Best Children's Hospitals: Understanding the Methodology

For routine ailments and procedures, hometown care is fine. For a child like Steven McDonough, who developed a rare spinal cord tumor, or Celli Sears, who needed new lungs because of an unusual condition, only a stand-alone children's hospital or a large multispecialty pediatric service in a medical center will do.

C.S. Mott Children's Hospital cardiologists and a patient.

Good luck finding out which ones truly excel. Children's facilities are way behind their adult counterparts in weighing individual performance against consensus benchmarks. Experts can't agree on how to do it. That is why U.S. News is now gathering data to rank children's hospitals by specialty, as adult institutions are in "America's Best Hospitals."

Last year, in a first step, the top 30 pediatric centers overall were ranked using death rates, information about technology in place, and other data supplied by the hospitals, as well as a survey of board-certified pediatricians. Now the 30 top-ranked hospitals in cancer, digestive disorders, heart and heart surgery, neonatal care, neurology and neurosurgery, and respiratory disorders are highlighted along with the top 30 in general pediatrics.

For advice on the information to get and how to use it, we turned to a roster of nearly 50 experts—medical directors, department chairs, and quality-assessment specialists at top U.S. children's hospitals. They were divided into specialty panels and about two months later had developed recommendations for each specialty. The proposals were reviewed by RTI International, which oversees the Best Hospitals methodology and collects the data for the rankings, and an online survey was prepared.

Hospitals were asked for information that ranged from volume of cystic fibrosis patients (in respiratory disorders) to the number of children who had moderately to very difficult heart surgery. Responses were received from 113 of the 143 hospitals invited to complete the survey. Most are members of the National Association of Children's Hospitals and Related Institutions, although several non-NACHRI members were added because of known expertise or other factors.

The basis for all of the rankings is a blend of reputation, outcome (where available), and care-related measures such as volume, nursing, and credentialing. (Terms are defined on here; a complete methodology report is available at health.usnews.com/pediatrics (.pdf).) A quick overview:

Reputation. A total of 1,100 pediatricians were surveyed by mail and asked to name up to five centers they believe to be most successful with difficult cases in their specialty. They were chosen at random from the database of the American Board of Medical Specialties, 200 for general pediatrics and 150 for each of the six specialties. In cancer, for example, only pediatric oncologists and hematologists were surveyed. Slightly more than half responded. Depending on specialty, reputation counted for 50 or 60 percent of the final score.

Outcome. Inpatient death rate is employed in heart and heart surgery and in neurology and neurosurgery. It does not work for specialties that focus on outpatients—cancer, digestive disorders, and respiratory disorders. Neonatal measures are in progress. Where available, outcome counted for 10 percent of the final score.

Care-related measures. Examples include nurse staffing, the availability of important technologies, and patient volume. These measures counted for 40 percent of a hospital's score.