One of the strengths of the multi-part series, “Too big to care,” that The Guardian US published in October, was the wealth of data the reporting team used to document the problems patients and employers faced when Parkview Health raised its prices over more than 10 years. The Guardian published the first of several articles on Indiana’s Parkview Health on Oct. 17, “‘Unlimited dollars:’ how an Indiana hospital chain took over a region and jacked up prices.”
Three reporters contributed to that first article: George Joseph, an investigative journalist; Will Craft, a data editor on the investigations team; and Jessica Glenza, who covers health care. In the series, the reporting team examined how hospital consolidation limits patients’ options while saddling them with big bills, medical debt and worse care, they reported.
All three contributed to the first article. Joseph wrote the next two: “Patients and employers accuse not-for-profit Indiana hospital of price gouging,” published Oct. 24, and “Indiana bill seeks to end tax breaks for non-profit hospitals that overcharge patients,” published Jan. 22.
We covered this series, but we did not go into detail about the reporting behind how Parkview Health’s prices became among the highest, harming consumers and employers. “Over more than a decade, Parkview Health has demanded that the people of north-eastern Indiana and north-western Ohio pay some of the highest prices of any hospital system in the country,” the reporters wrote.
Overcoming data challenges
In an interview, Craft explained the challenges journalists face when gathering data on health care prices. “For this story, we needed to be able to do two things,” he said. “We needed to show how much care costs at this hospital and how that changed over the years.”
Simply collecting hospital pricing data is challenging, Craft said. Even though hospitals and health insurers are required under a rule the federal Centers for Medicare and Medicaid (CMS) issued to post their prices publicly, not all insurers or hospitals have done so.
“Insurance companies want some of the data released because it helps them negotiate with hospitals, but also they don’t want to tick off the hospitals,” Craft explained. Still, insurers make some pricing data available only to hospitals and to a select few researchers, he added.
“They don’t want anyone touching that data because it’s the bread and butter behind the power health insurers have in negotiations,” he said. Health insurers use that leverage when contracting with employers, almost all of whom pay all the bills on behalf of their workers.
Showing how prices rose over time
Craft’s next challenge was showing how Parkview Health’s prices increased over time. He did a literature review and interviewed researchers who were familiar with what employers pay (called commercial rates) compared with what Medicare pays. The resulting comparison between what employers pay with what Medicare pays is called the commercial-to-Medicare ratio.
“Essentially, this ratio shows how much more expensive the cost of care is for a private payer compared with what the government pays each hospital,” Craft said.
One source of such numbers is surveys the RAND Corporation has done since 2017 on behalf of a coalition of businesses in the state, the Employers Forum of Indiana, and other employer coalitions nationwide. In May, RAND published the most recent report on what employers pay, as we reported here: “Employers pay 254% more for hospital care than Medicare, report says.”
“Rand makes available for purchase this big data set that includes the commercial to Medicare ratio,” Craft explained. “A simple way to explain that is that they look at how much money is coming in through different channels to estimate what a commercial payer, such as an employer, would pay.”
From that data, the reporting team showed that for 10 of the past 13 years, the hospitals in the Parkview Health system have been among the top 10% most expensive in the country, on average. That finding was based on their analysis of cost estimates from CMS data.
Also, the reporters explained that over two decades, Parkview Health took over six former rival hospitals and developed a network of nearly 300 sites for its physicians and other providers. That level of consolidation allowed Parkview Health to control patient referrals to its facilities and its primary care and specialist physicians. Also, Parkview Health could increase its negotiating leverage with health insurers who contract with employers, they added.
Adopting researchers’ methods
“I am a big proponent of adapting methodology from researchers,” Craft said. “I’ve done that in a number of different investigative stories.” Researchers have years of experience working with complex data sets and know how to avoid the problems reporters could encounter, he added.
“For example, when I looked at the commercial-to-Medicare ratio, I got a lot of nonsense numbers because I hadn’t adopted some of their methodology to clean and narrow the data set,” he explained. Hospital data from Medicare include different kinds of hospitals, including small facilities and nursing homes that can skew the overall estimates, Craft added.
At the end of that first article in the Guardian series, the reporting team added two paragraphs explaining their methodology. Using an approach adapted from a research paper that Health Affairs published in April 2022, they wrote, the Guardian found the 90th percentile commercial-to-Medicare estimate by analyzing every general hospital, an approach adapted from that article, “Trends In Hospital Prices Paid By Private Health Plans Varied Substantially Across The US.”
“When you use researchers’ methodology and the cleaning steps they use that have been peer-reviewed, you get a much better and more defensible methodology,” Craft commented. “By saying we’ve adopted this methodology from these researchers whose work is peer-reviewed, you give yourself a level of defense against criticism.”
