This quarterly update will discuss five issues:
- The direction of the Patient Safety Authority project to prevent wrong-site surgery and considerations of its recommendations to prevent wrong-site surgery
- Estimates of the incidence of wrong-site anesthetic blocks and ureteral stent insertions
- Informative near-miss reports
- Preventing misinformation from the surgeon’s office
- The relationship between wrong-site surgery and the operating room (OR) culture of safety
Direction of the Authority’s Project to Prevent Wrong-site Surgery and Considerations of Recommendations to Prevent Wrong-Site Surgery
There were another 15 wrong-site surgeries this quarter, bringing the total since reporting began in July 2004 to 469, in a seemingly inevitable weekly march to an unenviable 500. However, the project to prevent wrong-site surgery, starting with the initial identification of evidence-based best-practice principles in September 2007,1 shows progress. The Authority previously reported on the commitments of eight (now nine) facilities to implement wrong-site surgery prevention programs on their own.2 It has also described the results of two collaborations totaling 49 facilities to implement evidence-based best practices to prevent wrong-site surgery. The first collaboration of 30 facilities resulted in a 73% reduction of wrong-site surgery.3 The second collaboration resulted in no wrong-site surgeries in any of the facilities’ ORs for more than one year.4 Extensive analysis has shown no difference in recidivism rates between facilities exhibiting a commitment to self-improvement and facilities participating in collaborations, suggesting that both can be equally successful.
All of the progress in preventing wrong-site surgery has occurred in the 58 facilities that made a serious effort to prevent wrong-site surgery. The remaining facilities in Pennsylvania have actually shown an increase in wrong-site surgeries (see Figure). These opposite trends support the notion that institutional commitment is critical to preventing this “never” event. The Authority encourages the remaining facilities in Pennsylvania to make the institutional commitment to join in the collaborations to prevent wrong-site surgery.
Figure. Wrong-Site Surgery Trends by Intervention
The variation—or volatility—of the reported number of events per quarter (or of the time between events) has an inverse relationship to the degree of standardization or consistency of processes within the groups: the more the volatility, the less the consistency in the processes. Volatility can be measured by the standard deviation of the time between events: the higher the standard deviation, as a percentage of the mean, the greater the volatility and the less likely that the processes to prevent wrong-site surgery are consistently followed. The Authority examined the time between events from July 2004 through September 2011 (to avoid open-ended reporting at the end of the current quarter). The results confirm that demonstrated efforts to prevent wrong-site surgery are associated not only with better results, but also with less volatility, implying that the improvements are not due to random variation, but to more consistency in following procedures (see Table 1).
Table 1. Volatility of the Time between Wrong-Site Surgery Events
for Different Initiatives to Prevent Wrong-Site Surgery, Pre- and Post-Intervention
The Authority is circulating draft recommendations for preventing wrong-site surgery. It invites comments about the potential barriers to implementation of the recommendations from facilities that provide surgical services. The Authority specifically encourages the OR managers of all facilities in Pennsylvania to indicate facility-specific barriers to possible implementation, as well as provide stories about successful implementation for each of the draft recommendations. To solicit this information, the Authority has circulated a survey to all of the Pennsylvania OR mangers through their patient safety officers.
Estimates of the Incidence of Some Frequent Types of Wrong-site Procedures
Wrong-Site Anesthetic Blocks
There were another three wrong-site anesthetic blocks reported this quarter, despite the attention given to this problem in recent issues of the Advisory.5,6 A report by Stanton et al.7 allows one to infer an incidence of wrong-site anesthetic blocks. The authors reported on their experience with a verification protocol to prevent wrong-site blocks. Despite their prevention protocol, they experienced two events over three years in a hospital practice in which peripheral nerve blocks accounted for 44% of approximately 19,500 anesthetics per year. These estimates produce an incidence of approximately one wrong-site block for every 12,870 peripheral nerve blocks.
Wrong-Site Ureteral Stent Insertions
There were three wrong-site procedures involving ureteral stents reported this quarter, bringing the total number for these wrong-site procedures to 25. As mentioned in an earlier update, procedures involving stenting of the ureters account for about 80% of all wrong-site urological procedures.6
The Authority obtained the number of insertions of ureteral stents from the Pennsylvania Health Care Cost Containment Council to calculate an incidence of insertion of ureteral stents, as a primary procedure, on the wrong side. The techniques for estimating the number of single ureteral stent insertions as primary procedures was not straightforward. The method for obtaining a point estimate is explained in the “Calculation of the Number of Ureteral Stents Inserted as a Primary Procedure.” The median point estimate was 81,317 insertions of a single ureteral stent as a primary procedure from the beginning of the third quarter of 2004 through the end of the first quarter of 2011. During that period, the Authority received 19 reports of a ureteral stent being inserted in the incorrect ureter as a primary procedure without realization and reinsertion in the correct ureter before the end of the procedure, for an incidence of one wrong-site procedure for every 4,280 procedures. Assuming the upper range of the estimates for the number of unilateral procedures, the incidence is at least one wrong-site procedure for every 4,765 procedures.
Informative Near-Miss Reports
A Proposed Addition to Critical Near-Miss Reports
In the December 2011 update,4 the Authority listed three types of wrong-site near-miss events identified by the World Health Organization’s High 5s project as critical near misses worthy of root-cause analyses, perhaps using the Authority’s standard form: wrong-site near misses resulting in the following:
- Procedures that are done correctly on the correct patient despite incorrect information
- Errors caught by the last step of the Universal Protocol, the time-out
- Near-miss situations resulting in cancellation of the procedure
A report this quarter suggests a fourth category:
- Medically indicated procedures done, with prior approval, that differ from the originally scheduled procedure because of a near-miss event caught during the preparation of the patient for surgery. An example of a report that would fall into this category is as follows:
Left eye was marked with the word “right” over it. Drops were put in the left eye prior to surgery. Error noted prior to patient having procedure. The patient did require the surgery on the left side, so the surgeon spoke with the patient and his wife, got a new consent, and proceeded with the left eye.
As previously noted,4 treating critical near misses as seriously as wrong-site events should help maintain awareness of the constant risk of wrong-site events.
Reminders from Other Near-Miss Reports This Quarter
All verbal verification should be done using questions that require an active response of specific information rather than a passive agreement, such as in the following example:
The patient was identified in preoperative [area]. The patient was asked if name and physician were correct, and the patient answered yes. The bracelet was checked for identification and allergies. Upon entering the OR, during bracelet check and correct surgery confirmation, it was determined that it was the incorrect patient. The patient was for another physician. The patient was returned to the preoperative [area], and the correct patient was taken to the OR.
Even after the completion of the procedure, attention must be paid to proper documentation regarding the pathology specimen, postoperative orders, and dictation of the operative note. This report shows improper documentation in the postoperative orders:
Eye surgery orders were completed and sent to pharmacy with the wrong eye indicated (written as “right” when it should have been “left”). The surgery center called the pharmacy and gave a new copy of the orders with the correct eye indicated.
Preventing Misinformation from the Surgeon’s Office
The two major causes of wrong-site surgery are misinformation and misperception. Ensuring the completeness and accuracy of information from the surgeon’s office is the first step in preventing misinformation from leading to wrong-site surgery.
Three wrong-site surgeries, one near miss with cancellation of the operation, and three other reports of near misses during the past quarter underline the importance of the surgeon’s office providing accurate information. As usual, all cases have been edited to provide contextual de-identification.
A patient was scheduled for needle localization and sentinel node injection, and then surgery to remove sentinel nodes. The surgeon scheduled the surgery for Nov. 8, but scheduled the sentinel node injection in nuclear medicine for Nov. 9. The patient had the needle localization on Nov. 8 and then was taken to the OR without first going to nuclear medicine. It was then discovered that the patient was not injected in nuclear medicine.
A patient was on the OR schedule for total right-hip arthroplasty. The consent signed by the patient in the physician’s office was for left-shoulder surgery that he had completed in June. A new consent was signed by the patient.
A patient was scheduled for a procedure on the left foot. The surgery was actually to be done on the right foot. The consent was incorrect, as was all dictated information on the chart. The correct consent was obtained by the physician in the holding area. The handwritten history and physical from the physician was correct. Most incorrect areas have been corrected by appropriate personnel. Follow-up with involved hospital staff was done.
A patient with back and right-leg pain was scheduled for L4-5 discectomy. No laterality was identified by the physician when scheduling, nor was laterality identified on the consent. In the preoperative holding [area], the site was marked (lumbar area of the back). The surgeon performed a left L4-5 hemilaminotomy and excision of herniated disk. The next morning, the surgeon saw the patient, and when questioned about how his left leg was feeling, the patient stated his left leg was always fine, the pain was in his right leg. The surgeon realized the procedure was done on the incorrect side; the patient was taken back to the OR and underwent a right hemilaminotomy and excision of herniated disk.
The patient information received from the physician’s office identified the wrong side.
A patient was scheduled for L total hip replacement. While setting up, the doctor heard a [sales] representative talking to a technician about instruments for L hip. The doctor stated that it was not L but R hip. The office sheet was checked, and it was verified that it stated L hip. Implants and instrumentation present were for L hip. The patient [had discussed with surgeon doing] R hip. The case was cancelled and rescheduled for a later date.
A middle-aged man was scheduled by the surgeon for insertion of a left ureteral stent. The imaging studies were performed at another institution. The report was not present in patient information received from the office. The procedure was performed for ureteral calculus. The consent, reservation form, and registration all stated the left ureter. A time-out was conducted with the information as stated, with participation of the surgeon. The pyelogram [showed no irregularities]. After the procedure, when the surgeon consulted office notes, the surgeon determined that it should have been a right ureteral stent because the stone was present on the right. A right ureteral stent was inserted [later].
This year, the Authority had the opportunity to visit a facility that had requested a consultation as part of its initiative to prevent wrong-site surgery after experiencing four events within one year. The facility stated that its preadmission testing (PAT) personnel identified incomplete or inconsistent information on the schedule, consent, and/or history and physical from the surgeon’s office. These deficiencies were corrected with calls to the surgeon’s office. The PAT staff estimated that they caught inconsistencies on an average of four patients every day. That means the offices were putting incorrect information on the schedules, consents, and/or histories and physicians of 1,000 patients per year. Another approximately 500 patients per year were operated on without passing through PAT.
In contrast, AnMed Health Women’s and Children’s Hospital in Anderson, South Carolina, is reported to have never had a wrong-site surgery. Nevertheless, when staff of the 72-bed hospital looked at their system, they found 15 scheduling errors per day. A project team created a manual for each surgeon’s office and reduced scheduling errors by approximately 93%.8 Ms. Martha Rush, in a personal communication to the Authority, was gracious in sharing the following critical advice provided to AnMed surgeons’ offices:
The schedule must include the patient’s legal name and two other identifiers, the name of the procedure—including laterality if pertinent—without abbreviations and with all numbers written out, and the preoperative diagnosis.
The surgical procedure consent should be completed in the surgeon’s office, if possible. It should include the patient’s legal name and the name of the procedure, including laterality if pertinent, without abbreviations and with all numbers written out. The name of the procedure on the consent should match the surgeon’s order and the procedure scheduled. All spaces should be completed, including date, patient’s signature, and witness’s signature.
The history and physical should include the preoperative diagnosis and the planned surgical procedure.
The Authority advocates that facilities make a goal of 100% completeness, accuracy, and consistency of the information on the schedule, consent, and history and physical when the information leaves the surgeon’s office, prior to verification in PAT or the preoperative holding area. Achieving this goal would eliminate information errors and require more uniformity of processes from surgeon offices.
Surgeon offices and surgical facilities should establish systems to verify that the following information is complete and accurate before leaving the surgeon’s office once the decision to operate is made and the procedure is scheduled, during PAT, and in preoperative admissions, as well as before entering the OR. At minimum, this information includes the following:
- History and physical examination, including office notes if relevant
- Laboratory results
- Imaging studies
- Pathology reports when relevant
The information required for scheduling the procedure should include the side or site, if pertinent. The suggestion that the procedure be scheduled without abbreviations and with all numbers written out is consistent with general principles for preventing errors.9
The description of the procedure on the consent should also include the side or site, if pertinent, with the patient’s signature to indicate verification by the patient. The suggestion that the procedure be scheduled without abbreviations and with all numbers written out is again consistent with general principles for preventing errors.
The history and physical should include the preoperative diagnosis and planned procedure, including the side or site, if pertinent. Again, the suggestion that the procedure be scheduled without abbreviations and with all numbers written out is consistent with general principles for preventing errors. Pertinent supporting information uniquely found in the office records should be included, or copies attached, with the history and physical.
Everyone reviewing the schedule, consent, history and physical examination, and/or reports documenting the diagnosis should check for discrepancies among all those parts of the patient’s record and reconcile any discrepancies with the surgeon whenever noted.
A sample checklist and a sample monitoring tool are available from the Authority to help in achieving that goal.
The Relationship Between Wrong-Site Surgery and the OR Culture of Safety
Eleven facilities in Pennsylvania that happened to use the same culture of safety survey instrument have cooperated with the Authority by sharing the results of their culture of safety survey responses for OR personnel to create an aggregate report; eight of the facilities had experienced a wrong-site surgery and three of the facilities had not. They agreed to cooperate on the condition of confidentiality. The eight facilities with wrong-site surgery reported aggregate responses from 328 respondents. The three OR facilities without wrong-site surgery reported aggregate responses from 105 OR respondents. The results revealed statistically significant relationships between perceptions of the culture of safety by operating room providers and the histories of the facilities regarding wrong-site surgery (see Table 2).The results reinforce the importance of conducting proper briefings, being engaged, and expressing concerns.
Table 2. Differences in Agreement with Statements on a
Culture of Safety Survey by Operating Room Personnel
between Facilities with No History of Wrong-Site Surgery
and a History of Wrong-Site Surgery
Clarke JR, Johnston J, Finley ED. Getting surgery right. Ann Surg 2007 Sep;246(3):395-403.
Pelczarski KM, Braun PA, Young E. Hospitals collaborate to prevent wrong-site surgery. Patient Saf Qual Healthc. Sep/Oct 2010:20-26.
Stanton MA, Tong-Ngork S, Liguori GA, et al. A new approach to preanesthetic site verification after 2 cases of wrong site peripheral nerve blocks. Reg Anesth Pain Med 2008 Mar-Apr;33(2):174-7.
Calculation of the Number of Ureteral Stents Inserted as a Primary Procedure
The Authority obtained the number of insertions of ureteral stents between the beginning of the third quarter of 2004 and the end of the first quarter of 2011 from the Pennsylvania Health Care Cost Containment Council in order to calculate an incidence of insertion of ureteral stents, as a primary procedure, on the wrong side.
In-patient procedures were determined from International
Classification of Diseases-Ninth Revision (ICD-9) codes for insertion of ureteral stents as primary procedures.
Outpatient procedures were determined from the same ICD-9 codes from the third quarter of 2004 through the second quarter of 2007. Subsequently, only Current Procedural Terminology (CPT) codes for ureteral stents were available for outpatient procedures through the first quarter of 2011. The Authority extrapolated the number of outpatient stents done as primary procedures by four methods:
- Calculate the percentage of stents as primary procedures from ICD-9 codes compared to all stents from CPT codes for the times (until mid-2007) when both were available and then use that percentage as the conversion factor for each subsequent quarter.
- Calculate the percentage of stents as primary procedures from ICD-9 codes compared to all stents from CPT codes for each quarter (until mid-2007) when both were available and then take the average of those percentages as the conversion factor for each subsequent quarter.
- Calculate the percentage of stents as primary procedures from ICD-9 codes compared to all stents from CPT codes for each quarter (until mid-2007) when both were available and then extrapolate the subsequent percentages over time using linear regression.
- Use the correlation of the number of stents as primary procedures from ICD-9 codes compared to all stents from CPT codes for each quarter (until mid-2007) when both were available and then extrapolate the subsequent number of stents as primary procedures over time using linear regression.
The number of patients who received bilateral stents, with no potential for right-left confusion, was estimated by calculating the difference between the counts of primary ureteral stent procedures on all the records and the number of records with primary ureteral stent procedures. This difference was then subtracted from the number of records with primary ureteral stent procedures to estimate the lower limit on the number of unilateral stents. The upper limit was estimated by assuming all of the stents were unilateral, placed sequentially, if there was more than one count per record.
These calculations resulted in eight point estimates, ranging from 72,782 to 90,582, for the number of ureteral stents done as primary procedures from July 2004 through March 2011. The median point estimate was 81,317 insertions of a single ureteral stent as a primary procedure. In that period, the Authority received 19 reports of a ureteral stent being inserted in the incorrect ureter as a primary procedure, without timely realization and reinsertion in the correct ureter, for an incidence of 1 every 4,280 procedures. Assuming the upper range of the estimates for the number of unilateral procedures, the incidence is at least 1 every 4,765 procedures.