Here are the stories behind the stories: how I reported them, what I learned and why they are among my favorites. They aren’t all my best stories, just some of the ones that stick with me.
Dying for Care
The initial question posed by my colleague Letitia Stein was simple: Could we measure which of the nation’s big operating chains had the worst COVID-19 outcomes then report out what set them apart?
No one – not the federal government, not academics, not the media – had attempted to measure those outcomes at the level of operating chains or by corporate landlords (aka Real Estate Investment Trusts). First, I built a unique database of the ownership webs behind the nation’s more than 15,000 nursing homes. Then, I adapted a Poisson regression initially built by the federal Health Resources Administration to award billions of dollars in performance payments, using it to identify facilities where death rates were higher than usual given local outbreaks and resident characteristics. My results were published in a consumer database.
Within 24 hours of the first story publishing, the White House reached out. The Biden administration had a proposal for nursing home reforms in the works, and staffers wanted to incorporate our findings.
In a second phase of the project, I analyzed more than 25 million rows of daily staffing data that nursing homes submit to federal regulators. We found that thousands of U.S. nursing homes went without a single nurse for a full day and night last year a trend that existed long before the pandemic. Yet the government cited fewer than 1 in 20 of these homes for substandard staffing.
This project challenges the breadth of my skill set: iterative data analysis to drive decisions, project management with multiple collaborators, the largest data set I’ve ever analyzed, translating dense federal regulations and understanding how they’re practically applied on the ground, and hundreds of difficult conversations with current nursing homes residents, their relatives, nurses, aides, administrators and inspectors.
America’s largest corporations inch forward on diversity pledges made after George Floyd’s murder
I’m lucky that veteran tech reporter Jessica Guynn invited me to partner with her for an ongoing series of stories that uses two unique datasets, built by us from company records, to monitor demographic trends from laborers to CEOs. I can now tell you the difference between a 10-K and a DEF 14A filed with the Securities Exchange Commission. And I’ve learned so much from the people who have been willing to share their stories of life in corporate America. In this piece, we identified some progress on pledges to diversify leadership made by many familiar brands after the killing of George Floyd. And for this one, Asian women explained the barriers some face breaking into the executive ranks of companies despite being well represented in middle management.
Housing assistance for veterans delayed by voucher count
While Houston leaders celebrated programs to house veterans living on the street, officials quietly battled two federal agencies over how to count the number of available vouchers. Analyzing records, I discovered that more than 1,090 months of rent went unused in one year even as hundreds of homeless veterans were put on wait lists. My reporting uncovered additional layers of a national problem that a regional GAO audit had missed.
My story spurred changes to federal regulations and local practices, which moved veterans off the streets and into housing faster. Personally, I was moved by the resilience of a former Marine as he showed me the benches, bushes and bridges he made home while unnecessarily waiting for help. I turned that tour into a narrative, interactive map with Marie De Jesus, one of my favorite photographers for capturing storytelling images and connecting with subjects.
Shaken baby conviction of father who insisted his innocence relied on science, expert
Could Robert Wilkes actually be innocent?
Several details shared during jury selection and early stages of the trial seemed unusual to me, so I dug in. My story did not definitively answer the question, but it did highlight problematic parts of the trial and put them into the national context of wrongful convictions in similar cases. It also explained the ways the evolving conclusions of science can be an ill fit for the finality of decisions made in courts.
In hindsight, I should have asked more questions about the adequacy of the defense mounted by Wilkes’ public defender and waited to publish my story until I received documentation that showed whether a Minnesota doctor had, in fact, excluded other explanations for the death.
My reporting brought the case to the attention of the Montana Innocence Project. They later found new medical evidence that the child’s death was the result of a rare disease and not murder. In 2018, a judge reversed Wilkes’ conviction.
Untreated: Addicted and Expecting
People kept talking about the growing number of babies born into withdrawals and the skyrocketing number of kids removed from their families because of drug use. In Montana, the numbers were among the highest and fastest growing in the nation. Why?
That was the driving question of a first-time investigative collaboration between the Missoulian and the University of Montana School of Journalism. I worked alongside a dozen students guided by Professor Joe Eaton and me. We identified best practices for treating mothers-to-be who use drugs, finding Montana falls short on several measures. The nine-part series documented the failures of leaders in state government and health care that, while common nationwide, were more pronounced in Big Sky Country.
Our work led to reformed state policy and the passage of a new safe harbor law, as well as fueling a boom in doctors licensed to prescribe buprenorphine. An Esquire writer responded by typing a few critical words for a state legislator.
Bridging the gaps left by Indian Health Service
20 years: That’s how much sooner Native Americans, on average, die than white Montanans. Poor access to basic, let alone quality, health care is a critical reason. Montana has long ranked as having the highest uninsured rate among Native Americans. The Affordable Care Act and then Montana’s expanded Medicaid program were touted by federal and state leaders as ways to close that gap.
But how well have they done?
In a four-part series, I explored how expanded access to insurance can dramatically improve the care options for Native Montanans and increase revenue for hospitals that serve them, the practical and regulatory barriers that hinder those improvements, the federal outreach failures in Indian Country, why some Natives are skeptical about signing up, the work of IHS hospital leaders to expand services (hindered by the federal agency being slow to provide support for increased billing), and numerous tribal programs that are finding ways to make their communities healthier outside federal and state systems — often by growing insurance revenues faster than those government programs. The project relied on extensive in-person interviews, data analysis and government records that were difficult to secure. (In some cases, I had to convince sources to share information superiors had directed them not to release and do so in a way that would not jeopardize their jobs.)
For me, the series emphasized the rewards of tedious reporting to understand complex regulations and the trust it can create with sources wary of parachute journalists, who sometimes focus on the problems and not the people finding solutions despite them. The expertise I developed allowed me to move quickly on an enterprise story a few months later. Outside of indigenous media, I wrote the most comprehensive analysis of how the GOP plan to replace the Affordable Care Act would impact Indian Country. The project also helps me write fuller stories about other health challenges spanning Montana communities, including a shortage of mental health treatment facilities, a spike in meth and opioid use, and the fact the state has one of the nation’s highest suicide rates.
Montana students, on average, did not improve scores on standardized tests
There is nothing special about this brief. We have done better reporting on the significance of these scores. But it does include an interactive database of school-level test results that most parents would never see otherwise. For me, this chart is an example of the long-term value of some interactive designs. I built it years ago and takes only a few minutes to update it each year with the new results.
At NICAR 2018, I organized a panel with Rachel Alexander and Kai Teoh titled “Recycle dataviz.” We talked about how to maximize the value of your work by doing things that can be used again and again with quick updates, sharing ideas for data sets that can fuel multiple stories, visualizations that can have a long lifespan and how to reuse code for a variety of applications.
Sugar Land reinvents the refinery site that built the town
This story was one of highs and lows. I was proud of my narrative writing, interactive elements and video. But I relied too heavily on white historical sources so I missed a critical component of the story.
In short, Sugar Land was, from the start, a company town. The closure of the sugar refinery began a new era for the city that would ultimately reshape the industrial site into a multi-use development. What I didn’t learn from my interviews with city leaders, development officials, neighborhood residents and hours at the library was that the refinery had for decades relied on sugar cane produced by enslaved Black people and then with convicted Black people, which was the same practice by a different name. The error makes me reluctant to share the story. It was a searing reminder to reach out to diverse group of residents, particularly when writing about a community’s history. I corrected the record a little (just a little) with a follow-up blog post that included the personal memories of several residents.
JAYME FRASER 