Improving surgical outcomes with data tool
Keep your finger on the pulse with objective data
The National Surgical Quality Improvement Program (NSQIP) began in 1994 in response to concern over the quality of care, specifically operative mortality rates, in VA hospitals. Since then it has expanded to all hospital settings and come under the auspices of the American College of Surgeons (ACS).
When Pierre Saldinger, MD, FACS, chairman of the department of surgery at Danbury Hospital (CT) came across the NSQIP booth at an ACS annual hospital, he decided to compel his administration to sign up for it and the hospital became one of the first outside of the initial pilot to participate. His intention was to find "a database or system that would help me do objective data measurement" and he says he found it with NSQIP.
When Danbury sends its data to NSQIP, they are all blinded and are benchmarked against data the organization already has on the 200 or so participating hospitals. Saldinger can choose to run reports "on any permutation" he wants — surgeon-specific, procedure-specific or by hospital category according to admission, scope, or bedside. And twice a year, he gets a risk-adjusted assessment of the data. This is what makes the difference, he says, setting the data apart as more objective, more useful.
When you go through the peer review process, he says, the "bottom line" of the resulting data is that "we don't know because it may be a blip or it may be real, but with those data there's really no way to tell."
The risk-adjustment, he says, eliminates that variable. "I think honestly this is the only way to track performance," Saldinger says. "It's independent, it's objective, all the complications are defined." The ACS audits the data once a year, and hospitals are allowed only a 5% discrepancy.
Crucial to the program is what NSQIP refers to as a clinical nurse reviewer. Only the nurse and Saldinger are privy to the data returned from ACS. Instead of placing her in the department of surgery, Saldinger decided to place her in the performance improvement department, where she would have more team support. She is jointly accountable to Saldinger and the chief risk officer. The three meet monthly, and he says the collaboration has been key.
Currently, Danbury has a module for both the general vascular surgery department and its bariatric surgery program. Saldinger culls data on "anything from wound infection, which has a lot of traction these days; to DVTs; to renal failure; reintubation; prolonged intubation; cardiac events" and all of these within 30 days.
The other part, he says, is it is combined with the hospital's peer review process. For instance, he says, "we had a surgeon who had a cluster of complications in a short period of time, which throws everyone into a loop and everyone gets nervous and excited about it." They compared that surgeon with a similar profile in terms of cases and complication profile. "And other than the fact that they came clustered in time, they had an almost identical profile. So now you can say, we'll keep monitoring but there doesn't seem to be a problem. It's just bad luck."
In searching through the data he and the nurse reviewer receive, Saldinger says they come across other discrepancies that wouldn't have been revealed. "If you present that to the hospital administrator," he says, "that will resonate right away because of Joint Commission readiness and the stuff that goes along with that."
Keeping your 'finger on the pulse'
Because he can pull data on any time for any measure he deems necessary, Saldinger says participation in the program allows him to "constantly keep his finger on the pulse."
One of the things his team is looking at is if they should get a cardiology workup for surgery and if it's currently applied in the right way. Who makes that decision and whose discretion it is are the indicators he's looking at.
He also says "now there's a big emphasis on normathermia so we've got an initiative to maintain body temperature, particularly in colorectal surgery. The whole initiative that pertains to glycemic control is all part of the surgical site infection and even though we're low, we want to stay that way."
The NSQIP program is closely tied to the Surgical Care Improvement Project (SCIP) measures. Saldinger says his team found they weren't doing as well as they would have liked on those. So they went to the data gatherers to find out how data were collected and noticed "for the most part, it was a lack of documentation and and processes."
In response, they created an order form for preop orders according to the SCIP measures. Only one antibiotic is listed for a variety of surgeries and if a doctor selects another, he or she has to explain why that decision was made. All the DVT prophylactic measures are included as well as beta-blockers. Documentation during surgery as to the DVT prophylaxis, one of the measures, is now also done.
Now they are reviewing their preop processes as their data show part of the problem is they have a low threshold in getting duplex ultrasounds postoperatively.
The cost factor
Beyond the annual fee, the program calls for an FTE, "which [together] generously calculated," Saldinger says, "could go anywhere from $150,000 to $200,000 a year." As the physician champion, it was Saldinger's task to sell the program to administration. While he admits it is difficult to show direct returns, he can show administrators how, for example, if you can lower your wound infection incidents that will lower your costs, especially as Medicare no longer will pay for infections acquired in the hospital.
The private and faculty surgeons have agreed to let him look at their records, "which I felt was a big step and show of trust." Knowing that only Saldinger and the nurse reviewer have access to those data also helps in making physicians feel secure. With the multifactorial problems they face, he says, "you need to know how to navigate the system so as not to alienate people but at the same time be able to convey a problem in an objective fashion that will lead to people stepping up to the plate to contribute to performance improvement."
"I really had to convey to my department and the surgeons who participate that this is not big brother watching you." You don't want to force anyone into participating, and you don't want anyone to feel that the data will be used in a punitive fashion, he adds.
He says quality improvement personnel can be viewed as "bad people" or as "spies."
"It's a very sensitive topic, very sensitive if you give surgeons any inkling that [the data] could be used in any other way than advertised, they will not participate and your project is damned," he says.