For WNYC series on disparities in maternity care, roadblocks force a reporter to stay flexible

“Have a plan, but expect to ditch it,” a news mentor drilled into my mind 25 years ago. “If you’re well prepared but open to wherever the story leads you, the journalism gods will reward you.”

It’s obvious but durable guidance that has served me well and that I’ve passed on countless times to up-and-comers.

Still, when you’ve spent what turned out to be years researching a proposal before beginning reporting, you don’t expect it to end up in a far place from where it started.

I began by looking at hospitals’ different approaches to childbirth and what that meant for both quality of care and costs. I ended up focused almost exclusively on potentially deadly complications — and why they’re endemic at certain hospitals serving particular communities.   

The project — one of the more satisfying in my career — ended up being an object lesson in being flexible, dealing with stumbling blocks, digging deep, and getting lucky.

***

In 2014, I came across two intriguing graphs on vaginal delivery and C-sections, posted by the New York State Health Department. They showed costs and claims for each procedure for every hospital in the state, over the course of several years. At some hospitals, the cost and claim were both low; in others they were both high; and in many there was a wide spread between cost and claim.

I showed the graphs to a health economist, and together we hatched a project to see what patients (and their insurers and employers) got for their money: Did high-cost places have lower complication rates (adjusted for patient complexity) that justify what they charge? We planned to dive deep into the state’s rich and byzantine hospital data. It took us a year of persistence to get access it.

We were plotting how best to unspool the millions of rows of patient data, when the Center for Health Journalism invited me to apply for the 2016 National Fellowship. With the money, we planned to hire a grad student to spend dozens of hours cleaning up the database, so we could analyze it. 

And then the health economist decided to focus on other projects and “broke up” with me.

***

I quickly had to pivot. I took one of the subplots I had planned to investigate and made it into my new focus.

I had previously covered reports by the city health department on mothers dying during childbirth — or coming close. These were epidemiologic studies that looked at the subject, along racial, demographic and geographic lines. I had hoped the state data would let me pinpoint which specific hospitals were the deadliest.

But without the data, that approach seemingly went out the window.  I could explore the subject, of course, but not in a way that compared the track records of individual hospitals. But no matter, there was still plenty to write about, especially since one of the city studies broke down near-fatal complications by neighborhood.

Maternal mortality has long gotten much attention, and deservedly so: rates in the United States are scandalously high, and getting worse, while racial disparities are disgracefully wide. But, even so, maternal deaths are relatively rare: they occur around 700 times annually nationwide. In New York City, perinatal deaths occur a few dozen times a year. Again, that’s way too often, but it doesn’t necessarily provide enough data, by itself, to draw conclusions at the institutional level. 

Near-fatal complications, though, occur 100 times more frequently than deaths. These are problems like severe hemorrhage, infection and cardiac arrest that can kill women. The CDC and a handful of areas have increasingly been focusing on what’s formally known as Severe Maternal Mortality, or SMM, but it hasn’t received much news coverage. I had reported on city research showing that the city’s rate was much higher than the national one (232 per 10,000 deliveries versus 204 per 10,000), with black women suffering these complications three times more often whites.

I honed in on one Brooklyn neighborhood, Brownsville, where the complication rate was off the charts, at 500 per 10,000, or 5 percent. That’s an incredibly high proportion of patients coming near death as they bring new life into the world.

***

Who are these women, and why is their health so poor — even compared to other low-income populations? And how much of what goes wrong has to do with who they are and where they live, and how much of it is attributable to poor medical care?

I began reporting on all these dimensions. I spoke with numerous obstetricians, epidemiologists, nurses, midwives, doulas, community activists and malpractice attorneys. I reached out through many of these to locate specific mothers. To find those stories, I also got help from WNYC’s engagement editor, Elaine Chen, with whom I crafted questionnaires, Tweets and posts to Facebook and other social networks to find voices.

Mothers are often passionate about sharing their birth stories, and I was quickly deluged with recollections of terrible prenatal and obstetric care. For the community health piece, I was collecting a good mix of personal stories and data, but for the piece on problems with institutional medical care, I was data deprived. 

But then I got a couple lucky breaks. Not slam-dunk lucky, but lucky enough to put some meat on the bones of the story and name names.

The first was a newly published and unprecedentedly sophisticated analysis of severe complication rates at hospitals in New York City, by an obstetrician at Mt. Sinai Medical Center. She looked at 350,000 deliveries at 84 hospitals over two years. She found that about half of the complication rate is due to the hospital where women deliver and that hospitals which largely serve black women are significantly less safe than those that largely serve white women. At the most dangerous hospital, more than 10 percent of women experienced near-death complications, and even after adjusting for patient risk, it was close to 6 percent — compared with fewer than 1 percent of women at the safest hospital.

That study did not specify hospitals by name, however, as part of the researcher’s data use agreement with the state. She wasn’t about to divulge them to me. She was interested in using her findings for “QI” — quality improvement — not for “public reporting,” which many obstetricians I talked to largely don’t believe in.

My other partially lucky break was to find new data on the state health department’s website about Potentially Preventable Complications, or PPCs, a hospital-auditing system devised by and licensed from 3M, the company best known for Scotch Tape and Post-its. There are roughly 65 PPCs that cover numerous departments in hospitals, and eight of them are obstetric. The PPCs don’t fully capture all the deadly complications the CDC combines into “Severe Maternal Morbidity,” but they do include severe hemorrhage, which is what everyone told me I should focus on. A health data and medical error expert I came to trust at Consumer Reports had mixed feelings about the PPCs: he trusted the people behind the system, but he wasn’t sufficiently confident in them to incorporate them into his magazine’s hospital scorecards.

So I took a middle path. I didn’t trumpet the state’s hemorrhage data, but I did allude to it, since those hospitals with the worst PPC rates are also in the neighborhoods cited by the city with the most severe morbidity. I really wonder whether those are the same ultra-dangerous hospitals the Mt. Sinai research specified anonymously, and I wish there were a way to share that with the public, but that will have to wait for another project on another day.

***

I had wanted to score every hospital in the state for maternity care, using a system Consumer Reports and the state Department of Insurance developed for California, and take that report card to the next level, since it’s very much a work in progress. The data team at Consumer Reports was very receptive and helpful, but it quickly became clear that formal collaboration wouldn’t be possible within the timeframe of this project, especially with a new beat assignment looming (early in 2017, I became WNYC’s statehouse reporter). Meanwhile, my newsroom’s own data team reneged on a commitment to present maternity care data alongside my reporting, and developing the necessary interface was too heavy a lift for me to handle on my own.

In hindsight, I perhaps should have extracted stronger commitments from my collaborators before going as far as I did down several rabbit holes. But I’m glad to have worked with a wide range of statistical, medical and public health experts to put the intense personal stories I found into a broader context. 

Too many women still die or come near death during delivery. Racial disparities are still too wide. And information on hospitals’ complication rates is still largely unavailable. But I’m hopeful that by shining a light on these problems in our series, “What Hospitals Deliver,” WNYC helped advance the discussion and maybe even improve the quality of local maternity care in a small way.

[Photo by Bradley Gordon via Flickr.]