Lessons from efforts to reduce hospital-acquired infections

In health care reform discussions, talk inevitably turns to making hospitals and physicians accountable for patient outcomes. But in a commentary being published in the July 14 issue of the Journal of the American Medical Association, Johns Hopkins patient safety expert Peter Pronovost, M.D., Ph.D., argues that the health care industry doesn't yet have measurable, achievable and routine ways to prevent patient harm - and that, in many cases, there are too many barriers in the way to attain them.

One of the most important first steps, he says, is to eliminate the arrogance - of physicians who are overconfident about the quality of care they provide or always believe things will go right and aren't prepared when they don't, and of hospital officials who fail to aggressively address problems like hospital-acquired infections.

'It's unconscionable that so many people are dying because of these arrogance barriers,' says Pronovost, a professor of anaesthesiology and critical care medicine at the Johns Hopkins University School of Medicine. 'You can't have arrogance in a model for accountability.'

Annually, roughly 100,000 people die from health care-associated infections, another 44,000 to 98,000 die of other preventable mistakes and tens of thousands more die from diagnostic errors or failure to receive recommended therapies, he writes. Arrogance, he says, is responsible for too many of them.

Despite ongoing efforts to improve patient safety, there is limited evidence of improved patient outcomes, he says. The same scientific rigour applied to other areas of medicine needs to be applied to the study of patient safety. 'To be accountable for patient harms, health care needs valid and transparent measures, knowledge of how often harms are preventable, and interventions and incentives to improve performance,' Pronovost writes. But he also acknowledges that the science of patient safety is immature and underfunded. 'Few patient harms can be accurately measured, or the extent of preventability even known,' he writes.

One major success story, he notes, is central line-associated bloodstream infections, which are common and costly and kill 31,000 patients a year in the United States. These, however, have been proven to be accurately measured and largely preventable. Pronovost's research - which introduced a simple checklist into hospital ICUs at Johns Hopkins and then the entire state of Michigan - has shown that these infections can be brought to nearly zero. Once thought of as an inevitable risk associated with a hospital stay, Pronovost's work has shown that they can be largely avoided.

But it was not just the checklist that led to the dramatic improvements in patient safety in these ICUs, he says. Equally important was the changing of the prevailing medical cultures of each institution. In this new culture, nurses are allowed - even encouraged - to question doctors who may have skipped a step or otherwise violated safety protocols. Feedback is given constantly on infection rates so everyone knows the extent of the problem. Patient safety is put ahead of individual egos.

It is an example of how hospitals and physicians can indeed be held accountable for patient safety. Many hospitals won't report their infection rates publicly. Without knowing how big the problem is, Pronovost argues, how can it be suitably addressed?

The work to reduce these bloodstream infections is spreading to other states and there is a federal mandate to reduce them by 75 percent over three years - the 'first quantifiable patient safety goal in the U.S.,' he writes.

So why aren't all hospitals and physicians getting on board? Hospital enrolment in the program has been slow. In some states fewer than 20 percent of hospitals have volunteered to participate.

'Some hospitals have reduced infections, most have not,' Pronovost writes. 'Some hospitals claim they use the checklist, despite having high or unknown infection rates. Some hospitals are content to meet the national average, despite evidence that these rates may be reduced by half. Some hospital administrators say their patients are too sick; these infections are inevitable. Yet, intensive care units in several large academic hospitals have nearly eliminated CLASBIs, or central-line associated bloodstream infections. Some hospitals blame competing priorities for their inattention to these infections. If these lethal, expensive, measurable, and largely preventable infections are not a priority, what is?'

Working together - holding hospital leaders accountable for infection rates, getting financial incentives from insurers for reducing infections and, when needed, imposing regulatory sanctions - Pronovost says, 'we can remedy this pandemic and move on to other types of preventable harm.'