Research Activities, August 2013
State Spotlight: New York
Drivers refer to their dashboards to check data on their speed and distance. Data-driven researchers and clinicians look at AHRQ's online dashboards from State Snapshots to gauge the overall status of health care in their States.
The dashboard at the top of the Web page for each State provides a needle gauge ranging from weak to strong, which summarizes more than 100 care quality measures. Scrolling down the page, just like looking under the hood of a car to identify problems, the Snapshots offer specific information on how a State is doing in areas such as types of care, settings of care, common clinical conditions, and special areas, including diabetes, asthma, and Healthy People 2020.
In June, Research Activities began a bimonthly column to shine a spotlight on individual States that use information from their State Snapshots. Our first column profiled Iowa. In this issue, we hit the road with the Empire State.
A New York State of health
Compared to other States, New York is in an "average category" for overall health quality, as reported on its State Snapshot. New York's weakest measures include hospital care measures and respiratory disease measures. New York's strongest measures include preventive care and nursing home care. There are larger racial and ethnic disparities in hospital care and avoidable hospitalizations but smaller disparities in nursing home care.
"I found that the information in the State Snapshots gives you the context of the State and tells you something about racial and ethnic composition, income, and other areas. It's helpful for me to not only see where New York is compared to elsewhere, but to really understand and be able to more deeply see differences, for example, between New York and Utah, which have huge differences," says Foster Gesten, M.D., medical director of the State's Office of Quality and Patient Safety. "I could tell you what my biases about other States would be, but when I go to the State context, I can actually look at the dials and see why we're different."
The Snapshots help Gesten gauge progress and develop plans.
"Some of our investments in primary care or in chronic disease management flow from being able to see where in New York we are outliers," says Gesten. "Certainly data, whether it's from AHRQ or other sources that show where New York is relative to avoidable hospitalizations and readmissions has been a driver for us in our reform efforts, whether it's patient-centered medical homes (PCMHs), moving patients into managed care, or developing health-home programs focusing on individuals at high risk of avoidable hospitalization."
The data serve as a benchmark for New York. "We see areas where we're lagging or need to focus to help provide justification to the legislature, to the budget office, and to ourselves. Looking at some of the data around where New York stands on diabetes measures, for example, or asthma measures, led us to develop a benefit policy a few years ago to have Medicaid pay for certified asthma educators and certified diabetic educators," explains Gesten.
More recently, he has used data from different sources to make investments in primary care and in PCMHs. He saw this as a "real opportunity to try to institutionalize the chronic care model through creating incentives for practices to transform along the lines of the PCMH."
"One of our biggest challenges is our diversity—from the very rural farming counties that aren't very populated to New York City, which is a focus for immigration—so the health department thinks about all kinds of infectious diseases from all around the world, different cultures' traditions, and being able to provide care to all these diverse populations," says Jonathan P. Curtin, M.D., medical director in the Division of Provider Relations and Utilization Management at the Office of Health Insurance Programs at the New York State Department of Health.
"To try to summarize the state of health care in New York is always perilous," says Gesten. "New York is a multi-volume story."
As president of the Niagara Health Quality Coalition and developer of an independent Web site for patients, Bruce Boissonnault busts what he calls "myths" about publishing performance measurements: patients wouldn't know how to use the data, patients wouldn't understand the data, and publishing the data wouldn't make a difference.
"When we began, patients believed that all hospitals were about the same or they sometimes had the erroneous belief that if a hospital was good at heart attack care or open heart surgery care or something else they were famous for, there would be a halo effect and people would assume that they were good at everything," Boissonnault told Research Activities. "I think we've successfully shown that a hospital can be statistically significantly better, have better results consistently year after year in heart attack care, but, for example, might be average or below average for congestive heart failure."
Another myth is that patients won't know how to use the information. Says Boissonnault, "We did research. The way I describe it is if you ask a classroom full of 5-year-olds what is albuterol, if none of them have asthma, none of them will know what you're talking about. But if it's a class of kids who have asthma, they'll usually pull out their inhalers and, say, ‘This is albuterol.' So, if it's your laparoscopic cholecystectomy that you're facing, you're likely to know what that means, or if it's your acute myocardial infarction that you're at higher risk of, you're liable to know what that means. This notion that people won't know what to do with the data is false. The emphasis of all of our work, however, is to help you to have a more informed discussion with your doctor, not for you to play doctor."
Since publishing his State's hospital performance data, Boissonnault points to improvements. "Statewide, mortality has improved for the measures that we publish that have never been published before. Mortality rates have improved by an average of more than 50 percent," he explains. "That does not mean that it's gone from 60 down to 10; it means that the average mortality rate for heart attack was 4 percent, and now it's 2 percent in the State."
We know that health care itself is improving, but some of the improvement is plausibly related to the fact that the measures are out there. He notes that 12 out of 14 hospital error rates the State measures have improved. There's been a 25 percent reduction in the number of hospitals doing procedures below the volume thresholds recommended in the scientific literature [considered to correlate with good outcomes].Essentially these are the thresholds that underpin the AHRQ Volume Indicators.
A few years ago, Boissonnault got a call from a reporter at a business paper questioning the value of posting hospital performance information for the public. "The reporter said, ‘Gee, we just saw that only five percent of consumers use your report when making a decision on where to be hospitalized,' and I said, ‘That sounds like a home run. I'm thrilled.'" But Boissonnault also warns: "Our research suggests that a lot of this progress could be reversed if the measures go away. The thing we have done right is do it every year."