the second opinion

How to find stories in the new hospital data

A healthcare correspondent suggests angles, sources, and database tips
October 15, 2014

America’s hospitals traditionally have been free of financial repercussions for mediocre care. But now, hospitals have their quality evaluated annually by the federal government—and that can provide good fodder for journalists. 

The assessments, and the penalties that come with them, can yield new data on how hospitals are performing, prompt closer looks at how federal regulations are affecting hospital revenues and changing the way care is delivered, and highlight disparities in the healthcare system.

To see how, let’s consider the release of information earlier this month about the latest “readmission penalties” levied by Medicare. The federal program announced it will reduce payments to 2,610 hospitals across 49 states where patients frequently return for a second stay within a month of leaving.

Those readmissions occur 2 million times a year and afflict nearly one of every five Medicare hospital patients. They also cost Medicare $26 billion a year, and the government estimates $17 billion of those charges might be avoidable with better care both in the hospital and afterward.

The readmissions penalties, created by the Affordable Care Act and now in their third year, are intended to claw back some of that money. As front-page articles in the Chicago Tribune, Columbus Dispatch, Rochester Democrat and Chronicle, and elsewhere attest, these penalties are good subjects given the prominence of hospitals in local communities and reader demand for information about how they’re performing. The stories can also give a glimpse of the dysfunction of the medical system in general, since under the traditional way hospitals are paid, the institutions were not punished but rewarded with a second payment.

And for journalists, the fines can be jumping-off places for other stories. For example, how are local hospitals trying to avert readmissions? While hospitals in the past often washed their hands of patients after they left, many now identify patients most likely to come back and try to avoid it by giving them free medicine, setting up appointments for doctors, and even sending nurses to their homes to oversee their recoveries.

Sign up for CJR's daily email

Another angle: Hospitals that care for a lot of low-income patients tend to have the hardest time in reducing readmissions. Following some of them can show why there’s such a health disparity in this country. 

Tackling these stories in depth means digging into the data. In analyzing the penalties, it’s worthwhile to look at the data in several ways:

  • What percent of local hospitals received fines compared to other areas of your state, other states, and the nation;
  • What the average fine was and how it compared nationally and to the state;
  • Which hospital or hospitals had the highest and lowest fines; and
  • Which hospitals saw the largest increase or decrease in fines.

At Kaiser Health News, where I’m a senior correspondent, we’ve compiled all three years of readmissions fines in a spreadsheet that you can sort and filter by city, state, county, ZIP code, or hospital name. You can download it into Excel as a CSV file. We also have a primer on how the program works, with more detail on which conditions are monitored and how fines are levied. And we have a chart with the average fines and portion of hospitals receiving one in each state. All of this is free for other organizations and individuals to use. (With a little more work, you can pull the data directly from the Centers for Medicare & Medicaid Services. See the note at the bottom about how to do that.)

And once you’ve looked into the data, there are a number of smart experts who can help you interpret it and round out your story :

Useful as this data is, it’s important to know there are some basic details CMS doesn’t provide. One is how much these fines translate to in dollar figures. Because the fines are prospective, no one knows for sure, but most hospitals have made their own estimates and some will share them with you. In comparing the penalty impact on hospitals, the percentages are the fairer way to do it anyway; a $100,000 fine for a small hospital is more significant than a $100,000 fine for an academic medical center.

A second frustration is that CMS has not yet updated the actual readmission data on its Hospital Compare website, so you can’t tell readers what percentage of patients currently return within 30 days of discharge. The readmission rates currently on Hospital Compare are based on patients who were in the hospital from July 2009 through June 2012, a year older than the period CMS looked at to determine this year’s fines. We expect CMS to update the readmission rates in December, and then they’ll be available for download at data.Medicare.gov.

Even without the actual rates, you can figure out which condition or conditions tracked by Medicare triggered the readmission penalty, using the Hospital Readmissions Reduction Program Supplemental Data file. A note on how to use that data set is below.

Along with making a good story, readmission data—and hospital data in general—provide a good place for reporters to get experience building stories from data analysis. Medicare’s doctor payment database contains 9 million records, so many that you need to be handy with statistical software. By comparison, the hospital set is small, with fewer than 5,000 records, so all you need is Excel. The skills developed while researching local hospitals can be applied when you write about other subjects like municipal budgets, sports, and business. It’s just one more reason to dig in.

*****

How to work with the data sets from the Centers for Medicare & Medicaid Services: 

  • To look up the penalty applied to a particular hospital, you’ll need to use two files. The first is Medicare’s Hospital Readmissions Reduction Program Supplemental Data File for the 2015 federal fiscal year. (Make sure you are using the file from the Final Rule, not the Proposed Rule. Also, CMS often updates the file, so use the tab labeled with the most recent correction, e.g. “Final FY15-CN Oct 2014”.) The Supplemental Data File has each hospital’s Readmission Adjustment Factor. (Medicare describes how it calculates the Adjustment Factor here.) You’ll need to convert the adjustment factor into a percentage by subtracting it from 1.
  • The Supplemental Data File does not include the hospital name or location, so you need to link it to Medicare’s Hospital General Information File. This file has hospital names and addresses and some other basic information. The best way to pull together the data is the VLOOKUP function, using the Medicare Provider ID fields. When you are looking at hospitals, pay attention not just to their names but their locations, because hospitals in different regions can share common names. Around the country there are 11 Memorials, eight Good Samaritans, and seven Mercys and St. Josephs—pick the wrong Samaritan and there’s a correction in your future.
  • The Supplemental Data file also tells you which conditions trigged a penalty for each hospital. It lists the “excess readmission ratio” for each of the five conditions Medicare evaluated this year. A ratio greater than 1 means that the hospital had more readmissions than CMS thought appropriate, given its mix of patients. You can turn the ratio into something useful for readers by describing a ratio of 1.14 as a hospital’s readmission rate being 14 percent higher than the government believes it should have been. If a hospital is penalized in one area, CMS does not reduce the penalty if a hospital had fewer readmissions than expected (a ratio below 1) for another condition.
Jordan Rau is a senior correspondent for Kaiser Health News, a nonprofit news service that is an editorially independent program of the Henry J. Kaiser Family Foundation.