Like Mission, Statistical Analysis Can Improve - TribPapers

Like Mission, Statistical Analysis Can Improve

Mission continues to iterate its way toward compliance.

Asheville – When searching for the number of people who die each year from preventable hospital deaths, three numbers keep cropping up. In 1999, the Institute of Medicine, extrapolating New York State hospital data from a 1984 Harvard study, raised popular awareness when it reported that 44,000–98,000 people died each year from preventable hospital deaths. The institute said this was about equivalent to a jumbo jet crashing each day. A quarter century later, even though estimates of the US population have grown from 279,182,000 to 341,814,000, this rate is still frequently referenced.

In 2020, researchers at Yale performed a meta-analysis of eight studies on the subject, all performed after 2007, and put the number just over 22,000. The Yale group defended their number, saying all eight studies took a “more direct” approach by searching among hospital deaths for preventable causes, such as misdiagnosis, hand slips, inadequate monitoring, or avoidable delays. The 1999 study searched among patients who experienced an adverse event (a healthcare mistake) for preventable mishaps that caused harm and then for fatalities. The Yale group believed that by working backward, they reduced the number of false positives in the very subjective art of judging whether or not mishaps were avoidable.

The researchers also found that most people who died as a consequence of “avoidable” adverse events had a life expectancy of less than three months. They also acknowledged that the seminal 1999 study ushered in an era of escalated hospital safety precautions, which should have reduced preventable deaths.

Illustrating the subjectivity in calling lethal adverse events, the third number frequently cited, over 250,000, comes from a 2016 article in the British Medical Journal. That rate, the authors said, made adverse hospital events the third-leading cause of death in the US. Regardless, it is likely that all numbers should be corrected upward. Not only are some people faced with losing their jobs likely to underestimate their errors for the record (a 1995 Harvard study measured adverse drug events to be 20 times higher than numbers voluntarily reported), but really careless errors could go totally unnoticed.

To make a point, the conservative estimate of 22,000 will be used. It could be divided by the number of hospitals, but it wouldn’t be fair to compare a hospital that sees a few thousand patients a year to Mission, which in 2020 reported over 2.1 million patient encounters. According to Statista, 33.7 million people were admitted to US hospitals in 2022, and 865 million outpatients visited hospitals in 2021.

Dividing incidents by estimated patient counts shows that about 0.0024% of people visiting US hospitals died of adverse events. Meanwhile, local headlines are screaming about four very valuable people dying of avoidable causes at Mission. Mission’s immediate jeopardy surveys highlighted events that took place between April 5, 2022, and October 31, 2023. For purposes of overestimation, four preventable deaths in a year and a half would give Mission an annual rate of around 0.00013%. Mission surely has more than four preventable deaths a year, but the Centers for Medicare and Medicaid Services (CMS) still rank the hospital’s rates for complications and deaths at or better than the national average in all categories.

The above is a flyby newspaper analysis that didn’t even ask questions about how different facilities defined adverse events and how some were deemed preventable, how these numbers were standardized across hospitals, how emergency room rates vary from whole hospital rates, what consideration was given to differentiate facilities that treat high volumes of the frail elderly with multiple late-stage comorbidities from facilities that treat mostly walk-ins, and so on.

Exploring non-lethal adverse events that leave people sicker nonetheless is another can of worms. A recent study published in the New England Journal of Medicine found adverse events in 24% of admissions, preventable adverse events in 7%, and preventable “serious or higher” adverse events in 1%.

What matters, though, is the perception that patients are getting shoddy care at Mission. Patients, as well as nurses, have stories, and the number of anecdotes has piled high enough to catch the attention of authorities. Hot off the press, Asheville Watchdog released a letter from the CMS to Mission’s CEO Chad Patrick, dated March 8.

It clarified for members of the public that, although Mission’s immediate jeopardy designation had been lifted, noncompliance with conditions for participating in the Medicare and Medicaid programs continued in six areas. Mission now has until March 13 to submit another plan of correction that must include explanations for why identified shortcomings occurred in the first place, implementation schedules for corrective actions with dates, metrics for monitoring adherence to corrections and assessing the effectiveness of the plan, and names of persons charged with oversight.