X
Thirty-Day Outcomes Support Implementation of a Surgical Safety Checklist
Background
Thirty-day postoperative complications from unintended harm adversely affect patients and their families and increase institutional health care costs. A surgical checklist is an inexpensive tool that will facilitate effective communication and teamwork. Surgical team training has demonstrated the opportunity for stakeholders to professionally engage one another through leveling of the authority gradient to prevent patient harm. The American College of Surgeons National Surgical Quality Improvement Program database is an outcomes reporting tool capable of validating the use of surgical checklists.

Study Design
Three 60-minute team training sessions were conducted and participants were oriented to the use of a comprehensive surgical checklist. The surgical team used the checklist for high-risk procedures selected from those analyzed for the American College of Surgeons National Surgical Quality Improvement Program. Trained observers assessed the checklist completion and collected data about perioperative communication and safety-compromising events.

Results
Data from the American College of Surgeons National Surgical Quality Improvement Program were compared for 2,079 historical control cases, 246 cases without checklist use, and 73 cases with checklist use. Overall completion of the checklist sections was 97.26%. Comparison of 30-day morbidity demonstrated a statistically significant (p = 0.000) reduction in overall adverse event rates from 23.60% for historical control cases and 15.90% in cases with only team training, to 8.20% in cases with checklist use.

Conclusions
Use of a comprehensive surgical safety checklist and implementation of a structured team training curriculum produced a statistically significant decrease in 30-day morbidity. Adoption of a comprehensive checklist is feasible with team training intervention and can produce measurable improvements in patient outcomes.

Complications from unintended harm adversely affect patients and their families and increase institutional health care costs.1 Postoperative 30-day morbidity continues to be a vexing problem in the discipline of surgery.2 In June 2010, a white paper released by the Society of Actuaries estimates that postoperative infections are 1 of the 5 most expensive complications related to medical error, averaging $14,500 in excess costs per case.3 Although the WHO checklist has been used globally to reduce postoperative morbidity and mortality, no formal study has used an evidence-based, standardized surgical outcomes database for validation.4 The American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database is an outcomes reporting tool capable of validating the use of surgical safety checklists.

The surgical checklist is an inexpensive tool capable of shifting the hierarchical culture in the operating room (OR).5, 6, 7 Without the checklist, systems failures portend poor communication, increased team tension, resource waste, inefficiency, and procedural error.6, 8 The Association of Perioperative Registered Nurses Comprehensive Surgical Checklist released in April 2010 incorporates mandated clinical practice paradigms required by the WHO, the Joint Commission, and the Centers for Medicare and Medicaid Services.9 The one-page document compartmentalizes needed information to facilitate documentation throughout the perioperative process.10, 11 In addition, the inclusion of a debriefing component to the checklist encourages the OR team to acknowledge concerns and plan care for the patient to ensure safe handoff to the recovery room staff.

Previous studies have shown that use of a comprehensive surgical checklist enhances communication and reduces postoperative complications and death.4, 7, 12 For 30 years, the aviation industry has used team training and checklists as part of crew resource management principles and has become the safest industry for its consumers.13, 14 Accordingly, surgical team training has demonstrated the opportunity for stakeholders to professionally engage one another through leveling of the authority gradient to prevent patient harm.14, 15, 16, 17, 18 With the adoption of aviation industry teamwork principles with the checklist as a point of reference, the OR can become a safer environment for patients and staff.14

Methods

Study design
This is a prospective cohort design with historical controls conducted at a 600-bed tertiary care facility and major teaching hospital located in the Northeast. As the team training intervention was provided to all OR personnel before initiation of the study, historical as well as concurrent controls were used to avoid bias. The study was approved through the hospital's institutional review board.


Intervention
Before implementation of the standardized protocol using preoperative briefing and postoperative debriefing checklists, surgical services staff participated in a 3-session team-based training program. Sessions were conducted by internal organizational leadership staff based on the text Crucial Conversations: Tools for Talking When Stakes Are High, including session themes “Crucial Conversations,” “Getting What You Want: Communication Strategies That Help You Get What You Need” and “When the Going Gets Tough: Achieving a Positive Outcome.”19 Topics addressed included differences between introversion and extroversion, creation of a shared pool of meaning, encouragement of dialogue among all OR personnel, and avoidance of obstructive behaviors, including “silence” and “violence.” Participants were oriented to the use of the Association of Perioperative Registered Nurses Comprehensive Surgical Checklist and barriers to checklist use were discussed at the third team training session. After the initial team training session for perioperative personnel, use of the Association of Perioperative Registered Nurses Comprehensive Surgical Checklist was introduced into 73 general surgery cases.


Case selection
Operative cases were selected from those analyzed for the ACS NSQIP between December 2010 and March 2011. Eligible cases included specific high-risk procedures that required the use of general, spinal, or epidural anesthesia with the exclusion of traumatic injuries, organ transplantations, and patients younger than 18 years of age. Included procedures involved the small bowel, colon, stomach, esophagus, pancreas, and spleen, and elective amputations as determined by CPT codes. Of this sample, electively scheduled cases were examined based on the availability of observers. Although a power analysis determined 150 cases were necessary to maximize the likelihood of identifying statistical significance, the limited availability of trained observers restricted the sample size. No systematic randomization occurred. Cases performed by attending surgeons not represented in the historical control data were excluded from analysis. The historical control group included all ACS NSQIP cases meeting the study inclusion criteria and completed before the first team training session, between January 2007 and June 2010. A contemporary control group included all cases occurring during the period of data collection that met inclusion criteria and in which the checklist was not used. No observers were present during the contemporary control cases.


Data collection
One trained observer remained present for the full duration of each study case to assess the checklist completion and use and the number of circulating nurse exits during the case. No analysis of inter-observer or intra-observer reliability was performed, but the majority of cases were observed by a single individual. Observer documentation also reflected any safety-compromising events that occurred throughout the perioperative process. These events were grouped by the nature of the deficiency into the following categories: communication, decision making, equipment availability, equipment malfunction, disruptive behavior, patient flow and process, and sterility. Definitions and representative examples are included in Table 1.

Table 1
Definitions of Qualitative Observation Categories and Representative Examples Drawn from Observer Notation of Intraoperative Events
Categories    Definitions and examples
Communication    Absence of professional and appropriate exchange of information (eg, no blood ordered; issue not addressed during sign-in; ICD/pacer in place; not discussed and Medtronic not called until induction)
Decision making    Deviation from plan of care (eg, converted to open … [c]irculator exits related to gathering supplies for open case. Many supplies for open conversion not in room)
Equipment availability    Lack of ready access to necessary equipment (eg, surgeon changed mesh desire and informed team … surgery stopped waiting for mesh; wrong mesh delivered from ambulatory [operating rooms]; surgeon unscrubbed, went to ASU and picked up mesh himself; procedure resumed)
Equipment malfunction    Malfunction of necessary equipment (eg, bone blade detached and shot across the room)
Disruptive behavior    Departure from respectful and fair interactions with patient, family member, or colleague (eg, joking by surgeon at expense of female personnel)
Process/flow    Failure of timely progression of the patient through perioperative process (eg, patient had to wait in hallway for recovery assignment; no nurse ready for patient when patient arrived in PACU)
Sterility    Failure to maintain a sterile environment (eg, CRNA brought open cup of coffee into case, raised sheet to cover view of anesthesia area)
ASU, ambulatory surgery unit; CRNA, certified registered nurse anesthetist; ICD, implantable cardioverter-defibrillator; PACU, post anesthesia care unit.

A trained Surgical Clinical Reviewer collected data for each ACS NSQIP case. Data included 45 preoperative risk factors, 13 preoperative laboratory values, 20 intraoperative variables, 30-day postoperative mortality, and 20 categories of 30-day morbidity. Morbidity was classified as major—bleeding requiring transfusion, ventilator use for longer than 48 hours, pneumonia, thromboembolic event, acute renal failure, cardiac event, coma, deep surgical site infection, CVA, sepsis, unplanned intubation—or minor—urinary tract infection, superficial surgical site infection, peripheral nerve injury. After input into the secured ACS NSQIP website, data were reviewed, validated, and statistically analyzed by the ACS. The hospital's ACS NSQIP data were subsequently returned in a Microsoft Access file and combined with data collected by the trained observers and extracted from the GE Healthcare Centricity Perioperative system.


Statistical analysis and study oversight
Statistical analysis was completed with SPSS software (version 18.0, SPSS, Inc). Group differences in patient demographics, case characteristics, and morbidity were assessed using chi-square tests. Morbidity included in the analysis of total complications included all 30-day outcomes collected in the ACS NSQIP database. Operative time and circulating nurse exits were assessed with 2-tailed t-tests (bivariate comparisons).


Results
Data from ACS NSQIP were compared for 2,079 historical control cases, 246 cases without checklist use and 73 cases with checklist use. Patient demographics, preoperative comorbidities and procedure types were comparable among the 3 groups and are included in Table 2. Overall completion of the WHO checklist columns was 97.26%, with completion of individual checklist items varying from 24.7% to 100.00% as displayed in Table 3.

Table 2
Comparison of Patient Demographics, Preoperative Risk Factors, and Operative Details among Historical Control Cases, Contemporary Control Cases, and Checklist Cases
Cases without checklist    Cases with checklist    Historical control cases    p Value
Age, mean ± SD, y    54.50 ± 17.63    54.83 ± 15.72    54.61 ± 17.68    0.877
Sex, n (%)                0.971
 Male    99?(40.2)    30?(42.9)    827?(39.9)    
 Female    147?(59.8)    40?(57.1)    1,248?(60.1)    
Race, n (%)                0.531
 White    193?(78.8)    57?(80.3)    1,590?(76.6)    
 Black    42?(17.1)    10?(14.1)    350?(16.9)    
 Native Hawaiian or Pacific Islander    1?(0.4)    0?(0)    49?(2.4)    
 Asian    1?(0.4)    1?(1.4)    15?(0.7)    
 Unknown    8?(3.3)    3?(4.2)    73?(3.5)    
Transfer status, n (%)                0.352
 Not transferred    230?(93.5)    69?(94.5)    1,992?(95.8)    
 From acute care hospital    1?(0.4)    1?(1.4)    15?(0.7)    
 From nursing home, chronic care, intermediate care    12?(4.9)    3?(4.1)    64?(3.1)    
 From other    3?(1.2)    0?(0)    6?(0.3)    
 From outside emergency department    0?(0)    0?(0)    2?(0.1)    
Diabetes, n (%)                0.898
 None    207?(84.1)    64?(87.7)    1,788?(86.0)    
 Non−insulin dependent    23?(9.3)    6?(8.2)    181?(8.7)    
 Insulin dependent    16?(6.5)    3?(4.1)    110?(5.3)    
Smoker, n (%)                0.199
 Yes    37?(15.0)    9?(12.3)    382?(18.4)    
 No    209?(85.0)    64?(87.7)    1,697?(81.6)    
Alcohol abuse, n (%)                0.453
 Yes    5?(2.0)    0?(0)    31?(1.5)    
 No    241?(98.0)    73?(100)    2,048?(98.5)    
Functional status, n (%)                0.075
 Independent    221?(89.8)    71?(97.3)    1,887?(90.8)    
 Partial dependent    19?(7.7)    2?(2.7)    107?(5.1)    
 Totally dependent    6?(2.4)    0?(0)    85?(4.1)    
Ventilator dependent, n (%)                0.641
 Yes    3?(1.2)    0?(0)    25?(1.2)    
 No    243?(98.8)    73?(100)    2,054?(98.8)    
COPD, n (%)                0.021
 Yes    16?(6.5)    1?(1.4)    68?(3.3)    
 No    230?(93.5)    72?(98.6)    2,011?(96.7)    
Pneumonia, n (%)                0.432
 Yes    3?(1.2)    0?(0)    13?(0.6)    
 No    243?(98.8)    73?(100)    2,066?(99.4)    
Ascites, n (%)                0.369
 Yes    3?(1.2)    0?(0)    40?(1.9)    
 No    243?(98.8)    73?(100)    2,039?(98.1)    
Congestive heart failure, n (%)                0.366
 Yes    7?(2.8)    0?(0)    53?(2.5)    
 No    239?(97.2)    73?(100)    2,026?(97.5)    
Renal failure, n (%)                0.432
 Yes    3?(1.2)    0?(100)    13?(0.6)    
 No    243?(98.8)    73?(100)    2,066?(99.4)    
Dialysis, n (%)                0.3
 Yes    1?(0.4)    0?(100)    27?(1.3)    
 No    245?(99.6)    73?(100)    2,052?(98.7)    
Hemiplegia, n (%)                0.74
 Yes    2?(0.8)    0?(0)    17?(0.8)    
 No    244?(99.2)    73?(100)    2,062?(99.2)    
Cancer, n (%)                0.382
 Yes    9?(3.7)    1?(1.4)    49?(2.4)    
 No    237?(96.3)    72?(98.6)    2,030?(97.6)    
Open wound, n (%)                0.719
 Yes    12?(4.9)    2?(2.7)    99?(4.8)    
 No    234?(95.1)    71?(97.3)    1,980?(95.2)    
Steroids, n (%)                0.616
 Yes    9?(3.7)    1?(1.4)    71?(3.4)    
 No    237?(96.3)    72?(98.6)    2,008?(96.6)    
Bleeding, n (%)                0.424
 Yes    12?(4.9)    3?(4.1)    137?(6.6)    
 No    234?(95.1)    70?(95.9)    1,942?(93.4)    
Transfusion, n (%)                0.003
 Yes    8?(3.3)    0?(0)    19?(0.9)    
 No    238?(96.7)    73?(100)    2,060?(99.1)    
Sepsis, n (%)                0.099
 Sepsis    9?(3.7)    0?(0)    42?(2.1)    
 None    221?(89.8)    72?(98.6)    1,815?(90.0)    
 Septic shock    5?(2.0)    0?(0)    28?(1.4)    
 SIRS    11?(4.5)    1?(1.4)    131?(6.5)    
Myocardial ischemia, n (%)                0.396
 Yes    0?(0)    0?(0)    12?(0.6)    
 No    246?(100)    73?(100)    2,067?(99.4)    
Peripheral vascular disease, n (%)                0.822
 Yes    3?(1.2)    1?(1.4)    36?(1.7)    
 No    243?(98.8)    72?(98.6)    2,043?(98.3)    
Paraplegia, n (%)                
 Yes    1?(0.4)    0?(0)    11?(0.5)    0.8
 No    245?(99.6)    73?(100)    2,068?(99.5)    
Quadriplegia, n (%)                0.355
 Yes    2?(0.8)    0?(0)    6?(0.3)    
 No    244?(99.2)    73?(100)    2,073?(99.7)    
Pregnant, n (%)                0.867
 Yes    0?(0)    0?(0)    3?(0.2)    
 No    147?(100)    43?(100)    1,992?(99.8)    
Earlier operation, n (%)                0.315
 Yes    8?(3.3)    0?(0)    61?(2.9)    
 No    238?(96.7)    73?(100)    2,018?(97.1)    
Operative case information                
 Time in operating room, mean ± SD, min    155.17?±?94.87    144.75?±?80.30    153.26?±?92.91    1.000
 Operative priority, n (%)                0.006
 Emergency    53?(21.5)    4?(5.5)    422?(20.3)    
 Elective    193?(78.5)    69?(94.5)    1,657?(79.7)    
Wound class, n (%)                0.387
 I Clean    80?(32.5)    25?(34.2)    717?(34.5)    
 II Clean/contaminated    117?(47.6)    39?(53.4)    940?(45.2)    
 III Contaminated    33?(13.4)    8?(11)    256?(12.3)    
 IV Dirty/infected    16?(6.5)    1?(1.4)    166?(8.0)    
ASA class, n (%)                0.680
 1    19?(7.7)    5?(6.8)    176?(8.5)    
 2    129?(52.4)    47?(64.4)    1,094?(52.6)    
 3    78?(31.7)    20?(27.4)    677?(32.6)    
 4    17?(6.9)    1?(1.4)    116?(5.6)    
 5    1?(0.4)    0?(0)    8?(0.4)    
ASA, American Society of Anesthesiologists; SIRS, systemic inflammatory response syndrome.

Table 3
Frequency of Checklist Component Completion in Cases with Checklist Use
Frequencies checklist    %
Preprocedure check-in completed?    100.0
 Identity filled out?    98.6
 Procedure and procedure site filled out?    95.9
 Consent(s) filled out?    93.2
 Site marked filled out?    98.6
 History and physical filled out?    97.3
 Pre-anesthesia assessment filled out?    93.2
 Diagnostic and radiologic test results filled out?    91.8
 Blood products filled out?    95.9
 Any special equipment, devices, implants filled out?    93.2
 Beta blocker medication given (SCIP) filled out?    24.7
 Venous thromboembolism prophylaxis ordered (SCIP) filled out?    34.2
 Normothermia measures (SCIP) filled out?    34.2
Sign in completed?    97.3
 Confirmation of: identity, procedure, procedure site and consent(s) filled out?    26.0
 Site marked filled out?    91.8
 Patient allergies filled out?    98.6
 Difficult airway or aspiration risk? Filled out?    93.2
 Risk of blood loss (>500 mL) filled out?    79.5
 Anesthesia safety check completed filled out?    76.7
 Briefing filled out?    56.2
Time-out completed?    98.6
 Introduction of team members filled out?    74.0
 Confirmation of the following: identity, procedure, incision site, consent(s) filled out?    94.5
 Site is marked and visible filled out?    100.0
 Relevant images properly labeled and displayed filled out?    100.0
 Any equipment concerns? Filled out?    100.0
 Antibiotic prophylaxis within one hour before incision filled out?    95.9
 Sterilization indicators have been confirmed filled out?    95.9
Sign-out completed?    93.2
 Name of operative procedure, completion of sponge, sharp, and instrument counts filled out?    95.9
 Specimens identified and labeled filled out?    98.6
 Any equipment problems to be addressed? Filled out?    97.3
SCIP, Surgical Care Improvement Project.

Comparison of 30-day morbidity demonstrated a statistically significant (p = 0.000) reduction in overall adverse event rates from 23.60% for historical control cases and 15.90% in cases with only team training, to 8.20% in cases with checklist use. Checklist use was correlated with a decrease in all measured areas of 30-day morbidity, although these relationships failed to achieve statistical significance. All ACS NSQIP 30-day postsurgical morbidity variables were analyzed. Figure 1, Figure 2 reflect variations in 30-day morbidity among the 3 groups. The relationship between checklist component completion and 30-day morbidity rates demonstrates a decrease in many individual adverse outcomes associated with checklist completion, although only 3 checklist components achieve statistically significant changes in morbidity. Lack of confirmation of patient identity and failure to address procedure and procedure site during the preprocedure check-in section of the checklist were both associated with higher occurrences of deep surgical site infections. Cases without documentation of the introduction of all team members were more likely to include major morbidity and infectious events. These relationships are reflected in Figure 3.

Thumbnail image of Figure 1. Opens large image
Figure 1
Total and specific 30-day morbidity rates of historical control cases, cases without checklist use, and cases with checklist use. DVT, deep venous thrombus; PE, pulmonary embolism; SSI, surgical site infection; UTI, urinary tract infection.

View Large Image | View Hi-Res Image | Download PowerPoint Slide
Thumbnail image of Figure 2. Opens large image
Figure 2
Comparison of other individual 30-day morbidity occurrences among historical control cases, cases without checklist use, and cases with checklist use. DVT, deep venous thrombus; MI, myocardial infarction; PE, pulmonary embolism; SSI, surgical site infection.

View Large Image | View Hi-Res Image | Download PowerPoint Slide
Thumbnail image of Figure 3. Opens large image
Figure 3
Statistically significant relationships between individual checklist component completion and specific 30-day morbidities in cases with checklist use.

View Large Image | View Hi-Res Image | Download PowerPoint Slide
Data from ACS NSQIP were supplemented with information collected by trained observers present for cases with checklist use. Circulating nurse exits for observed cases varied from 0 to 25 per case. Fewer circulating nurse exits and shorter durations of time in the operating room were both associated with lower 30-day morbidity, as reflected in Table 4, Table 5.

Table 4
Comparison of the Number of Circulating Nurse Exits between Cases with 30-Day Morbidity and without 30-Day Morbidity
Circulating nurse exits, n
Present?    Absent?    p Value
Any morbidity    10.7?±?8.4    4.7?±?4.0    0.002
Major morbidity    14.3?±?8.1    4.7?±?3.9    0.000
Minor morbidity    7.7?±?7.2    5.1?±?4.6    0.354
Infectious event    11.4?±?9.2    4.7?±?3.9    0.002
Surgical site infection    13.5?±?9.1    4.71?±?3.9    0.000
Urinary tract infection    9.5?±?9.2    5.1?±?4.6    0.190
Sepsis    12.5?±?5.0    5.0?±?4.6    0.024
Superficial surgical site infection    4.00    5.2?±?4.7    0.800
Deep surgical site infection    25.00    4.9?±?4.1    0.000
Bleeding requiring blood transfusion    11.5?±?6.4    5.0?±?4.6    0.053
low asteriskValues are mean ± SD.
Table 5
Comparison of the Operating Room Time between Cases with 30-Day Morbidity and without 30-Day Morbidity
Time in operating room, min
Present?    Absent?    p Value
Any morbidity    259.16?±?165.29    134.51?±?60.56    0.000
Major morbidity    318.25?±?172.34    134.70?±?60.36    0.000
Minor morbidity    270.33?±?230.20    139.37?±?66.91    0.005
Infectious event    274.40?±?180.03    135.22?±?60.39    0.000
Surgical site infection    294.50?±?201.30    136.07?±?60.36    0.000
Urinary tract infection    361.50?±?236.88    138.65?±?66.71    0.000
Sepsis    459.00?±?98.99    135.90?±?59.92    0.000
Superficial surgical site infection    88.00    145.54?±?80.57    0.481
Deep surgical site infection    172.00    144.38?±?80.79    0.735
Bleeding requiring blood transfusion    356.00?±?244.66    138.80?±?66.83    0.000
low asteriskValues are mean ± SD.
Overall, 511 safety-compromising events were possible for the observed 73 surgical cases, with an observed total of 186 (37%) recorded. Eight surgical cases had no safety-compromising events. Figure 4 reflects the percentage of cases in which any given category of safety-compromising event occurred with evidence of impaired communication and deviation from sterile technique identified in more than half of all cases. Twenty-seven cases had 4 or more safety-compromising events. Of those 27 cases, 4 patients developed surgical site infections requiring readmission to the hospital. A general trend reflects that cases with safety-compromising events had higher rates of 30-day morbidity, as reflected in Figure 5. Individual relationships displayed in Figure 5 did not demonstrate statistical significance.

Thumbnail image of Figure 4. Opens large image
Figure 4
Frequency of observed intraoperative events based on categorization of observer notations.

View Large Image | View Hi-Res Image | Download PowerPoint Slide
Thumbnail image of Figure 5. Opens large image
Figure 5
Comparison between 30-day morbidity rates in cases with (A) observed intra-operative events and (B) without observed intra-operative events. SSI, surgical site infection; UTI, urinary tract infection.

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Discussion
As led by the airline industry, perioperative services demand high organizational reliability and commitment to reduction of safety-compromising events. The airline industry has a long history of checklist implementation to reduce risk and avoid catastrophic outcomes. Existing research demonstrates that checklist use can have a similar transformative impact on the perioperative arena facilitating organizational reliability. This study demonstrated that use of a comprehensive surgical safety checklist and implementation of a structured team training curriculum produced a measurable and statistically significant decrease in 30-day morbidity among high-risk general surgery cases. This represents the first study to document this improvement using the ACS NSQIP to compare postoperative morbidity. In addition, use of specific checklist items can be correlated with decreased morbidity rates. The team training sessions introduced the concept of a safety checklist in the context of addressing communication strategies and professional, patient-centered collaboration in the perioperative setting. Despite limited instruction, compliance with the checklist was 97.26%, with most individual component completion rates >90%. Individual components with low completion include those that staff viewed as redundant, such as introduction of all team members in an environment in which everyone present was already acquainted, and tasks that lacked already established accountability, such as the preoperative administration of β-blockers. This suggests adoption of a comprehensive checklist is feasible with minimal intervention and can produce measurable improvements in patient outcomes, as reflected by ACS NSQIP data. Checklist adoption in isolation fails to maximize the potential impact. Team training sessions represent a translatable, generalizable strategy for improving communication and providing OR personnel with strategies for engagement. Training allows OR personnel to use the checklist as a tool to facilitate communication. This prevents checklist completion from becoming an administrative task and transforms completion of the checklist into an integral part of perioperative care plan development.

When compared with historical controls, cases with checklist use showed a small reduction of time in the OR, and decreased OR time is correlated with a reduction in numerous adverse patient events. Lower frequencies of circulating nurse exits from the OR during cases are correlated with decreased rates of morbidity. Although a causal relationship is not addressed in this study, this correlation might reflect the bacterial transit that occurs during exit from and entrance to the OR, as well as the impact of time without an available circulating nurse. This absence could facilitate safety-compromising events and delay progression of the case when additional supplies are needed for the procedure or nonsterile tasks must be completed. The relationship between morbidity rates and circulating nurse exits or time spent in the OR might reflect the case complexity rather than the influence of checklist use.

Observations of safety-compromising events during cases with checklist implementation showed the need for focus on deficiencies in communication, equipment availability and functioning, patient flow, and adherence to sterile technique, as well as disruptive behavior. The presence of safety-compromising events did result in worse patient outcomes. In addition to identifying deficiencies, observer findings supported the theory that team training and checklist implementation together changed the OR microenvironment. With rigorous team training centered on the checklist as a compliance tool, stakeholders could freely advocate for patients. In addition, the checklist and the strategies to engage team members in a collegial framework helped to improve patient outcomes by establishing a plan of care to ensure proper handoffs throughout the perioperative process. Delineation and universal knowledge of the plan of care allows for improved preparation as well as anticipation of potential adverse events or complications, shortening operative time, increasing safety, improving efficiency, and reducing cost. These findings represent important steps toward eliminating blame, carelessness, and lack of accountability in the OR and creation of a more just culture where personnel can advocate for the patient without fear of retribution.

The strength of these findings suffers due to several study limitations. Although use of the ACS NSQIP database provided a robust historical control population of 2,079 cases, the small number of cases with checklist use and observation after team training hindered identification of trends in morbidity rates and reduced the likelihood of establishing statistically significant relationships. A follow-up multicenter study would allow for the increased power and scalability necessary to elucidate statistically significant relationships. The proportion of emergent cases varied considerably among the 3 groups, with fewer emergency cases captured in cases with checklist implementation. This may have been due to observer availability. Statistical analysis correcting for this confounding factor was not performed and emergent cases may have represented a less healthy patient population. The presence of trained observers in preoperative areas and ORs during cases with checklist use may have influenced the actions of perioperative staff and contributed to some of the improvements reflected in 30-day morbidity reduction. In addition, team training sessions did not capture all members of the perioperative team, with under-representation of surgeons and anesthesia providers. This may have undermined the new communication dynamics other staff tried to establish using the team training curriculum.

Based on the results of this initial study, future research efforts will focus on determination of which checklist components offer the largest opportunity for improving patient safety. The length of the comprehensive checklist can reduce compliance in an environment without observers present or at a time more distant from the training experience. Reducing the checklist to only the most essential elements can increase long-term adherence to checklist use. The pilot data presented in this study will be used to support the universal adoption of a surgical safety checklist at this medical center. Specifically, the correlation between completion of individual WHO Comprehensive Surgical Safety Checklist components with lower morbidity rates can be used to refine the checklist. After adoption of a perioperative checklist, ACS NSQIP data will be revisited to determine if other statistically significant relationships are identified with a larger sample size. A multicenter study will also be pursued to increase the statistical power and identify specific statistically significant relationships.

Additional qualitative investigations might include an ethnographic assessment of OR safety attitudes and communication based on focus groups and semi-structured interviews with the full spectrum of perioperative team members. Ongoing interventions associated with this research include development of a follow-up team training curriculum and expansion of team training efforts to capture more staff members and disciplines. Reinforcement of communication strategies and a reduction in the authority gradient will be achieved through aggressive enforcement of a medical staff professional code of conduct. This multidimensional strategy of intervention and reflective analysis of both patient outcomes and team-member perception will lead to an improvement in the institutional culture of safety.


Conclusions
Despite the limitations of this study, it demonstrates a considerable improvement in risk-adjusted 30-day postoperative morbidity. The implementation of team training and accountability measures, such as the comprehensive checklist, are inexpensive interventions that can contribute to cost savings in the expanding environment of pay for performance and increase patient satisfaction with the perioperative experience.


Author Contributions
Study conception and design: Bliss, Ross-Richardson, Sanzari, Shapiro, Lukianoff, Bernstein, Ellner

Acquisition of data: Bliss, Ross-Richardson, Sanzari, Ellner

Analysis and interpretation of data: Bliss, Ross-Richardson, Sanzari, Ellner

Drafting of manuscript: Bliss, Ellner

Critical revision: Bliss, Ross-Richardson, Sanzari, Shapiro, Ellner


Acknowledgement
We would like to acknowledge Jeffrey Steinberg, MD, FACS, Steven Ruby, MD, FACS and Rebecca Crowell, PhD, for institutional support of these research efforts, Gregory Makoul, PhD, Nancy Krafcik-Rousseau, PhD, Elizabeth Lunt, MSW, and Timothy Michaels, for assistance with team training curriculum development and implementation, Olga Clark, PhD, for data analysis assistance, and Kristina Ziegler, MD, Patricia Davis, MD, Mary Long, BSN, APRN, and Maria Trigg, PA-C, for data collection assistance.


References
in: L. Kohn, J. Corrigan, M. Donaldson (Eds.) To Err is Human: Building a Safer Health System. National Academy Press, Washington, DC; 2000
View in Article
Leape, L.L. Errors in medicine. Clin Chim Acta. 2009; 404: 2–5
View in Article | CrossRef | PubMed | Scopus (82)
Shreve, J., van Den Bos, J., Gray, T. et al. The Economic Measurement of Medical Errors. Society of Actuaries and Milliman, Inc, Schaumburg, IL; 2010
View in Article
Haynes, A., Weiser, T., Berry, W. et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009; 360: 491–499
View in Article | CrossRef | PubMed | Scopus (1814)
Schimpff, S.C. Improving operating room and perioperative safety: background and specific recommendations. Surg Innov. 2007; 14: 127–135
View in Article | CrossRef | PubMed | Scopus (16)
Carney, B., West, P., Neily, J. et al. Differences in nurse and surgeon perceptions of teamwork: implications for use of a briefing checklist in the OR. AORN J. 2010; 91: 722–729
View in Article | PubMed | Scopus (40)
Weiser, T., Haynes, A., Dziekan, G. et al. Effect of a 19-item surgical safety checklist during urgent operations in a global patient population. Ann Surg. 2010; 251: 976–980
View in Article | CrossRef | PubMed | Scopus (177)
Makary, M., Sexton, J., Freischlag, J. et al. Patient safety in surgery. Ann Surg. 2006; 243: 628–632
View in Article | CrossRef | PubMed | Scopus (1)
Association of Perioperative Registered Nurses. Comprehensive surgical checklist. (Denver, CO)http://www.aorn.org/Clinical_Practice/ToolKits/Correct_Site_Surgery_Tool_Kit/Comprehensive_checklist.aspx. (Accessed April 25, 2012)
View in Article
Wilson, I. and Walker, I. The WHO surgical safety checklist: the evidence. J Perioper Pract. 2009; 19: 362–364
View in Article | PubMed
Gawande, A. The checklist: if something so simple can transform intensive care, what else can it do?. New Yorker. 2007; 10: 86–101
View in Article
Verdaasdonk, E., Stassen, L., Widhiasmara, P., and Dankelman, J. Requirements for the design and implementation of checklists for surgical processes. Surg Endosc. 2009; 23: 715–726
View in Article | CrossRef | PubMed | Scopus (44)
Dunn, E., Mills, P., Neily, J. et al. Medical team training: applying crew resource management in the Veterans Health Administration. Jt Comm J Qual Patient Saf. 2007; 33: 317–325
View in Article | PubMed
Sax, H., Browne, P., Mayewski, R. et al. Can aviation-based team training elicit sustainable behavioral change?. Arch Surg. 2009; 144: 1133–1137
View in Article | CrossRef | PubMed | Scopus (37)
Mazzocco, K., Petitti, D., Fong, K. et al. Surgical team behaviors and patient outcomes. Am J Surg. 2009; 197: 678–685
View in Article | PubMed | Scopus (183)
Lingard, L., Regehr, G., Orser, B. et al. Evaluation of a preoperative checklist and team briefing among surgeons, nurses, and anesthesiologists to reduce failures in communication. Arch Surg. 2008; 143: 12–17
View in Article | CrossRef | PubMed | Scopus (284)
Neilly, J., Mills, P., Lee, P. et al. Medical team training and coaching in the Veterans Health Administration; assessment and impact on the first 32 facilities in the programme. Qual Saf Health Care. 2010; 19: 360–364
View in Article | CrossRef | PubMed | Scopus (22)
Lingard, L., Garwood, S., and Poenaru, D. Tensions influencing operating room team function: does institutional context make a difference?. Med Educ. 2004; 38: 691–699
View in Article | CrossRef | PubMed | Scopus (63)
Patterson, K., Grenny, J., McMillan, R. et al. Crucial Conversations: Tools for Talking When Stakes Are High. McGraw-Hill, New York; 2002


Source: http://goo.gl/2czEAp

Share This Posting
Hospitals use checklists to reduce errors