mayo clinic study

12
www.jcrinc.com Eliminating Retained Surgical Sponges Features Performance Improvement Using a Data-Matrix–Coded Sponge Counting System Across a Surgical Practice: Impact After 18 Months Organizational Change and Learning A Strategic Approach for Managing Conflict in Hospitals: Responding to the Joint Commission Leadership Standard, Part 1 A Strategic Approach for Managing Conflict in Hospitals: Responding to the Joint Commission Leadership Standard, Part 2 Adverse Events The Cost of Serious Fall-Related Injuries at Three Midwestern Hospitals Health Professions Education Trainees’ Perceptions of Patient Safety Practices: Recounting Failures of Supervision February 2011 Volume 37 Number 2 Improvement from Front Office to Front Line “After 18 months of continuous use of the data-matrix–coded DMS system throughout our surgical practice, in which some 1,862,373 sponges were counted, we had no retained-sponge events.” —Cima et al. (p. 55)

Upload: nicholas-bean

Post on 20-Jan-2017

42 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Mayo Clinic Study

wwwjcrinccom

Eliminating RetainedSurgical Sponges

FeaturesPerformance Improvement

Using a Data-MatrixndashCoded Sponge Counting System Across aSurgical Practice Impact After 18 Months

Organizational Change and Learning

A Strategic Approach for Managing Conflict in Hospitals Respondingto the Joint Commission Leadership Standard Part 1

A Strategic Approach for Managing Conflict in Hospitals Respondingto the Joint Commission Leadership Standard Part 2

Adverse Events

The Cost of Serious Fall-Related Injuries at Three MidwesternHospitals

Health Professions Education

Traineesrsquo Perceptions of Patient Safety Practices Recounting Failuresof Supervision

February 2011Volume 37 Number 2

Improvement fromFront Office to Front Line

ldquoAfter 18 months of continuous use of the

data-matrixndashcoded DMS system throughout our

surgical practice in whichsome 1862373 sponges were counted we had noretained-sponge eventsrdquo

mdashCima et al (p 55)

51February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Robert R Cima MD MA Anantha Kollengode PhD MBA James Clark RN MS Sarah Pool RN MSCheryl Weisbrod RN MS Gwendolyn J Amstutz MHA Claude Deschamps MD

The publication of the Institute of Medicine (IOM) reportTo Err Is Human in 1999 highlighted the state of patient

safety in the hospital environment1 Following the IOM reportthe National Quality Forum released a list of ldquo27 NeverEventsrdquo which described those medical errors that should neveroccur in a United States hospital2 Four of the 27 events direct-ly related to patient safety in the operating room (OR) opera-tions involving the wrong patient wrong-site and wrong-sidesurgery and retained surgical items (RSIs) after surgeryPatient-related hospital errors especially those occurring in theOR result in intense and often critical public scrutiny A dra-matic example of OR errors are RSIs after surgery3

The true incidence of RSIs after intraabdominal surgery inthe United States is unknown Estimates range from 1 in every1000 to 1500 abdominal operations to 1 in every 8000 to18000 inpatient operations45 Risk factors for RSIs on the basisof insurance claims data include emergency procedures un -planned change in operation and body mass index5 We previ-ously reported our experience of RSIs at the Mayo ClinicRochester (MCR)6 During the four-year analysis period191168 operations were performed and 34 RSIs were discov-ered resulting in an overall incidence of 0178 RSIs1000 oper-ations or approximately 1 RSI per 5500 cases In ourexperience none of the RSIs occurred in patients with the riskfactors described in other studies5 However similar to other re -ports in the literature our most common RSI was a cottonsponge productmdashwith 23 sponges (68 of RSIs) retained Alsoconsistent with othersrsquo findings the majority of our RSIsoccurred despite a ldquocorrectrdquo count being reported before comple-tion of the operation Surprisingly those items that are routine-ly counted most frequently during an operation cotton spongesare the items most commonly miscounted and retained

Although a multidisciplinary effort can significantly reducethe incidence of RSIs such efforts have not been shown tocompletely eliminate them for a sustained period of time7 Anumber of technologies have recently been introduced in an

Performance Improvement

Using a Data-MatrixndashCoded Sponge Counting System Across aSurgical Practice Impact After 18 Months

Article-at-a-Glance

Background Retained surgical items (RSIs) most com-monly sponges are infrequent Yet despite sponge-countingstandards failure to maintain an accurate count is a com-mon error To improve counting performance technologysolutions have been developed A data-matrixndashcoded sponge(DMS) system was evaluated and implemented in a high-volume academic surgical practice at Mayo Clinic Rochester(MCR) The primary end point was prevention of spongeRSIs after 18 monthsMethods Two trials were conducted before implementa-tion A randomized-controlled trial assessed the systemrsquosfunction efficiency and ergonomics The second largertrial was conducted to validate the prior findings and testproduct improvements After the trials the system wasimplemented in all 128 operatingprocedure rooms acrossthe MCR campus on February 2 2009 The institutionwideimplementation was intended to avoid the possibility ofhaving standard unmarked sponges and DMSs in the oper-ating room suite concurrently Results Before implementation a retained spongeoccurred on average every 64 days Between February 2009and July 2010 87404 procedures were performed and1862373 DMSs were used without an RSI (p lt 001)After four cases the average time to count a DMS decreasedfrom 11 to 4 seconds Total sponge counting timeoperationin creased without any increase in overall operative time Conclusions After 18 months a DMS system eliminat-ed sponge RSIs from a high-volume surgical practice TheDMS system caused no work-flow disruption or increases incase duration Staff satisfaction was acceptable with a highdegree of trust in the system The DMS system is a reliableand cost-effective technology that improves patient safety

Copyright 2011 copy The Joint Commission

52 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

attempt specifically to prevent retention ofcotton sponge products during surgical pro-cedures8ndash10 There are two major types of tech-nologies (1) sponges with unique identifyingdata-matrix codes (previously known as barcodes) annealed onto the sponge and (2)sponges with a radiofrequency (RF) chipembedded into them The data-matrixndashcodedsystem is solely a counting system for spongeproducts If used properly it records thenumber of sponges scanned into the case andat the end reconciles the number of spongesscanned out of the case Depending on theproduct the RF technology can act solely as acounting technology or additionally as adetecting technology with the use of a hand-held device to scan the patient for the pres-ence of a sponge if one is missing from thefinal count The few studies regarding theseproducts are quite limited in scope and donot address the real-world effectiveness ofthese technologies being implemented acrossan entire OR practice Specifically there areno reports of long-term results on sponge RSIreduction after implementation of a technol-ogy in a surgical practice

In this study we describe the technologyassessment of a data-matrixndashcoded sponge(DMS) system through two internal trials and after full imple-mentation in the operating and surgical procedure rooms atMCR The system includes individually unique data-matrixndashcoded cotton sponge products (Figure 1 right also available inonline article) and a data-matrixndashreading scanner that keeps arunning ledger of the sponges scanned onto and off of the ster-ile field (Figure 2 page 53 also available in online article)Safety performance after implementation as measured by theincidence of retained sponge products and staff satisfaction wasevaluated

MethodsCOUNTING OF SPONGES

The surgical staff at MCR follow the Association ofperiOperative Registered Nurses (AORN) recommended prac-tices for counting sponges11 Our standard sponge-countingpolicy is to have the certified surgical technician and circulatingnurse perform concurrent visual and audible sponge countsthroughout the procedure According to policy counts had to

be performed at the beginning of the case with any addition orremoval of sponges from the sterile field at the closure of anybody cavity with change in nursing personnel and before thelast stitch is placed For the DMS system (pilot cases and afterfull implementation) the standard counting practices asdescribed were performed and in addition the sponges werescanned using a pole-mounted data-matrix tag reader beforeadding sponges to or when sponges were removed from thesterile surgical field

TRAINED OBSERVERS

Quality management services staff trained surgical servicenursing staff in proper observational techniques and use ofstandardized data intake forms In the first phase(FebruaryndashMarch 2008) two trained observers were present ineach OR collecting data independently to determine the inter-observer reliability In the second phase (August 2008) onetrained observer collected the data in each OR during the pilotsin the surgical specialties

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cottonsurgical sponge products Each sponge or towel has a unique data-matrix tag annealed to the itemThis tag contains the unique code identifying the individual sponge After the sponge is scanned intothe scanner it must be scanned out to remove it from the case sponge ledger The sponge packs aresecured with a band labeled with a ldquomaster bandrdquo This data-matrixndashcoded band has the data foreach individual sponge within the package Scanning this band loads into the scanner all the indi-vidual sponges avoiding the need to scan each individual sponge onto the field However eachsponge must be scanned out at the end of the procedure (Figure 1 is available in color in onlinearticle)

The Data-MatrixndashCoded Sponge System

Copyright 2011 copy The Joint Commission

53February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

STATISTICAL ANALYSIS

Statistical analysis was performed using Minitab software(version 142 State College Pennsylvania)

TRIAL PHASE 1 METHODS AND METRICS

All testing and evaluation of the DMS system were carriedout at MCR a tertiary-referral academic medical center in theupper Midwest in 2008 There are 117 inpatient and outpa-tient ORs 3 obstetrical ORs and 8 labor amp delivery (LampD)birthing rooms on the MCR campus distributed between twoacute care hospitals All operating and LampD rooms are staffedby MCR physicians nurses and allied health staff All staffmembers are employed under one organizational leadershipstructure with a unified policy and procedure manual for OR

conduct including sponge countingA Plan-Do-Study-Act (PDSA) methodology was adapted to

evaluate the DMS system in our ORs In the first PDSA cyclea total of 16 cases was randomly assigned to either the DMSs(Pilot A) or standard surgical sponges (Control A) on the basisof the availability of the observer teams Patients undergoing anelective weekday procedure in cardiovascular surgery or colonand rectal surgery were eligible for the randomization duringthe one-week trial period A trial was performed in two ORs forcardiovascular surgery and two ORs for colon and rectal sur-gery Surgical staff in the specialties were trained on the use ofthe DMS system equipment during the week before the trialTwo trained observers were in the room to independently eval-uate system performance and collect appropriate data

The following technical performance metrics were collected Interobserver reliability Average time spent counting a sponge (time spent to

count sponges at the beginning of the case during the case foraddition or removal of sponges from the sterile field duringpermanent personnel changes and before the last stitch isplaced divided by the number of sponges used)

Learning curve for scanning DMSs Total time spent to count sponges Battery life of the scannerIn addition the staff rsquos impression of product features and

work flow were assessed with a standardized new product survey

TRIAL PHASE 2 METHODS AND METRICS

In the second PDSA cycle a trial was performed in fourORs in each of the following specialties colon and rectal car-diovascular and general surgery DMSs were used in all electivecases in each of the four ORs for a period of one week Trainedobservers were present in the surgical suite to help with anytechnical difficulties experienced by the staff The observerswere not assigned to specific cases but moved from case to case

Data were gathered from a total of 57 surgeries in the singleweek In this phase upgrades to the DMS system software weremade on the basis of feedback from staff from the first phase inregard to the patient data entry screens In addition to the met-rics used for first phase data were collected to verify accuracyof the ldquomaster bandsrdquo which was a new packaging style intro-duced by the manufacturer Master bands contained uniquesponge-identifying information of the 5 or 10 sponges includ-ed in the pack thereby eliminating the need to scan in eachsponge from the package However each individual spongemust be scanned out when it is removed from the sterile field

Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number ofsponges still on the sterile field and the number of sponges that have beenremoved from the sterile field (Figure 2 is available in color in online article)

Copyright 2011 copy The Joint Commission

54 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

IMPLEMENTATION PHASE 3 METHODS

AND METRICS

In November 2008 after review of thetrial data it was decided to implement theDMS system in all operating and LampDrooms at MCR After education and trainingof all surgical services staff use of the DMSsystem was begun across the MCR campus onFebruary 2 2009 Evaluation of use per-formance and staff satisfaction were assessedone year after implementation InstitutionalReview Board approval was obtained for datacollection staff survey data analysis andpublication

ResultsTRIAL PHASE ONE

Interobserver Reliability The interobserverreliability for sponge counting in the trialrooms was determined to be excellent withthe mean count times recorded by the twoobservers not being statistically different (p lt 650)

Time to Count The average time to count a sponge for con-trol sponges was 40 seconds (n = 335) versus 114 seconds forthe DMSs (n = 365 p lt 05) The average time to count wassignificantly different (p = 002) between the colon and rectal(8 seconds) and cardiovascular services teams (13 seconds) Thetimes to count at three different stages (at start of the case dur-ing the case and at the end of the case) were also significantlydifferent (p = 006) The average time to count a sponge for theDMSs at the start of the case was 6 seconds whereas the timeto count in a sponge during the case was 12 seconds and theaverage time to count out a sponge was 16 seconds The aver-age time to count a sponge in the same OR for four days (learn-ing curve effect) is shown in Figure 3 (right) The average timeto count a sponge decreased from 11 seconds on day 1 of thetrial to 5 seconds on day 4mdashor about the same as for countingunmarked spongesmdashfor the colon and rectal surgery ORs

Battery Performance To assess the duration of scanner bat-tery performance during longer procedures the percent chargeleft in the battery over the duration of the case without placingit in the charger was measured The units retained a charge of40 or higher after five hours of use

TRIAL PHASE TWO

The goals of the second phase trial were to (a) conduct thestudy in a larger setting assess the impact on work flow across

more diverse case types and validate the Phase 1 results (b)assess the changes made to the scanner-user interface on thebasis of the staff feedback and (c) study the effectiveness of thenew master-band packaging released by the company to speedup the count-in process (especially in emergent situations)

Time to Count The two specialties colon rectal and cardio-vascular surgery that participated in the first pilot showed adecline in the mean count times for the DMSs from 80 sec-onds to 45 seconds and from 132 seconds to 112 secondsrespectively In the general surgery ORsmdasha new specialty forthe studymdashmean counting times were similar to those observedin Trial 1 for the other specialties

IMPLEMENTATION PHASE THREE

The DMS system was implemented across the MCR cam-pus on February 2 2009 An institutionwide implementationwas performed rather than a multiphased approach to avoid thepossibility of having standard unmarked sponges and DMSs inthe OR suite concurrently Implementation followed a two-month period of staff education and training

After 18 months of continuous use in 87404 operationsprocedures 1862373 DMSs were used without a singlesponge retained During this period our institution continuedthe practice of postoperative high-resolution x-rays in a dedicat-ed radiology unit at the completion of surgery In the preced-ing six years MCR was averaging a cotton sponge RSI every 64

Figure 3 The learning curve for use of the sponge system in a single operating room demonstratedthat after four cases the time to count individual sponges decreased from 101 seconds on Day 1 to45 seconds on Day 4

Average Time to Count a Sponge with Use of the Data-Matrix Coded Sponge (DMS) System in a

Single Operating Room (OR)

Copyright 2011 copy The Joint Commission

55February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

days (Figure 4 page 55 also available in online article) Thischange in event frequency represents a significant reduction (p lt 001) related to the implementation of the DMS systemSystem performance on RSI prevention is monitored daily byour surgical safety team [RRC JC SP CW] in collabora-tion with the Institutional Sentinel Event group To date therehave been three reported events where the manual count hadreconciled all the sponge counts as correct but the DMS systemwas not reconciled Intraoperative x-rays or wound explorationrevealed the sponge before completion of the procedure thusavoiding an RSI

STAFF SURVEY

After each of the two trials and after one year of system useOR staff were electronically mailed a multiquestion surveyReturn of the survey was voluntary No effort was made toensure that staff had completed both surveys or had been in theOR during the entire year after implementation Each questionused a 5-point Likert scale with 1 indicating ldquostrong dislikerdquoand 5 indicating a ldquostrong likingrdquo The summary of the resultsfrom the survey are provided in Table 1 (page 56) Additionalcomments from the staff were documented in the commentssection of the survey The percentage of 4 and 5 scores (4

ldquosomewhat like the DMS systemrdquo 5ldquostrongly like the DMS systemrdquo) increasedduring the first year of use Confidence inreliability of the system was relativelyunchanged Given the limits of the surveymethodology a statistical analysis of thesurvey results was not conducted

DiscussionRSIs after surgical or invasive procedurescontinue to challenge OR staff as signifi-cant medical errors Cotton sponges arethe most common retained items despitenational standards for sponge counting5612

In a multiphase technology assessment of aDMS system at MCR which entailed twointernal randomized controlled trials thesystem performed within the specifiedtechnical parameters and did not disruptthe flow of the surgical case After 18months of continuous use of the DMS sys-tem throughout our surgical practice inwhich some 1862373 sponges werecounted we had no retained-sponge

events There was no increase in overall operative time relatedto this sponge-counting technology The learning curve for usewas extremely short (lt four cases) Staff satisfaction with thesystem was acceptable with a high degree of confidence in thereliability of the system

Although many retained sponges are detected in the earlypostoperative period reports of retained sponges causingchronic symptoms or incidentally being found years to decadesafter the index operation are unfortunately quite frequent13ndash16

RSIs including sponges are associated with significant morbid-ity (small-bowel fistula obstruction visceral perforation re-operations to remove the object) and rarely death12 Despitenational practice standards related to sponge and instrumentcounting these protocols are apparently not sufficient to avoidRSIs11 As noted by Christian et al the counting procedures arewell designed and nearly universally performed and account foras much as 14 of total operative time17 Unfortunately thisseemingly ldquoeasyrdquo counting task is error prone because it is per-formed in a discontinuous fashion throughout a complex pro-cedure with multiple interruptions competing demands andtasks and possibly numerous participants which leads toerrors

As our previous experience has demonstrated reliance on

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July2010 at Mayo Clinic Rochester The data-matrix coded sponge system was introduced in February2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit (Figure4 is available in color in online article)

Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Copyright 2011 copy The Joint Commission

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 2: Mayo Clinic Study

51February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Robert R Cima MD MA Anantha Kollengode PhD MBA James Clark RN MS Sarah Pool RN MSCheryl Weisbrod RN MS Gwendolyn J Amstutz MHA Claude Deschamps MD

The publication of the Institute of Medicine (IOM) reportTo Err Is Human in 1999 highlighted the state of patient

safety in the hospital environment1 Following the IOM reportthe National Quality Forum released a list of ldquo27 NeverEventsrdquo which described those medical errors that should neveroccur in a United States hospital2 Four of the 27 events direct-ly related to patient safety in the operating room (OR) opera-tions involving the wrong patient wrong-site and wrong-sidesurgery and retained surgical items (RSIs) after surgeryPatient-related hospital errors especially those occurring in theOR result in intense and often critical public scrutiny A dra-matic example of OR errors are RSIs after surgery3

The true incidence of RSIs after intraabdominal surgery inthe United States is unknown Estimates range from 1 in every1000 to 1500 abdominal operations to 1 in every 8000 to18000 inpatient operations45 Risk factors for RSIs on the basisof insurance claims data include emergency procedures un -planned change in operation and body mass index5 We previ-ously reported our experience of RSIs at the Mayo ClinicRochester (MCR)6 During the four-year analysis period191168 operations were performed and 34 RSIs were discov-ered resulting in an overall incidence of 0178 RSIs1000 oper-ations or approximately 1 RSI per 5500 cases In ourexperience none of the RSIs occurred in patients with the riskfactors described in other studies5 However similar to other re -ports in the literature our most common RSI was a cottonsponge productmdashwith 23 sponges (68 of RSIs) retained Alsoconsistent with othersrsquo findings the majority of our RSIsoccurred despite a ldquocorrectrdquo count being reported before comple-tion of the operation Surprisingly those items that are routine-ly counted most frequently during an operation cotton spongesare the items most commonly miscounted and retained

Although a multidisciplinary effort can significantly reducethe incidence of RSIs such efforts have not been shown tocompletely eliminate them for a sustained period of time7 Anumber of technologies have recently been introduced in an

Performance Improvement

Using a Data-MatrixndashCoded Sponge Counting System Across aSurgical Practice Impact After 18 Months

Article-at-a-Glance

Background Retained surgical items (RSIs) most com-monly sponges are infrequent Yet despite sponge-countingstandards failure to maintain an accurate count is a com-mon error To improve counting performance technologysolutions have been developed A data-matrixndashcoded sponge(DMS) system was evaluated and implemented in a high-volume academic surgical practice at Mayo Clinic Rochester(MCR) The primary end point was prevention of spongeRSIs after 18 monthsMethods Two trials were conducted before implementa-tion A randomized-controlled trial assessed the systemrsquosfunction efficiency and ergonomics The second largertrial was conducted to validate the prior findings and testproduct improvements After the trials the system wasimplemented in all 128 operatingprocedure rooms acrossthe MCR campus on February 2 2009 The institutionwideimplementation was intended to avoid the possibility ofhaving standard unmarked sponges and DMSs in the oper-ating room suite concurrently Results Before implementation a retained spongeoccurred on average every 64 days Between February 2009and July 2010 87404 procedures were performed and1862373 DMSs were used without an RSI (p lt 001)After four cases the average time to count a DMS decreasedfrom 11 to 4 seconds Total sponge counting timeoperationin creased without any increase in overall operative time Conclusions After 18 months a DMS system eliminat-ed sponge RSIs from a high-volume surgical practice TheDMS system caused no work-flow disruption or increases incase duration Staff satisfaction was acceptable with a highdegree of trust in the system The DMS system is a reliableand cost-effective technology that improves patient safety

Copyright 2011 copy The Joint Commission

52 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

attempt specifically to prevent retention ofcotton sponge products during surgical pro-cedures8ndash10 There are two major types of tech-nologies (1) sponges with unique identifyingdata-matrix codes (previously known as barcodes) annealed onto the sponge and (2)sponges with a radiofrequency (RF) chipembedded into them The data-matrixndashcodedsystem is solely a counting system for spongeproducts If used properly it records thenumber of sponges scanned into the case andat the end reconciles the number of spongesscanned out of the case Depending on theproduct the RF technology can act solely as acounting technology or additionally as adetecting technology with the use of a hand-held device to scan the patient for the pres-ence of a sponge if one is missing from thefinal count The few studies regarding theseproducts are quite limited in scope and donot address the real-world effectiveness ofthese technologies being implemented acrossan entire OR practice Specifically there areno reports of long-term results on sponge RSIreduction after implementation of a technol-ogy in a surgical practice

In this study we describe the technologyassessment of a data-matrixndashcoded sponge(DMS) system through two internal trials and after full imple-mentation in the operating and surgical procedure rooms atMCR The system includes individually unique data-matrixndashcoded cotton sponge products (Figure 1 right also available inonline article) and a data-matrixndashreading scanner that keeps arunning ledger of the sponges scanned onto and off of the ster-ile field (Figure 2 page 53 also available in online article)Safety performance after implementation as measured by theincidence of retained sponge products and staff satisfaction wasevaluated

MethodsCOUNTING OF SPONGES

The surgical staff at MCR follow the Association ofperiOperative Registered Nurses (AORN) recommended prac-tices for counting sponges11 Our standard sponge-countingpolicy is to have the certified surgical technician and circulatingnurse perform concurrent visual and audible sponge countsthroughout the procedure According to policy counts had to

be performed at the beginning of the case with any addition orremoval of sponges from the sterile field at the closure of anybody cavity with change in nursing personnel and before thelast stitch is placed For the DMS system (pilot cases and afterfull implementation) the standard counting practices asdescribed were performed and in addition the sponges werescanned using a pole-mounted data-matrix tag reader beforeadding sponges to or when sponges were removed from thesterile surgical field

TRAINED OBSERVERS

Quality management services staff trained surgical servicenursing staff in proper observational techniques and use ofstandardized data intake forms In the first phase(FebruaryndashMarch 2008) two trained observers were present ineach OR collecting data independently to determine the inter-observer reliability In the second phase (August 2008) onetrained observer collected the data in each OR during the pilotsin the surgical specialties

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cottonsurgical sponge products Each sponge or towel has a unique data-matrix tag annealed to the itemThis tag contains the unique code identifying the individual sponge After the sponge is scanned intothe scanner it must be scanned out to remove it from the case sponge ledger The sponge packs aresecured with a band labeled with a ldquomaster bandrdquo This data-matrixndashcoded band has the data foreach individual sponge within the package Scanning this band loads into the scanner all the indi-vidual sponges avoiding the need to scan each individual sponge onto the field However eachsponge must be scanned out at the end of the procedure (Figure 1 is available in color in onlinearticle)

The Data-MatrixndashCoded Sponge System

Copyright 2011 copy The Joint Commission

53February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

STATISTICAL ANALYSIS

Statistical analysis was performed using Minitab software(version 142 State College Pennsylvania)

TRIAL PHASE 1 METHODS AND METRICS

All testing and evaluation of the DMS system were carriedout at MCR a tertiary-referral academic medical center in theupper Midwest in 2008 There are 117 inpatient and outpa-tient ORs 3 obstetrical ORs and 8 labor amp delivery (LampD)birthing rooms on the MCR campus distributed between twoacute care hospitals All operating and LampD rooms are staffedby MCR physicians nurses and allied health staff All staffmembers are employed under one organizational leadershipstructure with a unified policy and procedure manual for OR

conduct including sponge countingA Plan-Do-Study-Act (PDSA) methodology was adapted to

evaluate the DMS system in our ORs In the first PDSA cyclea total of 16 cases was randomly assigned to either the DMSs(Pilot A) or standard surgical sponges (Control A) on the basisof the availability of the observer teams Patients undergoing anelective weekday procedure in cardiovascular surgery or colonand rectal surgery were eligible for the randomization duringthe one-week trial period A trial was performed in two ORs forcardiovascular surgery and two ORs for colon and rectal sur-gery Surgical staff in the specialties were trained on the use ofthe DMS system equipment during the week before the trialTwo trained observers were in the room to independently eval-uate system performance and collect appropriate data

The following technical performance metrics were collected Interobserver reliability Average time spent counting a sponge (time spent to

count sponges at the beginning of the case during the case foraddition or removal of sponges from the sterile field duringpermanent personnel changes and before the last stitch isplaced divided by the number of sponges used)

Learning curve for scanning DMSs Total time spent to count sponges Battery life of the scannerIn addition the staff rsquos impression of product features and

work flow were assessed with a standardized new product survey

TRIAL PHASE 2 METHODS AND METRICS

In the second PDSA cycle a trial was performed in fourORs in each of the following specialties colon and rectal car-diovascular and general surgery DMSs were used in all electivecases in each of the four ORs for a period of one week Trainedobservers were present in the surgical suite to help with anytechnical difficulties experienced by the staff The observerswere not assigned to specific cases but moved from case to case

Data were gathered from a total of 57 surgeries in the singleweek In this phase upgrades to the DMS system software weremade on the basis of feedback from staff from the first phase inregard to the patient data entry screens In addition to the met-rics used for first phase data were collected to verify accuracyof the ldquomaster bandsrdquo which was a new packaging style intro-duced by the manufacturer Master bands contained uniquesponge-identifying information of the 5 or 10 sponges includ-ed in the pack thereby eliminating the need to scan in eachsponge from the package However each individual spongemust be scanned out when it is removed from the sterile field

Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number ofsponges still on the sterile field and the number of sponges that have beenremoved from the sterile field (Figure 2 is available in color in online article)

Copyright 2011 copy The Joint Commission

54 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

IMPLEMENTATION PHASE 3 METHODS

AND METRICS

In November 2008 after review of thetrial data it was decided to implement theDMS system in all operating and LampDrooms at MCR After education and trainingof all surgical services staff use of the DMSsystem was begun across the MCR campus onFebruary 2 2009 Evaluation of use per-formance and staff satisfaction were assessedone year after implementation InstitutionalReview Board approval was obtained for datacollection staff survey data analysis andpublication

ResultsTRIAL PHASE ONE

Interobserver Reliability The interobserverreliability for sponge counting in the trialrooms was determined to be excellent withthe mean count times recorded by the twoobservers not being statistically different (p lt 650)

Time to Count The average time to count a sponge for con-trol sponges was 40 seconds (n = 335) versus 114 seconds forthe DMSs (n = 365 p lt 05) The average time to count wassignificantly different (p = 002) between the colon and rectal(8 seconds) and cardiovascular services teams (13 seconds) Thetimes to count at three different stages (at start of the case dur-ing the case and at the end of the case) were also significantlydifferent (p = 006) The average time to count a sponge for theDMSs at the start of the case was 6 seconds whereas the timeto count in a sponge during the case was 12 seconds and theaverage time to count out a sponge was 16 seconds The aver-age time to count a sponge in the same OR for four days (learn-ing curve effect) is shown in Figure 3 (right) The average timeto count a sponge decreased from 11 seconds on day 1 of thetrial to 5 seconds on day 4mdashor about the same as for countingunmarked spongesmdashfor the colon and rectal surgery ORs

Battery Performance To assess the duration of scanner bat-tery performance during longer procedures the percent chargeleft in the battery over the duration of the case without placingit in the charger was measured The units retained a charge of40 or higher after five hours of use

TRIAL PHASE TWO

The goals of the second phase trial were to (a) conduct thestudy in a larger setting assess the impact on work flow across

more diverse case types and validate the Phase 1 results (b)assess the changes made to the scanner-user interface on thebasis of the staff feedback and (c) study the effectiveness of thenew master-band packaging released by the company to speedup the count-in process (especially in emergent situations)

Time to Count The two specialties colon rectal and cardio-vascular surgery that participated in the first pilot showed adecline in the mean count times for the DMSs from 80 sec-onds to 45 seconds and from 132 seconds to 112 secondsrespectively In the general surgery ORsmdasha new specialty forthe studymdashmean counting times were similar to those observedin Trial 1 for the other specialties

IMPLEMENTATION PHASE THREE

The DMS system was implemented across the MCR cam-pus on February 2 2009 An institutionwide implementationwas performed rather than a multiphased approach to avoid thepossibility of having standard unmarked sponges and DMSs inthe OR suite concurrently Implementation followed a two-month period of staff education and training

After 18 months of continuous use in 87404 operationsprocedures 1862373 DMSs were used without a singlesponge retained During this period our institution continuedthe practice of postoperative high-resolution x-rays in a dedicat-ed radiology unit at the completion of surgery In the preced-ing six years MCR was averaging a cotton sponge RSI every 64

Figure 3 The learning curve for use of the sponge system in a single operating room demonstratedthat after four cases the time to count individual sponges decreased from 101 seconds on Day 1 to45 seconds on Day 4

Average Time to Count a Sponge with Use of the Data-Matrix Coded Sponge (DMS) System in a

Single Operating Room (OR)

Copyright 2011 copy The Joint Commission

55February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

days (Figure 4 page 55 also available in online article) Thischange in event frequency represents a significant reduction (p lt 001) related to the implementation of the DMS systemSystem performance on RSI prevention is monitored daily byour surgical safety team [RRC JC SP CW] in collabora-tion with the Institutional Sentinel Event group To date therehave been three reported events where the manual count hadreconciled all the sponge counts as correct but the DMS systemwas not reconciled Intraoperative x-rays or wound explorationrevealed the sponge before completion of the procedure thusavoiding an RSI

STAFF SURVEY

After each of the two trials and after one year of system useOR staff were electronically mailed a multiquestion surveyReturn of the survey was voluntary No effort was made toensure that staff had completed both surveys or had been in theOR during the entire year after implementation Each questionused a 5-point Likert scale with 1 indicating ldquostrong dislikerdquoand 5 indicating a ldquostrong likingrdquo The summary of the resultsfrom the survey are provided in Table 1 (page 56) Additionalcomments from the staff were documented in the commentssection of the survey The percentage of 4 and 5 scores (4

ldquosomewhat like the DMS systemrdquo 5ldquostrongly like the DMS systemrdquo) increasedduring the first year of use Confidence inreliability of the system was relativelyunchanged Given the limits of the surveymethodology a statistical analysis of thesurvey results was not conducted

DiscussionRSIs after surgical or invasive procedurescontinue to challenge OR staff as signifi-cant medical errors Cotton sponges arethe most common retained items despitenational standards for sponge counting5612

In a multiphase technology assessment of aDMS system at MCR which entailed twointernal randomized controlled trials thesystem performed within the specifiedtechnical parameters and did not disruptthe flow of the surgical case After 18months of continuous use of the DMS sys-tem throughout our surgical practice inwhich some 1862373 sponges werecounted we had no retained-sponge

events There was no increase in overall operative time relatedto this sponge-counting technology The learning curve for usewas extremely short (lt four cases) Staff satisfaction with thesystem was acceptable with a high degree of confidence in thereliability of the system

Although many retained sponges are detected in the earlypostoperative period reports of retained sponges causingchronic symptoms or incidentally being found years to decadesafter the index operation are unfortunately quite frequent13ndash16

RSIs including sponges are associated with significant morbid-ity (small-bowel fistula obstruction visceral perforation re-operations to remove the object) and rarely death12 Despitenational practice standards related to sponge and instrumentcounting these protocols are apparently not sufficient to avoidRSIs11 As noted by Christian et al the counting procedures arewell designed and nearly universally performed and account foras much as 14 of total operative time17 Unfortunately thisseemingly ldquoeasyrdquo counting task is error prone because it is per-formed in a discontinuous fashion throughout a complex pro-cedure with multiple interruptions competing demands andtasks and possibly numerous participants which leads toerrors

As our previous experience has demonstrated reliance on

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July2010 at Mayo Clinic Rochester The data-matrix coded sponge system was introduced in February2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit (Figure4 is available in color in online article)

Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Copyright 2011 copy The Joint Commission

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 3: Mayo Clinic Study

52 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

attempt specifically to prevent retention ofcotton sponge products during surgical pro-cedures8ndash10 There are two major types of tech-nologies (1) sponges with unique identifyingdata-matrix codes (previously known as barcodes) annealed onto the sponge and (2)sponges with a radiofrequency (RF) chipembedded into them The data-matrixndashcodedsystem is solely a counting system for spongeproducts If used properly it records thenumber of sponges scanned into the case andat the end reconciles the number of spongesscanned out of the case Depending on theproduct the RF technology can act solely as acounting technology or additionally as adetecting technology with the use of a hand-held device to scan the patient for the pres-ence of a sponge if one is missing from thefinal count The few studies regarding theseproducts are quite limited in scope and donot address the real-world effectiveness ofthese technologies being implemented acrossan entire OR practice Specifically there areno reports of long-term results on sponge RSIreduction after implementation of a technol-ogy in a surgical practice

In this study we describe the technologyassessment of a data-matrixndashcoded sponge(DMS) system through two internal trials and after full imple-mentation in the operating and surgical procedure rooms atMCR The system includes individually unique data-matrixndashcoded cotton sponge products (Figure 1 right also available inonline article) and a data-matrixndashreading scanner that keeps arunning ledger of the sponges scanned onto and off of the ster-ile field (Figure 2 page 53 also available in online article)Safety performance after implementation as measured by theincidence of retained sponge products and staff satisfaction wasevaluated

MethodsCOUNTING OF SPONGES

The surgical staff at MCR follow the Association ofperiOperative Registered Nurses (AORN) recommended prac-tices for counting sponges11 Our standard sponge-countingpolicy is to have the certified surgical technician and circulatingnurse perform concurrent visual and audible sponge countsthroughout the procedure According to policy counts had to

be performed at the beginning of the case with any addition orremoval of sponges from the sterile field at the closure of anybody cavity with change in nursing personnel and before thelast stitch is placed For the DMS system (pilot cases and afterfull implementation) the standard counting practices asdescribed were performed and in addition the sponges werescanned using a pole-mounted data-matrix tag reader beforeadding sponges to or when sponges were removed from thesterile surgical field

TRAINED OBSERVERS

Quality management services staff trained surgical servicenursing staff in proper observational techniques and use ofstandardized data intake forms In the first phase(FebruaryndashMarch 2008) two trained observers were present ineach OR collecting data independently to determine the inter-observer reliability In the second phase (August 2008) onetrained observer collected the data in each OR during the pilotsin the surgical specialties

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cottonsurgical sponge products Each sponge or towel has a unique data-matrix tag annealed to the itemThis tag contains the unique code identifying the individual sponge After the sponge is scanned intothe scanner it must be scanned out to remove it from the case sponge ledger The sponge packs aresecured with a band labeled with a ldquomaster bandrdquo This data-matrixndashcoded band has the data foreach individual sponge within the package Scanning this band loads into the scanner all the indi-vidual sponges avoiding the need to scan each individual sponge onto the field However eachsponge must be scanned out at the end of the procedure (Figure 1 is available in color in onlinearticle)

The Data-MatrixndashCoded Sponge System

Copyright 2011 copy The Joint Commission

53February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

STATISTICAL ANALYSIS

Statistical analysis was performed using Minitab software(version 142 State College Pennsylvania)

TRIAL PHASE 1 METHODS AND METRICS

All testing and evaluation of the DMS system were carriedout at MCR a tertiary-referral academic medical center in theupper Midwest in 2008 There are 117 inpatient and outpa-tient ORs 3 obstetrical ORs and 8 labor amp delivery (LampD)birthing rooms on the MCR campus distributed between twoacute care hospitals All operating and LampD rooms are staffedby MCR physicians nurses and allied health staff All staffmembers are employed under one organizational leadershipstructure with a unified policy and procedure manual for OR

conduct including sponge countingA Plan-Do-Study-Act (PDSA) methodology was adapted to

evaluate the DMS system in our ORs In the first PDSA cyclea total of 16 cases was randomly assigned to either the DMSs(Pilot A) or standard surgical sponges (Control A) on the basisof the availability of the observer teams Patients undergoing anelective weekday procedure in cardiovascular surgery or colonand rectal surgery were eligible for the randomization duringthe one-week trial period A trial was performed in two ORs forcardiovascular surgery and two ORs for colon and rectal sur-gery Surgical staff in the specialties were trained on the use ofthe DMS system equipment during the week before the trialTwo trained observers were in the room to independently eval-uate system performance and collect appropriate data

The following technical performance metrics were collected Interobserver reliability Average time spent counting a sponge (time spent to

count sponges at the beginning of the case during the case foraddition or removal of sponges from the sterile field duringpermanent personnel changes and before the last stitch isplaced divided by the number of sponges used)

Learning curve for scanning DMSs Total time spent to count sponges Battery life of the scannerIn addition the staff rsquos impression of product features and

work flow were assessed with a standardized new product survey

TRIAL PHASE 2 METHODS AND METRICS

In the second PDSA cycle a trial was performed in fourORs in each of the following specialties colon and rectal car-diovascular and general surgery DMSs were used in all electivecases in each of the four ORs for a period of one week Trainedobservers were present in the surgical suite to help with anytechnical difficulties experienced by the staff The observerswere not assigned to specific cases but moved from case to case

Data were gathered from a total of 57 surgeries in the singleweek In this phase upgrades to the DMS system software weremade on the basis of feedback from staff from the first phase inregard to the patient data entry screens In addition to the met-rics used for first phase data were collected to verify accuracyof the ldquomaster bandsrdquo which was a new packaging style intro-duced by the manufacturer Master bands contained uniquesponge-identifying information of the 5 or 10 sponges includ-ed in the pack thereby eliminating the need to scan in eachsponge from the package However each individual spongemust be scanned out when it is removed from the sterile field

Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number ofsponges still on the sterile field and the number of sponges that have beenremoved from the sterile field (Figure 2 is available in color in online article)

Copyright 2011 copy The Joint Commission

54 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

IMPLEMENTATION PHASE 3 METHODS

AND METRICS

In November 2008 after review of thetrial data it was decided to implement theDMS system in all operating and LampDrooms at MCR After education and trainingof all surgical services staff use of the DMSsystem was begun across the MCR campus onFebruary 2 2009 Evaluation of use per-formance and staff satisfaction were assessedone year after implementation InstitutionalReview Board approval was obtained for datacollection staff survey data analysis andpublication

ResultsTRIAL PHASE ONE

Interobserver Reliability The interobserverreliability for sponge counting in the trialrooms was determined to be excellent withthe mean count times recorded by the twoobservers not being statistically different (p lt 650)

Time to Count The average time to count a sponge for con-trol sponges was 40 seconds (n = 335) versus 114 seconds forthe DMSs (n = 365 p lt 05) The average time to count wassignificantly different (p = 002) between the colon and rectal(8 seconds) and cardiovascular services teams (13 seconds) Thetimes to count at three different stages (at start of the case dur-ing the case and at the end of the case) were also significantlydifferent (p = 006) The average time to count a sponge for theDMSs at the start of the case was 6 seconds whereas the timeto count in a sponge during the case was 12 seconds and theaverage time to count out a sponge was 16 seconds The aver-age time to count a sponge in the same OR for four days (learn-ing curve effect) is shown in Figure 3 (right) The average timeto count a sponge decreased from 11 seconds on day 1 of thetrial to 5 seconds on day 4mdashor about the same as for countingunmarked spongesmdashfor the colon and rectal surgery ORs

Battery Performance To assess the duration of scanner bat-tery performance during longer procedures the percent chargeleft in the battery over the duration of the case without placingit in the charger was measured The units retained a charge of40 or higher after five hours of use

TRIAL PHASE TWO

The goals of the second phase trial were to (a) conduct thestudy in a larger setting assess the impact on work flow across

more diverse case types and validate the Phase 1 results (b)assess the changes made to the scanner-user interface on thebasis of the staff feedback and (c) study the effectiveness of thenew master-band packaging released by the company to speedup the count-in process (especially in emergent situations)

Time to Count The two specialties colon rectal and cardio-vascular surgery that participated in the first pilot showed adecline in the mean count times for the DMSs from 80 sec-onds to 45 seconds and from 132 seconds to 112 secondsrespectively In the general surgery ORsmdasha new specialty forthe studymdashmean counting times were similar to those observedin Trial 1 for the other specialties

IMPLEMENTATION PHASE THREE

The DMS system was implemented across the MCR cam-pus on February 2 2009 An institutionwide implementationwas performed rather than a multiphased approach to avoid thepossibility of having standard unmarked sponges and DMSs inthe OR suite concurrently Implementation followed a two-month period of staff education and training

After 18 months of continuous use in 87404 operationsprocedures 1862373 DMSs were used without a singlesponge retained During this period our institution continuedthe practice of postoperative high-resolution x-rays in a dedicat-ed radiology unit at the completion of surgery In the preced-ing six years MCR was averaging a cotton sponge RSI every 64

Figure 3 The learning curve for use of the sponge system in a single operating room demonstratedthat after four cases the time to count individual sponges decreased from 101 seconds on Day 1 to45 seconds on Day 4

Average Time to Count a Sponge with Use of the Data-Matrix Coded Sponge (DMS) System in a

Single Operating Room (OR)

Copyright 2011 copy The Joint Commission

55February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

days (Figure 4 page 55 also available in online article) Thischange in event frequency represents a significant reduction (p lt 001) related to the implementation of the DMS systemSystem performance on RSI prevention is monitored daily byour surgical safety team [RRC JC SP CW] in collabora-tion with the Institutional Sentinel Event group To date therehave been three reported events where the manual count hadreconciled all the sponge counts as correct but the DMS systemwas not reconciled Intraoperative x-rays or wound explorationrevealed the sponge before completion of the procedure thusavoiding an RSI

STAFF SURVEY

After each of the two trials and after one year of system useOR staff were electronically mailed a multiquestion surveyReturn of the survey was voluntary No effort was made toensure that staff had completed both surveys or had been in theOR during the entire year after implementation Each questionused a 5-point Likert scale with 1 indicating ldquostrong dislikerdquoand 5 indicating a ldquostrong likingrdquo The summary of the resultsfrom the survey are provided in Table 1 (page 56) Additionalcomments from the staff were documented in the commentssection of the survey The percentage of 4 and 5 scores (4

ldquosomewhat like the DMS systemrdquo 5ldquostrongly like the DMS systemrdquo) increasedduring the first year of use Confidence inreliability of the system was relativelyunchanged Given the limits of the surveymethodology a statistical analysis of thesurvey results was not conducted

DiscussionRSIs after surgical or invasive procedurescontinue to challenge OR staff as signifi-cant medical errors Cotton sponges arethe most common retained items despitenational standards for sponge counting5612

In a multiphase technology assessment of aDMS system at MCR which entailed twointernal randomized controlled trials thesystem performed within the specifiedtechnical parameters and did not disruptthe flow of the surgical case After 18months of continuous use of the DMS sys-tem throughout our surgical practice inwhich some 1862373 sponges werecounted we had no retained-sponge

events There was no increase in overall operative time relatedto this sponge-counting technology The learning curve for usewas extremely short (lt four cases) Staff satisfaction with thesystem was acceptable with a high degree of confidence in thereliability of the system

Although many retained sponges are detected in the earlypostoperative period reports of retained sponges causingchronic symptoms or incidentally being found years to decadesafter the index operation are unfortunately quite frequent13ndash16

RSIs including sponges are associated with significant morbid-ity (small-bowel fistula obstruction visceral perforation re-operations to remove the object) and rarely death12 Despitenational practice standards related to sponge and instrumentcounting these protocols are apparently not sufficient to avoidRSIs11 As noted by Christian et al the counting procedures arewell designed and nearly universally performed and account foras much as 14 of total operative time17 Unfortunately thisseemingly ldquoeasyrdquo counting task is error prone because it is per-formed in a discontinuous fashion throughout a complex pro-cedure with multiple interruptions competing demands andtasks and possibly numerous participants which leads toerrors

As our previous experience has demonstrated reliance on

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July2010 at Mayo Clinic Rochester The data-matrix coded sponge system was introduced in February2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit (Figure4 is available in color in online article)

Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Copyright 2011 copy The Joint Commission

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 4: Mayo Clinic Study

53February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

STATISTICAL ANALYSIS

Statistical analysis was performed using Minitab software(version 142 State College Pennsylvania)

TRIAL PHASE 1 METHODS AND METRICS

All testing and evaluation of the DMS system were carriedout at MCR a tertiary-referral academic medical center in theupper Midwest in 2008 There are 117 inpatient and outpa-tient ORs 3 obstetrical ORs and 8 labor amp delivery (LampD)birthing rooms on the MCR campus distributed between twoacute care hospitals All operating and LampD rooms are staffedby MCR physicians nurses and allied health staff All staffmembers are employed under one organizational leadershipstructure with a unified policy and procedure manual for OR

conduct including sponge countingA Plan-Do-Study-Act (PDSA) methodology was adapted to

evaluate the DMS system in our ORs In the first PDSA cyclea total of 16 cases was randomly assigned to either the DMSs(Pilot A) or standard surgical sponges (Control A) on the basisof the availability of the observer teams Patients undergoing anelective weekday procedure in cardiovascular surgery or colonand rectal surgery were eligible for the randomization duringthe one-week trial period A trial was performed in two ORs forcardiovascular surgery and two ORs for colon and rectal sur-gery Surgical staff in the specialties were trained on the use ofthe DMS system equipment during the week before the trialTwo trained observers were in the room to independently eval-uate system performance and collect appropriate data

The following technical performance metrics were collected Interobserver reliability Average time spent counting a sponge (time spent to

count sponges at the beginning of the case during the case foraddition or removal of sponges from the sterile field duringpermanent personnel changes and before the last stitch isplaced divided by the number of sponges used)

Learning curve for scanning DMSs Total time spent to count sponges Battery life of the scannerIn addition the staff rsquos impression of product features and

work flow were assessed with a standardized new product survey

TRIAL PHASE 2 METHODS AND METRICS

In the second PDSA cycle a trial was performed in fourORs in each of the following specialties colon and rectal car-diovascular and general surgery DMSs were used in all electivecases in each of the four ORs for a period of one week Trainedobservers were present in the surgical suite to help with anytechnical difficulties experienced by the staff The observerswere not assigned to specific cases but moved from case to case

Data were gathered from a total of 57 surgeries in the singleweek In this phase upgrades to the DMS system software weremade on the basis of feedback from staff from the first phase inregard to the patient data entry screens In addition to the met-rics used for first phase data were collected to verify accuracyof the ldquomaster bandsrdquo which was a new packaging style intro-duced by the manufacturer Master bands contained uniquesponge-identifying information of the 5 or 10 sponges includ-ed in the pack thereby eliminating the need to scan in eachsponge from the package However each individual spongemust be scanned out when it is removed from the sterile field

Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number ofsponges still on the sterile field and the number of sponges that have beenremoved from the sterile field (Figure 2 is available in color in online article)

Copyright 2011 copy The Joint Commission

54 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

IMPLEMENTATION PHASE 3 METHODS

AND METRICS

In November 2008 after review of thetrial data it was decided to implement theDMS system in all operating and LampDrooms at MCR After education and trainingof all surgical services staff use of the DMSsystem was begun across the MCR campus onFebruary 2 2009 Evaluation of use per-formance and staff satisfaction were assessedone year after implementation InstitutionalReview Board approval was obtained for datacollection staff survey data analysis andpublication

ResultsTRIAL PHASE ONE

Interobserver Reliability The interobserverreliability for sponge counting in the trialrooms was determined to be excellent withthe mean count times recorded by the twoobservers not being statistically different (p lt 650)

Time to Count The average time to count a sponge for con-trol sponges was 40 seconds (n = 335) versus 114 seconds forthe DMSs (n = 365 p lt 05) The average time to count wassignificantly different (p = 002) between the colon and rectal(8 seconds) and cardiovascular services teams (13 seconds) Thetimes to count at three different stages (at start of the case dur-ing the case and at the end of the case) were also significantlydifferent (p = 006) The average time to count a sponge for theDMSs at the start of the case was 6 seconds whereas the timeto count in a sponge during the case was 12 seconds and theaverage time to count out a sponge was 16 seconds The aver-age time to count a sponge in the same OR for four days (learn-ing curve effect) is shown in Figure 3 (right) The average timeto count a sponge decreased from 11 seconds on day 1 of thetrial to 5 seconds on day 4mdashor about the same as for countingunmarked spongesmdashfor the colon and rectal surgery ORs

Battery Performance To assess the duration of scanner bat-tery performance during longer procedures the percent chargeleft in the battery over the duration of the case without placingit in the charger was measured The units retained a charge of40 or higher after five hours of use

TRIAL PHASE TWO

The goals of the second phase trial were to (a) conduct thestudy in a larger setting assess the impact on work flow across

more diverse case types and validate the Phase 1 results (b)assess the changes made to the scanner-user interface on thebasis of the staff feedback and (c) study the effectiveness of thenew master-band packaging released by the company to speedup the count-in process (especially in emergent situations)

Time to Count The two specialties colon rectal and cardio-vascular surgery that participated in the first pilot showed adecline in the mean count times for the DMSs from 80 sec-onds to 45 seconds and from 132 seconds to 112 secondsrespectively In the general surgery ORsmdasha new specialty forthe studymdashmean counting times were similar to those observedin Trial 1 for the other specialties

IMPLEMENTATION PHASE THREE

The DMS system was implemented across the MCR cam-pus on February 2 2009 An institutionwide implementationwas performed rather than a multiphased approach to avoid thepossibility of having standard unmarked sponges and DMSs inthe OR suite concurrently Implementation followed a two-month period of staff education and training

After 18 months of continuous use in 87404 operationsprocedures 1862373 DMSs were used without a singlesponge retained During this period our institution continuedthe practice of postoperative high-resolution x-rays in a dedicat-ed radiology unit at the completion of surgery In the preced-ing six years MCR was averaging a cotton sponge RSI every 64

Figure 3 The learning curve for use of the sponge system in a single operating room demonstratedthat after four cases the time to count individual sponges decreased from 101 seconds on Day 1 to45 seconds on Day 4

Average Time to Count a Sponge with Use of the Data-Matrix Coded Sponge (DMS) System in a

Single Operating Room (OR)

Copyright 2011 copy The Joint Commission

55February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

days (Figure 4 page 55 also available in online article) Thischange in event frequency represents a significant reduction (p lt 001) related to the implementation of the DMS systemSystem performance on RSI prevention is monitored daily byour surgical safety team [RRC JC SP CW] in collabora-tion with the Institutional Sentinel Event group To date therehave been three reported events where the manual count hadreconciled all the sponge counts as correct but the DMS systemwas not reconciled Intraoperative x-rays or wound explorationrevealed the sponge before completion of the procedure thusavoiding an RSI

STAFF SURVEY

After each of the two trials and after one year of system useOR staff were electronically mailed a multiquestion surveyReturn of the survey was voluntary No effort was made toensure that staff had completed both surveys or had been in theOR during the entire year after implementation Each questionused a 5-point Likert scale with 1 indicating ldquostrong dislikerdquoand 5 indicating a ldquostrong likingrdquo The summary of the resultsfrom the survey are provided in Table 1 (page 56) Additionalcomments from the staff were documented in the commentssection of the survey The percentage of 4 and 5 scores (4

ldquosomewhat like the DMS systemrdquo 5ldquostrongly like the DMS systemrdquo) increasedduring the first year of use Confidence inreliability of the system was relativelyunchanged Given the limits of the surveymethodology a statistical analysis of thesurvey results was not conducted

DiscussionRSIs after surgical or invasive procedurescontinue to challenge OR staff as signifi-cant medical errors Cotton sponges arethe most common retained items despitenational standards for sponge counting5612

In a multiphase technology assessment of aDMS system at MCR which entailed twointernal randomized controlled trials thesystem performed within the specifiedtechnical parameters and did not disruptthe flow of the surgical case After 18months of continuous use of the DMS sys-tem throughout our surgical practice inwhich some 1862373 sponges werecounted we had no retained-sponge

events There was no increase in overall operative time relatedto this sponge-counting technology The learning curve for usewas extremely short (lt four cases) Staff satisfaction with thesystem was acceptable with a high degree of confidence in thereliability of the system

Although many retained sponges are detected in the earlypostoperative period reports of retained sponges causingchronic symptoms or incidentally being found years to decadesafter the index operation are unfortunately quite frequent13ndash16

RSIs including sponges are associated with significant morbid-ity (small-bowel fistula obstruction visceral perforation re-operations to remove the object) and rarely death12 Despitenational practice standards related to sponge and instrumentcounting these protocols are apparently not sufficient to avoidRSIs11 As noted by Christian et al the counting procedures arewell designed and nearly universally performed and account foras much as 14 of total operative time17 Unfortunately thisseemingly ldquoeasyrdquo counting task is error prone because it is per-formed in a discontinuous fashion throughout a complex pro-cedure with multiple interruptions competing demands andtasks and possibly numerous participants which leads toerrors

As our previous experience has demonstrated reliance on

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July2010 at Mayo Clinic Rochester The data-matrix coded sponge system was introduced in February2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit (Figure4 is available in color in online article)

Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Copyright 2011 copy The Joint Commission

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 5: Mayo Clinic Study

54 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

IMPLEMENTATION PHASE 3 METHODS

AND METRICS

In November 2008 after review of thetrial data it was decided to implement theDMS system in all operating and LampDrooms at MCR After education and trainingof all surgical services staff use of the DMSsystem was begun across the MCR campus onFebruary 2 2009 Evaluation of use per-formance and staff satisfaction were assessedone year after implementation InstitutionalReview Board approval was obtained for datacollection staff survey data analysis andpublication

ResultsTRIAL PHASE ONE

Interobserver Reliability The interobserverreliability for sponge counting in the trialrooms was determined to be excellent withthe mean count times recorded by the twoobservers not being statistically different (p lt 650)

Time to Count The average time to count a sponge for con-trol sponges was 40 seconds (n = 335) versus 114 seconds forthe DMSs (n = 365 p lt 05) The average time to count wassignificantly different (p = 002) between the colon and rectal(8 seconds) and cardiovascular services teams (13 seconds) Thetimes to count at three different stages (at start of the case dur-ing the case and at the end of the case) were also significantlydifferent (p = 006) The average time to count a sponge for theDMSs at the start of the case was 6 seconds whereas the timeto count in a sponge during the case was 12 seconds and theaverage time to count out a sponge was 16 seconds The aver-age time to count a sponge in the same OR for four days (learn-ing curve effect) is shown in Figure 3 (right) The average timeto count a sponge decreased from 11 seconds on day 1 of thetrial to 5 seconds on day 4mdashor about the same as for countingunmarked spongesmdashfor the colon and rectal surgery ORs

Battery Performance To assess the duration of scanner bat-tery performance during longer procedures the percent chargeleft in the battery over the duration of the case without placingit in the charger was measured The units retained a charge of40 or higher after five hours of use

TRIAL PHASE TWO

The goals of the second phase trial were to (a) conduct thestudy in a larger setting assess the impact on work flow across

more diverse case types and validate the Phase 1 results (b)assess the changes made to the scanner-user interface on thebasis of the staff feedback and (c) study the effectiveness of thenew master-band packaging released by the company to speedup the count-in process (especially in emergent situations)

Time to Count The two specialties colon rectal and cardio-vascular surgery that participated in the first pilot showed adecline in the mean count times for the DMSs from 80 sec-onds to 45 seconds and from 132 seconds to 112 secondsrespectively In the general surgery ORsmdasha new specialty forthe studymdashmean counting times were similar to those observedin Trial 1 for the other specialties

IMPLEMENTATION PHASE THREE

The DMS system was implemented across the MCR cam-pus on February 2 2009 An institutionwide implementationwas performed rather than a multiphased approach to avoid thepossibility of having standard unmarked sponges and DMSs inthe OR suite concurrently Implementation followed a two-month period of staff education and training

After 18 months of continuous use in 87404 operationsprocedures 1862373 DMSs were used without a singlesponge retained During this period our institution continuedthe practice of postoperative high-resolution x-rays in a dedicat-ed radiology unit at the completion of surgery In the preced-ing six years MCR was averaging a cotton sponge RSI every 64

Figure 3 The learning curve for use of the sponge system in a single operating room demonstratedthat after four cases the time to count individual sponges decreased from 101 seconds on Day 1 to45 seconds on Day 4

Average Time to Count a Sponge with Use of the Data-Matrix Coded Sponge (DMS) System in a

Single Operating Room (OR)

Copyright 2011 copy The Joint Commission

55February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

days (Figure 4 page 55 also available in online article) Thischange in event frequency represents a significant reduction (p lt 001) related to the implementation of the DMS systemSystem performance on RSI prevention is monitored daily byour surgical safety team [RRC JC SP CW] in collabora-tion with the Institutional Sentinel Event group To date therehave been three reported events where the manual count hadreconciled all the sponge counts as correct but the DMS systemwas not reconciled Intraoperative x-rays or wound explorationrevealed the sponge before completion of the procedure thusavoiding an RSI

STAFF SURVEY

After each of the two trials and after one year of system useOR staff were electronically mailed a multiquestion surveyReturn of the survey was voluntary No effort was made toensure that staff had completed both surveys or had been in theOR during the entire year after implementation Each questionused a 5-point Likert scale with 1 indicating ldquostrong dislikerdquoand 5 indicating a ldquostrong likingrdquo The summary of the resultsfrom the survey are provided in Table 1 (page 56) Additionalcomments from the staff were documented in the commentssection of the survey The percentage of 4 and 5 scores (4

ldquosomewhat like the DMS systemrdquo 5ldquostrongly like the DMS systemrdquo) increasedduring the first year of use Confidence inreliability of the system was relativelyunchanged Given the limits of the surveymethodology a statistical analysis of thesurvey results was not conducted

DiscussionRSIs after surgical or invasive procedurescontinue to challenge OR staff as signifi-cant medical errors Cotton sponges arethe most common retained items despitenational standards for sponge counting5612

In a multiphase technology assessment of aDMS system at MCR which entailed twointernal randomized controlled trials thesystem performed within the specifiedtechnical parameters and did not disruptthe flow of the surgical case After 18months of continuous use of the DMS sys-tem throughout our surgical practice inwhich some 1862373 sponges werecounted we had no retained-sponge

events There was no increase in overall operative time relatedto this sponge-counting technology The learning curve for usewas extremely short (lt four cases) Staff satisfaction with thesystem was acceptable with a high degree of confidence in thereliability of the system

Although many retained sponges are detected in the earlypostoperative period reports of retained sponges causingchronic symptoms or incidentally being found years to decadesafter the index operation are unfortunately quite frequent13ndash16

RSIs including sponges are associated with significant morbid-ity (small-bowel fistula obstruction visceral perforation re-operations to remove the object) and rarely death12 Despitenational practice standards related to sponge and instrumentcounting these protocols are apparently not sufficient to avoidRSIs11 As noted by Christian et al the counting procedures arewell designed and nearly universally performed and account foras much as 14 of total operative time17 Unfortunately thisseemingly ldquoeasyrdquo counting task is error prone because it is per-formed in a discontinuous fashion throughout a complex pro-cedure with multiple interruptions competing demands andtasks and possibly numerous participants which leads toerrors

As our previous experience has demonstrated reliance on

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July2010 at Mayo Clinic Rochester The data-matrix coded sponge system was introduced in February2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit (Figure4 is available in color in online article)

Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Copyright 2011 copy The Joint Commission

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 6: Mayo Clinic Study

55February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

days (Figure 4 page 55 also available in online article) Thischange in event frequency represents a significant reduction (p lt 001) related to the implementation of the DMS systemSystem performance on RSI prevention is monitored daily byour surgical safety team [RRC JC SP CW] in collabora-tion with the Institutional Sentinel Event group To date therehave been three reported events where the manual count hadreconciled all the sponge counts as correct but the DMS systemwas not reconciled Intraoperative x-rays or wound explorationrevealed the sponge before completion of the procedure thusavoiding an RSI

STAFF SURVEY

After each of the two trials and after one year of system useOR staff were electronically mailed a multiquestion surveyReturn of the survey was voluntary No effort was made toensure that staff had completed both surveys or had been in theOR during the entire year after implementation Each questionused a 5-point Likert scale with 1 indicating ldquostrong dislikerdquoand 5 indicating a ldquostrong likingrdquo The summary of the resultsfrom the survey are provided in Table 1 (page 56) Additionalcomments from the staff were documented in the commentssection of the survey The percentage of 4 and 5 scores (4

ldquosomewhat like the DMS systemrdquo 5ldquostrongly like the DMS systemrdquo) increasedduring the first year of use Confidence inreliability of the system was relativelyunchanged Given the limits of the surveymethodology a statistical analysis of thesurvey results was not conducted

DiscussionRSIs after surgical or invasive procedurescontinue to challenge OR staff as signifi-cant medical errors Cotton sponges arethe most common retained items despitenational standards for sponge counting5612

In a multiphase technology assessment of aDMS system at MCR which entailed twointernal randomized controlled trials thesystem performed within the specifiedtechnical parameters and did not disruptthe flow of the surgical case After 18months of continuous use of the DMS sys-tem throughout our surgical practice inwhich some 1862373 sponges werecounted we had no retained-sponge

events There was no increase in overall operative time relatedto this sponge-counting technology The learning curve for usewas extremely short (lt four cases) Staff satisfaction with thesystem was acceptable with a high degree of confidence in thereliability of the system

Although many retained sponges are detected in the earlypostoperative period reports of retained sponges causingchronic symptoms or incidentally being found years to decadesafter the index operation are unfortunately quite frequent13ndash16

RSIs including sponges are associated with significant morbid-ity (small-bowel fistula obstruction visceral perforation re-operations to remove the object) and rarely death12 Despitenational practice standards related to sponge and instrumentcounting these protocols are apparently not sufficient to avoidRSIs11 As noted by Christian et al the counting procedures arewell designed and nearly universally performed and account foras much as 14 of total operative time17 Unfortunately thisseemingly ldquoeasyrdquo counting task is error prone because it is per-formed in a discontinuous fashion throughout a complex pro-cedure with multiple interruptions competing demands andtasks and possibly numerous participants which leads toerrors

As our previous experience has demonstrated reliance on

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July2010 at Mayo Clinic Rochester The data-matrix coded sponge system was introduced in February2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit (Figure4 is available in color in online article)

Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Copyright 2011 copy The Joint Commission

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 7: Mayo Clinic Study

56 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

policy and repeated education efforts does not lead to zero-sponge RSIs67 Even with those efforts during this study peri-od three near-miss events occurred that would not have beendetected until the patient had left the OR and had his or herpostoperative x-ray if it had not been for use of the DMS sys-tem In the absence of a highly reliable and accurate accountingsystem for sponges interventions based solely on educatingstaff or expecting them to strictly adhere to all counting poli-cies will likely not be completely successful7

In other industries where repetitive tasks are performed orproducts need reliable tracking in a high-throughput systemthere is a strong reliance on technology In regard to account-ing for surgical sponge products there is limited literature onthe use of RF sponge-tracking systems Macario et al report-ing on the initial clinical experience in the use of RF sponge-tracking systems in a proof-of-concept trial in eight patientsfound 100 detection of a single sponge placed in theabdomen9 At present there is no published trial or real-worldclinical experience with an RF system through which to evalu-ate its effectiveness Bar-code technology which has been usedfor decades in numerous industries has an established trackrecord of success in improving safety in the health care environ-ment1819 Greenberg et al evaluated the performance of theDMS system in a randomized controlled trial in 300 generalsurgery operations10 Use of the DMS system as compared withthe standard process led to the identification of significantlymore count discrepancies Greenberg et al also found that totaltime devoted to sponge counting increased significantly from24 to 53 minutes Furthermore some type of difficulty with

the technology was reported for 17 of the 150 cases In ourstudies we did not assess the number of counting discrepan-cies if any that required resolution We also found that theDMS system increased sponge-counting time However thelearning curve was quite short and after four cases using thetechnology the time that staff spent in counting sponges (meanfive seconds) was very similar to the time associated with thestandard counting process of nonndashDMS sponges (mean fourseconds) Although we did find that time devoted to spongecounting with the DMS system did increase it did not increasethe overall case length sponge counting is a parallel process thatoccurs during the course of the operation rather than just at theend of the procedure Furthermore the time to countmdashthat isscanning timemdashwas reduced by 50 by the introduction ofmaster tags for each pack of sponges

Much as found by Greenberg et al10 a survey of our staff atthe time of testing and implementation demonstrated nostrong preference for use of the bar-coding technology butrecognition of the improved accuracy However after a year ofuse staff impressions substantially improved Staff clearly aremore comfortable with the system and associated processImportantly a sizable number of staff believe that the DMS sys-tem has reduced the stress of counting in our ORs

COST-BENEFIT

Although the addition of any new technology requires ananalysis of the economic impact only limited data are availableon the economic cost of RSIs In the analysis of RSI casesreported by Gawande et al the cost per event averaged

Before DMS System One Year After

Question Implementation (2 Trial Periods) DMS Implementation

How do you rate the DMSndashcounting process 41 60

How comfortable do you feel with the DMS process 65 82

Compared with the manual sponge counting process how would you rate

the DMSndashcounting process 18 52

Compared with the manual sponge counting process did the DMSndashcounting

process decrease stress associated with counting 0 31

The DMSndashcounting process is very efficient 53 59

Do you have a high level of confidence in the accuracy of the DMSndashcounting

process 88 81

The responses are from a convenience sample of staff who voluntarily completed the survey The number of respondents varied from 65 in trial phase one to 60

in trial phase two and 204 one-year postimplementation

Table 1 Percent of 4 (ldquoSomewhat Likerdquo) or 5 (ldquoStrongly Like) Scores on a Voluntary Staff Survey Regarding the Data-MatrixndashCoded Sponge (DMS) System (N = )

Copyright 2011 copy The Joint Commission

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 8: Mayo Clinic Study

57February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

$52581 for compensation and legal-defense expenses5

However others have suggested that total legal costs are muchhigher exceeding $200000 per event2021 Furthermore recentMedicare changes in 2008 made RSIs nonreimbursed eventswhich needs to be factored into the institutional cost of RSIs22

Because RSIs are fortunately a rare event assessing the econom-ic impact of prevention strategies is difficult Modeling hasbeen used to estimate the cost-effectiveness for different modal-ities to detectprevent RSIs Dossett et al who modeled the useof routine intraoperative x-rays for all emergent open-cavitycases found that it was an economically appropriateapproach23 Although emergent cases are one risk factor forRSIs others include obesity high blood-loss procedureschanges in operative plan and multiple-team cases524 Targetedscreening for RSIs on the basis of risk factors is one approachHowever in an evaluation of nearly a quarter million opera-tions at a single institution none of the 34 RSIs occurred withone of the reported risk factors6 Given the uncertainty of tar-geted screening technology applied universally needs to beconsidered Regenbogen et al used standard counting a DMSsystem mandatory x-rays and RFID to model selective anduniversal screening for prevention of sponge RSIs20 They con-cluded that data-matrix coding was the most cost-effectivemodality when balancing prevention of a retained sponge andthe cost of implementation whereas selective and universal x-rays were considered prohibitively expensive with RFID fallingsomewhere in between The implementation costs and one-yearutilization at our institution has resulted in an average case costincrease of $1163 directly related to the DMS system

Cost factors associated with the prevention of sponge RSIsor dealing with sequela of such events must be quantified byeach institution to determine the cost-benefit of implementingthis OR patient safety technology For example the additionalOR time required in resolving miscounts during the case hasbeen shown to average 13 minutes per event10 The time spentin meetings for the involved staff and completion of a rootcause analysis for such events the reporting process and imple-menting solutions all need to be considered Finally in an eraof mandatory public reporting of major adverse patient eventsthe institutionrsquos reputation might be negatively influenced bysuch publicity

LIMITATIONS

This study represents a single institutionrsquos experience withthe DMS system OR processes vary widely and introducingany new technology requires a detailed analysis of the affectedwork flows and institutionrsquos needs In addition the highly cen-

tralized organizational structure of our institution may limit thegeneralizability of our experience to other organizations Allstakeholders in the OR are employees of the Mayo Clinic Anessential element of our organizational culture is to build con-sensus for process change However if organizational leadershipdecides that a specific process change is aligned with institu-tional goals then that process will be implemented In otherorganizations the influence of individuals or groups mighthamper the successful implementation of the DMS systemFinally implementation of the DMS system was part of a mul-tiyear staff education and quality improvement process whichmost likely significantly contributed to the successful imple-mentation

Summary and ConclusionA high-volume academic surgical practice eliminated cottonsponge RSIs through the implementation of a DMS system andhas maintained this performance level for more than 18months Through a multiphase trial we determined that thetechnology was simple to use and could be implemented acrossa large multispecialty surgical practice with minimal work-flowdisruption The technology was acceptable to the users and theeconomic impact of implementation on a per-case cost wasconsidered acceptable given the significant improvement inpatient safety This technology which appears to be reliable andeffective should be considered as an adjunct to standard ORsponge-counting practices J

Online-Only Content

See the online version of this article for

Figure 1 The Data-MatrixndashCoded Sponge System (color version)

Figure 2 Data-Matrix Scanner (color version)

Figure 4 Interval Between Retained Sponge Items (RSIs) January

2003ndashJuly 2010 (color version)

8

Robert R Cima MD MA is Associate Professor and Vice-

Chairman Surgical Services Mayo Clinic College of Medicine

Mayo Clinic Rochester Minnesota Anantha Kollengode PhD

MBA is a Quality Analyst Quality Management Services Mayo

Clinic James Clark RN MS and Sarah Pool RN MS are

Nursing Educators and Cheryl Weisbrod RN MS is Director

Surgical Services Mayo Clinic Gwendolyn J Amstutz MHA is

an Associate Administrator Administrative Services Mayo Clinic

Claude Deschamps MD is Professor and Chairman

Department of Surgery Mayo Clinic Please address correspon-

dence to Robert R Cima CimaRobertMayoedu

Copyright 2011 copy The Joint Commission

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 9: Mayo Clinic Study

58 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

1 Institute of Medicine To Err Is Human Building a Safer Health SystemWashington DC National Academy Press 19992 National Quality Forum (NQF) Serious Reportable Events in Healthcare AConsensus Report Washington DC NQF 20023 Shetty R Towel Found in Dead Womanrsquos Body Aug 6 2004httpwwwcbsnewscomstories20040806healthmain634558shtml (lastacces sed Dec 7 2010)4 Sowka MP Executive summary The medical malpractice closed claimsstudy Conn Med 45109ndash115 Feb 19815 Gawande AA et al Risk factors for retained instruments and spongesafter surgery N Engl J Med 348229ndash235 Jan 16 20036 Cima RR et al Incidence and characteristics of potential and actualretained foreign object events in surgical patients J Am Coll Surg 20780ndash87Jul 2008 Epub May 23 20087 Cima RR et al A multidisciplinary team approach to retained foreignobjects Jt Comm J Qual Patient Saf 35123ndash132 Mar 20098 Rogers A Jones E Oleynikov D Radio frequency identification (RFID)applied to surgical sponges Surg Endosc 211235ndash1237 Jul 2007 Epub May5 20079 Macario A Morris D Morris S Initial clinical evaluation of a handhelddevice for detecting retained surgical gauze sponges using radiofrequency iden-tification technology Arch Surg 141659ndash662 Jul 200610 Greenberg CC et al Bar-coding surgical sponges to improve safety Arandomized controlled trial Ann Surg 247612ndash616 Apr 200811 Association of periOperative Registered Nurses (AORN) Recommendedpractices for sponge sharp and instrument counts In Perioperative Standardsand Recommended Practices Denver AORN Inc 2008 pp 293ndash30212 Gonzaacutelez-Ojeda A et al Retained foreign bodies following intra-abdominal surgery Hepatogastroenterology 46808ndash812 MarndashApr 199913 Hyslop JW Maull KI Natural history of the retained surgical spongeSouth Med J 75657ndash660 Jun 1982

14 Yildirim S et al Retained surgical sponge (gossypiboma) after intraab-dominal or retroperitoneal surgery 14 cases treated at a single centerLangenbecks Arch Surg 391390ndash395 Aug 2006 Epub Sep 17 200515 Lauwers PR Van Hee RH Intraperitoneal gossypibomas The need tocount sponges World J Surg 24521ndash527 May 200016 Yeung KW Chang MS Huang JF Imaging of transmural migration ofa retained surgical sponge A case report Kaohsiung J Med Sci 20567ndash571Nov 200417 Christian CK et al A prospective study of patient safety in the operat-ing room Surgery 139159ndash173 Feb 2006 18 Anderson S Wittwer W Using bar-code point-of-care technology forpatient safety J Healthc Qual 265ndash11 NovndashDec 200419 Cochran GL et al Errors prevented by and associated with bar-codemedication administration systems Jt Comm J Qual Patient Saf 33293ndash301May 200720 Regenbogen SE et al Prevention of retained surgical sponges A decision-analytic model predicting relative cost-effectiveness Surgery 145527ndash535 May 2009 Epub Mar 21 200921 Kaiser CW et al The retained surgical sponge Ann Surg 22479ndash84 Jul199622 Centers for Medicare amp Medicaid Services (CMS) Fact Sheet CMSImproves Patient Safety for Medicare and Medicaid by Addressing Never EventshttpwwwcmshhsgovappsmediapressfactsheetaspCounter=3224 (lastaccessed Dec 7 2010)23 Dossett LA et al Cost-effectiveness of routine radiographs after emer-gent open cavity operations Surgery 144317ndash321 Aug 2008 Epub Jun 5200824 Lincourt AE et al Retained foreign bodies after surgery J Surg Res138170ndash174 Apr 2007 Epub Feb 1 2007

References

Consider a site license

Having a Journal site license means anyone in your

organization gets affordable electronic delivery of all

of the great new Journal content directly to their

computer every month Plus theyrsquoll have complete

searchable electronic access to nearly 1000

archived Journal articles dating back to the year

2000

Call 800-746-6578 today for more information on a

site license for the Journal or any other Joint

Commission Resources periodical

Does everyone in your organization need The JointCommission Journal on Quality and Patient Safety

Copyright 2011 copy The Joint Commission

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 10: Mayo Clinic Study

AP1 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8The Data-Matrix Coded Sponge System

Figure 1 The data-matrixndashcoded sponge (DMS) system includes a wide variety of labeled cotton surgical sponge products Each sponge or towel has aunique data-matrix tag annealed to the item This tag contains the unique code identifying the individual sponge After the sponge is scanned into thescanner it must be scanned out to remove it from the case sponge ledger The sponge packs are secured with a band labeled with a ldquomaster bandrdquo Thisdata-matrixndashcoded band has the data for each individual sponge within the package Scanning this band loads into the scanner all the individualsponges avoiding the need to scan each individual sponge onto the field However each sponge must be scanned out at the end of the procedure

Copyright 2011 copy The Joint Commission

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 11: Mayo Clinic Study

AP2February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Data-Matrix Scanner

Figure 2 The data-matrix scanner shows the type of and the number of sponges still on the sterile field and the number of sponges that have beenremoved from the sterile field

Copyright 2011 copy The Joint Commission

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission

Page 12: Mayo Clinic Study

AP3 February 2011 Volume 37 Number 2

The Joint Commission Journal on Quality and Patient Safety

Online-Only Content8Interval Between Retained Sponge Items (RSIs) January 2003ndashJuly 2010

Figure 4 This G-chart presents the interval between sponge RSIs from January 2003 through July 2010 at Mayo Clinic Rochester The data-matrixcoded sponge system was introduced in February 2009 DMS data-matrixndashcoded sponge LCL lower control limit UCL upper control limit

Copyright 2011 copy The Joint Commission