Recently, the U.S. Centers for Disease Control and Prevention issued two reports that are simultaneously scary and encouraging.

First, the scary news: A national survey conducted in 2011 found that one in every 25 U.S. hospital patients experienced a healthcare-associated infection. That’s 648,000 patients with a combined 722,000 infections. About 75,000 of those patients died during their hospitalizations, although it’s unknown how many of those deaths resulted from the infections, the CDC researchers reported in the New England Journal of Medicine.

On the bright side, those numbers are less than half the number of hospital-acquired infections that a national survey estimated in 2007. And a second report issued this week found significant decreases in several infection types that have seen the most focused prevention efforts on a national scale. Noteworthy was a 44 percent decrease in central line-associated bloodstream infections (CLABSI) between 2008 and 2012, as well as a 20 percent reduction in infections related to 10 surgical procedures over the same time period.

These infections were once thought to be inevitable, resulting from patients who were too old, too sick or just plain unlucky. We now know that we can put a significant dent in these events, and even achieve zero infections among the most vulnerable patients. At Johns Hopkins, we created a program that combated CLABSI in intensive care units through a multi-pronged approach — implementing a simple checklist of evidence-based measures while changing culture and caregivers’ attitudes through an approach called the Comprehensive Unit-based Safety Program (CUSP). The success was replicated on a larger scale across 103 Michigan ICUs and then later across most U.S. states, with impressive results.

These and similar successes have changed caregivers’ beliefs about what is possible, and inspired more efforts to reach zero infections.

What will it take to attain this goal — or at least get much closer?

We need policymakers to continue providing support so that we can mature the science of improving patient safety. We need their help to create valid and widely accepted performance measures, as well as advance implementation science so that we can learn how best to translate medical evidence into everyday bedside practice.

Hospitals have a big role, of course. As organizations, they must focus on the safety and quality of care with the same rigor and accountability that they bring to their financial performance. Almost without fail, hospital CEOs can tell you if their organization is meeting its budget goals. There are financial specialists at various levels of the organization, and there are consequences for poor performance. When it comes to patient safety, however, those structures rarely exist, even when the desire to reduce harm is strong. Some hospital CEOs I’ve met didn’t know the infection rates at their facilities. Sometimes those rates are known only by the infection prevention department.

What we need are chains of accountability that link everyone in a hospital — from the board to the frontline staff — so that everyone has a shared understanding of their organizational goals, knows their role in meeting them, and gets feedback (such as dashboards) on how they are performing. Those organizations also need the internal capacity — health care professionals with the appropriate training — to carry out their roles in this chain. It sounds simple, but clearly it’s not. Over the past year, Armstrong Institute researchers worked with the VHA hospital engagement network on a demonstration project that sought to create those accountability structures at 10 U.S. hospitals. The initial results are encouraging, with 92 percent of hospitals reporting that they have made improvements in targeted areas, such as surgical site infections (SSIs). It’s breathtaking what we can accomplish when everyone is working toward the same goal.

While leadership and accountability are crucial, we also need to empower frontline caregivers to embrace their central role in preventing infections and other harms. The root causes of patient safety problems so often are local in nature. Several years ago at The Johns Hopkins Hospital, our colorectal surgery CUSP team sought to unearth the factors that contributed to surgical site infections. Like many hospitals, they were following Surgical Care Improvement Project measures to reduce the infection risk. For instance, they were meeting the standard for delivering antibiotics within one hour of surgery. As they dug deeper, however, they found that they sometimes weren’t giving high enough doses of the antibiotics. By looking at their unique setting, and fixing the local defects they discovered in their care, they reduced SSIs by one third. The lessons from that effort have led to a federally funded project to reduce SSIs in hospitals across the nation.

There are no quick fixes to hospital-acquired infections. It’s not about the checklist, but about continually improving how we work, whether at the level of government, hospital, care team, or individual clinician. Even with the recently reported improvements, every day there are 200 people with hospital-acquired infections who die in U.S. hospitals. It is time our efforts match the magnitude of the problem.

Peter Pronovost is an anesthesiologist and director, Armstrong Institute for Patient Safety and Quality.  He blogs at Points from Pronovost.