harvard Pilgrim site to pick an obstetrician. She wanted one who delivered babies at a top-ranked hospital. Brigham, Mass. General, and Tufts-New England Medical Center rated high, but when she looked at the hospitals' websites, she found Brigham had a strong midwifery program, so she settled on that.
"I would definitely use this again," she said.
Health plan executives advise consumers not to rely solely on the new ranking products in picking a hospital. They say patients should discuss the results with their own doctor, who can provide a more comprehensive picture of a hospital's strengths and weaknesses.
Despite these caveats, many hospitals and doctors say the information is sometimes misleading. In fact, while Blue Cross planned to roll out its new program last year, hospital executives who test-drove the service raised so many objections, the insurer delayed it for eight months to make changes.
"It's better to have no data than data that leads to the wrong conclusion," said Dr. David Torchiana, a heart surgeon at Mass. General and chief executive of the Massachusetts General Physicians Organization, the hospital's doctors group, in an e-mail.
Many health plans and large employers across the country already have started to use these tools, but it's not clear how quickly consumers will begin to use them in Massachusetts.
"These sites may be a small factor," said Alice Polley, vice president for clinical services at Sturdy Memorial Hospital in Attleboro. "Patients don't pick their hospital, they pick their doctor. And they pick them based on reputation, and pedigree, and whether they like them."
The Internet sites work like this: Consumers choose an illness or surgery from a drop-down menu. Then they pick the characteristics that are important to them in choosing a hospital. The websites suggest that consumers mark as very important a hospital's experience treating patients with the particular illness, lower complication rates, and lower death rates. The computer software then offers an overall ranking of hospitals on those characteristics.
But some results simply don't make sense, some doctors said. For example, on the HealthShare Technology Inc. website, which Tufts and Blue Cross will customize, a search for the best place to have a total hip replacement within 20 miles of Boston ranked Salem Hospital, a small community hospital north of the city, above esteemed academic medical centers like Mass. General, the Brigham, Tufts-NEMC, and Beth Israel Deaconess Medical Center. The reason: Salem tied for first place for mortality -- no patients died during surgery in 2002, the year from which HealthShare gathers data. Plus, Salem ranked fourth for the absence of complications -- better than most of the teaching hospitals -- because patients had fewer surgery-related problems like blood clots and ulcers.
Harvard Pilgrim purchased a product made by Subimo, based in Chicago, called Healthcare Advisor.
On these websites, community hospitals sometimes rank above teaching hospitals for various surgeries and illnesses. But the teaching hospitals often perform surgery on far sicker patients who are more likely to die during surgery and more susceptible to life-threatening complications.
HealthShare, based in Acton, and other companies look at data extracted from hospital bills, rather than reviewing patient charts and medical records, so they can't fully account for sicker patients. The bills include some information about sicker patients, and the companies consider this when computing hospitals' ranking. But the companies can't tell the difference, for example, between a patient in the initial stages of cancer and one who has advanced cancer and is at higher risk for certain surgical complications.
Health plan executives and employers say these services are a beginning.
"Don't make the perfect the enemy of the good," said Dr. John Freedman, Tufts Health Plan's medical director for clinical quality. "This information is not the be-all and end-all; we would never see it as replacing reasonable clinical judgment. But it really enables people to research their own specific condition. People want this information and we should make it available."
A community hospital may have a superb surgeon who elevates its rankings, Freedman said. Also, said Rick Siegrist, HealthShare's chief executive, small hospitals sometimes come out on top if they perform a small number of a particular procedure, too few for complications to show up.
For example, in HealthShare's website, Norwood Hospital ties for second, with Beth Israel Deaconess, for the treatment of brain cancer. Mass. General is ranked third in the category. But, Siegrist cautioned, Norwood had just 24 cases, while the larger hospitals each had more than 100.
In response to hospitals' concerns, Blue Cross and Tufts are customizing the program so that hospitals with very small numbers of cases can't be compared. They are also changing the program to ensure that differences between hospitals meet a high standard of statistical significance.
But in other cases, Siegrist said, "community hospitals can provide very good care. Our tool is not biased toward teaching hospitals or toward community hospitals."
Liz Kowalczyk can be reached at kowalczyk@globe.com. 