Are hospitals safer now?
Are hospitals safer now?
TJC data may show yes, but OIG isn't sure
When you talk to officials at The Joint Commission (TJC) about the annual sentinel event reports released in July, they are quick to tell you that the data represent just a portion of sentinel events in the country, because reporting them to the organization is voluntary. But it's hard not to draw conclusions. Look at some of the rarer events reported, such as discharging an infant to the wrong family — it hasn't happened since 2010, when there was a single case. Or ventilator deaths, which went from 39 reported events two years ago to just one thus far this year.
Even those that are most common among the reports, such as wrong site/wrong person/wrong procedure errors, have declined precipitously, from 846 in 2010 to an annualized rate of about 120 this year.
While the bulk of sentinel events have been self-reported for as long as TJC has created these reports, the percentage has increased from just below 64% in 2004 to 72% in the first quarter of this year. That increase partly explains the climb in the number of events reported in the data: 740 five years ago, 810 in 2010 and 1,243 last year. The annualized rate for 2012, based on the 225 events reported in the first quarter of this year, looks to be around 900.
Still, don't read into the rates, says Gerard Castro, MPH, project director for patient safety initiatives at TJC. "This is all voluntary data, and is just a small portion of what's going on out there," he says. "This is just the proverbial tip of the iceberg, and I don't think we can draw any conclusions about the frequency of events."
OIG report on adverse events
Indeed, the HHS Office of Inspector General (OIG) found that of all the adverse events in U.S. hospitals, just 12% meet reporting requirements for states — just half of which require adverse event reporting — and only 1% of events are reported. If those numbers are used to extrapolate from TJC data, then the 900 events that might be the total reported to the commission this year may really be more like 90,000.
The OIG report looked at a sample of 780 Medicare patients in 2008, and found that 13.5% had an adverse event happen to them, and 13.5% experienced temporary harm. Of the incidents that the OIG discovered, it found 32 that should have been reported to states but were not; only one of those was actually reported to the hospital itself. The rest, the report notes, may not have even been recognized by staff as being adverse events, even though six were associated with patient death, and others required serious medical intervention.
That proves what Castro says about the need to take the data part of the report with a grain of salt. But he notes that it still provides insight into the kinds of events reported and the emphasis that hospitals are putting on certain areas to address them. For example, if they are working on reducing medication errors — down from 345 in 2010 to 44 in 2011 and 45 in the first quarter of this year — then there might be more attention paid to that kind of event, and thus more reporting.
That there are a lot of reports on retained foreign objects may be due to the increased complexity of medical devices, Castro says. "It might be a guide wire associated with a catheter, rather than a medical instrument," he says.
No uniform solutions found
That increasing complexity — throughout the entire healthcare industry — means that there are more things that can go wrong, says Paul Schyve, MD, senior advisor for healthcare improvement at TJC.
"The kinds of retained objects we see reports of now didn't exist a few years ago," he says. "And we see issues related to electronic health records now that we didn't see before. There are a lot of safety advantages related to them for patients — avoiding the perils of misreading someone's handwriting, providing information on drug-drug interactions — but as we have introduced them, more and more people identify risks that occur because of them." Take the example of computerized physician order entry. If someone programs in a mistake about a standing order for a particular drug in a particular class of patients, you can end up providing a wrong dose not just for a single patient, but for dozens.
Alerts that pop up in the electronic records may become less alarming and more pestering, leading people to become immune to them, Schyve says. "Then they won't pay attention to them, and a significant risk might be ignored." What happens when a system goes down and there is no back up? These issues weren't even considered to be risk factors for sentinel events before, but they are now. "Changes in technology that can be good for patients can also change where the risk is."
What Schyve says you can tell from the data is that no uniform solutions have been found to the things that happen most often to harm patients. And that gives quality managers a list of things they can look at to see what they have in place and what might need to be addressed. "This won't tell you what to do, but it will tell you what to pay attention to," Schyve says.
He adds that many organizations have tried valiantly to solve these issues and made amazing progress — the kind of progress that leads them to share their solutions with their peers. But the results can't always be replicated. The first few years of the reports, the root cause analyses showed that one of the main issues resulting in a sentinel event was imperfect communication. "So we told everyone to pay more attention," says Schyve. But if it was that simple, then the problems would have all been solved in all situations. It doesn't work like that, though. Every situation is different; the problem may be communication between nurses and doctors in one hospital, and between nurses changing shift at another. What works in one hospital may not make any difference in another.
There are some commonalities, though, and you can see those in the data collected on root cause analyses for sentinel events. Certain items turn up again and again in a variety of sentinel event situations.
Schyve says you should determine the causes that relate to your events, and then craft solutions that work for you. Your peers may have great ideas to share — and that's a great resource to tap — but don't just ask them about what worked, because it might not work for you. Also, find out what didn't pan out. You might get some other good ideas.
Using sentinel event data
One resource that Schyve says is underutilized is the Center for Transforming Healthcare, which was created just for the purpose of helping hospitals determine the causes of their problems and specific solutions that fit their needs.
Take the case of hand hygiene, he says. The center had eight hospitals pilot a project. Together, their hand hygiene rate averaged an abysmal 48%. No hospitals had the same pattern. No single cause was the same in every single hospital (although seven did share at least one cause). In a handoff project, Schyve says that only one of more than 20 causes appeared in every hospital involved. "We have used that project to develop a tool for organizations to use that steps them through how to make an accurate diagnosis of causes and then find solutions related to those causes in a database we have."
Whatever the issue, the improvement rates by those who have used the tool have been impressive, he says. The hand hygiene group improved hand-washing to 82% in just eight months.
The reason so few have made use of the center and its tools, Schyve thinks, is that you can get access only through the accreditation coordinator at your hospital. "People don't know about this because it isn't filtering down from that office to others. The coordinator can help other appropriate people gain access and use the tools."
The website for the center, along with some publicly available information on the hygiene and handoffs projects, as well as one on wrong-site surgery, is www.centerfortransforminghealthcare.org/tst.aspx.
The sentinel event data and associated root cause analysis information are something that quality and safety staff should look at and use. Even if TJC doesn't want to make any overarching comments about what the data mean, they are useful, Schyve says. "If you have a particular kind of event that happens — or even if you are just doing a prospective analysis of a kind of event, you can look at the causes related to them and use that as a starting place for your investigation."
And Castro notes that the root cause analysis data are more scientifically reliable, since they're a sample of the reports TJC receives, not an imperfect self-reported sample of what exists.
So use the report, Schyve concludes. If you see a spike in the number of a particular kind of sentinel event, you might use that as an impetus to look at your own systems and procedures around it. TJC itself can use an increase as rationale for putting new emphasis on a topic. "If it's at the top of your mind, you naturally pay more attention and put it higher on the list of priorities."
The Joint Commission reports are available online at www.jointcommission.org/assets/1/18/General_information_1995-1Q2012.pdf for the general information and http://www.jointcommission.org/assets/1/18/Root_Causes_by_Event_Type_2004-1Q2012.pdf for root causes. The OIG report can be found at oig.hhs.gov/oei/reports/oei-06-09-00092.pdf.
For more information on this story, contact Paul Schyve, MD, senior adviser for healthcare improvement, and Gerard Castro, MPH, Project Director, Patient Safety Initiatives, The Joint Commission, Oakbrook Terrace, IL. Email: [email protected] and [email protected].
When you talk to officials at The Joint Commission (TJC) about the annual sentinel event reports released in July, they are quick to tell you that the data represent just a portion of sentinel events in the country, because reporting them to the organization is voluntary.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.