The Metric and the Body Count: How Mass General Brigham Learned to Look Like It’s Saving Lives

The Metric and the Body Count: How Mass General Brigham Learned to Look Like It's Saving Lives

When Dr. Giles Boland stood before Mass General Brigham doctors in November 2024 and pointed to the system's falling mortality ratio as evidence of success, he had a point. The observed-to-expected metric, which compares actual hospital deaths against projections based on patient illness levels, had dropped sharply. MGH had gone from a ratio of 1.1 to an all-time low of 0.62. The system was saving lives, its leaders insisted, and the numbers proved it.

Nearly two years later, MGB executives are still citing those numbers, and they have multiplied them dramatically. Vizient rankings that once placed MGH at 96th and the Brigham at 72nd among academic medical centers have climbed to 37th and 36th, per Vizient data cited in the Boston Globe, which reported the latest results alongside detailed skepticism from clinicians inside the hospital system. More than 1,400 additional lives saved at MGH and Brigham and Women's Hospital from January 2023 through December 2024, according to MGB's own reporting. The quality transformation, led by a unified approach under the system's For Every Patient initiative, is working.

But ask the doctors inside the hospitals what they see, and a different picture emerges.

"It's a propaganda campaign," said Dr. Mark Eisenberg, a primary care physician and addiction specialist at Massachusetts General Hospital, speaking to the Boston Globe (and reprinted by STAT News).

The skepticism runs through nearly every conversation with clinicians at MGB who agreed to speak on condition of anonymity, citing fear of professional retaliation. They describe administrative changes that improved the numbers without noticeable changes at the bedside. And they point to something specific: a goal, set internally, to move dying patients out of the hospital and into hospice care, where their deaths would stop being counted.

That goal appears explicitly in internal MGB presentation slides reviewed by the Boston Globe. The slides, from an October 2024 MGH doctors' meeting, list as a priority the increase of hospice enrollment from two to three patients per week to 10 to 15 per week. A bullet point in the deck, bolded for emphasis, explains the logic directly: increasing hospice enrollment, it states, may also result in improved inpatient mortality performance, because a hospice death does not count as an inpatient death.

Hospice deaths are excluded from the observed-to-expected mortality calculations that Vizient and other analytics firms use to rank hospitals. Moving a patient from the hospital floor to hospice removes them from the numerator of the mortality equation while simultaneously potentially lowering the denominator if their illness severity was elevated in the documentation. The metric improves. The patient is still dead.

Dr. Rachel Sisodia, MGB's chief quality officer, did not dispute the substance of the slides when reached for comment. She called the work remarkable and said it was being shared because it warranted sharing. She attributed the mortality improvement to multiple factors: better early warning systems that identify deteriorating patients, improved infection control, more rigorous clinical documentation, and expanded hospice enrollment that genuinely better served dying patients and their families.

The documentation change is itself a significant detail. Beginning in January 2023, MGB asked clinicians to document patient illness more thoroughly, an effort Sisodia said was meant to capture the true baseline of the system's patient population. But the effect on the mortality metric is mathematical: sicker-seeming patients raise the expected death rate in the denominator. If observed deaths stay flat while expected deaths rise, the ratio drops. Sisodia said that initial documentation work explained some early movement but that subsequent improvements came from genuine clinical changes.

The early warning system, which uses predictive analytics to flag at-risk patients, launched in late 2024. Changes to sepsis protocols followed in October 2025. Those interventions, Sisodia said, drove the steepest mortality improvements.

Some clinicians who spoke with the Globe and others interviewed for this article said they believed the quality work was real in certain areas. Reducing infections had meaningfully improved outcomes, one specialist at MGH said, and the collaboration between physicians and nurses on the HRO quality huddles was genuine. That specialist asked not to be named because they were not authorized to speak with the media.

But none of the clinicians who described skepticism believed the integration itself was responsible for the mortality gains, and several said they had not noticed any dramatic change in the number of patients dying on their floors.

The stakes of the dispute extend beyond MGB's reputation. Vizient rankings are used by insurers, employers, and increasingly by patients choosing where to seek care. The observed-to-expected mortality metric is a cornerstone of how quality is defined and compared across the industry. If the metric can be meaningfully improved through hospice diversion and documentation strategy rather than clinical care, every hospital competing on the same rankings has incentive to do the same.

MGB's CEO, Dr. Anne Klibanski, cited the Vizient results in a recent message to staff, noting that Brigham ranked fourth and MGH ranked 15th out of 122 peer academic medical centers. The system has also highlighted its performance in its December 2025 annual financial filing and in multiple posts on its public website, including the March 2025 summary of two years of quality progress and the July 2025 description of its high-reliability quality huddles.

Whether the 1,400 lives claimed represents lives extended, lives saved in the intensive care sense, or simply deaths reclassified outside the counting window is the question that remains unresolved. The hospice enrollment strategy, documented in the system's own slides, is the mechanism that makes that question urgent.

Not every patient enrolled in hospice at MGB was appropriate for aggressive acute intervention, and hospice enrollment can reflect genuine goals of comfort over cure. Sisodia said hospice expansion was done with patient benefit as the primary consideration. But the internal framing of the hospice goal as a mortality metric lever, separate from and in addition to its clinical rationale, is what the clinicians found most pointed.

When asked what portion of the claimed lives saved could be attributed to the recent quality initiatives versus the enrollment shift, the MGH specialist said it was genuinely difficult to know. That uncertainty is the most honest answer the data currently permits.