You can't improve what you don't measure. That quality-improvement
mantra rolls easily off the tongue. But when it comes to reducing preventable harm in hospital settings, it turns out that the existing measurement system is woefully inadequate to the task of helping hospitals identify and fix problems that are unnecessarily injuring and killing patients by the tens of thousands each year.
There are two aspects to the problem. One involves deciding what to measure and how to measure it—a hugely difficult issue as even the most strident patient-safety
advocates will concede.
The other issue is the partial, fragmented and toothless oversight system. The nation's patchwork quilt of patient-safety regulators have neither the data nor the enforcement tools for educating and incentivizing hospital officials to make more rapid progress in reducing unnecessary accidents in hospitals.
The reporting problem begins with the basic task of identifying how many people are killed each year by preventable medical errors. The Institute of Medicine
, in its 1999 landmark report “To Err Is Human,” estimated that figure at 98,000 patients. More recent estimates have pegged the number from 40,000 to 400,000—a tenfold difference.
Patient-safety experts such as physicians Peter Pronovost at Johns Hopkins Medicine and Robert Wachter at UCSF Medical Center noted in a paper last year some of the reasons for this wide variation in reports of iatrogenic patient deaths and injuries.
Few preventable incidents are solely due to faulty hospital processes or medical error, so collecting data is subject to human judgment. Hospitals with better surveillance systems could, paradoxically, wind up with the highest error rates, even as thousands of preventable deaths go unreported from less vigilant institutions. And much harm comes from misdiagnoses, which no one as yet knows how to measure.
But a few specific causes of patient harm have been well categorized, well measured and well reported. The CMS
, as part of its value-based purchasing program, now requires hospitals to report central line-associated bloodstream infections and catheter-associated urinary tract infections using definitions provided by the Centers for Disease Control and Prevention.
When they were able to accurately measure these harm incidences from one year to the next, hospitals across the country began adopting specific practices that reduced and in some cases eliminated the problems. CLABSI incidence, for instance, has fallen 41% in recent years, a recent report showed.
But not all hospital-acquired harms are like that. The CMS also bases its VBP rewards and penalties on hospital-acquired conditions such as hip fractures and sepsis rates. Many situations—not all of them under the control of hospital personnel—can lead to the falls and infections that trigger those conditions.
The National Quality Forum, with participation from major hospital associations, has developed a list of more than two dozen events that are preventable and ought to be reported. But then the question becomes to whom and for what purpose?
Last week, Modern Healthcare reported on just one part of the patchwork of regulatory oversight that exists to identify and fix unnecessary hospital errors: patients who self-report their experiences to the CMS. That triggers an investigation that too often leads to the equivalent of a bureaucratic slap on the wrist.
The CMS has only one real penalty in its tool box: cutting off funding, which could ultimately shut down a hospital. No wonder it is almost never used. Its other tools—the penalties and rewards in the value-based purchasing program and readmissions penalties—are not directly linked to reducing unnecessary patient harm.
More than half the states have created their own reporting systems for hospital-acquired conditions and harms. But their reporting requirements vary. And while some systems have been useful for triggering investigations by state health departments, others are far from rigorous. More than 20 states require nothing at all.
The time has come for a national reporting system for hospital-acquired conditions that stakeholders—operating through a group such as the National Quality Forum—agree are worth tracking because reducing their incidence will lead directly to fewer unnecessary injuries and deaths in hospital settings. Results should be publicly posted.
It will allow hospital leaders and the public to measure their performance against national benchmarks. More significantly, it will give hospital officials a valuable tool for measuring their own performance year to year, the first step in reducing patient harm.Follow Merrill Goozner on Twitter: @MHgoozner