Increase accuracy in your organization before sharing surgical infection data

Prevent inaccurate data with work on the front end’

(Editor’s note: This is the first of a two-part series on surgical infection prevention. This month, we address ways to improve core measure data. Next month, quality leaders share strategies to reduce surgical infections.) 

If your organization was found to have a high rate of surgical infections, you could expect to face bad publicity, potential malpractice lawsuits, and major problems with the Joint Commission. But are the data telling an accurate story?

In July 2004, the Joint Commission, the Centers for Medicare & Medicaid Services (CMS), and the Hospital Quality Alliance introduced a standardized set of surgical infection prevention (SIP) measures designed to help hospitals continually assess and decrease surgical infections.

Several pay-for-performance programs now are giving payment incentives for lower surgical infection rates. These include the CMS/Premier Hospital Quality Incentive demonstration project and also some programs in the private sector.

"Pay-for-performance creates incentives to achieve high rates of performance on all of the quality indicators that are part of the demonstration projects," says Dale W. Bratzler, DO, MPH, principal clinical coordinator for the Oklahoma Foundation for Medical Quality, based in Oklahoma City, and immediate past president of the American Health Quality Association. "These demonstrations have also brought close scrutiny to the specifications of each performance measure."

To ensure your organization’s core measure data include only infections actually acquired during surgical procedures, take the time upfront to identify whether there was a diagnosis of a potential infection at the time the patient came into the hospital, or if the infection was developing during the hospitalization but prior to the surgical procedure, recommends Frank Zibrat, JCAHO’s associate director of ORYX implementation.

"If it was not an infection acquired from the surgical procedure, you need to exclude those to get valid data," he underscores.

"We don’t want hospital staff held accountable for something they had no control over as it relates to the surgical procedure," Zibrat points out. "That has the potential for making an organization look worse than they might otherwise look."

Surveyors will want to see that you have determined whether infections were acquired before or after the patient was admitted, says Darlene Christiansen, director of JCAHO’s standards interpretation and office of quality monitoring.

"The surveyors will look globally across the continuum about where that infection initiated from and what the follow-up analysis was," she explains.

The Joint Commission currently is working with third-party vendors that provide patient-level data, asking them to perform interrater reliability analysis to make sure the data collected by the organization are valid.

"In fact, we are going to be asking them to submit to us the results of the data reliability analysis, because we want to be assured that the data are accurate," says Zibrat.

SIP key focus during JCAHO surveys

There is increasing demand from consumer organizations for more transparency in health care, and surgical infections are a key focus.

"Some are calling for public accountability for infection rates. Others are promoting transparency related to the processes of care shown to reduce infections," Bratzler says.

The Hospital Quality Alliance has called for voluntary public reporting of the three SIP performance measures starting later this year: Prophylactic antibiotic administered within one hour of incision, discontinuation of antibiotics within 24 hours, and whether the agent used was consistent with published guidelines.

Surgical-site infections are the second-most common cause of hospital-associated infections, according to the Centers for Disease Control and Prevention, and are a major cause of mortality and morbidity for hospitalized patients.

Of 15 million inpatient surgeries each year, about 300,000 patients develop surgical-site infections, at an estimated cost of $1.5 billion.

If there is a trigger for the surveyor to review surgical infections based on the organization’s Priority Focus Data, that may be one of the initial patient tracers they conduct during the survey, Christiansen says.

However, even if infection control is identified as a Priority Focus Area, that doesn’t necessarily mean the organization has a risk issue in that area — it could simply mean there is the potential for risk based on the patient population cared for, she adds.

If surveyors detect infection control issues during patient tracers, a system tracer for infection control will be done, and surveyors will want to know the surgical department’s role in the overall hospitalwide infection control program, points out Christiansen.

JCAHO surveyors will be looking for the following:

  • Staff know how to protect patients and staff from infections, with knowledge not only about infection control processes but also about the specific patient they are caring for.

Although surveyors will not go in and interrupt a surgical procedure, they will observe from a doorway to make sure processes are being followed, as identified by staff.

"The surveyor may not have looked at the actual written policies, but after speaking to staff about what their process is, they will observe to see that what the staff described is what is actually being done," Christiansen says.

If any discrepancy is noted, the surveyor may ask to see written policies and procedures, so they can speak with leadership about their observations.

  • Appropriate processes for central sterilization are followed after a surgical procedure is completed.

"They will take a close look at those procedures to make sure safe practices are in place to prevent recontamination," Christiansen notes.

Surveyors will want to see that use of flash sterilization is minimized, with risk areas identified and staff appropriately trained.

  • Appropriate protocols are followed by environment-of-care staff when cleaning rooms for the next patient.

At some organizations, problems have been identified involving failure to follow appropriate protocols because housekeeping staff feel pressured to turn rooms over quickly, she reports.

"In several cases, they said that the physician likes the room turned around in 10 minutes, and as a result, they weren’t following the policies for appropriate decontamination," Christiansen points out.

  • Staff are aware of the chain of command to follow if they’re not comfortable with infection control practices.

"All professionals have equal autonomy in the surgical area, because that is one of the most critical areas in reduction of infection," she says.

"So at any point, anyone has the authority to put a halt to a procedure until they are comfortable that all the processes have been followed," Christiansen notes.

  • Ancillary services and all professional staff with access to the operating suite areas follow infection control practices.

"Sometimes, we are so geared to the caregivers taking care of the patient that we don’t consider all the other ancillary departments who may not be as knowledgeable about infection control, including pharmacy, laboratory, and radiology," she says.

  • Action is taken if data reveal opportunities for improvement.

The main thing surveyors will want to know? What your organization actually is doing with the data once they’re collected, Zibrat says.

The problem was that some organizations were focusing on data collection but falling short in terms of corrective action, he explains.

"But this is getting significantly better," Zibrat continues. "The focus has moved from I’m doing the best I can to collect the data’ to Now that I have it, I better start looking at it.’ Organizations are now really starting to sit up and take a look at their data because it’s being publicly reported."

He points to the Joint Commission’s Quality Check web site (; Hospital Compare, a web site created by CMS and the Hospital Quality Alliance (; and a growing number of health insurers that are factoring quality data into reimbursement.

SIP measures used internally

Not many hospitals actually are reporting SIP data to the JCAHO currently, Zibrat reports.

Organizations were required to report data for a third set of core measures as of Jan. 1, 2004, so a third core measure set already had been selected before the SIP measures became available on July 1, 2004.

He also points to Medicare’s pay-for-performance initiative, Reporting Hospital Quality Data for Annual Payment Update. The program — part of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 — gives hospitals financial incentives to submit data on 10 quality indicators, which didn’t include SIP.

"So there have been disincentives for picking up SIP, but when we increase our requirements for the use of a fourth core measure set, we will see a lot of hospitals picking this up," Zibrat predicts.

The additional requirement is expected to become effective by the beginning of 2006.

However, even if hospitals aren’t submitting SIP data to JCAHO, many are collecting and tracking the measures internally, he says.

"This is a key focus for organizations; they have been waiting for the SIP core measure set for a long time. They liked what they saw, and they are using the measures. But until we up the requirement, there is not much of an incentive to them to submit the data to us," Zibrat adds.

Collecting data internally gives organizations a chance to fine-tune their data collection processes to ensure accurate data, he emphasizes.

"This gives them breathing room to put together the processes internally to be sure they are collecting reliable data, so the analysis is valid pertaining to their actual infection rates," Zibrat says.

"There are approximately 35 pages on antibiotics to look at, and also 30 pages on different infectious processes that might be going on prior to surgery," he continues.

Strategies to improve SIP analysis

To improve your analysis of SIP data, use these effective strategies:

1. Do a careful analysis when a surgical-site infection does occur.

When a patient develops a surgical-site infection, audit the chart to see if all known practices that can prevent surgical-site infections were done, Bratzler says.

If everything was done right, the infection should be termed "apparently unavoidable," but if some processes were missed, such as the antibiotic dose not given until after incision, then the infection was "potentially preventable," he points out.

2. Use real-time data.

SIP measures currently publicly reported on the Hospital Compare web site reflect events in the first and second quarter of 2004, notes Terry Hill, MD, medical director for quality improvement at San Francisco-based Lumetra.

"As excited as I am about the site, quality managers need data that is closer to real time," he says.

"Real-time data collection is important to identify the vulnerabilities in processes for giving antibiotics. It doesn’t have to be research-quality data, and sampling is perfectly appropriate. But in order to make credible suggestions to medical staff, you need enough data, and it needs to be pretty close to real time," Hill explains.

Once the medical staff know that data will be publicly reported and are given a credible amount of data to show their performance is less than optimum, they are likely to take dramatic leadership action, he says.

"The role of the quality manager is to be a champion for change and to give enough credible data to the surgical leaders to make a difference," Hill adds.

3. Identify problems and take appropriate action.

The following actions were taken based on analysis of SIP data at Louisiana State University Health Sciences Center in Shreveport, says Leisa Oglesby, assistant hospital administrator of quality:

  • Pocket cards listing core measure indicators were given to all residents and faculty.
  • New guidelines were implemented for ordering antibiotics preoperatively, with antibiotics given at induction of anesthesia. Inservices on the guidelines were given to involved staff and departments.
  • Quarterly updates were sent to inform involved department chairs of results, with action plans to be implemented when thresholds are not met.

Developing an action plan

When problems are identified, an action plan is developed and implemented to improve performance, Oglesby says. For example, inadequate nursing documentation was addressed by giving inservices, "bathroom blitzes," and ongoing monitoring to improve performance.

"They have also included this in the nurses’ self-learning packet, which must be completed annually," she notes.

The process also revealed that Code 99 forms were not being completed; or in some cases, the documentation on the form was different from what was found in the medical record.

To address this, quality leaders asked a team to review the entire process and make recommendations for improvement.

These included revising orientation for nursing staff, educating physicians and members of the code team, defining responsibilities for all code team members, redesigning the form, revising policies, and establishing ongoing monitoring.

The first step is to identify who owns the process of making sure antibiotics are delivered, then determine how problems occur, notes Hill, who gives the example of one hospital where anesthesia residents sometimes gave antibiotics before they knew when the surgery actually would occur.

"They were very eager to do the right thing and were going down their checklist, but they didn’t really know whether the incision was about to happen," he adds. "As a result, the antibiotics often wound up being given further out than they should have been."

The quality manager’s role is to trace the steps that occurred and explain what went wrong with the process, Hill explains. "You need to make it clear, in a very simple and visual way, that the process either has redundancies, lack of ownership, or other problems."

Getting a description from one or two people or observing on one occasion can be enough to draft a flowchart, which the quality manager can then use to facilitate further individual or group discussions, he says.

"People will generally be quite willing to point out vulnerabilities or common variations in the process," he says.

Good opportunity for an FMEA

4. Do a failure mode and effects analysis.

"This is a wonderful opportunity to take a look at current processes from when the patient is admitted until they are discharged from the [post-anesthesia care unit] to whatever unit they are going into," Christiansen notes.

"This can become very burdensome if you try to look at everything at once, but if you strictly take a look at the processes within the surgical area related to infection control, you can identify where the gaps are. Then you can begin to evaluate how to correct that and implement new strategies," she adds.

Recommended reading

  • Bratzler DW. Use of antimicrobial prophylaxis for major surgery: Baseline results from the national surgical infection prevention project. Arch Surg 2005; 140:174-182.

[For more information, contact:

  • Dale W. Bratzler,DO, MPH, Principal Clinical Coordinator, Oklahoma Foundation for Medical Quality, 14000 Quail Springs Parkway, Suite 400, Oklahoma City, OK 73134. Phone: (405) 840-2891, ext. 209. Fax: (405) 840-1343. E-mail:
  • Terry Hill,MD, Medical Director for Quality Improvement, Lumetra, One Sansome St., Suite 600, San Francisco, CA 94104-4448. Phone: (415) 677-2000. Fax: (415) 677-2195. E-mail:
  • Leisa Oglesby, Assistant Hospital Administrator of Quality, Louisiana State University Health Sciences Center, 1541 Kings Highway, Shreveport, LA 71130. Phone: (318) 675-5030. Fax: (318) 675-4646. E-mail]