Welcome to Second Opinion

Where the U.S. News health rankings editors open the floor to discuss the use and misuse of health data.


The tone and much of the message of this new blog may ring a bell with those who visited Comarow on Quality during its several-year run. As CoQ did, Second Opinion will focus on how U.S. News and other institutions evaluate healthcare quality. We will offer a behind-the-curtains view of Best Hospitals and other U.S. News health rankings—why we include certain data and exclude other types.

We will devote more space in Second Opinion, however, to discussions of what might be termed demonstrative data. These are the metrics that reveal important facets of healthcare performance, quality, and value. Finding, analyzing, and presenting such numbers—even creating them when necessary—are vital to all U.S. News health-related rankings and ratings.

We will talk about new data categories we are exploring, and how we might use them in future rankings. And we will offer our views on the data collection and dissemination efforts of others. Reader feedback will be crucial to the viability of Second Opinion. We welcome attention and comments from all, and we expect healthcare professionals in particular to have much to say about the ideas and opinions we share. If you are among them, your reactions will carry special weight in helping us refine our understanding and shaping the rankings and ratings we make public. 

During more than a quarter century at this job, I have learned the surprising elusiveness of good data. From their launch in 1990 through 1992, the Best Hospitals rankings were generated solely from reputation, using responses from specialists who were asked where they would send their hardest cases. That survey still plays an important role in the rankings, but reputation now accounts for less than one-third of a hospital's score in most adult specialties. The rest is data. The proportions could continue to shift in favor of data.


That unfortunate "if" is because trustworthy, broad-based, meaningful, on-message data are hard to find. Junk data, limited data, pointless data, abound. My colleague Steve Sternberg, who joined us a few months ago, and I all but cry in frustration about statistics that could be so much better. To cite a pet example, we would love to build into our Best Hospitals methodology solid, comprehensive data for every hospital on the incidence of sentinel events that kill and harm tens of thousands of hospital patients, but such events are critically undercounted in the data currently available. 

Too often valuable data are locked up because they are proprietary, or government agencies that collect them don't permit access, or because professional groups worry about how the data might be interpreted by the media or the public. That said, some forward-looking medical organizations and state and federal agencies appear to be moving toward greater data-openness. That trend is one we at U.S. News strongly encourage and support. 

The scarcity of good, available data has thrown a temporary roadblock into the path of every health-related product and service U.S. News has evaluated, most recently group health insurance plans, children's hospitals, diet programs, and doctors. Now we want to evaluate private health insurance plans marketed to individuals and their families and are again in uncharted territory, as Steve spells out in the next post

We could wait. The Affordable Care Act requires that plans sold on private and state exchanges be rated, based on stipulated criteria. But that mandate doesn't take effect until 2014, and in our view, that's not soon enough. Consumers deserve guidance without waiting nearly two more years. Moreover, we believe that publicly recognizing plans that offer consumers good value (and those that don't) will encourage insurers to compete to offer the best benefits for the lowest cost. Health insurance coverage needs to emerge from behind the curtains. Let's pull them back—using the right data in the right way.