Sentinel event data show little change

Could tech lead to more surgical left-behinds?

Given the lack of reporting of errors and potential errors, it should be seen as good news that the number of sentinel events reported to The Joint Commission has gone up, right? The organization figures that voluntary reporting brings it maybe 1% of the total of what's out there, but the numbers are steadily rising.

There are more reports of unintentional foreign bodies left in patients — 69 in the second quarter of this year, compared with 188 in all of last year. That might be due to more and smaller items associated with technologically advanced surgeries, says Gerard M. Castro, MPH, project director of patient safety initiatives for The Joint Commission. They've dug a little deeper into the data and found that sponges are still the number-one thing left behind, but "other" is gaining ground. And that's where the new things associated with new surgery techniques might be having an impact.

Wrong site/wrong patient/wrong procedure is number two on the list of "what bad thing is reported most" — 60 incidents during the second quarter, compared with 152 last year. He says that it is hard to think that someone would wonder whether these kind of events need to be reported. They either happened or they did not, says Castro.

But not everything is that straightforward. Paul Schyve, MD, senior director for healthcare improvement at TJC, says that he thinks delay in treatment is very underreported, with 55 reports during this quarter, and probably outstrips the number of times someone operates on the wrong body part or person. "You might not realize there was a delay, or you can rationalize it away," he says. "There are many people involved in patient care, and you might not see the delay, or the impact of the delay, even if in retrospect you can find evidence of it on a chart."

There are three things that affect the reporting of events, whether they are relatively common (at least in terms of reporting) or as rare as infant abduction (one all of last year, two in the second quarter of 2012). One is if the definition of an incident is changing, says Schyve. The second is if it is easily identifiable, and third is what kind of attention is being paid to it. "Maybe there are four, because there is also the issue of the motivation for reporting it." There isn't a lot of motivation for reporting it to The Joint Commission, for instance. And there's a lot of attention put on things like infant abductions.

We may only see a small sliver of what's out there in this report, but Schyve says you can note whether the numbers go up or down. "Every year, we put out an alert on something and the number of reports goes up. And it isn't because people are doing it more, but because it is at the front of their minds. They recognize it more." Similarly, a change in definition can impact it the other way, as in 2010, when a change in the definition of wrong-site surgery led to a drop in the reported cases from 149 in 2009 to 93.

Castro notes that if you look at the number of suicides reported, that jumped from 67 to 131 between 2010 and 2011. "Is it an increase in incidence or just reporting?" It is hard to say, he says, and hard to do more than be happy, since everyone wants to see an increase in reporting of adverse events.

The sentinel event statistics show that about 60% of occurrences since 2004 result in death, and that a further 9% result in some loss of function.

What's more interesting than looking at the number of reports in a given year and how that changes over time is looking at the root cause data available when there are enough of a particular kind of event to draw conclusions, says Schyve. "We can look at something and maybe see something that people hadn't considered before or recognized earlier." Like recognizing that there are more small bits and pieces associated with new technology in surgeries that might be easier to lose track of than a rib spreader, for example. "The really helpful information is in the deeper analysis of what we know about the factors related to these events," he says.

The root cause analysis report notes that the most common root causes of reported events are human factors, followed by issues related to leadership, communication, and assessment. After those top four, the root causes drop off precipitously in commonality. While there were between about 250 and 300 of the first four causes named in the analyses, medication use has just 42 mentions.

In events that led to death, anesthesia was the most common root cause, and communication problems were the most common root cause of delays in treatment. Falls are most often the result of assessment problems. The lessons are obvious from some of these: Do better assessments and you'll have fewer falls; improve communications and hand-offs and there might be fewer delays.

It is probable that the sentinel event report will never have everything bad in it that happens in the healthcare world. But it will have more than it does now because people are putting an emphasis on creating a just culture where reporting of events is applauded.

"It is clear that having a culture of safety, where there is trust enough that people are reporting, can be instrumental in making improvements to safety," Schyve says. "And there are two ways you can look at the major elements behind that culture of safety. First is what are the policies, procedures and expectations set by the leaders. By that, I mean the governing body, the C-suite executives, and the clinical staff leadership. They have to establish those expectations, policies and procedures that create a culture of safety."

The second piece is the personal behavior of everyone in the organization, he continues. Many organizations have leaders who develop the culture of safety but struggle with the personal behavior element. "How do you make sure that people do not act in an intimidating way or other ways that can undermine that?" Schyve asks. "How do you make sure that someone will not ignore a report, that someone will not get physically attacked or have someone throw stuff at them when they bring up a problem? Disruptive behavior may be the result of extreme pressure, but it interferes with even internal reporting of events. Forget about the external reporting — it impacts internal reporting."

Schyve says that TJC is taking "every opportunity" to talk about this particular problem and how it is disruptive and corrosive to a culture of safety, that it will lead to failure to report internally and a failure to learn. "People can feel it is too risky to speak up and raise a concern. They will watch the error occur rather than speak up if they think someone will throw something."

The four statistical reports related to sentinel events are available at

For more information on this topic, contact:

  • Paul Schyve, MD, Senior Director, Healthcare Improvement, The Joint Commission, Oakbrook Terrace, IL. Email:
  • Gerard M. Castro, MPH, Project Director of Patient Safety Initiatives, The Joint Commission, Oakbrook Terrace, IL. Email: