Effectiveness of a RF Sponge Detection

Background:

Retained surgical items (e.g. sponges, needles, and instruments) remain one of the sentinel events most frequently reported to The Joint Commission, and are estimated to occur in 1: 5500 surgeries. These serious adverse events have resulted in negative patient outcomes (reoperation, readmission/prolonged hospital stay, infection or sepsis, fistulas/ bowel obstructions, visceral perforation, and death). Cotton gauze sponges account for 48-69% of retained surgical items, and result in more serious tissue reaction than metal items. The national standard for prevention of retained surgical sponges relies heavily on manual counting, a process that occurs prior to incision, at the time of closing a cavity within a cavity, closing the first layer of tissue, and closing skin. The immediate outcome of the closing counts is either correct or unresolved. If unresolved, a series of steps are taken to reconcile the count (surgeon notified, wound searched, sterile field searched, room searched, count repeated). These steps require extra time and effort, pulling personnel away from other competing priorities. If the final closing count remains incorrect, it is common practice to obtain an intraoperative radiograph to rule out retention of a surgical sponge. The practice of searching for misplaced sponges and using radiography to rule out the presence of a retained sponge extend the operating time and add to costs associated care of the patient in surgery. Novel technology is now available for detecting and preventing retained surgical sponges. The technology uses a radiofrequency (RF) chip in the sponge, and a mat or wand that scans the patient for a sponge. Only one clinical trial has evaluated the in-use effectiveness of this RF technology. No retained sponges occurred during the 18 month trial. However the sample size was not large enough to determine significance.

Primary Objective:

The objective in this Health Services Research is to evaluate the in-use effectiveness of a radiofrequency (RF) surgical sponge detection system for reducing the cost of searching for sponges and prevention of incorrect counts.

Specific Aims are:

1. To compare the estimated time and cost of searching for sponges and using radiography to rule out the presence of retained sponges before and after implementation of a RF sponge detection system.
2. To compare the incidence of incorrect final closing counts before and after implementation of a RF sponge detection system.

Study Design:

Using deidentified existing data, investigators propose a quasiexperimental study evaluating the outcomes of using RF surgical sponge detection system. The setting is the University of Iowa Hospitals and Clinics, a 711-bed academic medical center, a Level 1 trauma center, located in Iowa City, Iowa. The University of Iowa Institutional Review Board has determined that this study does not meet the regulatory definition of human subjects research.

Primary Endpoints

1. The estimated time and cost of searching for sponges and using radiography to rule out the presence of retained surgical sponges.
2. The frequency of incorrect final closing sponge counts.

Data for individual patients will be recorded by the circulating nurses in a questionnaire embedded in the surgical log of Epic Systems®, module Optime®. The questionnaire will consist of three individual counts: initial closing, final closing, and additional count. If the first question, "was a search required" is answered "yes", four subsequent questions will be asked: type of sponge, actions taken, number of minutes required, and the location of the sponge if found. Data will be retrieved from the Health Information System (HIS) using S.A.P. ® Business Objects™ (surgical volume, surgical service) and Crystal reports (details about searches for sponges). Incorrect final closing counts are documented in the Patient Safety Net® electronic event reporting system. Data will be retrieved from events reported within the category of "count incomplete or not done" and "count incorrect". Data will be retrieved by HIS personnel and provided via a report to the Director of Clinical Functions responsible for the Main Operating Room. This report is part of the quality improvement program. Electronic data will be stored in the College of Nursing, the University of Iowa, on the college's secure server. Data will be cleaned by a data manager and analyzed by the college's statistician.

Analysis:

Aim 1: Minutes of time spent searching for misplaced sponges will be analyzed using ranges, means, and standard deviations. Comparison before and after implementation of the RF detection technology will be described as percentages and analyzed using chi square with Yates correction. To provide the best estimate of the resources used in searching for sponges, investigators will use published data on operating room costs per minute to estimate search costs ($17.23). However, there are significant differences across institutions. To accommodate this broad range of potential adopters of this RF detection technology and show differences in potential cost-effectiveness, in addition to using the average estimate of operating room time cost from the literature, investigators will construct lower bound and upper bound estimates of the operating room time costs to provide a range of cost estimates to use in assessing the RF detection technology intervention.

Investigators further adjust search costs for variation in practices during searches. At times surgeons continue with the procedure (e.g. closing the wound) while in other cases surgeons will assist with efforts to locate the sponge. As investigators are interested in time spent on non-productive searches, investigators will use a range of estimates of non-productive search costs. Investigators start with 100% non-productive and also estimate based on 75% non-productive (25% of the search time moving forward with the procedure) and 50% non-productive (50% of the search time moving forward with the procedure). Investigators will use the average radiography costs from published literature ($286) in addition to the operating room time associated with obtaining an intraoperative radiograph (30minutes). All times and costs will be analyzed using ranges, means, and standard deviations. Comparison before and after implementation of the RF detection technology will be described as percentages and analyzed using chi square with Yates correction.

Aim 2: To compare the incidence of incorrect final closing counts The number of incorrect final closing counts before and after implementation of a RF sponge detection system will be analyzed using ranges, means, and standard deviations. Comparison before and after implementation of the RF detection technology will be described as percentages and analyzed using chi square with Yates correction.